diff --git a/Makefile.am b/Makefile.am index 4300273..7d269d1 100644 --- a/Makefile.am +++ b/Makefile.am @@ -275,19 +275,19 @@ COMMON_SOURCES = \ src/pcre2_jit_misc.c \ src/pcre2_maketables.c \ src/pcre2_match_data.c \ + src/pcre2_newline.c \ + src/pcre2_ord2utf.c \ src/pcre2_pattern_info.c \ src/pcre2_string_utils.c \ src/pcre2_substring.c \ src/pcre2_tables.c \ src/pcre2_ucd.c \ src/pcre2_ucp.h \ + src/pcre2_valid_utf.c \ src/pcre2_version.c -# src/pcre2_newline.c \ -# src/pcre2_ord2utf8.c \ # src/pcre2_refcount.c \ # src/pcre2_study.c \ -# src/pcre2_valid_utf8.c \ # src/pcre2_xclass.c diff --git a/configure.ac b/configure.ac index 4c72cd5..3130962 100644 --- a/configure.ac +++ b/configure.ac @@ -2,13 +2,13 @@ dnl Process this file with autoconf to produce a configure script. dnl NOTE FOR MAINTAINERS: Do not use minor version numbers 08 or 09 because dnl the leading zeros may cause them to be treated as invalid octal constants -dnl if a PCRE user writes code that uses PCRE_MINOR as a number. There is now +dnl if a PCRE2 user writes code that uses PCRE2_MINOR as a number. There is now dnl a check further down that throws an error if 08 or 09 are used. -dnl The PCRE_PRERELEASE feature is for identifying release candidates. It might +dnl The PCRE2_PRERELEASE feature is for identifying release candidates. It might dnl be defined as -RC2, for example. For real releases, it should be empty. -m4_define(pcre2_major, [9]) +m4_define(pcre2_major, [10]) m4_define(pcre2_minor, [00]) m4_define(pcre2_prerelease, [-DEV]) m4_define(pcre2_date, [2014-99-99]) @@ -125,11 +125,11 @@ AC_ARG_ENABLE(jit, [enable Just-In-Time compiling support]), , enable_jit=no) -# Handle --disable-pcregrep-jit (enabled by default) -AC_ARG_ENABLE(pcregrep-jit, - AS_HELP_STRING([--disable-pcregrep-jit], - [disable JIT support in pcregrep]), - , enable_pcregrep_jit=yes) +# Handle --disable-pcre2grep-jit (enabled by default) +AC_ARG_ENABLE(pcre2grep-jit, + AS_HELP_STRING([--disable-pcre2grep-jit], + [disable JIT support in pcre2grep]), + , enable_pcre2grep_jit=yes) # Handle --enable-rebuild-chartables AC_ARG_ENABLE(rebuild-chartables, @@ -144,28 +144,28 @@ AC_ARG_ENABLE(utf, , enable_utf=unset) # Handle newline options -ac_pcre_newline=lf +ac_pcre2_newline=lf AC_ARG_ENABLE(newline-is-cr, AS_HELP_STRING([--enable-newline-is-cr], [use CR as newline character]), - ac_pcre_newline=cr) + ac_pcre2_newline=cr) AC_ARG_ENABLE(newline-is-lf, AS_HELP_STRING([--enable-newline-is-lf], [use LF as newline character (default)]), - ac_pcre_newline=lf) + ac_pcre2_newline=lf) AC_ARG_ENABLE(newline-is-crlf, AS_HELP_STRING([--enable-newline-is-crlf], [use CRLF as newline sequence]), - ac_pcre_newline=crlf) + ac_pcre2_newline=crlf) AC_ARG_ENABLE(newline-is-anycrlf, AS_HELP_STRING([--enable-newline-is-anycrlf], [use CR, LF, or CRLF as newline sequence]), - ac_pcre_newline=anycrlf) + ac_pcre2_newline=anycrlf) AC_ARG_ENABLE(newline-is-any, AS_HELP_STRING([--enable-newline-is-any], [use any valid Unicode newline sequence]), - ac_pcre_newline=any) -enable_newline="$ac_pcre_newline" + ac_pcre2_newline=any) +enable_newline="$ac_pcre2_newline" # Handle --enable-bsr-anycrlf AC_ARG_ENABLE(bsr-anycrlf, @@ -191,35 +191,35 @@ AC_ARG_ENABLE(stack-for-recursion, [don't use stack recursion when matching]), , enable_stack_for_recursion=yes) -# Handle --enable-pcregrep-libz -AC_ARG_ENABLE(pcregrep-libz, - AS_HELP_STRING([--enable-pcregrep-libz], - [link pcregrep with libz to handle .gz files]), - , enable_pcregrep_libz=no) +# Handle --enable-pcre2grep-libz +AC_ARG_ENABLE(pcre2grep-libz, + AS_HELP_STRING([--enable-pcre2grep-libz], + [link pcre2grep with libz to handle .gz files]), + , enable_pcre2grep_libz=no) -# Handle --enable-pcregrep-libbz2 -AC_ARG_ENABLE(pcregrep-libbz2, - AS_HELP_STRING([--enable-pcregrep-libbz2], - [link pcregrep with libbz2 to handle .bz2 files]), - , enable_pcregrep_libbz2=no) +# Handle --enable-pcre2grep-libbz2 +AC_ARG_ENABLE(pcre2grep-libbz2, + AS_HELP_STRING([--enable-pcre2grep-libbz2], + [link pcre2grep with libbz2 to handle .bz2 files]), + , enable_pcre2grep_libbz2=no) -# Handle --with-pcregrep-bufsize=N -AC_ARG_WITH(pcregrep-bufsize, - AS_HELP_STRING([--with-pcregrep-bufsize=N], - [pcregrep buffer size (default=20480, minimum=8192)]), - , with_pcregrep_bufsize=20480) +# Handle --with-pcre2grep-bufsize=N +AC_ARG_WITH(pcre2grep-bufsize, + AS_HELP_STRING([--with-pcre2grep-bufsize=N], + [pcre2grep buffer size (default=20480, minimum=8192)]), + , with_pcre2grep_bufsize=20480) -# Handle --enable-pcretest-libedit -AC_ARG_ENABLE(pcretest-libedit, - AS_HELP_STRING([--enable-pcretest-libedit], - [link pcretest with libedit]), - , enable_pcretest_libedit=no) +# Handle --enable-pcre2test-libedit +AC_ARG_ENABLE(pcre2test-libedit, + AS_HELP_STRING([--enable-pcre2test-libedit], + [link pcre2test with libedit]), + , enable_pcre2test_libedit=no) -# Handle --enable-pcretest-libreadline -AC_ARG_ENABLE(pcretest-libreadline, - AS_HELP_STRING([--enable-pcretest-libreadline], - [link pcretest with libreadline]), - , enable_pcretest_libreadline=no) +# Handle --enable-pcre2test-libreadline +AC_ARG_ENABLE(pcre2test-libreadline, + AS_HELP_STRING([--enable-pcre2test-libreadline], + [link pcre2test with libreadline]), + , enable_pcre2test_libreadline=no) # Handle --with-link-size=N AC_ARG_WITH(link-size, @@ -298,11 +298,11 @@ fi # agree with the PCRE2_NEWLINE_xxx values in pcre2.h. case "$enable_newline" in - cr) ac_pcre_newline_value=0 ;; - lf) ac_pcre_newline_value=1 ;; - crlf) ac_pcre_newline_value=2 ;; - any) ac_pcre_newline_value=3 ;; - anycrlf) ac_pcre_newline_value=4 ;; + cr) ac_pcre2_newline_value=1 ;; + lf) ac_pcre2_newline_value=2 ;; + crlf) ac_pcre2_newline_value=3 ;; + any) ac_pcre2_newline_value=4 ;; + anycrlf) ac_pcre2_newline_value=5 ;; *) AC_MSG_ERROR([invalid argument \"$enable_newline\" to --enable-newline option]) ;; @@ -314,7 +314,7 @@ if test "x$enable_ebcdic_nl25" = "xyes"; then fi # Make sure that if enable_ebcdic is set, rebuild_chartables is also enabled. -# Also check that UTF support is not requested, because PCRE cannot handle +# Also check that UTF support is not requested, because PCRE2 cannot handle # EBCDIC and UTF in the same build. To do so it would need to use different # character constants depending on the mode. # @@ -334,13 +334,13 @@ case "$with_link_size" in esac AH_TOP([ -/* PCRE is written in Standard C, but there are a few non-standard things it +/* PCRE2 is written in Standard C, but there are a few non-standard things it can cope with, allowing it to run on SunOS4 and other "close to standard" systems. In environments that support the GNU autotools, config.h.in is converted into config.h by the "configure" script. In environments that use CMake, -config-cmake.in is converted into config.h. If you are going to build PCRE "by +config-cmake.in is converted into config.h. If you are going to build PCRE2 "by hand" without using "configure" or CMake, you should copy the distributed config.h.generic to config.h, and edit the macro definitions to be the way you need them. You must then add -DHAVE_CONFIG_H to all of your compile commands, @@ -357,7 +357,7 @@ macros are listed as a commented #undef in config.h.generic. Macros such as MATCH_LIMIT, whose actual value is relevant, have defaults defined, but are surrounded by #ifndef/#endif lines so that the value can be overridden by -D. -PCRE uses memmove() if HAVE_MEMMOVE is defined; otherwise it uses bcopy() if +PCRE2 uses memmove() if HAVE_MEMMOVE is defined; otherwise it uses bcopy() if HAVE_BCOPY is defined. If your system has neither bcopy() nor memmove(), make sure both macros are undefined; an emulation function will then be used. */]) @@ -370,7 +370,7 @@ AC_CHECK_HEADERS([windows.h], [HAVE_WINDOWS_H=1]) AM_CONDITIONAL(WITH_PCRE8, test "x$enable_pcre8" = "xyes") AM_CONDITIONAL(WITH_PCRE16, test "x$enable_pcre16" = "xyes") AM_CONDITIONAL(WITH_PCRE32, test "x$enable_pcre32" = "xyes") -AM_CONDITIONAL(WITH_PCRE_CPP, test "x$enable_cpp" = "xyes") +#AM_CONDITIONAL(WITH_PCRE2_CPP, test "x$enable_cpp" = "xyes") AM_CONDITIONAL(WITH_REBUILD_CHARTABLES, test "x$enable_rebuild_chartables" = "xyes") AM_CONDITIONAL(WITH_JIT, test "x$enable_jit" = "xyes") AM_CONDITIONAL(WITH_UTF, test "x$enable_utf" = "xyes") @@ -400,7 +400,7 @@ AC_CHECK_LIB([z], [gzopen], [HAVE_LIBZ=1]) # therefore missing the function definition. # - The compiler thus generates a "C" signature for the test function. # - The linker fails to find the "C" function. -# - PCRE fails to configure if asked to do so against libbz2. +# - PCRE2 fails to configure if asked to do so against libbz2. # # Solution: # @@ -426,7 +426,7 @@ LIBS="$OLD_LIBS" # Check for the availabiity of libreadline -if test "$enable_pcretest_libreadline" = "yes"; then +if test "$enable_pcre2test_libreadline" = "yes"; then AC_CHECK_HEADERS([readline/readline.h], [HAVE_READLINE_H=1]) AC_CHECK_HEADERS([readline/history.h], [HAVE_HISTORY_H=1]) AC_CHECK_LIB([readline], [readline], [LIBREADLINE="-lreadline"], @@ -459,7 +459,7 @@ fi # Check for the availability of libedit. Different distributions put its # headers in different places. Try to cover the most common ones. -if test "$enable_pcretest_libedit" = "yes"; then +if test "$enable_pcre2test_libedit" = "yes"; then AC_CHECK_HEADERS([editline/readline.h], [HAVE_EDITLINE_READLINE_H=1], [AC_CHECK_HEADERS([edit/readline/readline.h], [HAVE_READLINE_READLINE_H=1], [AC_CHECK_HEADERS([readline/readline.h], [HAVE_READLINE_READLINE_H=1])])]) @@ -477,21 +477,21 @@ if test "x$enable_shared" = "xno" ; then fi AC_SUBST(PCRE2_STATIC_CFLAG) -# Here is where pcre specific defines are handled +# Here is where PCRE2-specific defines are handled if test "$enable_pcre8" = "yes"; then AC_DEFINE([SUPPORT_PCRE8], [], [ - Define to any value to enable the 8 bit PCRE library.]) + Define to any value to enable the 8 bit PCRE2 library.]) fi if test "$enable_pcre16" = "yes"; then AC_DEFINE([SUPPORT_PCRE16], [], [ - Define to any value to enable the 16 bit PCRE library.]) + Define to any value to enable the 16 bit PCRE2 library.]) fi if test "$enable_pcre32" = "yes"; then AC_DEFINE([SUPPORT_PCRE32], [], [ - Define to any value to enable the 32 bit PCRE library.]) + Define to any value to enable the 32 bit PCRE2 library.]) fi # Unless running under Windows, JIT support requires pthreads. @@ -506,87 +506,87 @@ if test "$enable_jit" = "yes"; then AC_DEFINE([SUPPORT_JIT], [], [ Define to any value to enable support for Just-In-Time compiling.]) else - enable_pcregrep_jit="no" + enable_pcre2grep_jit="no" fi -if test "$enable_pcregrep_jit" = "yes"; then - AC_DEFINE([SUPPORT_PCREGREP_JIT], [], [ - Define to any value to enable JIT support in pcregrep.]) +if test "$enable_pcre2grep_jit" = "yes"; then + AC_DEFINE([SUPPORT_PCRE2GREP_JIT], [], [ + Define to any value to enable JIT support in pcre2grep.]) fi if test "$enable_utf" = "yes"; then AC_DEFINE([SUPPORT_UTF], [], [ Define to any value to enable support for the UTF-8/16/32 Unicode encoding. This will work even in an EBCDIC environment, but it is incompatible - with the EBCDIC macro. That is, PCRE can support *either* EBCDIC + with the EBCDIC macro. That is, PCRE2 can support *either* EBCDIC code *or* ASCII/UTF-8/16/32, but not both at once.]) fi if test "$enable_stack_for_recursion" = "no"; then AC_DEFINE([NO_RECURSE], [], [ - PCRE uses recursive function calls to handle backtracking while + PCRE2 uses recursive function calls to handle backtracking while matching. This can sometimes be a problem on systems that have stacks of limited size. Define NO_RECURSE to any value to get a version that doesn't use recursion in the match() function; instead - it creates its own stack by steam using pcre_recurse_malloc() to obtain - memory from the heap. For more detail, see the comments and other stuff - just above the match() function.]) + it creates its own stack by steam using memory from the heap. For more + detail, see the comments and other stuff just above the match() function.]) fi -if test "$enable_pcregrep_libz" = "yes"; then +if test "$enable_pcre2grep_libz" = "yes"; then AC_DEFINE([SUPPORT_LIBZ], [], [ - Define to any value to allow pcregrep to be linked with libz, so that it is + Define to any value to allow pcre2grep to be linked with libz, so that it is able to handle .gz files.]) fi -if test "$enable_pcregrep_libbz2" = "yes"; then +if test "$enable_pcre2grep_libbz2" = "yes"; then AC_DEFINE([SUPPORT_LIBBZ2], [], [ - Define to any value to allow pcregrep to be linked with libbz2, so that it + Define to any value to allow pcre2grep to be linked with libbz2, so that it is able to handle .bz2 files.]) fi -if test $with_pcregrep_bufsize -lt 8192 ; then - AC_MSG_WARN([$with_pcregrep_bufsize is too small for --with-pcregrep-bufsize; using 8192]) - with_pcregrep_bufsize="8192" +if test $with_pcre2grep_bufsize -lt 8192 ; then + AC_MSG_WARN([$with_pcre2grep_bufsize is too small for --with-pcre2grep-bufsize; using 8192]) + with_pcre2grep_bufsize="8192" else if test $? -gt 1 ; then - AC_MSG_ERROR([Bad value for --with-pcregrep-bufsize]) + AC_MSG_ERROR([Bad value for --with-pcre2grep-bufsize]) fi fi -AC_DEFINE_UNQUOTED([PCREGREP_BUFSIZE], [$with_pcregrep_bufsize], [ - The value of PCREGREP_BUFSIZE determines the size of buffer used by pcregrep +AC_DEFINE_UNQUOTED([PCRE2GREP_BUFSIZE], [$with_pcre2grep_bufsize], [ + The value of PCRE2GREP_BUFSIZE determines the size of buffer used by pcre2grep to hold parts of the file it is searching. This is also the minimum value. - The actual amount of memory used by pcregrep is three times this number, + The actual amount of memory used by pcre2grep is three times this number, because it allows for the buffering of "before" and "after" lines.]) -if test "$enable_pcretest_libedit" = "yes"; then +if test "$enable_pcre2test_libedit" = "yes"; then AC_DEFINE([SUPPORT_LIBEDIT], [], [ - Define to any value to allow pcretest to be linked with libedit.]) + Define to any value to allow pcre2test to be linked with libedit.]) LIBREADLINE="$LIBEDIT" -elif test "$enable_pcretest_libreadline" = "yes"; then +elif test "$enable_pcre2test_libreadline" = "yes"; then AC_DEFINE([SUPPORT_LIBREADLINE], [], [ - Define to any value to allow pcretest to be linked with libreadline.]) + Define to any value to allow pcre2test to be linked with libreadline.]) fi -AC_DEFINE_UNQUOTED([NEWLINE], [$ac_pcre_newline_value], [ - The value of NEWLINE determines the default newline character sequence. PCRE - client programs can override this by selecting other values at run time. The - valid values are 0 (CR), 1 (LF), 2 (CRLF), 3 (ANY), and 4 (ANYCRLF).]) +AC_DEFINE_UNQUOTED([NEWLINE_DEFAULT], [$ac_pcre2_newline_value], [ + The value of NEWLINE_DEFAULT determines the default newline character + sequence. PCRE2 client programs can override this by selecting other values + at run time. The valid values are 1 (CR), 2 (LF), 3 (CRLF), 4 (ANY), + and 5 (ANYCRLF).]) if test "$enable_bsr_anycrlf" = "yes"; then AC_DEFINE([BSR_ANYCRLF], [], [ By default, the \R escape sequence matches any Unicode line ending character or sequence of characters. If BSR_ANYCRLF is defined (to any value), this is changed so that backslash-R matches only CR, LF, or CRLF. - The build-time default can be overridden by the user of PCRE at runtime.]) + The build-time default can be overridden by the user of PCRE2 at runtime.]) fi AC_DEFINE_UNQUOTED([LINK_SIZE], [$with_link_size], [ The value of LINK_SIZE determines the number of bytes used to store links as offsets within the compiled regex. The default is 2, which allows for compiled patterns up to 64K long. This covers the vast - majority of cases. However, PCRE can also be compiled to use 3 or 4 + majority of cases. However, PCRE2 can also be compiled to use 3 or 4 bytes instead. This allows for longer patterns in extreme cases.]) AC_DEFINE_UNQUOTED([PARENS_NEST_LIMIT], [$with_parens_nest_limit], [ @@ -597,7 +597,7 @@ AC_DEFINE_UNQUOTED([PARENS_NEST_LIMIT], [$with_parens_nest_limit], [ AC_DEFINE_UNQUOTED([MATCH_LIMIT], [$with_match_limit], [ The value of MATCH_LIMIT determines the default number of times the internal match() function can be called during a single execution of - pcre_exec(). There is a runtime interface for setting a different + pcre2_match(). There is a runtime interface for setting a different limit. The limit exists in order to catch runaway regular expressions that take for ever to determine that they do not match. The default is set very large so that it does not accidentally catch @@ -639,10 +639,10 @@ AH_VERBATIM([PCRE2_EXP_DEFN], [ if test "$enable_ebcdic" = "yes"; then AC_DEFINE_UNQUOTED([EBCDIC], [], [ If you are compiling for a system that uses EBCDIC instead of ASCII - character codes, define this macro to any value. When EBCDIC is set, PCRE + character codes, define this macro to any value. When EBCDIC is set, PCRE2 assumes that all input strings are in EBCDIC. If you do not define this - macro, PCRE will assume input strings are ASCII or UTF-8/16/32 Unicode. It - is not possible to build a version of PCRE that supports both EBCDIC and + macro, PCRE2 will assume input strings are ASCII or UTF-8/16/32 Unicode. It + is not possible to build a version of PCRE2 that supports both EBCDIC and UTF-8/16/32.]) fi @@ -695,65 +695,65 @@ AC_SUBST(EXTRA_LIBPCRE2_POSIX_LDFLAGS) DISTCHECK_CONFIGURE_FLAGS="CFLAGS='' CXXFLAGS='' --enable-pcre16 --enable-pcre32 --enable-jit --enable-utf" AC_SUBST(DISTCHECK_CONFIGURE_FLAGS) -# Check that, if --enable-pcregrep-libz or --enable-pcregrep-libbz2 is +# Check that, if --enable-pcre2grep-libz or --enable-pcre2grep-libbz2 is # specified, the relevant library is available. -if test "$enable_pcregrep_libz" = "yes"; then +if test "$enable_pcre2grep_libz" = "yes"; then if test "$HAVE_ZLIB_H" != "1"; then - echo "** Cannot --enable-pcregrep-libz because zlib.h was not found" + echo "** Cannot --enable-pcre2grep-libz because zlib.h was not found" exit 1 fi if test "$HAVE_LIBZ" != "1"; then - echo "** Cannot --enable-pcregrep-libz because libz was not found" + echo "** Cannot --enable-pcre2grep-libz because libz was not found" exit 1 fi LIBZ="-lz" fi AC_SUBST(LIBZ) -if test "$enable_pcregrep_libbz2" = "yes"; then +if test "$enable_pcre2grep_libbz2" = "yes"; then if test "$HAVE_BZLIB_H" != "1"; then - echo "** Cannot --enable-pcregrep-libbz2 because bzlib.h was not found" + echo "** Cannot --enable-pcre2grep-libbz2 because bzlib.h was not found" exit 1 fi if test "$HAVE_LIBBZ2" != "1"; then - echo "** Cannot --enable-pcregrep-libbz2 because libbz2 was not found" + echo "** Cannot --enable-pcre2grep-libbz2 because libbz2 was not found" exit 1 fi LIBBZ2="-lbz2" fi AC_SUBST(LIBBZ2) -# Similarly for --enable-pcretest-readline +# Similarly for --enable-pcre2test-readline -if test "$enable_pcretest_libedit" = "yes"; then - if test "$enable_pcretest_libreadline" = "yes"; then - echo "** Cannot use both --enable-pcretest-libedit and --enable-pcretest-readline" +if test "$enable_pcre2test_libedit" = "yes"; then + if test "$enable_pcre2test_libreadline" = "yes"; then + echo "** Cannot use both --enable-pcre2test-libedit and --enable-pcre2test-readline" exit 1 fi if test "$HAVE_EDITLINE_READLINE_H" != "1" -a \ "$HAVE_READLINE_READLINE_H" != "1"; then - echo "** Cannot --enable-pcretest-libedit because neither editline/readline.h" + echo "** Cannot --enable-pcre2test-libedit because neither editline/readline.h" echo "** nor readline/readline.h was found." exit 1 fi if test -z "$LIBEDIT"; then - echo "** Cannot --enable-pcretest-libedit because libedit library was not found." + echo "** Cannot --enable-pcre2test-libedit because libedit library was not found." exit 1 fi fi -if test "$enable_pcretest_libreadline" = "yes"; then +if test "$enable_pcre2test_libreadline" = "yes"; then if test "$HAVE_READLINE_H" != "1"; then - echo "** Cannot --enable-pcretest-readline because readline/readline.h was not found." + echo "** Cannot --enable-pcre2test-readline because readline/readline.h was not found." exit 1 fi if test "$HAVE_HISTORY_H" != "1"; then - echo "** Cannot --enable-pcretest-readline because readline/history.h was not found." + echo "** Cannot --enable-pcre2test-readline because readline/history.h was not found." exit 1 fi if test -z "$LIBREADLINE"; then - echo "** Cannot --enable-pcretest-readline because readline library was not found." + echo "** Cannot --enable-pcre2test-readline because readline library was not found." exit 1 fi fi @@ -868,12 +868,12 @@ $PACKAGE-$VERSION configuration summary: Match limit recursion ........... : ${with_match_limit_recursion} Build shared libs ............... : ${enable_shared} Build static libs ............... : ${enable_static} - Use JIT in pcregrep ............. : ${enable_pcregrep_jit} - Buffer size for pcregrep ........ : ${with_pcregrep_bufsize} - Link pcregrep with libz ......... : ${enable_pcregrep_libz} - Link pcregrep with libbz2 ....... : ${enable_pcregrep_libbz2} - Link pcretest with libedit ...... : ${enable_pcretest_libedit} - Link pcretest with libreadline .. : ${enable_pcretest_libreadline} + Use JIT in pcre2grep ............ : ${enable_pcre2grep_jit} + Buffer size for pcre2grep ....... : ${with_pcre2grep_bufsize} + Link pcre2grep with libz ........ : ${enable_pcre2grep_libz} + Link pcre2grep with libbz2 ...... : ${enable_pcre2grep_libbz2} + Link pcre2test with libedit ..... : ${enable_pcre2test_libedit} + Link pcre2test with libreadline . : ${enable_pcre2test_libreadline} Valgrind support ................ : ${enable_valgrind} Code coverage ................... : ${enable_coverage} diff --git a/doc/pcre2test.1 b/doc/pcre2test.1 index bbfd151..fc4e202 100644 --- a/doc/pcre2test.1 +++ b/doc/pcre2test.1 @@ -420,6 +420,7 @@ about the pattern: flipbytes flip endianness /BB fullbincode show binary code with lengths /I info show info about compiled pattern + hex pattern is coded in hexadecimal jit[=] use JIT locale= use this locale memory show memory used @@ -430,6 +431,7 @@ about the pattern: save= save compiled pattern stackguard= test the stackguard feature tables=[0|1|2] select internal tables + use_length use the pattern's length .sp The effects of these modifiers are described in the following sections. FIXME: Give more examples. @@ -481,6 +483,27 @@ specified. See also the section about saving and reloading compiled patterns below. . . +.SS "Specifying a pattern in hex" +.rs +.sp +The \fBhex\fP modifier specifies that the characters of the pattern are to be +interpreted as pairs of hexadecimal digits. White space is permitted between +pairs. For example: +.sp + /ab 32 59/hex +.sp +This feature is provided as a way of creating patterns that contain binary zero +characters. When \fBhex\fP is set, it implies \fBuse_length\fP. +. +. +.SS "Using the pattern's length" +.rs +.sp +By default, \fBpcre2test\fP passes patterns as zero-terminated strings to +\fBpcre2_compile()\fP, giving the length as -1. If \fBuse_length\fP is set, the +length of the pattern is passed. This is implied if \fBhex\fP is set. +. +. .SS "JIT compilation" .rs .sp @@ -595,38 +618,6 @@ letters, digits, spaces, etc. Setting alternate character tables and a locale are mutually exclusive. . . -.SS "Locking out certain modifiers" -.rs -.sp -FIXME FIXME -PCRE can be compiled with or without support for certain features such as -UTF-8/16/32 or Unicode properties. Accordingly, the standard tests are split up -into a number of different files that are selected for running depending on -which features are available. When updating the tests, it is all too easy to -put a new test into the wrong file by mistake; for example, to put a test that -requires UTF support into a file that is used when it is not available. To help -detect such mistakes as early as possible, there is a facility for locking out -specific modifiers. If an input line for \fBpcre2test\fP starts with the string -"< forbid " the following sequence of characters is taken as a list of -forbidden modifiers. For example, in the test files that must not use UTF or -Unicode property support, this line appears: -.sp - < forbid 8W -.sp -This locks out the /8 and /W modifiers. An immediate error is given if they are -subsequently encountered. If the character string contains < but not >, all the -multi-character modifiers that begin with < are locked out. Otherwise, such -modifiers must be explicitly listed, for example: -.sp - < forbid -.sp -There must be a single space between < and "forbid" for this feature to be -recognised. If there is not, the line is interpreted either as a request to -re-load a pre-compiled pattern (see "SAVING AND RELOADING COMPILED PATTERNS" -below) or, if there is a another < character, as a pattern that uses < as its -delimiter. -. -. .SS "Setting certain match controls" .rs .sp @@ -653,6 +644,7 @@ defaults, set them in a \fB#subject\fP command. The modifiers that can appear in subject lines and the \fB#subject\fP command are of two types. . +. .SS "Setting match options" .rs .sp @@ -1199,6 +1191,6 @@ Cambridge CB2 3QH, England. .rs .sp .nf -Last updated: 13 May 2014 +Last updated: 08 June 2014 Copyright (c) 1997-2014 University of Cambridge. .fi diff --git a/libpcre2-posix.pc b/libpcre2-posix.pc index bbbaf42..5844ddf 100644 --- a/libpcre2-posix.pc +++ b/libpcre2-posix.pc @@ -7,7 +7,7 @@ includedir=${prefix}/include Name: libpcre2-posix Description: Posix compatible interface to libpcre2-8 -Version: 9.00-DEV +Version: 10.00-DEV Libs: -L${libdir} -lpcre2-posix Cflags: -I${includedir} @PCRE_STATIC_CFLAG@ Requires.private: libpcre2-8 diff --git a/src/config.h b/src/config.h index 06262c9..b81449e 100644 --- a/src/config.h +++ b/src/config.h @@ -2,13 +2,13 @@ /* src/config.h.in. Generated from configure.ac by autoheader. */ -/* PCRE is written in Standard C, but there are a few non-standard things it +/* PCRE2 is written in Standard C, but there are a few non-standard things it can cope with, allowing it to run on SunOS4 and other "close to standard" systems. In environments that support the GNU autotools, config.h.in is converted into config.h by the "configure" script. In environments that use CMake, -config-cmake.in is converted into config.h. If you are going to build PCRE "by +config-cmake.in is converted into config.h. If you are going to build PCRE2 "by hand" without using "configure" or CMake, you should copy the distributed config.h.generic to config.h, and edit the macro definitions to be the way you need them. You must then add -DHAVE_CONFIG_H to all of your compile commands, @@ -25,21 +25,22 @@ macros are listed as a commented #undef in config.h.generic. Macros such as MATCH_LIMIT, whose actual value is relevant, have defaults defined, but are surrounded by #ifndef/#endif lines so that the value can be overridden by -D. -PCRE uses memmove() if HAVE_MEMMOVE is defined; otherwise it uses bcopy() if +PCRE2 uses memmove() if HAVE_MEMMOVE is defined; otherwise it uses bcopy() if HAVE_BCOPY is defined. If your system has neither bcopy() nor memmove(), make sure both macros are undefined; an emulation function will then be used. */ /* By default, the \R escape sequence matches any Unicode line ending character or sequence of characters. If BSR_ANYCRLF is defined (to any value), this is changed so that backslash-R matches only CR, LF, or CRLF. - The build-time default can be overridden by the user of PCRE at runtime. */ + The build-time default can be overridden by the user of PCRE2 at runtime. + */ /* #undef BSR_ANYCRLF */ /* If you are compiling for a system that uses EBCDIC instead of ASCII - character codes, define this macro to any value. When EBCDIC is set, PCRE + character codes, define this macro to any value. When EBCDIC is set, PCRE2 assumes that all input strings are in EBCDIC. If you do not define this - macro, PCRE will assume input strings are ASCII or UTF-8/16/32 Unicode. It - is not possible to build a version of PCRE that supports both EBCDIC and + macro, PCRE2 will assume input strings are ASCII or UTF-8/16/32 Unicode. It + is not possible to build a version of PCRE2 that supports both EBCDIC and UTF-8/16/32. */ /* #undef EBCDIC */ @@ -126,8 +127,8 @@ sure both macros are undefined; an emulation function will then be used. */ /* The value of LINK_SIZE determines the number of bytes used to store links as offsets within the compiled regex. The default is 2, which allows for compiled patterns up to 64K long. This covers the vast majority of cases. - However, PCRE can also be compiled to use 3 or 4 bytes instead. This allows - for longer patterns in extreme cases. */ + However, PCRE2 can also be compiled to use 3 or 4 bytes instead. This + allows for longer patterns in extreme cases. */ #define LINK_SIZE 2 /* Define to the sub-directory in which libtool stores uninstalled libraries. @@ -136,7 +137,7 @@ sure both macros are undefined; an emulation function will then be used. */ /* The value of MATCH_LIMIT determines the default number of times the internal match() function can be called during a single execution of - pcre_exec(). There is a runtime interface for setting a different limit. + pcre2_match(). There is a runtime interface for setting a different limit. The limit exists in order to catch runaway regular expressions that take for ever to determine that they do not match. The default is set very large so that it does not accidentally catch legitimate cases. */ @@ -162,19 +163,18 @@ sure both macros are undefined; an emulation function will then be used. */ overflow caused by enormously large patterns. */ #define MAX_NAME_SIZE 32 -/* The value of NEWLINE determines the default newline character sequence. - PCRE client programs can override this by selecting other values at run - time. The valid values are 0 (CR), 1 (LF), 2 (CRLF), 3 (ANY), and 4 +/* The value of NEWLINE_DEFAULT determines the default newline character + sequence. PCRE2 client programs can override this by selecting other values + at run time. The valid values are 1 (CR), 2 (LF), 3 (CRLF), 4 (ANY), and 5 (ANYCRLF). */ -#define NEWLINE 1 +#define NEWLINE_DEFAULT 2 -/* PCRE uses recursive function calls to handle backtracking while matching. +/* PCRE2 uses recursive function calls to handle backtracking while matching. This can sometimes be a problem on systems that have stacks of limited size. Define NO_RECURSE to any value to get a version that doesn't use recursion in the match() function; instead it creates its own stack by - steam using pcre_recurse_malloc() to obtain memory from the heap. For more - detail, see the comments and other stuff just above the match() function. - */ + steam using memory from the heap. For more detail, see the comments and + other stuff just above the match() function. */ /* #undef NO_RECURSE */ /* Name of package */ @@ -187,7 +187,7 @@ sure both macros are undefined; an emulation function will then be used. */ #define PACKAGE_NAME "PCRE2" /* Define to the full name and version of this package. */ -#define PACKAGE_STRING "PCRE2 9.00-DEV" +#define PACKAGE_STRING "PCRE2 10.00-DEV" /* Define to the one symbol short name of this package. */ #define PACKAGE_TARNAME "pcre2" @@ -196,13 +196,20 @@ sure both macros are undefined; an emulation function will then be used. */ #define PACKAGE_URL "" /* Define to the version of this package. */ -#define PACKAGE_VERSION "9.00-DEV" +#define PACKAGE_VERSION "10.00-DEV" /* The value of PARENS_NEST_LIMIT specifies the maximum depth of nested parentheses (of any kind) in a pattern. This limits the amount of system stack that is used while compiling a pattern. */ #define PARENS_NEST_LIMIT 250 +/* The value of PCRE2GREP_BUFSIZE determines the size of buffer used by + pcre2grep to hold parts of the file it is searching. This is also the + minimum value. The actual amount of memory used by pcre2grep is three times + this number, because it allows for the buffering of "before" and "after" + lines. */ +#define PCRE2GREP_BUFSIZE 20480 + /* to make a symbol visible */ #define PCRE2POSIX_EXP_DECL extern __attribute__ ((visibility ("default"))) @@ -227,13 +234,6 @@ sure both macros are undefined; an emulation function will then be used. */ /* Define to any value if linking statically (TODO: make nice with Libtool) */ #define PCRE2_STATIC 1 -/* The value of PCREGREP_BUFSIZE determines the size of buffer used by - pcregrep to hold parts of the file it is searching. This is also the - minimum value. The actual amount of memory used by pcregrep is three times - this number, because it allows for the buffering of "before" and "after" - lines. */ -#define PCREGREP_BUFSIZE 20480 - /* Define to necessary symbol if this constant uses a non-standard name on your system. */ /* #undef PTHREAD_CREATE_JOINABLE */ @@ -244,35 +244,35 @@ sure both macros are undefined; an emulation function will then be used. */ /* Define to any value to enable support for Just-In-Time compiling. */ /* #undef SUPPORT_JIT */ -/* Define to any value to allow pcregrep to be linked with libbz2, so that it +/* Define to any value to allow pcre2grep to be linked with libbz2, so that it is able to handle .bz2 files. */ #define SUPPORT_LIBBZ2 /**/ -/* Define to any value to allow pcretest to be linked with libedit. */ +/* Define to any value to allow pcre2test to be linked with libedit. */ /* #undef SUPPORT_LIBEDIT */ -/* Define to any value to allow pcretest to be linked with libreadline. */ +/* Define to any value to allow pcre2test to be linked with libreadline. */ #define SUPPORT_LIBREADLINE /**/ -/* Define to any value to allow pcregrep to be linked with libz, so that it is - able to handle .gz files. */ +/* Define to any value to allow pcre2grep to be linked with libz, so that it + is able to handle .gz files. */ #define SUPPORT_LIBZ /**/ -/* Define to any value to enable the 16 bit PCRE library. */ +/* Define to any value to enable the 16 bit PCRE2 library. */ #define SUPPORT_PCRE16 /**/ -/* Define to any value to enable the 32 bit PCRE library. */ +/* Define to any value to enable JIT support in pcre2grep. */ +/* #undef SUPPORT_PCRE2GREP_JIT */ + +/* Define to any value to enable the 32 bit PCRE2 library. */ #define SUPPORT_PCRE32 /**/ -/* Define to any value to enable the 8 bit PCRE library. */ +/* Define to any value to enable the 8 bit PCRE2 library. */ #define SUPPORT_PCRE8 /**/ -/* Define to any value to enable JIT support in pcregrep. */ -/* #undef SUPPORT_PCREGREP_JIT */ - /* Define to any value to enable support for the UTF-8/16/32 Unicode encoding. This will work even in an EBCDIC environment, but it is incompatible with - the EBCDIC macro. That is, PCRE can support *either* EBCDIC code *or* + the EBCDIC macro. That is, PCRE2 can support *either* EBCDIC code *or* ASCII/UTF-8/16/32, but not both at once. */ #define SUPPORT_UTF /**/ @@ -280,7 +280,7 @@ sure both macros are undefined; an emulation function will then be used. */ #define SUPPORT_VALGRIND /**/ /* Version number of package */ -#define VERSION "9.00-DEV" +#define VERSION "10.00-DEV" /* Define to empty if `const' does not conform to ANSI C. */ /* #undef const */ diff --git a/src/config.h.in b/src/config.h.in index a512e78..ec5365f 100644 --- a/src/config.h.in +++ b/src/config.h.in @@ -1,13 +1,13 @@ /* src/config.h.in. Generated from configure.ac by autoheader. */ -/* PCRE is written in Standard C, but there are a few non-standard things it +/* PCRE2 is written in Standard C, but there are a few non-standard things it can cope with, allowing it to run on SunOS4 and other "close to standard" systems. In environments that support the GNU autotools, config.h.in is converted into config.h by the "configure" script. In environments that use CMake, -config-cmake.in is converted into config.h. If you are going to build PCRE "by +config-cmake.in is converted into config.h. If you are going to build PCRE2 "by hand" without using "configure" or CMake, you should copy the distributed config.h.generic to config.h, and edit the macro definitions to be the way you need them. You must then add -DHAVE_CONFIG_H to all of your compile commands, @@ -24,21 +24,22 @@ macros are listed as a commented #undef in config.h.generic. Macros such as MATCH_LIMIT, whose actual value is relevant, have defaults defined, but are surrounded by #ifndef/#endif lines so that the value can be overridden by -D. -PCRE uses memmove() if HAVE_MEMMOVE is defined; otherwise it uses bcopy() if +PCRE2 uses memmove() if HAVE_MEMMOVE is defined; otherwise it uses bcopy() if HAVE_BCOPY is defined. If your system has neither bcopy() nor memmove(), make sure both macros are undefined; an emulation function will then be used. */ /* By default, the \R escape sequence matches any Unicode line ending character or sequence of characters. If BSR_ANYCRLF is defined (to any value), this is changed so that backslash-R matches only CR, LF, or CRLF. - The build-time default can be overridden by the user of PCRE at runtime. */ + The build-time default can be overridden by the user of PCRE2 at runtime. + */ #undef BSR_ANYCRLF /* If you are compiling for a system that uses EBCDIC instead of ASCII - character codes, define this macro to any value. When EBCDIC is set, PCRE + character codes, define this macro to any value. When EBCDIC is set, PCRE2 assumes that all input strings are in EBCDIC. If you do not define this - macro, PCRE will assume input strings are ASCII or UTF-8/16/32 Unicode. It - is not possible to build a version of PCRE that supports both EBCDIC and + macro, PCRE2 will assume input strings are ASCII or UTF-8/16/32 Unicode. It + is not possible to build a version of PCRE2 that supports both EBCDIC and UTF-8/16/32. */ #undef EBCDIC @@ -125,8 +126,8 @@ sure both macros are undefined; an emulation function will then be used. */ /* The value of LINK_SIZE determines the number of bytes used to store links as offsets within the compiled regex. The default is 2, which allows for compiled patterns up to 64K long. This covers the vast majority of cases. - However, PCRE can also be compiled to use 3 or 4 bytes instead. This allows - for longer patterns in extreme cases. */ + However, PCRE2 can also be compiled to use 3 or 4 bytes instead. This + allows for longer patterns in extreme cases. */ #undef LINK_SIZE /* Define to the sub-directory in which libtool stores uninstalled libraries. @@ -135,7 +136,7 @@ sure both macros are undefined; an emulation function will then be used. */ /* The value of MATCH_LIMIT determines the default number of times the internal match() function can be called during a single execution of - pcre_exec(). There is a runtime interface for setting a different limit. + pcre2_match(). There is a runtime interface for setting a different limit. The limit exists in order to catch runaway regular expressions that take for ever to determine that they do not match. The default is set very large so that it does not accidentally catch legitimate cases. */ @@ -161,19 +162,18 @@ sure both macros are undefined; an emulation function will then be used. */ overflow caused by enormously large patterns. */ #undef MAX_NAME_SIZE -/* The value of NEWLINE determines the default newline character sequence. - PCRE client programs can override this by selecting other values at run - time. The valid values are 0 (CR), 1 (LF), 2 (CRLF), 3 (ANY), and 4 +/* The value of NEWLINE_DEFAULT determines the default newline character + sequence. PCRE2 client programs can override this by selecting other values + at run time. The valid values are 1 (CR), 2 (LF), 3 (CRLF), 4 (ANY), and 5 (ANYCRLF). */ -#undef NEWLINE +#undef NEWLINE_DEFAULT -/* PCRE uses recursive function calls to handle backtracking while matching. +/* PCRE2 uses recursive function calls to handle backtracking while matching. This can sometimes be a problem on systems that have stacks of limited size. Define NO_RECURSE to any value to get a version that doesn't use recursion in the match() function; instead it creates its own stack by - steam using pcre_recurse_malloc() to obtain memory from the heap. For more - detail, see the comments and other stuff just above the match() function. - */ + steam using memory from the heap. For more detail, see the comments and + other stuff just above the match() function. */ #undef NO_RECURSE /* Name of package */ @@ -202,6 +202,13 @@ sure both macros are undefined; an emulation function will then be used. */ stack that is used while compiling a pattern. */ #undef PARENS_NEST_LIMIT +/* The value of PCRE2GREP_BUFSIZE determines the size of buffer used by + pcre2grep to hold parts of the file it is searching. This is also the + minimum value. The actual amount of memory used by pcre2grep is three times + this number, because it allows for the buffering of "before" and "after" + lines. */ +#undef PCRE2GREP_BUFSIZE + /* to make a symbol visible */ #undef PCRE2POSIX_EXP_DECL @@ -226,13 +233,6 @@ sure both macros are undefined; an emulation function will then be used. */ /* Define to any value if linking statically (TODO: make nice with Libtool) */ #undef PCRE2_STATIC -/* The value of PCREGREP_BUFSIZE determines the size of buffer used by - pcregrep to hold parts of the file it is searching. This is also the - minimum value. The actual amount of memory used by pcregrep is three times - this number, because it allows for the buffering of "before" and "after" - lines. */ -#undef PCREGREP_BUFSIZE - /* Define to necessary symbol if this constant uses a non-standard name on your system. */ #undef PTHREAD_CREATE_JOINABLE @@ -243,35 +243,35 @@ sure both macros are undefined; an emulation function will then be used. */ /* Define to any value to enable support for Just-In-Time compiling. */ #undef SUPPORT_JIT -/* Define to any value to allow pcregrep to be linked with libbz2, so that it +/* Define to any value to allow pcre2grep to be linked with libbz2, so that it is able to handle .bz2 files. */ #undef SUPPORT_LIBBZ2 -/* Define to any value to allow pcretest to be linked with libedit. */ +/* Define to any value to allow pcre2test to be linked with libedit. */ #undef SUPPORT_LIBEDIT -/* Define to any value to allow pcretest to be linked with libreadline. */ +/* Define to any value to allow pcre2test to be linked with libreadline. */ #undef SUPPORT_LIBREADLINE -/* Define to any value to allow pcregrep to be linked with libz, so that it is - able to handle .gz files. */ +/* Define to any value to allow pcre2grep to be linked with libz, so that it + is able to handle .gz files. */ #undef SUPPORT_LIBZ -/* Define to any value to enable the 16 bit PCRE library. */ +/* Define to any value to enable the 16 bit PCRE2 library. */ #undef SUPPORT_PCRE16 -/* Define to any value to enable the 32 bit PCRE library. */ +/* Define to any value to enable JIT support in pcre2grep. */ +#undef SUPPORT_PCRE2GREP_JIT + +/* Define to any value to enable the 32 bit PCRE2 library. */ #undef SUPPORT_PCRE32 -/* Define to any value to enable the 8 bit PCRE library. */ +/* Define to any value to enable the 8 bit PCRE2 library. */ #undef SUPPORT_PCRE8 -/* Define to any value to enable JIT support in pcregrep. */ -#undef SUPPORT_PCREGREP_JIT - /* Define to any value to enable support for the UTF-8/16/32 Unicode encoding. This will work even in an EBCDIC environment, but it is incompatible with - the EBCDIC macro. That is, PCRE can support *either* EBCDIC code *or* + the EBCDIC macro. That is, PCRE2 can support *either* EBCDIC code *or* ASCII/UTF-8/16/32, but not both at once. */ #undef SUPPORT_UTF diff --git a/src/pcre2.h b/src/pcre2.h index 91d33bb..5df9861 100644 --- a/src/pcre2.h +++ b/src/pcre2.h @@ -41,7 +41,7 @@ POSSIBILITY OF SUCH DAMAGE. /* The current PCRE version information. */ -#define PCRE2_MAJOR 9 +#define PCRE2_MAJOR 10 #define PCRE2_MINOR 00 #define PCRE2_PRERELEASE -DEV #define PCRE2_DATE 2014-99-99 @@ -138,86 +138,83 @@ D is inspected during pcre2_dfa_exec() execution /* Newline and \R settings, for use in the compile context. */ -#define PCRE2_NEWLINE_DEFAULT 0 #define PCRE2_NEWLINE_CR 1 #define PCRE2_NEWLINE_LF 2 #define PCRE2_NEWLINE_CRLF 3 #define PCRE2_NEWLINE_ANY 4 #define PCRE2_NEWLINE_ANYCRLF 5 -#define PCRE2_BSR_DEFAULT 0 #define PCRE2_BSR_UNICODE 1 #define PCRE2_BSR_ANYCRLF 2 -/* Match-time and get/set-time error codes */ +/* Error codes: no match and partial match are "expected" errors. */ #define PCRE2_ERROR_NOMATCH (-1) +#define PCRE2_ERROR_PARTIAL (-2) -#define PCRE2_ERROR_BADCOUNT (-2) -#define PCRE2_ERROR_BADENDIANNESS (-3) -#define PCRE2_ERROR_BADLENGTH (-4) -#define PCRE2_ERROR_BADMAGIC (-5) -#define PCRE2_ERROR_BADMODE (-6) -#define PCRE2_ERROR_BADOFFSET (-7) -#define PCRE2_ERROR_BADOPTION (-8) -#define PCRE2_ERROR_BADUTF (-9) -#define PCRE2_ERROR_BADUTF_OFFSET (-10) -#define PCRE2_ERROR_CALLOUT (-11) /* Never used by PCRE2 itself */ -#define PCRE2_ERROR_INTERNAL (-12) -#define PCRE2_ERROR_JIT_BADOPTION (-13) -#define PCRE2_ERROR_JIT_STACKLIMIT (-14) -#define PCRE2_ERROR_MATCHLIMIT (-15) -#define PCRE2_ERROR_NOMEMORY (-16) -#define PCRE2_ERROR_NOSUBSTRING (-17) -#define PCRE2_ERROR_NULL (-18) -#define PCRE2_ERROR_PARTIAL (-19) -#define PCRE2_ERROR_RECURSELOOP (-20) -#define PCRE2_ERROR_RECURSIONLIMIT (-21) -#define PCRE2_ERROR_UNKNOWN_OPCODE (-22) -#define PCRE2_ERROR_UNSET (-23) +/* Error codes for UTF-8 validity checks */ -#define PCRE2_ERROR_DFA_BADRESTART (-30) -#define PCRE2_ERROR_DFA_RECURSE (-31) -#define PCRE2_ERROR_DFA_UCOND (-32) -#define PCRE2_ERROR_DFA_UITEM (-33) -#define PCRE2_ERROR_DFA_UMLIMIT (-34) -#define PCRE2_ERROR_DFA_WSSIZE (-35) +#define PCRE2_ERROR_UTF8_ERR1 (-3) +#define PCRE2_ERROR_UTF8_ERR2 (-4) +#define PCRE2_ERROR_UTF8_ERR3 (-5) +#define PCRE2_ERROR_UTF8_ERR4 (-6) +#define PCRE2_ERROR_UTF8_ERR5 (-7) +#define PCRE2_ERROR_UTF8_ERR6 (-8) +#define PCRE2_ERROR_UTF8_ERR7 (-9) +#define PCRE2_ERROR_UTF8_ERR8 (-10) +#define PCRE2_ERROR_UTF8_ERR9 (-11) +#define PCRE2_ERROR_UTF8_ERR10 (-12) +#define PCRE2_ERROR_UTF8_ERR11 (-13) +#define PCRE2_ERROR_UTF8_ERR12 (-14) +#define PCRE2_ERROR_UTF8_ERR13 (-15) +#define PCRE2_ERROR_UTF8_ERR14 (-16) +#define PCRE2_ERROR_UTF8_ERR15 (-17) +#define PCRE2_ERROR_UTF8_ERR16 (-18) +#define PCRE2_ERROR_UTF8_ERR17 (-19) +#define PCRE2_ERROR_UTF8_ERR18 (-20) +#define PCRE2_ERROR_UTF8_ERR19 (-21) +#define PCRE2_ERROR_UTF8_ERR20 (-22) +#define PCRE2_ERROR_UTF8_ERR21 (-23) +/* Error codes for UTF-16 validity checks */ -/* Specific error codes for UTF-8 validity checks */ +#define PCRE2_ERROR_UTF16_ERR1 (-24) +#define PCRE2_ERROR_UTF16_ERR2 (-25) +#define PCRE2_ERROR_UTF16_ERR3 (-26) -#define PCRE2_ERROR_UTF8_ERR1 (-41) -#define PCRE2_ERROR_UTF8_ERR2 (-42) -#define PCRE2_ERROR_UTF8_ERR3 (-43) -#define PCRE2_ERROR_UTF8_ERR4 (-44) -#define PCRE2_ERROR_UTF8_ERR5 (-45) -#define PCRE2_ERROR_UTF8_ERR6 (-46) -#define PCRE2_ERROR_UTF8_ERR7 (-47) -#define PCRE2_ERROR_UTF8_ERR8 (-48) -#define PCRE2_ERROR_UTF8_ERR9 (-49) -#define PCRE2_ERROR_UTF8_ERR10 (-50) -#define PCRE2_ERROR_UTF8_ERR11 (-51) -#define PCRE2_ERROR_UTF8_ERR12 (-52) -#define PCRE2_ERROR_UTF8_ERR13 (-53) -#define PCRE2_ERROR_UTF8_ERR14 (-54) -#define PCRE2_ERROR_UTF8_ERR15 (-55) -#define PCRE2_ERROR_UTF8_ERR16 (-56) -#define PCRE2_ERROR_UTF8_ERR17 (-57) -#define PCRE2_ERROR_UTF8_ERR18 (-58) -#define PCRE2_ERROR_UTF8_ERR19 (-59) -#define PCRE2_ERROR_UTF8_ERR20 (-60) -#define PCRE2_ERROR_UTF8_ERR21 (-61) +/* Error codes for UTF-32 validity checks */ -/* Specific error codes for UTF-16 validity checks */ +#define PCRE2_ERROR_UTF32_ERR1 (-27) +#define PCRE2_ERROR_UTF32_ERR2 (-28) -#define PCRE2_ERROR_UTF16_ERR1 (-62) -#define PCRE2_ERROR_UTF16_ERR2 (-63) -#define PCRE2_ERROR_UTF16_ERR3 (-64) +/* Error codes for pcre2[_dfa]_match() */ -/* Specific error codes for UTF-32 validity checks */ - -#define PCRE2_ERROR_UTF32_ERR1 (-65) -#define PCRE2_ERROR_UTF32_ERR3 (-66) +#define PCRE2_ERROR_BADCOUNT (-29) +#define PCRE2_ERROR_BADENDIANNESS (-30) +#define PCRE2_ERROR_BADLENGTH (-31) +#define PCRE2_ERROR_BADMAGIC (-32) +#define PCRE2_ERROR_BADMODE (-33) +#define PCRE2_ERROR_BADOFFSET (-34) +#define PCRE2_ERROR_BADOPTION (-35) +#define PCRE2_ERROR_BADUTF_OFFSET (-36) +#define PCRE2_ERROR_CALLOUT (-37) /* Never used by PCRE2 itself */ +#define PCRE2_ERROR_DFA_BADRESTART (-38) +#define PCRE2_ERROR_DFA_RECURSE (-39) +#define PCRE2_ERROR_DFA_UCOND (-40) +#define PCRE2_ERROR_DFA_UITEM (-41) +#define PCRE2_ERROR_DFA_UMLIMIT (-42) +#define PCRE2_ERROR_DFA_WSSIZE (-43) +#define PCRE2_ERROR_INTERNAL (-44) +#define PCRE2_ERROR_JIT_BADOPTION (-45) +#define PCRE2_ERROR_JIT_STACKLIMIT (-46) +#define PCRE2_ERROR_MATCHLIMIT (-47) +#define PCRE2_ERROR_NOMEMORY (-48) +#define PCRE2_ERROR_NOSUBSTRING (-49) +#define PCRE2_ERROR_NULL (-50) +#define PCRE2_ERROR_RECURSELOOP (-51) +#define PCRE2_ERROR_RECURSIONLIMIT (-52) +#define PCRE2_ERROR_UNKNOWN_OPCODE (-53) +#define PCRE2_ERROR_UNSET (-54) /* Request types for pcre2_pattern_info() */ diff --git a/src/pcre2.h.in b/src/pcre2.h.in index e284716..7741de2 100644 --- a/src/pcre2.h.in +++ b/src/pcre2.h.in @@ -138,86 +138,83 @@ D is inspected during pcre2_dfa_exec() execution /* Newline and \R settings, for use in the compile context. */ -#define PCRE2_NEWLINE_DEFAULT 0 #define PCRE2_NEWLINE_CR 1 #define PCRE2_NEWLINE_LF 2 #define PCRE2_NEWLINE_CRLF 3 #define PCRE2_NEWLINE_ANY 4 #define PCRE2_NEWLINE_ANYCRLF 5 -#define PCRE2_BSR_DEFAULT 0 #define PCRE2_BSR_UNICODE 1 #define PCRE2_BSR_ANYCRLF 2 -/* Match-time and get/set-time error codes */ +/* Error codes: no match and partial match are "expected" errors. */ #define PCRE2_ERROR_NOMATCH (-1) +#define PCRE2_ERROR_PARTIAL (-2) -#define PCRE2_ERROR_BADCOUNT (-2) -#define PCRE2_ERROR_BADENDIANNESS (-3) -#define PCRE2_ERROR_BADLENGTH (-4) -#define PCRE2_ERROR_BADMAGIC (-5) -#define PCRE2_ERROR_BADMODE (-6) -#define PCRE2_ERROR_BADOFFSET (-7) -#define PCRE2_ERROR_BADOPTION (-8) -#define PCRE2_ERROR_BADUTF (-9) -#define PCRE2_ERROR_BADUTF_OFFSET (-10) -#define PCRE2_ERROR_CALLOUT (-11) /* Never used by PCRE2 itself */ -#define PCRE2_ERROR_INTERNAL (-12) -#define PCRE2_ERROR_JIT_BADOPTION (-13) -#define PCRE2_ERROR_JIT_STACKLIMIT (-14) -#define PCRE2_ERROR_MATCHLIMIT (-15) -#define PCRE2_ERROR_NOMEMORY (-16) -#define PCRE2_ERROR_NOSUBSTRING (-17) -#define PCRE2_ERROR_NULL (-18) -#define PCRE2_ERROR_PARTIAL (-19) -#define PCRE2_ERROR_RECURSELOOP (-20) -#define PCRE2_ERROR_RECURSIONLIMIT (-21) -#define PCRE2_ERROR_UNKNOWN_OPCODE (-22) -#define PCRE2_ERROR_UNSET (-23) +/* Error codes for UTF-8 validity checks */ -#define PCRE2_ERROR_DFA_BADRESTART (-30) -#define PCRE2_ERROR_DFA_RECURSE (-31) -#define PCRE2_ERROR_DFA_UCOND (-32) -#define PCRE2_ERROR_DFA_UITEM (-33) -#define PCRE2_ERROR_DFA_UMLIMIT (-34) -#define PCRE2_ERROR_DFA_WSSIZE (-35) +#define PCRE2_ERROR_UTF8_ERR1 (-3) +#define PCRE2_ERROR_UTF8_ERR2 (-4) +#define PCRE2_ERROR_UTF8_ERR3 (-5) +#define PCRE2_ERROR_UTF8_ERR4 (-6) +#define PCRE2_ERROR_UTF8_ERR5 (-7) +#define PCRE2_ERROR_UTF8_ERR6 (-8) +#define PCRE2_ERROR_UTF8_ERR7 (-9) +#define PCRE2_ERROR_UTF8_ERR8 (-10) +#define PCRE2_ERROR_UTF8_ERR9 (-11) +#define PCRE2_ERROR_UTF8_ERR10 (-12) +#define PCRE2_ERROR_UTF8_ERR11 (-13) +#define PCRE2_ERROR_UTF8_ERR12 (-14) +#define PCRE2_ERROR_UTF8_ERR13 (-15) +#define PCRE2_ERROR_UTF8_ERR14 (-16) +#define PCRE2_ERROR_UTF8_ERR15 (-17) +#define PCRE2_ERROR_UTF8_ERR16 (-18) +#define PCRE2_ERROR_UTF8_ERR17 (-19) +#define PCRE2_ERROR_UTF8_ERR18 (-20) +#define PCRE2_ERROR_UTF8_ERR19 (-21) +#define PCRE2_ERROR_UTF8_ERR20 (-22) +#define PCRE2_ERROR_UTF8_ERR21 (-23) +/* Error codes for UTF-16 validity checks */ -/* Specific error codes for UTF-8 validity checks */ +#define PCRE2_ERROR_UTF16_ERR1 (-24) +#define PCRE2_ERROR_UTF16_ERR2 (-25) +#define PCRE2_ERROR_UTF16_ERR3 (-26) -#define PCRE2_ERROR_UTF8_ERR1 (-41) -#define PCRE2_ERROR_UTF8_ERR2 (-42) -#define PCRE2_ERROR_UTF8_ERR3 (-43) -#define PCRE2_ERROR_UTF8_ERR4 (-44) -#define PCRE2_ERROR_UTF8_ERR5 (-45) -#define PCRE2_ERROR_UTF8_ERR6 (-46) -#define PCRE2_ERROR_UTF8_ERR7 (-47) -#define PCRE2_ERROR_UTF8_ERR8 (-48) -#define PCRE2_ERROR_UTF8_ERR9 (-49) -#define PCRE2_ERROR_UTF8_ERR10 (-50) -#define PCRE2_ERROR_UTF8_ERR11 (-51) -#define PCRE2_ERROR_UTF8_ERR12 (-52) -#define PCRE2_ERROR_UTF8_ERR13 (-53) -#define PCRE2_ERROR_UTF8_ERR14 (-54) -#define PCRE2_ERROR_UTF8_ERR15 (-55) -#define PCRE2_ERROR_UTF8_ERR16 (-56) -#define PCRE2_ERROR_UTF8_ERR17 (-57) -#define PCRE2_ERROR_UTF8_ERR18 (-58) -#define PCRE2_ERROR_UTF8_ERR19 (-59) -#define PCRE2_ERROR_UTF8_ERR20 (-60) -#define PCRE2_ERROR_UTF8_ERR21 (-61) +/* Error codes for UTF-32 validity checks */ -/* Specific error codes for UTF-16 validity checks */ +#define PCRE2_ERROR_UTF32_ERR1 (-27) +#define PCRE2_ERROR_UTF32_ERR2 (-28) -#define PCRE2_ERROR_UTF16_ERR1 (-62) -#define PCRE2_ERROR_UTF16_ERR2 (-63) -#define PCRE2_ERROR_UTF16_ERR3 (-64) +/* Error codes for pcre2[_dfa]_match() */ -/* Specific error codes for UTF-32 validity checks */ - -#define PCRE2_ERROR_UTF32_ERR1 (-65) -#define PCRE2_ERROR_UTF32_ERR3 (-66) +#define PCRE2_ERROR_BADCOUNT (-29) +#define PCRE2_ERROR_BADENDIANNESS (-30) +#define PCRE2_ERROR_BADLENGTH (-31) +#define PCRE2_ERROR_BADMAGIC (-32) +#define PCRE2_ERROR_BADMODE (-33) +#define PCRE2_ERROR_BADOFFSET (-34) +#define PCRE2_ERROR_BADOPTION (-35) +#define PCRE2_ERROR_BADUTF_OFFSET (-36) +#define PCRE2_ERROR_CALLOUT (-37) /* Never used by PCRE2 itself */ +#define PCRE2_ERROR_DFA_BADRESTART (-38) +#define PCRE2_ERROR_DFA_RECURSE (-39) +#define PCRE2_ERROR_DFA_UCOND (-40) +#define PCRE2_ERROR_DFA_UITEM (-41) +#define PCRE2_ERROR_DFA_UMLIMIT (-42) +#define PCRE2_ERROR_DFA_WSSIZE (-43) +#define PCRE2_ERROR_INTERNAL (-44) +#define PCRE2_ERROR_JIT_BADOPTION (-45) +#define PCRE2_ERROR_JIT_STACKLIMIT (-46) +#define PCRE2_ERROR_MATCHLIMIT (-47) +#define PCRE2_ERROR_NOMEMORY (-48) +#define PCRE2_ERROR_NOSUBSTRING (-49) +#define PCRE2_ERROR_NULL (-50) +#define PCRE2_ERROR_RECURSELOOP (-51) +#define PCRE2_ERROR_RECURSIONLIMIT (-52) +#define PCRE2_ERROR_UNKNOWN_OPCODE (-53) +#define PCRE2_ERROR_UNSET (-54) /* Request types for pcre2_pattern_info() */ diff --git a/src/pcre2_compile.c b/src/pcre2_compile.c index 65754a9..19a1ffa 100644 --- a/src/pcre2_compile.c +++ b/src/pcre2_compile.c @@ -43,10 +43,627 @@ POSSIBILITY OF SUCH DAMAGE. #include "config.h" #endif +#define NLBLOCK cd /* Block containing newline information */ +#define PSSTART start_pattern /* Field containing processed string start */ +#define PSEND end_pattern /* Field containing processed string end */ + #include "pcre2_internal.h" -/* FIXME: this is currently a placeholder function */ +/* There are a few things that vary with different code unit sizes. Handle them +by defining macros in order to minimize #if usage. */ + +#if PCRE2_CODE_UNIT_WIDTH == 8 +#define STRING_UTFn_RIGHTPAR STRING_UTF8_RIGHTPAR, 5 +#define BAD_UTF_ERROR ERR44 +#define XDIGIT(c) xdigitab[c] + +#else /* Either 16-bit or 32-bit */ +#define XDIGIT(c) (MAX_255(c)? xdigitab[c] : 0xff) + +#if PCRE2_CODE_UNIT_WIDTH == 16 +#define STRING_UTFn_RIGHTPAR STRING_UTF16_RIGHTPAR, 6 +#define BAD_UTF_ERROR ERR74 + +#else +#define STRING_UTFn_RIGHTPAR STRING_UTF32_RIGHTPAR, 6 +#define BAD_UTF_ERROR ERR77 +#endif +#endif + + +/* Function definitions to allow mutual recursion */ + +static int + add_list_to_class(uint8_t *, PCRE2_UCHAR **, int, compile_data *, + const uint32_t *, unsigned int); + +static BOOL + compile_regex(uint32_t, PCRE2_UCHAR **, PCRE2_SPTR *, int *, BOOL, BOOL, int, + int, uint32_t *, int32_t *, uint32_t *, int32_t *, branch_chain *, + compile_data *, size_t *); + + + +/************************************************* +* Code parameters and static tables * +*************************************************/ + +/* This value specifies the size of stack workspace, which is used during the +pre-compile phase when determining how much memory is required. The regex is +partly compiled into this space, but the compiled parts are discarded as soon +as they can be, so that hopefully there will never be an overrun. The code +does, however, check for an overrun. The largest amount I've seen used is 218, +so this number is very generous. + +The same workspace is used during the second, actual compile phase for +remembering forward references to groups so that they can be filled in at the +end. Each entry in this list occupies LINK_SIZE bytes, so even when LINK_SIZE +is 4 there is plenty of room for most patterns. However, the memory can get +filled up by repetitions of forward references, for example patterns like +/(?1){0,1999}(b)/, and one user did hit the limit. The code has been changed so +that the workspace is expanded in this situation. The value below is therefore +a minimum, and we put a maximum on it for safety. The minimum is now also +defined in terms of LINK_SIZE so that the size increase kicks in at the same +number of forward references in all cases. */ + +#define COMPILE_WORK_SIZE (2048*LINK_SIZE) +#define COMPILE_WORK_SIZE_MAX (100*COMPILE_WORK_SIZE) + +/* The overrun tests check for a slightly smaller size so that they detect the +overrun before it actually does run off the end of the data block. */ + +#define WORK_SIZE_SAFETY_MARGIN (100) + +/* This value determines the size of the initial vector that is used for +remembering named groups during the pre-compile. It is allocated on the stack, +but if it is too small, it is expanded, in a similar way to the workspace. The +value is the number of slots in the list. */ + +#define NAMED_GROUP_LIST_SIZE 20 + +/* The original PCRE required patterns to be zero-terminated, and it simplifies +the compiling code if it is guaranteed that there is a zero code unit at the +end of the pattern, because this means that tests for coding sequences such as +(*SKIP) or even just (?<= can check a sequence of code units without having to +keep checking for the end of the pattern. The new PCRE2 API allows zero code +units within patterns if a positive length is given, but in order to keep most +of the compiling code as it was, we copy such patterns and add a zero on the +end. This value determines the size of space on the stack that is used if the +pattern fits; if not, heap memory is used. */ + +#define COPIED_PATTERN_SIZE 1024 + +/* Maximum length value to check against when making sure that the variable +that holds the compiled pattern length does not overflow. We make it a bit less +than INT_MAX to allow for adding in group terminating bytes, so that we don't +have to check them every time. */ + +#define OFLOW_MAX (INT_MAX - 20) + +/* Macro for setting individual bits in class bitmaps. */ + +#define SETBIT(a,b) a[(b)/8] |= (1 << ((b)&7)) + +/* Private flags added to firstcu and reqcu. */ + +#define REQ_CASELESS (1 << 0) /* Indicates caselessness */ +#define REQ_VARY (1 << 1) /* reqcu followed non-literal item */ +/* Negative values for the firstcu and reqcu flags */ +#define REQ_UNSET (-2) +#define REQ_NONE (-1) + +/* This bit (which is greater than any UTF value) is used to indicate that a +variable contains a number of code units instead of an actual code point. */ + +#define UTF_LENGTH 0x10000000l + +/* This simple test for a decimal digit works for both ASCII/Unicode and EBCDIC +and is fast (a good compiler can turn it into a subtraction and unsigned +comparison). */ + +#define IS_DIGIT(x) ((x) >= CHAR_0 && (x) <= CHAR_9) + +/* Table to identify hex digits. The tables in chartables are dependent on the +locale, and may mark arbitrary characters as digits. We want to recognize only +0-9, a-z, and A-Z as hex digits, which is why we have a private table here. It +costs 256 bytes, but it is a lot faster than doing character value tests (at +least in some simple cases I timed), and in some applications one wants PCRE to +compile efficiently as well as match efficiently. The value in the table is +the binary hex digit value, or 0xff for non-hex digits. */ + +/* This is the "normal" case, for ASCII systems, and EBCDIC systems running in +UTF-8 mode. */ + +#ifndef EBCDIC +static const uint8_t xdigitab[] = + { + 0xff,0xff,0xff,0xff,0xff,0xff,0xff,0xff, /* 0- 7 */ + 0xff,0xff,0xff,0xff,0xff,0xff,0xff,0xff, /* 8- 15 */ + 0xff,0xff,0xff,0xff,0xff,0xff,0xff,0xff, /* 16- 23 */ + 0xff,0xff,0xff,0xff,0xff,0xff,0xff,0xff, /* 24- 31 */ + 0xff,0xff,0xff,0xff,0xff,0xff,0xff,0xff, /* - ' */ + 0xff,0xff,0xff,0xff,0xff,0xff,0xff,0xff, /* ( - / */ + 0x00,0x01,0x02,0x03,0x04,0x05,0x06,0x07, /* 0 - 7 */ + 0x08,0x09,0xff,0xff,0xff,0xff,0xff,0xff, /* 8 - ? */ + 0xff,0x0a,0x0b,0x0c,0x0d,0x0e,0x0f,0xff, /* @ - G */ + 0xff,0xff,0xff,0xff,0xff,0xff,0xff,0xff, /* H - O */ + 0xff,0xff,0xff,0xff,0xff,0xff,0xff,0xff, /* P - W */ + 0xff,0xff,0xff,0xff,0xff,0xff,0xff,0xff, /* X - _ */ + 0xff,0x0a,0x0b,0x0c,0x0d,0x0e,0x0f,0xff, /* ` - g */ + 0xff,0xff,0xff,0xff,0xff,0xff,0xff,0xff, /* h - o */ + 0xff,0xff,0xff,0xff,0xff,0xff,0xff,0xff, /* p - w */ + 0xff,0xff,0xff,0xff,0xff,0xff,0xff,0xff, /* x -127 */ + 0xff,0xff,0xff,0xff,0xff,0xff,0xff,0xff, /* 128-135 */ + 0xff,0xff,0xff,0xff,0xff,0xff,0xff,0xff, /* 136-143 */ + 0xff,0xff,0xff,0xff,0xff,0xff,0xff,0xff, /* 144-151 */ + 0xff,0xff,0xff,0xff,0xff,0xff,0xff,0xff, /* 152-159 */ + 0xff,0xff,0xff,0xff,0xff,0xff,0xff,0xff, /* 160-167 */ + 0xff,0xff,0xff,0xff,0xff,0xff,0xff,0xff, /* 168-175 */ + 0xff,0xff,0xff,0xff,0xff,0xff,0xff,0xff, /* 176-183 */ + 0xff,0xff,0xff,0xff,0xff,0xff,0xff,0xff, /* 184-191 */ + 0xff,0xff,0xff,0xff,0xff,0xff,0xff,0xff, /* 192-199 */ + 0xff,0xff,0xff,0xff,0xff,0xff,0xff,0xff, /* 2ff-207 */ + 0xff,0xff,0xff,0xff,0xff,0xff,0xff,0xff, /* 208-215 */ + 0xff,0xff,0xff,0xff,0xff,0xff,0xff,0xff, /* 216-223 */ + 0xff,0xff,0xff,0xff,0xff,0xff,0xff,0xff, /* 224-231 */ + 0xff,0xff,0xff,0xff,0xff,0xff,0xff,0xff, /* 232-239 */ + 0xff,0xff,0xff,0xff,0xff,0xff,0xff,0xff, /* 240-247 */ + 0xff,0xff,0xff,0xff,0xff,0xff,0xff,0xff};/* 248-255 */ + +#else + +/* This is the "abnormal" case, for EBCDIC systems not running in UTF-8 mode. */ + +static const uint8_t xdigitab[] = + { + 0xff,0xff,0xff,0xff,0xff,0xff,0xff,0xff, /* 0- 7 0 */ + 0xff,0xff,0xff,0xff,0xff,0xff,0xff,0xff, /* 8- 15 */ + 0xff,0xff,0xff,0xff,0xff,0xff,0xff,0xff, /* 16- 23 10 */ + 0xff,0xff,0xff,0xff,0xff,0xff,0xff,0xff, /* 24- 31 */ + 0xff,0xff,0xff,0xff,0xff,0xff,0xff,0xff, /* 32- 39 20 */ + 0xff,0xff,0xff,0xff,0xff,0xff,0xff,0xff, /* 40- 47 */ + 0xff,0xff,0xff,0xff,0xff,0xff,0xff,0xff, /* 48- 55 30 */ + 0xff,0xff,0xff,0xff,0xff,0xff,0xff,0xff, /* 56- 63 */ + 0xff,0xff,0xff,0xff,0xff,0xff,0xff,0xff, /* - 71 40 */ + 0xff,0xff,0xff,0xff,0xff,0xff,0xff,0xff, /* 72- | */ + 0xff,0xff,0xff,0xff,0xff,0xff,0xff,0xff, /* & - 87 50 */ + 0xff,0xff,0xff,0xff,0xff,0xff,0xff,0xff, /* 88- 95 */ + 0xff,0xff,0xff,0xff,0xff,0xff,0xff,0xff, /* - -103 60 */ + 0xff,0xff,0xff,0xff,0xff,0xff,0xff,0xff, /* 104- ? */ + 0xff,0xff,0xff,0xff,0xff,0xff,0xff,0xff, /* 112-119 70 */ + 0xff,0xff,0xff,0xff,0xff,0xff,0xff,0xff, /* 120- " */ + 0xff,0x0a,0x0b,0x0c,0x0d,0x0e,0x0f,0xff, /* 128- g 80 */ + 0xff,0xff,0xff,0xff,0xff,0xff,0xff,0xff, /* h -143 */ + 0xff,0xff,0xff,0xff,0xff,0xff,0xff,0xff, /* 144- p 90 */ + 0xff,0xff,0xff,0xff,0xff,0xff,0xff,0xff, /* q -159 */ + 0xff,0xff,0xff,0xff,0xff,0xff,0xff,0xff, /* 160- x A0 */ + 0xff,0xff,0xff,0xff,0xff,0xff,0xff,0xff, /* y -175 */ + 0xff,0xff,0xff,0xff,0xff,0xff,0xff,0xff, /* ^ -183 B0 */ + 0xff,0xff,0xff,0xff,0xff,0xff,0xff,0xff, /* 184-191 */ + 0xff,0x0a,0x0b,0x0c,0x0d,0x0e,0x0f,0xff, /* { - G C0 */ + 0xff,0xff,0xff,0xff,0xff,0xff,0xff,0xff, /* H -207 */ + 0xff,0xff,0xff,0xff,0xff,0xff,0xff,0xff, /* } - P D0 */ + 0xff,0xff,0xff,0xff,0xff,0xff,0xff,0xff, /* Q -223 */ + 0xff,0xff,0xff,0xff,0xff,0xff,0xff,0xff, /* \ - X E0 */ + 0xff,0xff,0xff,0xff,0xff,0xff,0xff,0xff, /* Y -239 */ + 0x00,0x01,0x02,0x03,0x04,0x05,0x06,0x07, /* 0 - 7 F0 */ + 0x08,0x09,0xff,0xff,0xff,0xff,0xff,0xff};/* 8 -255 */ +#endif /* EBCDIC */ + + +/* Table for handling alphanumeric escaped characters. Positive returns are +simple data values; negative values are for special things like \d and so on. +Zero means further processing is needed (for things like \x), or the escape is +invalid. */ + +/* This is the "normal" table for ASCII systems or for EBCDIC systems running +in UTF-8 mode. It runs from '0' to 'z'. */ + +#ifndef EBCDIC +#define ESCAPES_FIRST CHAR_0 +#define ESCAPES_LAST CHAR_z + +static const short int escapes[] = { + 0, 0, + 0, 0, + 0, 0, + 0, 0, + 0, 0, + CHAR_COLON, CHAR_SEMICOLON, + CHAR_LESS_THAN_SIGN, CHAR_EQUALS_SIGN, + CHAR_GREATER_THAN_SIGN, CHAR_QUESTION_MARK, + CHAR_COMMERCIAL_AT, -ESC_A, + -ESC_B, -ESC_C, + -ESC_D, -ESC_E, + 0, -ESC_G, + -ESC_H, 0, + 0, -ESC_K, + 0, 0, + -ESC_N, 0, + -ESC_P, -ESC_Q, + -ESC_R, -ESC_S, + 0, 0, + -ESC_V, -ESC_W, + -ESC_X, 0, + -ESC_Z, CHAR_LEFT_SQUARE_BRACKET, + CHAR_BACKSLASH, CHAR_RIGHT_SQUARE_BRACKET, + CHAR_CIRCUMFLEX_ACCENT, CHAR_UNDERSCORE, + CHAR_GRAVE_ACCENT, 7, + -ESC_b, 0, + -ESC_d, ESC_e, + ESC_f, 0, + -ESC_h, 0, + 0, -ESC_k, + 0, 0, + ESC_n, 0, + -ESC_p, 0, + ESC_r, -ESC_s, + ESC_tee, 0, + -ESC_v, -ESC_w, + 0, 0, + -ESC_z +}; + +#else + +/* This is the "abnormal" table for EBCDIC systems without UTF-8 support. +It runs from 'a' to '9'. */ + +#define ESCAPES_FIRST CHAR_a +#define ESCAPES_LAST CHAR_9 + +static const short int escapes[] = { +/* 80 */ 7, -ESC_b, 0, -ESC_d, ESC_e, ESC_f, 0, +/* 88 */-ESC_h, 0, 0, '{', 0, 0, 0, 0, +/* 90 */ 0, 0, -ESC_k, 'l', 0, ESC_n, 0, -ESC_p, +/* 98 */ 0, ESC_r, 0, '}', 0, 0, 0, 0, +/* A0 */ 0, '~', -ESC_s, ESC_tee, 0,-ESC_v, -ESC_w, 0, +/* A8 */ 0,-ESC_z, 0, 0, 0, '[', 0, 0, +/* B0 */ 0, 0, 0, 0, 0, 0, 0, 0, +/* B8 */ 0, 0, 0, 0, 0, ']', '=', '-', +/* C0 */ '{',-ESC_A, -ESC_B, -ESC_C, -ESC_D,-ESC_E, 0, -ESC_G, +/* C8 */-ESC_H, 0, 0, 0, 0, 0, 0, 0, +/* D0 */ '}', 0, -ESC_K, 0, 0,-ESC_N, 0, -ESC_P, +/* D8 */-ESC_Q,-ESC_R, 0, 0, 0, 0, 0, 0, +/* E0 */ '\\', 0, -ESC_S, 0, 0,-ESC_V, -ESC_W, -ESC_X, +/* E8 */ 0,-ESC_Z, 0, 0, 0, 0, 0, 0, +/* F0 */ 0, 0, 0, 0, 0, 0, 0, 0, +/* F8 */ 0, 0 +}; + +#endif + + +/* Table of special "verbs" like (*PRUNE). This is a short table, so it is +searched linearly. Put all the names into a single string, in order to reduce +the number of relocations when a shared library is dynamically linked. The +string is built from string macros so that it works in UTF-8 mode on EBCDIC +platforms. */ + +typedef struct verbitem { + int len; /* Length of verb name */ + int op; /* Op when no arg, or -1 if arg mandatory */ + int op_arg; /* Op when arg present, or -1 if not allowed */ +} verbitem; + +static const char verbnames[] = + "\0" /* Empty name is a shorthand for MARK */ + STRING_MARK0 + STRING_ACCEPT0 + STRING_COMMIT0 + STRING_F0 + STRING_FAIL0 + STRING_PRUNE0 + STRING_SKIP0 + STRING_THEN; + +static const verbitem verbs[] = { + { 0, -1, OP_MARK }, + { 4, -1, OP_MARK }, + { 6, OP_ACCEPT, -1 }, + { 6, OP_COMMIT, -1 }, + { 1, OP_FAIL, -1 }, + { 4, OP_FAIL, -1 }, + { 5, OP_PRUNE, OP_PRUNE_ARG }, + { 4, OP_SKIP, OP_SKIP_ARG }, + { 4, OP_THEN, OP_THEN_ARG } +}; + +static const int verbcount = sizeof(verbs)/sizeof(verbitem); + + +/* Substitutes for [[:<:]] and [[:>:]], which mean start and end of word in +another regex library. */ + +static const PCRE2_UCHAR sub_start_of_word[] = { + CHAR_BACKSLASH, CHAR_b, CHAR_LEFT_PARENTHESIS, CHAR_QUESTION_MARK, + CHAR_EQUALS_SIGN, CHAR_BACKSLASH, CHAR_w, CHAR_RIGHT_PARENTHESIS, '\0' }; + +static const PCRE2_UCHAR sub_end_of_word[] = { + CHAR_BACKSLASH, CHAR_b, CHAR_LEFT_PARENTHESIS, CHAR_QUESTION_MARK, + CHAR_LESS_THAN_SIGN, CHAR_EQUALS_SIGN, CHAR_BACKSLASH, CHAR_w, + CHAR_RIGHT_PARENTHESIS, '\0' }; + + +/* Tables of names of POSIX character classes and their lengths. The names are +now all in a single string, to reduce the number of relocations when a shared +library is dynamically loaded. The list of lengths is terminated by a zero +length entry. The first three must be alpha, lower, upper, as this is assumed +for handling case independence. The indices for graph, print, and punct are +needed, so identify them. */ + +static const char posix_names[] = + STRING_alpha0 STRING_lower0 STRING_upper0 STRING_alnum0 + STRING_ascii0 STRING_blank0 STRING_cntrl0 STRING_digit0 + STRING_graph0 STRING_print0 STRING_punct0 STRING_space0 + STRING_word0 STRING_xdigit; + +static const uint8_t posix_name_lengths[] = { + 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 4, 6, 0 }; + +#define PC_GRAPH 8 +#define PC_PRINT 9 +#define PC_PUNCT 10 + + +/* Table of class bit maps for each POSIX class. Each class is formed from a +base map, with an optional addition or removal of another map. Then, for some +classes, there is some additional tweaking: for [:blank:] the vertical space +characters are removed, and for [:alpha:] and [:alnum:] the underscore +character is removed. The triples in the table consist of the base map offset, +second map offset or -1 if no second map, and a non-negative value for map +addition or a negative value for map subtraction (if there are two maps). The +absolute value of the third field has these meanings: 0 => no tweaking, 1 => +remove vertical space characters, 2 => remove underscore. */ + +static const int posix_class_maps[] = { + cbit_word, cbit_digit, -2, /* alpha */ + cbit_lower, -1, 0, /* lower */ + cbit_upper, -1, 0, /* upper */ + cbit_word, -1, 2, /* alnum - word without underscore */ + cbit_print, cbit_cntrl, 0, /* ascii */ + cbit_space, -1, 1, /* blank - a GNU extension */ + cbit_cntrl, -1, 0, /* cntrl */ + cbit_digit, -1, 0, /* digit */ + cbit_graph, -1, 0, /* graph */ + cbit_print, -1, 0, /* print */ + cbit_punct, -1, 0, /* punct */ + cbit_space, -1, 0, /* space */ + cbit_word, -1, 0, /* word - a Perl extension */ + cbit_xdigit,-1, 0 /* xdigit */ +}; + +/* Table of substitutes for \d etc when PCRE2_UCP is set. They are replaced by +Unicode property escapes. */ + +#ifdef SUPPORT_UTF +static const PCRE2_UCHAR string_PNd[] = { + CHAR_BACKSLASH, CHAR_P, CHAR_LEFT_CURLY_BRACKET, + CHAR_N, CHAR_d, CHAR_RIGHT_CURLY_BRACKET, '\0' }; +static const PCRE2_UCHAR string_pNd[] = { + CHAR_BACKSLASH, CHAR_p, CHAR_LEFT_CURLY_BRACKET, + CHAR_N, CHAR_d, CHAR_RIGHT_CURLY_BRACKET, '\0' }; +static const PCRE2_UCHAR string_PXsp[] = { + CHAR_BACKSLASH, CHAR_P, CHAR_LEFT_CURLY_BRACKET, + CHAR_X, CHAR_s, CHAR_p, CHAR_RIGHT_CURLY_BRACKET, '\0' }; +static const PCRE2_UCHAR string_pXsp[] = { + CHAR_BACKSLASH, CHAR_p, CHAR_LEFT_CURLY_BRACKET, + CHAR_X, CHAR_s, CHAR_p, CHAR_RIGHT_CURLY_BRACKET, '\0' }; +static const PCRE2_UCHAR string_PXwd[] = { + CHAR_BACKSLASH, CHAR_P, CHAR_LEFT_CURLY_BRACKET, + CHAR_X, CHAR_w, CHAR_d, CHAR_RIGHT_CURLY_BRACKET, '\0' }; +static const PCRE2_UCHAR string_pXwd[] = { + CHAR_BACKSLASH, CHAR_p, CHAR_LEFT_CURLY_BRACKET, + CHAR_X, CHAR_w, CHAR_d, CHAR_RIGHT_CURLY_BRACKET, '\0' }; + +static PCRE2_SPTR substitutes[] = { + string_PNd, /* \D */ + string_pNd, /* \d */ + string_PXsp, /* \S */ /* Xsp is Perl space, but from 8.34, Perl */ + string_pXsp, /* \s */ /* space and POSIX space are the same. */ + string_PXwd, /* \W */ + string_pXwd /* \w */ +}; + +/* The POSIX class substitutes must be in the order of the POSIX class names, +defined above, and there are both positive and negative cases. NULL means no +general substitute of a Unicode property escape (\p or \P). However, for some +POSIX classes (e.g. graph, print, punct) a special property code is compiled +directly. */ + +static const PCRE2_UCHAR string_pL[] = { + CHAR_BACKSLASH, CHAR_p, CHAR_LEFT_CURLY_BRACKET, + CHAR_L, CHAR_RIGHT_CURLY_BRACKET, '\0' }; +static const PCRE2_UCHAR string_pLl[] = { + CHAR_BACKSLASH, CHAR_p, CHAR_LEFT_CURLY_BRACKET, + CHAR_L, CHAR_l, CHAR_RIGHT_CURLY_BRACKET, '\0' }; +static const PCRE2_UCHAR string_pLu[] = { + CHAR_BACKSLASH, CHAR_p, CHAR_LEFT_CURLY_BRACKET, + CHAR_L, CHAR_u, CHAR_RIGHT_CURLY_BRACKET, '\0' }; +static const PCRE2_UCHAR string_pXan[] = { + CHAR_BACKSLASH, CHAR_p, CHAR_LEFT_CURLY_BRACKET, + CHAR_X, CHAR_a, CHAR_n, CHAR_RIGHT_CURLY_BRACKET, '\0' }; +static const PCRE2_UCHAR string_h[] = { + CHAR_BACKSLASH, CHAR_h, '\0' }; +static const PCRE2_UCHAR string_pXps[] = { + CHAR_BACKSLASH, CHAR_p, CHAR_LEFT_CURLY_BRACKET, + CHAR_X, CHAR_p, CHAR_s, CHAR_RIGHT_CURLY_BRACKET, '\0' }; +static const PCRE2_UCHAR string_PL[] = { + CHAR_BACKSLASH, CHAR_P, CHAR_LEFT_CURLY_BRACKET, + CHAR_L, CHAR_RIGHT_CURLY_BRACKET, '\0' }; +static const PCRE2_UCHAR string_PLl[] = { + CHAR_BACKSLASH, CHAR_P, CHAR_LEFT_CURLY_BRACKET, + CHAR_L, CHAR_l, CHAR_RIGHT_CURLY_BRACKET, '\0' }; +static const PCRE2_UCHAR string_PLu[] = { + CHAR_BACKSLASH, CHAR_P, CHAR_LEFT_CURLY_BRACKET, + CHAR_L, CHAR_u, CHAR_RIGHT_CURLY_BRACKET, '\0' }; +static const PCRE2_UCHAR string_PXan[] = { + CHAR_BACKSLASH, CHAR_P, CHAR_LEFT_CURLY_BRACKET, + CHAR_X, CHAR_a, CHAR_n, CHAR_RIGHT_CURLY_BRACKET, '\0' }; +static const PCRE2_UCHAR string_H[] = { + CHAR_BACKSLASH, CHAR_H, '\0' }; +static const PCRE2_UCHAR string_PXps[] = { + CHAR_BACKSLASH, CHAR_P, CHAR_LEFT_CURLY_BRACKET, + CHAR_X, CHAR_p, CHAR_s, CHAR_RIGHT_CURLY_BRACKET, '\0' }; + +static PCRE2_SPTR posix_substitutes[] = { + string_pL, /* alpha */ + string_pLl, /* lower */ + string_pLu, /* upper */ + string_pXan, /* alnum */ + NULL, /* ascii */ + string_h, /* blank */ + NULL, /* cntrl */ + string_pNd, /* digit */ + NULL, /* graph */ + NULL, /* print */ + NULL, /* punct */ + string_pXps, /* space */ /* Xps is POSIX space, but from 8.34 */ + string_pXwd, /* word */ /* Perl and POSIX space are the same */ + NULL, /* xdigit */ + /* Negated cases */ + string_PL, /* ^alpha */ + string_PLl, /* ^lower */ + string_PLu, /* ^upper */ + string_PXan, /* ^alnum */ + NULL, /* ^ascii */ + string_H, /* ^blank */ + NULL, /* ^cntrl */ + string_PNd, /* ^digit */ + NULL, /* ^graph */ + NULL, /* ^print */ + NULL, /* ^punct */ + string_PXps, /* ^space */ /* Xps is POSIX space, but from 8.34 */ + string_PXwd, /* ^word */ /* Perl and POSIX space are the same */ + NULL /* ^xdigit */ +}; +#define POSIX_SUBSIZE (sizeof(posix_substitutes) / sizeof(PCRE2_UCHAR *)) +#endif /* SUPPORT_UTF */ + +/* Masks for checking option settings. */ + +#define PUBLIC_COMPILE_OPTIONS \ + (PCRE2_ANCHORED|PCRE2_ALLOW_EMPTY_CLASS|PCRE2_ALT_BSUX|PCRE2_AUTO_CALLOUT| \ + PCRE2_CASELESS|PCRE2_DOLLAR_ENDONLY|PCRE2_DOTALL|PCRE2_DUPNAMES| \ + PCRE2_EXTENDED|PCRE2_FIRSTLINE|PCRE2_MATCH_UNSET_BACKREF| \ + PCRE2_MULTILINE|PCRE2_NEVER_UCP|PCRE2_NEVER_UTF|PCRE2_NO_AUTO_CAPTURE| \ + PCRE2_NO_AUTO_POSSESS|PCRE2_NO_START_OPTIMIZE|PCRE2_NO_UTF_CHECK| \ + PCRE2_UCP|PCRE2_UNGREEDY|PCRE2_UTF) + +/* Compile time error code numbers. They are given names so that they can more +easily be tracked. When a new number is added, the tables called eint1 and +eint2 in pcre2posix.c must be updated. */ + +enum { ERR0 = COMPILE_ERROR_BASE, + ERR1, ERR2, ERR3, ERR4, ERR5, ERR6, ERR7, ERR8, ERR9, ERR10, + ERR11, ERR12, ERR13, ERR14, ERR15, ERR16, ERR17, ERR18, ERR19, ERR20, + ERR21, ERR22, ERR23, ERR24, ERR25, ERR26, ERR27, ERR28, ERR29, ERR30, + ERR31, ERR32, ERR33, ERR34, ERR35, ERR36, ERR37, ERR38, ERR39, ERR40, + ERR41, ERR42, ERR43, ERR44, ERR45, ERR46, ERR47, ERR48, ERR49, ERR50, + ERR51, ERR52, ERR53, ERR54, ERR55, ERR56, ERR57, ERR58, ERR59, ERR60, + ERR61, ERR62, ERR63, ERR64, ERR65, ERR66, ERR67, ERR68, ERR69, ERR70, + ERR71, ERR72, ERR73, ERR74, ERR75, ERR76, ERR77, ERR78, ERR79, ERR80, + ERR81, ERR82, ERR83, ERR84, ERR85, ERR86 }; + +/* This is a table of start-of-pattern options such as (*UTF) and settings such +as (*LIMIT_MATCH=nnnn) and (*CRLF). For completeness and backward +compatibility, (*UTFn) is supported in the relevant libraries, but (*UTF) is +generic and always supported. */ + +enum { PSO_OPT, /* Value is an option bit */ + PSO_NL, /* Value is a newline type */ + PSO_BSR, /* Value is a \R type */ + PSO_LIMM, /* Read integer value for match limit */ + PSO_LIMR }; /* Read integer value for recursion limit */ + +typedef struct pso { + const uint8_t *name; + uint16_t length; + uint16_t type; + uint32_t value; +} pso; + +/* NB: STRING_UTFn_RIGHTPAR contains the length as well */ + +static pso pso_list[] = { + { (uint8_t *)STRING_UTFn_RIGHTPAR, PSO_OPT, PCRE2_UTF }, + { (uint8_t *)STRING_UTF_RIGHTPAR, 4, PSO_OPT, PCRE2_UTF }, + { (uint8_t *)STRING_UCP_RIGHTPAR, 4, PSO_OPT, PCRE2_UCP }, + { (uint8_t *)STRING_NO_AUTO_POSSESS_RIGHTPAR, 16, PSO_OPT, PCRE2_NO_AUTO_POSSESS }, + { (uint8_t *)STRING_NO_START_OPT_RIGHTPAR, 13, PSO_OPT, PCRE2_NO_START_OPTIMIZE }, + { (uint8_t *)STRING_LIMIT_MATCH_EQ, 12, PSO_LIMM, 0 }, + { (uint8_t *)STRING_LIMIT_RECURSION_EQ, 16, PSO_LIMR, 0 }, + { (uint8_t *)STRING_CR_RIGHTPAR, 3, PSO_NL, PCRE2_NEWLINE_CR }, + { (uint8_t *)STRING_LF_RIGHTPAR, 3, PSO_NL, PCRE2_NEWLINE_LF }, + { (uint8_t *)STRING_CRLF_RIGHTPAR, 5, PSO_NL, PCRE2_NEWLINE_CRLF }, + { (uint8_t *)STRING_ANY_RIGHTPAR, 4, PSO_NL, PCRE2_NEWLINE_ANY }, + { (uint8_t *)STRING_ANYCRLF_RIGHTPAR, 8, PSO_NL, PCRE2_NEWLINE_ANYCRLF }, + { (uint8_t *)STRING_BSR_ANYCRLF_RIGHTPAR, 12, PSO_BSR, PCRE2_BSR_ANYCRLF }, + { (uint8_t *)STRING_BSR_UNICODE_RIGHTPAR, 12, PSO_BSR, PCRE2_BSR_UNICODE } +}; + + +/* This table is used when converting repeating opcodes into possessified +versions as a result of an explicit possessive quantifier such as ++. A zero +value means there is no possessified version - in those cases the item in +question must be wrapped in ONCE brackets. The table is truncated at OP_CALLOUT +because all relevant opcodes are less than that. */ + +static const uint8_t opcode_possessify[] = { + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, /* 0 - 15 */ + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, /* 16 - 31 */ + + 0, /* NOTI */ + OP_POSSTAR, 0, /* STAR, MINSTAR */ + OP_POSPLUS, 0, /* PLUS, MINPLUS */ + OP_POSQUERY, 0, /* QUERY, MINQUERY */ + OP_POSUPTO, 0, /* UPTO, MINUPTO */ + 0, /* EXACT */ + 0, 0, 0, 0, /* POS{STAR,PLUS,QUERY,UPTO} */ + + OP_POSSTARI, 0, /* STARI, MINSTARI */ + OP_POSPLUSI, 0, /* PLUSI, MINPLUSI */ + OP_POSQUERYI, 0, /* QUERYI, MINQUERYI */ + OP_POSUPTOI, 0, /* UPTOI, MINUPTOI */ + 0, /* EXACTI */ + 0, 0, 0, 0, /* POS{STARI,PLUSI,QUERYI,UPTOI} */ + + OP_NOTPOSSTAR, 0, /* NOTSTAR, NOTMINSTAR */ + OP_NOTPOSPLUS, 0, /* NOTPLUS, NOTMINPLUS */ + OP_NOTPOSQUERY, 0, /* NOTQUERY, NOTMINQUERY */ + OP_NOTPOSUPTO, 0, /* NOTUPTO, NOTMINUPTO */ + 0, /* NOTEXACT */ + 0, 0, 0, 0, /* NOTPOS{STAR,PLUS,QUERY,UPTO} */ + + OP_NOTPOSSTARI, 0, /* NOTSTARI, NOTMINSTARI */ + OP_NOTPOSPLUSI, 0, /* NOTPLUSI, NOTMINPLUSI */ + OP_NOTPOSQUERYI, 0, /* NOTQUERYI, NOTMINQUERYI */ + OP_NOTPOSUPTOI, 0, /* NOTUPTOI, NOTMINUPTOI */ + 0, /* NOTEXACTI */ + 0, 0, 0, 0, /* NOTPOS{STARI,PLUSI,QUERYI,UPTOI} */ + + OP_TYPEPOSSTAR, 0, /* TYPESTAR, TYPEMINSTAR */ + OP_TYPEPOSPLUS, 0, /* TYPEPLUS, TYPEMINPLUS */ + OP_TYPEPOSQUERY, 0, /* TYPEQUERY, TYPEMINQUERY */ + OP_TYPEPOSUPTO, 0, /* TYPEUPTO, TYPEMINUPTO */ + 0, /* TYPEEXACT */ + 0, 0, 0, 0, /* TYPEPOS{STAR,PLUS,QUERY,UPTO} */ + + OP_CRPOSSTAR, 0, /* CRSTAR, CRMINSTAR */ + OP_CRPOSPLUS, 0, /* CRPLUS, CRMINPLUS */ + OP_CRPOSQUERY, 0, /* CRQUERY, CRMINQUERY */ + OP_CRPOSRANGE, 0, /* CRRANGE, CRMINRANGE */ + 0, 0, 0, 0, /* CRPOS{STAR,PLUS,QUERY,RANGE} */ + + 0, 0, 0, /* CLASS, NCLASS, XCLASS */ + 0, 0, /* REF, REFI */ + 0, 0, /* DNREF, DNREFI */ + 0, 0 /* RECURSE, CALLOUT */ +}; + + +/* FIXME: this is currently incomplete (free JIT?) */ /************************************************* * Free compiled code * @@ -61,7 +678,6572 @@ if (code != NULL) code->memctl.free(code, code->memctl.memory_data); /************************************************* -* Compile a Regular Expression * +* Insert an automatic callout point * +*************************************************/ + +/* This function is called when the PCRE2_AUTO_CALLOUT option is set, to insert +callout points before each pattern item. + +Arguments: + code current code pointer + ptr current pattern pointer + cd general compile-time data + +Returns: new code pointer +*/ + +static PCRE2_UCHAR * +auto_callout(PCRE2_UCHAR *code, PCRE2_SPTR ptr, compile_data *cd) +{ +*code++ = OP_CALLOUT; +*code++ = 255; +PUT(code, 0, ptr - cd->start_pattern); /* Pattern offset */ +PUT(code, LINK_SIZE, 0); /* Default length */ +return code + 2 * LINK_SIZE; +} + + + +/************************************************* +* Complete a callout item * +*************************************************/ + +/* A callout item contains the length of the next item in the pattern, which +we can't fill in till after we have reached the relevant point. This is used +for both automatic and manual callouts. + +Arguments: + previous_callout points to previous callout item + ptr current pattern pointer + cd general compile-time data + +Returns: nothing +*/ + +static void +complete_callout(PCRE2_UCHAR *previous_callout, PCRE2_SPTR ptr, + compile_data *cd) +{ +size_t length = ptr - cd->start_pattern - GET(previous_callout, 2); +PUT(previous_callout, 2 + LINK_SIZE, length); +} + + + +/************************************************* +* Find the fixed length of a branch * +*************************************************/ + +/* Scan a branch and compute the fixed length of subject that will match it, if +the length is fixed. This is needed for dealing with backward assertions. In +UTF mode, the result is in code units rather than bytes. The branch is +temporarily terminated with OP_END when this function is called. + +This function is called when a backward assertion is encountered, so that if it +fails, the error message can point to the correct place in the pattern. +However, we cannot do this when the assertion contains subroutine calls, +because they can be forward references. We solve this by remembering this case +and doing the check at the end; a flag specifies which mode we are running in. + +Arguments: + code points to the start of the pattern (the bracket) + utf TRUE in UTF mode + atend TRUE if called when the pattern is complete + cd the "compile data" structure + +Returns: the fixed length, + or -1 if there is no fixed length, + or -2 if \C was encountered (in UTF-8 mode only) + or -3 if an OP_RECURSE item was encountered and atend is FALSE + or -4 if an unknown opcode was encountered (internal error) +*/ + +static int +find_fixedlength(PCRE2_UCHAR *code, BOOL utf, BOOL atend, compile_data *cd) +{ +int length = -1; + +register int branchlength = 0; +register PCRE2_UCHAR *cc = code + 1 + LINK_SIZE; + +/* Scan along the opcodes for this branch. If we get to the end of the +branch, check the length against that of the other branches. */ + +for (;;) + { + int d; + PCRE2_UCHAR *ce, *cs; + register PCRE2_UCHAR op = *cc; + + switch (op) + { + /* We only need to continue for OP_CBRA (normal capturing bracket) and + OP_BRA (normal non-capturing bracket) because the other variants of these + opcodes are all concerned with unlimited repeated groups, which of course + are not of fixed length. */ + + case OP_CBRA: + case OP_BRA: + case OP_ONCE: + case OP_ONCE_NC: + case OP_COND: + d = find_fixedlength(cc + ((op == OP_CBRA)? IMM2_SIZE : 0), utf, atend, cd); + if (d < 0) return d; + branchlength += d; + do cc += GET(cc, 1); while (*cc == OP_ALT); + cc += 1 + LINK_SIZE; + break; + + /* Reached end of a branch; if it's a ket it is the end of a nested call. + If it's ALT it is an alternation in a nested call. An ACCEPT is effectively + an ALT. If it is END it's the end of the outer call. All can be handled by + the same code. Note that we must not include the OP_KETRxxx opcodes here, + because they all imply an unlimited repeat. */ + + case OP_ALT: + case OP_KET: + case OP_END: + case OP_ACCEPT: + case OP_ASSERT_ACCEPT: + if (length < 0) length = branchlength; + else if (length != branchlength) return -1; + if (*cc != OP_ALT) return length; + cc += 1 + LINK_SIZE; + branchlength = 0; + break; + + /* A true recursion implies not fixed length, but a subroutine call may + be OK. If the subroutine is a forward reference, we can't deal with + it until the end of the pattern, so return -3. */ + + case OP_RECURSE: + if (!atend) return -3; + cs = ce = (PCRE2_UCHAR *)cd->start_code + GET(cc, 1); /* Start subpattern */ + do ce += GET(ce, 1); while (*ce == OP_ALT); /* End subpattern */ + if (cc > cs && cc < ce) return -1; /* Recursion */ + d = find_fixedlength(cs + IMM2_SIZE, utf, atend, cd); + if (d < 0) return d; + branchlength += d; + cc += 1 + LINK_SIZE; + break; + + /* Skip over assertive subpatterns */ + + case OP_ASSERT: + case OP_ASSERT_NOT: + case OP_ASSERTBACK: + case OP_ASSERTBACK_NOT: + do cc += GET(cc, 1); while (*cc == OP_ALT); + cc += PRIV(OP_lengths)[*cc]; + break; + + /* Skip over things that don't match chars */ + + case OP_MARK: + case OP_PRUNE_ARG: + case OP_SKIP_ARG: + case OP_THEN_ARG: + cc += cc[1] + PRIV(OP_lengths)[*cc]; + break; + + case OP_CALLOUT: + case OP_CIRC: + case OP_CIRCM: + case OP_CLOSE: + case OP_COMMIT: + case OP_CREF: + case OP_DEF: + case OP_DNCREF: + case OP_DNRREF: + case OP_DOLL: + case OP_DOLLM: + case OP_EOD: + case OP_EODN: + case OP_FAIL: + case OP_NOT_WORD_BOUNDARY: + case OP_PRUNE: + case OP_REVERSE: + case OP_RREF: + case OP_SET_SOM: + case OP_SKIP: + case OP_SOD: + case OP_SOM: + case OP_THEN: + case OP_WORD_BOUNDARY: + cc += PRIV(OP_lengths)[*cc]; + break; + + /* Handle literal characters */ + + case OP_CHAR: + case OP_CHARI: + case OP_NOT: + case OP_NOTI: + branchlength++; + cc += 2; +#ifdef SUPPORT_UTF + if (utf && HAS_EXTRALEN(cc[-1])) cc += GET_EXTRALEN(cc[-1]); +#endif + break; + + /* Handle exact repetitions. The count is already in characters, but we + need to skip over a multibyte character in UTF8 mode. */ + + case OP_EXACT: + case OP_EXACTI: + case OP_NOTEXACT: + case OP_NOTEXACTI: + branchlength += (int)GET2(cc,1); + cc += 2 + IMM2_SIZE; +#ifdef SUPPORT_UTF + if (utf && HAS_EXTRALEN(cc[-1])) cc += GET_EXTRALEN(cc[-1]); +#endif + break; + + case OP_TYPEEXACT: + branchlength += GET2(cc,1); + if (cc[1 + IMM2_SIZE] == OP_PROP || cc[1 + IMM2_SIZE] == OP_NOTPROP) + cc += 2; + cc += 1 + IMM2_SIZE + 1; + break; + + /* Handle single-char matchers */ + + case OP_PROP: + case OP_NOTPROP: + cc += 2; + /* Fall through */ + + case OP_HSPACE: + case OP_VSPACE: + case OP_NOT_HSPACE: + case OP_NOT_VSPACE: + case OP_NOT_DIGIT: + case OP_DIGIT: + case OP_NOT_WHITESPACE: + case OP_WHITESPACE: + case OP_NOT_WORDCHAR: + case OP_WORDCHAR: + case OP_ANY: + case OP_ALLANY: + branchlength++; + cc++; + break; + + /* The single-byte matcher isn't allowed. This only happens in UTF-8 mode; + otherwise \C is coded as OP_ALLANY. */ + + case OP_ANYBYTE: + return -2; + + /* Check a class for variable quantification */ + + case OP_CLASS: + case OP_NCLASS: +#ifdef SUPPORT_WIDE_CHARS + case OP_XCLASS: + /* The original code caused an unsigned overflow in 64 bit systems, + so now we use a conditional statement. */ + if (op == OP_XCLASS) + cc += GET(cc, 1); + else + cc += PRIV(OP_lengths)[OP_CLASS]; +#else + cc += PRIV(OP_lengths)[OP_CLASS]; +#endif + + switch (*cc) + { + case OP_CRSTAR: + case OP_CRMINSTAR: + case OP_CRPLUS: + case OP_CRMINPLUS: + case OP_CRQUERY: + case OP_CRMINQUERY: + case OP_CRPOSSTAR: + case OP_CRPOSPLUS: + case OP_CRPOSQUERY: + return -1; + + case OP_CRRANGE: + case OP_CRMINRANGE: + case OP_CRPOSRANGE: + if (GET2(cc,1) != GET2(cc,1+IMM2_SIZE)) return -1; + branchlength += (int)GET2(cc,1); + cc += 1 + 2 * IMM2_SIZE; + break; + + default: + branchlength++; + } + break; + + /* Anything else is variable length */ + + case OP_ANYNL: + case OP_BRAMINZERO: + case OP_BRAPOS: + case OP_BRAPOSZERO: + case OP_BRAZERO: + case OP_CBRAPOS: + case OP_EXTUNI: + case OP_KETRMAX: + case OP_KETRMIN: + case OP_KETRPOS: + case OP_MINPLUS: + case OP_MINPLUSI: + case OP_MINQUERY: + case OP_MINQUERYI: + case OP_MINSTAR: + case OP_MINSTARI: + case OP_MINUPTO: + case OP_MINUPTOI: + case OP_NOTMINPLUS: + case OP_NOTMINPLUSI: + case OP_NOTMINQUERY: + case OP_NOTMINQUERYI: + case OP_NOTMINSTAR: + case OP_NOTMINSTARI: + case OP_NOTMINUPTO: + case OP_NOTMINUPTOI: + case OP_NOTPLUS: + case OP_NOTPLUSI: + case OP_NOTPOSPLUS: + case OP_NOTPOSPLUSI: + case OP_NOTPOSQUERY: + case OP_NOTPOSQUERYI: + case OP_NOTPOSSTAR: + case OP_NOTPOSSTARI: + case OP_NOTPOSUPTO: + case OP_NOTPOSUPTOI: + case OP_NOTQUERY: + case OP_NOTQUERYI: + case OP_NOTSTAR: + case OP_NOTSTARI: + case OP_NOTUPTO: + case OP_NOTUPTOI: + case OP_PLUS: + case OP_PLUSI: + case OP_POSPLUS: + case OP_POSPLUSI: + case OP_POSQUERY: + case OP_POSQUERYI: + case OP_POSSTAR: + case OP_POSSTARI: + case OP_POSUPTO: + case OP_POSUPTOI: + case OP_QUERY: + case OP_QUERYI: + case OP_REF: + case OP_REFI: + case OP_DNREF: + case OP_DNREFI: + case OP_SBRA: + case OP_SBRAPOS: + case OP_SCBRA: + case OP_SCBRAPOS: + case OP_SCOND: + case OP_SKIPZERO: + case OP_STAR: + case OP_STARI: + case OP_TYPEMINPLUS: + case OP_TYPEMINQUERY: + case OP_TYPEMINSTAR: + case OP_TYPEMINUPTO: + case OP_TYPEPLUS: + case OP_TYPEPOSPLUS: + case OP_TYPEPOSQUERY: + case OP_TYPEPOSSTAR: + case OP_TYPEPOSUPTO: + case OP_TYPEQUERY: + case OP_TYPESTAR: + case OP_TYPEUPTO: + case OP_UPTO: + case OP_UPTOI: + return -1; + + /* Catch unrecognized opcodes so that when new ones are added they + are not forgotten, as has happened in the past. */ + + default: + return -4; + } + } +/* Control never gets here */ +} + + + +/************************************************* +* Find first significant op code * +*************************************************/ + +/* This is called by several functions that scan a compiled expression looking +for a fixed first character, or an anchoring op code etc. It skips over things +that do not influence this. For some calls, it makes sense to skip negative +forward and all backward assertions, and also the \b assertion; for others it +does not. + +Arguments: + code pointer to the start of the group + skipassert TRUE if certain assertions are to be skipped + +Returns: pointer to the first significant opcode +*/ + +static const PCRE2_UCHAR* +first_significant_code(PCRE2_SPTR code, BOOL skipassert) +{ +for (;;) + { + switch ((int)*code) + { + case OP_ASSERT_NOT: + case OP_ASSERTBACK: + case OP_ASSERTBACK_NOT: + if (!skipassert) return code; + do code += GET(code, 1); while (*code == OP_ALT); + code += PRIV(OP_lengths)[*code]; + break; + + case OP_WORD_BOUNDARY: + case OP_NOT_WORD_BOUNDARY: + if (!skipassert) return code; + /* Fall through */ + + case OP_CALLOUT: + case OP_CREF: + case OP_DNCREF: + case OP_RREF: + case OP_DNRREF: + case OP_DEF: + code += PRIV(OP_lengths)[*code]; + break; + + default: + return code; + } + } +/* Control never reaches here */ +} + + + +/************************************************* +* Scan compiled branch for non-emptiness * +*************************************************/ + +/* This function scans through a branch of a compiled pattern to see whether it +can match the empty string. It is called from could_be_empty() below and from +compile_branch() when checking for an unlimited repeat of a group that can +match nothing. Note that first_significant_code() skips over backward and +negative forward assertions when its final argument is TRUE. If we hit an +unclosed bracket, we return "empty" - this means we've struck an inner bracket +whose current branch will already have been scanned. + +Arguments: + code points to start of search + endcode points to where to stop + utf TRUE if in UTF mode + cd compile data + recurses chain of recurse_check to catch mutual recursion + +Returns: TRUE if what is matched could be empty +*/ + +typedef struct recurse_check { + struct recurse_check *prev; + PCRE2_SPTR group; +} recurse_check; + +static BOOL +could_be_empty_branch(PCRE2_SPTR code, PCRE2_SPTR endcode, BOOL utf, + compile_data *cd, recurse_check *recurses) +{ +register PCRE2_UCHAR c; +recurse_check this_recurse; + +for (code = first_significant_code(code + PRIV(OP_lengths)[*code], TRUE); + code < endcode; + code = first_significant_code(code + PRIV(OP_lengths)[c], TRUE)) + { + PCRE2_SPTR ccode; + + c = *code; + + /* Skip over forward assertions; the other assertions are skipped by + first_significant_code() with a TRUE final argument. */ + + if (c == OP_ASSERT) + { + do code += GET(code, 1); while (*code == OP_ALT); + c = *code; + continue; + } + + /* For a recursion/subroutine call, if its end has been reached, which + implies a backward reference subroutine call, we can scan it. If it's a + forward reference subroutine call, we can't. To detect forward reference + we have to scan up the list that is kept in the workspace. This function is + called only when doing the real compile, not during the pre-compile that + measures the size of the compiled pattern. */ + + if (c == OP_RECURSE) + { + PCRE2_SPTR scode = cd->start_code + GET(code, 1); + BOOL empty_branch; + + /* Test for forward reference or uncompleted reference. This is disabled + when called to scan a completed pattern by setting cd->start_workspace to + NULL. */ + + if (cd->start_workspace != NULL) + { + PCRE2_SPTR tcode; + for (tcode = cd->start_workspace; tcode < cd->hwm; tcode += LINK_SIZE) + if ((int)GET(tcode, 0) == (int)(code + 1 - cd->start_code)) return TRUE; + if (GET(scode, 1) == 0) return TRUE; /* Unclosed */ + } + + /* If we are scanning a completed pattern, there are no forward references + and all groups are complete. We need to detect whether this is a recursive + call, as otherwise there will be an infinite loop. If it is a recursion, + just skip over it. Simple recursions are easily detected. For mutual + recursions we keep a chain on the stack. */ + + else + { + recurse_check *r = recurses; + PCRE2_SPTR endgroup = scode; + + do endgroup += GET(endgroup, 1); while (*endgroup == OP_ALT); + if (code >= scode && code <= endgroup) continue; /* Simple recursion */ + + for (r = recurses; r != NULL; r = r->prev) + if (r->group == scode) break; + if (r != NULL) continue; /* Mutual recursion */ + } + + /* Completed reference; scan the referenced group, remembering it on the + stack chain to detect mutual recursions. */ + + empty_branch = FALSE; + this_recurse.prev = recurses; + this_recurse.group = scode; + + do + { + if (could_be_empty_branch(scode, endcode, utf, cd, &this_recurse)) + { + empty_branch = TRUE; + break; + } + scode += GET(scode, 1); + } + while (*scode == OP_ALT); + + if (!empty_branch) return FALSE; /* All branches are non-empty */ + continue; + } + + /* Groups with zero repeats can of course be empty; skip them. */ + + if (c == OP_BRAZERO || c == OP_BRAMINZERO || c == OP_SKIPZERO || + c == OP_BRAPOSZERO) + { + code += PRIV(OP_lengths)[c]; + do code += GET(code, 1); while (*code == OP_ALT); + c = *code; + continue; + } + + /* A nested group that is already marked as "could be empty" can just be + skipped. */ + + if (c == OP_SBRA || c == OP_SBRAPOS || + c == OP_SCBRA || c == OP_SCBRAPOS) + { + do code += GET(code, 1); while (*code == OP_ALT); + c = *code; + continue; + } + + /* For other groups, scan the branches. */ + + if (c == OP_BRA || c == OP_BRAPOS || + c == OP_CBRA || c == OP_CBRAPOS || + c == OP_ONCE || c == OP_ONCE_NC || + c == OP_COND) + { + BOOL empty_branch; + if (GET(code, 1) == 0) return TRUE; /* Hit unclosed bracket */ + + /* If a conditional group has only one branch, there is a second, implied, + empty branch, so just skip over the conditional, because it could be empty. + Otherwise, scan the individual branches of the group. */ + + if (c == OP_COND && code[GET(code, 1)] != OP_ALT) + code += GET(code, 1); + else + { + empty_branch = FALSE; + do + { + if (!empty_branch && could_be_empty_branch(code, endcode, utf, cd, NULL)) + empty_branch = TRUE; + code += GET(code, 1); + } + while (*code == OP_ALT); + if (!empty_branch) return FALSE; /* All branches are non-empty */ + } + + c = *code; + continue; + } + + /* Handle the other opcodes */ + + switch (c) + { + /* Check for quantifiers after a class. XCLASS is used for classes that + cannot be represented just by a bit map. This includes negated single + high-valued characters. The length in PRIV(OP_lengths)[] is zero; the + actual length is stored in the compiled code, so we must update "code" + here. */ + +#if defined SUPPORT_UTF || !defined COMPILE_PCRE8 + case OP_XCLASS: + ccode = code += GET(code, 1); + goto CHECK_CLASS_REPEAT; +#endif + + case OP_CLASS: + case OP_NCLASS: + ccode = code + PRIV(OP_lengths)[OP_CLASS]; + +#if defined SUPPORT_UTF || !defined COMPILE_PCRE8 + CHECK_CLASS_REPEAT: +#endif + + switch (*ccode) + { + case OP_CRSTAR: /* These could be empty; continue */ + case OP_CRMINSTAR: + case OP_CRQUERY: + case OP_CRMINQUERY: + case OP_CRPOSSTAR: + case OP_CRPOSQUERY: + break; + + default: /* Non-repeat => class must match */ + case OP_CRPLUS: /* These repeats aren't empty */ + case OP_CRMINPLUS: + case OP_CRPOSPLUS: + return FALSE; + + case OP_CRRANGE: + case OP_CRMINRANGE: + case OP_CRPOSRANGE: + if (GET2(ccode, 1) > 0) return FALSE; /* Minimum > 0 */ + break; + } + break; + + /* Opcodes that must match a character */ + + case OP_ANY: + case OP_ALLANY: + case OP_ANYBYTE: + + case OP_PROP: + case OP_NOTPROP: + case OP_ANYNL: + + case OP_NOT_HSPACE: + case OP_HSPACE: + case OP_NOT_VSPACE: + case OP_VSPACE: + case OP_EXTUNI: + + case OP_NOT_DIGIT: + case OP_DIGIT: + case OP_NOT_WHITESPACE: + case OP_WHITESPACE: + case OP_NOT_WORDCHAR: + case OP_WORDCHAR: + + case OP_CHAR: + case OP_CHARI: + case OP_NOT: + case OP_NOTI: + + case OP_PLUS: + case OP_PLUSI: + case OP_MINPLUS: + case OP_MINPLUSI: + + case OP_NOTPLUS: + case OP_NOTPLUSI: + case OP_NOTMINPLUS: + case OP_NOTMINPLUSI: + + case OP_POSPLUS: + case OP_POSPLUSI: + case OP_NOTPOSPLUS: + case OP_NOTPOSPLUSI: + + case OP_EXACT: + case OP_EXACTI: + case OP_NOTEXACT: + case OP_NOTEXACTI: + + case OP_TYPEPLUS: + case OP_TYPEMINPLUS: + case OP_TYPEPOSPLUS: + case OP_TYPEEXACT: + + return FALSE; + + /* These are going to continue, as they may be empty, but we have to + fudge the length for the \p and \P cases. */ + + case OP_TYPESTAR: + case OP_TYPEMINSTAR: + case OP_TYPEPOSSTAR: + case OP_TYPEQUERY: + case OP_TYPEMINQUERY: + case OP_TYPEPOSQUERY: + if (code[1] == OP_PROP || code[1] == OP_NOTPROP) code += 2; + break; + + /* Same for these */ + + case OP_TYPEUPTO: + case OP_TYPEMINUPTO: + case OP_TYPEPOSUPTO: + if (code[1 + IMM2_SIZE] == OP_PROP || code[1 + IMM2_SIZE] == OP_NOTPROP) + code += 2; + break; + + /* End of branch */ + + case OP_KET: + case OP_KETRMAX: + case OP_KETRMIN: + case OP_KETRPOS: + case OP_ALT: + return TRUE; + + /* In UTF-8 or UTF-16 mode, STAR, MINSTAR, POSSTAR, QUERY, MINQUERY, + POSQUERY, UPTO, MINUPTO, and POSUPTO and their caseless and negative + versions may be followed by a multibyte character. */ + +#ifdef MAYBE_UTF_MULTI + case OP_STAR: + case OP_STARI: + case OP_NOTSTAR: + case OP_NOTSTARI: + + case OP_MINSTAR: + case OP_MINSTARI: + case OP_NOTMINSTAR: + case OP_NOTMINSTARI: + + case OP_POSSTAR: + case OP_POSSTARI: + case OP_NOTPOSSTAR: + case OP_NOTPOSSTARI: + + case OP_QUERY: + case OP_QUERYI: + case OP_NOTQUERY: + case OP_NOTQUERYI: + + case OP_MINQUERY: + case OP_MINQUERYI: + case OP_NOTMINQUERY: + case OP_NOTMINQUERYI: + + case OP_POSQUERY: + case OP_POSQUERYI: + case OP_NOTPOSQUERY: + case OP_NOTPOSQUERYI: + + if (utf && HAS_EXTRALEN(code[1])) code += GET_EXTRALEN(code[1]); + break; + + case OP_UPTO: + case OP_UPTOI: + case OP_NOTUPTO: + case OP_NOTUPTOI: + + case OP_MINUPTO: + case OP_MINUPTOI: + case OP_NOTMINUPTO: + case OP_NOTMINUPTOI: + + case OP_POSUPTO: + case OP_POSUPTOI: + case OP_NOTPOSUPTO: + case OP_NOTPOSUPTOI: + + if (utf && HAS_EXTRALEN(code[1 + IMM2_SIZE])) code += GET_EXTRALEN(code[1 + IMM2_SIZE]); + break; +#endif + + /* MARK, and PRUNE/SKIP/THEN with an argument must skip over the argument + string. */ + + case OP_MARK: + case OP_PRUNE_ARG: + case OP_SKIP_ARG: + case OP_THEN_ARG: + code += code[1]; + break; + + /* None of the remaining opcodes are required to match a character. */ + + default: + break; + } + } + +return TRUE; +} + + + +/************************************************* +* Scan compiled regex for non-emptiness * +*************************************************/ + +/* This function is called to check for left recursive calls. We want to check +the current branch of the current pattern to see if it could match the empty +string. If it could, we must look outwards for branches at other levels, +stopping when we pass beyond the bracket which is the subject of the recursion. +This function is called only during the real compile, not during the +pre-compile. + +Arguments: + code points to start of the recursion + endcode points to where to stop (current RECURSE item) + bcptr points to the chain of current (unclosed) branch starts + utf TRUE if in UTF mode + cd compile data + +Returns: TRUE if what is matched could be empty +*/ + +static BOOL +could_be_empty(PCRE2_SPTR code, PCRE2_SPTR endcode, branch_chain *bcptr, + BOOL utf, compile_data *cd) +{ +while (bcptr != NULL && bcptr->current_branch >= code) + { + if (!could_be_empty_branch(bcptr->current_branch, endcode, utf, cd, NULL)) + return FALSE; + bcptr = bcptr->outer; + } +return TRUE; +} + + + +/************************************************* +* Expand the workspace * +*************************************************/ + +/* This function is called during the second compiling phase, if the number of +forward references fills the existing workspace, which is originally a block on +the stack. A larger block is obtained from the heap unless the ultimate limit +has been reached or the increase will be rather small. + +Argument: pointer to the compile data block +Returns: 0 if all went well, else an error number +*/ + +static int +expand_workspace(compile_data *cd) +{ +PCRE2_UCHAR *newspace; +int newsize = cd->workspace_size * 2; +if (newsize > COMPILE_WORK_SIZE_MAX) newsize = COMPILE_WORK_SIZE_MAX; +if (cd->workspace_size >= COMPILE_WORK_SIZE_MAX || + newsize - cd->workspace_size < WORK_SIZE_SAFETY_MARGIN) + return ERR72; +newspace = cd->cx->memctl.malloc(CU2BYTES(newsize), cd->cx->memctl.memory_data); +if (newspace == NULL) return ERR21; +memcpy(newspace, cd->start_workspace, cd->workspace_size * sizeof(PCRE2_UCHAR)); +cd->hwm = (PCRE2_UCHAR *)newspace + (cd->hwm - cd->start_workspace); +if (cd->workspace_size > COMPILE_WORK_SIZE) + cd->cx->memctl.free((void *)cd->start_workspace, cd->cx->memctl.memory_data); +cd->start_workspace = newspace; +cd->workspace_size = newsize; +return 0; +} + + + +/************************************************* +* Check for counted repeat * +*************************************************/ + +/* This function is called when a '{' is encountered in a place where it might +start a quantifier. It looks ahead to see if it really is a quantifier, that +is, one of the forms {ddd} {ddd,} or {ddd,ddd} where the ddds are digits. + +Argument: pointer to the first char after '{' +Returns: TRUE or FALSE +*/ + +static BOOL +is_counted_repeat(PCRE2_SPTR p) +{ +if (!IS_DIGIT(*p)) return FALSE; +p++; +while (IS_DIGIT(*p)) p++; +if (*p == CHAR_RIGHT_CURLY_BRACKET) return TRUE; + +if (*p++ != CHAR_COMMA) return FALSE; +if (*p == CHAR_RIGHT_CURLY_BRACKET) return TRUE; + +if (!IS_DIGIT(*p)) return FALSE; +p++; +while (IS_DIGIT(*p)) p++; + +return (*p == CHAR_RIGHT_CURLY_BRACKET); +} + + + +/************************************************* +* Handle escapes * +*************************************************/ + +/* This function is called when a \ has been encountered. It either returns a +positive value for a simple escape such as \d, or 0 for a data character, which +is placed in chptr. A backreference to group n is returned as negative n. On +entry, ptr is pointing at the \. On exit, it points the final code unit of the +escape sequence. + +Arguments: + ptrptr points to the pattern position pointer + chptr points to a returned data character + errorcodeptr points to the errorcode variable (containing zero) + options the current options bits + isclass TRUE if inside a character class + cd compile data block + +Returns: zero => a data character + positive => a special escape sequence + negative => a back reference + on error, errorcodeptr is set non-zero +*/ + +static int +check_escape(PCRE2_SPTR *ptrptr, uint32_t *chptr, int *errorcodeptr, + int options, BOOL isclass, compile_data *cd) +{ +BOOL utf = (options & PCRE2_UTF) != 0; +PCRE2_SPTR ptr = *ptrptr + 1; +register uint32_t c, cc; +int escape = 0; +int i; + +GETCHARINCTEST(c, ptr); /* Get character value, increment pointer */ +ptr--; /* Set pointer back to the last code unit */ + +/* If backslash is at the end of the pattern, it's an error. */ + +if (c == CHAR_NULL && ptr >= cd->end_pattern) *errorcodeptr = ERR1; + +/* Non-alphanumerics are literals, so we just leave the value in c. An initial +value test saves a memory lookup for code points outside the alphanumeric +range. Otherwise, do a table lookup. A non-zero result is something that can be +returned immediately. Otherwise further processing is required. */ + +else if (c < ESCAPES_FIRST || c > ESCAPES_LAST) {} /* Definitely literal */ + +else if ((i = escapes[c - ESCAPES_FIRST]) != 0) + { + if (i > 0) c = (uint32_t)i; /* Positive is a data character */ + else escape = -i; /* Else return a special escape */ + } + +/* Escapes that need further processing, including those that are unknown. */ + +else + { + PCRE2_SPTR oldptr; + BOOL braced, negated, overflow; + unsigned int s; + + switch (c) + { + /* A number of Perl escapes are not handled by PCRE. We give an explicit + error. */ + + case CHAR_l: + case CHAR_L: + *errorcodeptr = ERR37; + break; + + /* \u is unrecognized when PCRE2_ALT_BSUX is not set. When it is treated + specially, \u must be followed by four hex digits. Otherwise it is a + lowercase u letter. */ + + case CHAR_u: + if ((options & PCRE2_ALT_BSUX) == 0) *errorcodeptr = ERR37; else + { + uint32_t xc; + if ((cc = XDIGIT(ptr[1])) == 0xff) break; /* Not a hex digit */ + if ((xc = XDIGIT(ptr[2])) == 0xff) break; /* Not a hex digit */ + cc = (cc << 4) | xc; + if ((xc = XDIGIT(ptr[3])) == 0xff) break; /* Not a hex digit */ + cc = (cc << 4) | xc; + if ((xc = XDIGIT(ptr[4])) == 0xff) break; /* Not a hex digit */ + c = (cc << 4) | xc; + ptr += 4; + if (utf) + { + if (c > 0x10ffffU) *errorcodeptr = ERR76; + else if (c >= 0xd800 && c <= 0xdfff) *errorcodeptr = ERR73; + } + else if (c > MAX_NON_UTF_CHAR) *errorcodeptr = ERR76; + } + break; + + case CHAR_U: + /* \U is unrecognized unless PCRE2_ALT_BSUX is set, in which case it is an + upper case letter. */ + if ((options & PCRE2_ALT_BSUX) == 0) *errorcodeptr = ERR37; + break; + + /* In a character class, \g is just a literal "g". Outside a character + class, \g must be followed by one of a number of specific things: + + (1) A number, either plain or braced. If positive, it is an absolute + backreference. If negative, it is a relative backreference. This is a Perl + 5.10 feature. + + (2) Perl 5.10 also supports \g{name} as a reference to a named group. This + is part of Perl's movement towards a unified syntax for back references. As + this is synonymous with \k{name}, we fudge it up by pretending it really + was \k. + + (3) For Oniguruma compatibility we also support \g followed by a name or a + number either in angle brackets or in single quotes. However, these are + (possibly recursive) subroutine calls, _not_ backreferences. Just return + the ESC_g code (cf \k). */ + + case CHAR_g: + if (isclass) break; + if (ptr[1] == CHAR_LESS_THAN_SIGN || ptr[1] == CHAR_APOSTROPHE) + { + escape = ESC_g; + break; + } + + /* Handle the Perl-compatible cases */ + + if (ptr[1] == CHAR_LEFT_CURLY_BRACKET) + { + PCRE2_SPTR p; + for (p = ptr+2; *p != CHAR_NULL && *p != CHAR_RIGHT_CURLY_BRACKET; p++) + if (*p != CHAR_MINUS && !IS_DIGIT(*p)) break; + if (*p != CHAR_NULL && *p != CHAR_RIGHT_CURLY_BRACKET) + { + escape = ESC_k; + break; + } + braced = TRUE; + ptr++; + } + else braced = FALSE; + + if (ptr[1] == CHAR_MINUS) + { + negated = TRUE; + ptr++; + } + else negated = FALSE; + + /* The integer range is limited by the machine's int representation. */ + s = 0; + overflow = FALSE; + while (IS_DIGIT(ptr[1])) + { + if (s > INT_MAX / 10 - 1) /* Integer overflow */ + { + overflow = TRUE; + break; + } + s = s * 10 + (int)(*(++ptr) - CHAR_0); + } + if (overflow) /* Integer overflow */ + { + while (IS_DIGIT(ptr[1])) + ptr++; + *errorcodeptr = ERR61; + break; + } + + if (braced && *(++ptr) != CHAR_RIGHT_CURLY_BRACKET) + { + *errorcodeptr = ERR57; + break; + } + + if (s == 0) + { + *errorcodeptr = ERR58; + break; + } + + if (negated) + { + if (s > cd->bracount) + { + *errorcodeptr = ERR15; + break; + } + s = cd->bracount - (s - 1); + } + + escape = -s; + break; + + /* The handling of escape sequences consisting of a string of digits + starting with one that is not zero is not straightforward. Perl has changed + over the years. Nowadays \g{} for backreferences and \o{} for octal are + recommended to avoid the ambiguities in the old syntax. + + Outside a character class, the digits are read as a decimal number. If the + number is less than 8 (used to be 10), or if there are that many previous + extracting left brackets, then it is a back reference. Otherwise, up to + three octal digits are read to form an escaped byte. Thus \123 is likely to + be octal 123 (cf \0123, which is octal 012 followed by the literal 3). If + the octal value is greater than 377, the least significant 8 bits are + taken. \8 and \9 are treated as the literal characters 8 and 9. + + Inside a character class, \ followed by a digit is always either a literal + 8 or 9 or an octal number. */ + + case CHAR_1: case CHAR_2: case CHAR_3: case CHAR_4: case CHAR_5: + case CHAR_6: case CHAR_7: case CHAR_8: case CHAR_9: + + if (!isclass) + { + oldptr = ptr; + /* The integer range is limited by the machine's int representation. */ + s = (int)(c -CHAR_0); + overflow = FALSE; + while (IS_DIGIT(ptr[1])) + { + if (s > INT_MAX / 10 - 1) /* Integer overflow */ + { + overflow = TRUE; + break; + } + s = s * 10 + (int)(*(++ptr) - CHAR_0); + } + if (overflow) /* Integer overflow */ + { + while (IS_DIGIT(ptr[1])) + ptr++; + *errorcodeptr = ERR61; + break; + } + if (s < 8 || s <= cd->bracount) /* Check for back reference */ + { + escape = -s; + break; + } + ptr = oldptr; /* Put the pointer back and fall through */ + } + + /* Handle a digit following \ when the number is not a back reference. If + the first digit is 8 or 9, Perl used to generate a binary zero byte and + then treat the digit as a following literal. At least by Perl 5.18 this + changed so as not to insert the binary zero. */ + + if ((c = *ptr) >= CHAR_8) break; + + /* Fall through with a digit less than 8 */ + + /* \0 always starts an octal number, but we may drop through to here with a + larger first octal digit. The original code used just to take the least + significant 8 bits of octal numbers (I think this is what early Perls used + to do). Nowadays we allow for larger numbers in UTF-8 mode and 16-bit mode, + but no more than 3 octal digits. */ + + case CHAR_0: + c -= CHAR_0; + while(i++ < 2 && ptr[1] >= CHAR_0 && ptr[1] <= CHAR_7) + c = c * 8 + *(++ptr) - CHAR_0; +#ifdef COMPILE_PCRE8 + if (!utf && c > 0xff) *errorcodeptr = ERR51; +#endif + break; + + /* \o is a relatively new Perl feature, supporting a more general way of + specifying character codes in octal. The only supported form is \o{ddd}. */ + + case CHAR_o: + if (ptr[1] != CHAR_LEFT_CURLY_BRACKET) *errorcodeptr = ERR81; else + if (ptr[2] == CHAR_RIGHT_CURLY_BRACKET) *errorcodeptr = ERR86; else + { + ptr += 2; + c = 0; + overflow = FALSE; + while (*ptr >= CHAR_0 && *ptr <= CHAR_7) + { + cc = *ptr++; + if (c == 0 && cc == CHAR_0) continue; /* Leading zeroes */ +#ifdef COMPILE_PCRE32 + if (c >= 0x20000000l) { overflow = TRUE; break; } +#endif + c = (c << 3) + cc - CHAR_0 ; +#if defined COMPILE_PCRE8 + if (c > (utf ? 0x10ffffU : 0xffU)) { overflow = TRUE; break; } +#elif defined COMPILE_PCRE16 + if (c > (utf ? 0x10ffffU : 0xffffU)) { overflow = TRUE; break; } +#elif defined COMPILE_PCRE32 + if (utf && c > 0x10ffffU) { overflow = TRUE; break; } +#endif + } + if (overflow) + { + while (*ptr >= CHAR_0 && *ptr <= CHAR_7) ptr++; + *errorcodeptr = ERR34; + } + else if (*ptr == CHAR_RIGHT_CURLY_BRACKET) + { + if (utf && c >= 0xd800 && c <= 0xdfff) *errorcodeptr = ERR73; + } + else *errorcodeptr = ERR80; + } + break; + + /* \x is complicated. When PCRE2_ALT_BSUX is set, \x must be followed by + two hexadecimal digits. Otherwise it is a lowercase x letter. */ + + case CHAR_x: + if ((options & PCRE2_ALT_BSUX) != 0) + { + uint32_t xc; + if ((cc = XDIGIT(ptr[1])) == 0xff) break; /* Not a hex digit */ + if ((xc = XDIGIT(ptr[2])) == 0xff) break; /* Not a hex digit */ + c = (cc << 4) | xc; + ptr += 2; + } /* End PCRE2_ALT_BSUX handling */ + + /* Handle \x in Perl's style. \x{ddd} is a character number which can be + greater than 0xff in UTF-8 or non-8bit mode, but only if the ddd are hex + digits. If not, { used to be treated as a data character. However, Perl + seems to read hex digits up to the first non-such, and ignore the rest, so + that, for example \x{zz} matches a binary zero. This seems crazy, so PCRE + now gives an error. */ + + else + { + if (ptr[1] == CHAR_LEFT_CURLY_BRACKET) + { + ptr += 2; + if (*ptr == CHAR_RIGHT_CURLY_BRACKET) + { + *errorcodeptr = ERR86; + break; + } + c = 0; + overflow = FALSE; + + while ((cc = XDIGIT(*ptr)) != 0xff) + { + if (c == 0 && cc == 0) continue; /* Leading zeroes */ +#if PCRE2_CODE_UNIT_WIDTH == 32 + if (c >= 0x10000000l) { overflow = TRUE; break; } +#endif + c = (c << 4) | cc; + ptr++; + if ((utf && c > 0x10ffffU) || (!utf && c > MAX_NON_UTF_CHAR)) + { + overflow = TRUE; + break; + } + } + + if (overflow) + { + while (XDIGIT(*ptr) != 0xff) ptr++; + *errorcodeptr = ERR34; + } + else if (*ptr == CHAR_RIGHT_CURLY_BRACKET) + { + if (utf && c >= 0xd800 && c <= 0xdfff) *errorcodeptr = ERR73; + } + + /* If the sequence of hex digits does not end with '}', give an error. + We used just to recognize this construct and fall through to the normal + \x handling, but nowadays Perl gives an error, which seems much more + sensible, so we do too. */ + + else *errorcodeptr = ERR79; + } /* End of \x{} processing */ + + /* Read a single-byte hex-defined char (up to two hex digits after \x) */ + + else + { + c = 0; + if ((cc = XDIGIT(ptr[1])) == 0xff) break; /* Not a hex digit */ + ptr++; + c = cc; + if ((cc = XDIGIT(ptr[1])) == 0xff) break; /* Not a hex digit */ + ptr++; + c = (c << 4) | cc; + } /* End of \xdd handling */ + } /* End of Perl-style \x handling */ + break; + + /* For \c, a following letter is upper-cased; then the 0x40 bit is flipped. + An error is given if the byte following \c is not an ASCII character. This + coding is ASCII-specific, but then the whole concept of \cx is + ASCII-specific. (However, an EBCDIC equivalent has now been added.) */ + + case CHAR_c: + c = *(++ptr); + if (c == CHAR_NULL) + { + *errorcodeptr = ERR2; + break; + } +#ifndef EBCDIC /* ASCII/UTF-8 coding */ + if (c > 127) /* Excludes all non-ASCII in either mode */ + { + *errorcodeptr = ERR68; + break; + } + if (c >= CHAR_a && c <= CHAR_z) c -= 32; + c ^= 0x40; +#else /* EBCDIC coding */ + if (c >= CHAR_a && c <= CHAR_z) c += 64; + c ^= 0xC0; +#endif + break; + + /* Any other alphanumeric following \ is an error. Perl gives an error only + if in warning mode, but PCRE doesn't have a warning mode. */ + + default: + *errorcodeptr = ERR3; + break; + } + } + +/* Perl supports \N{name} for character names, as well as plain \N for "not +newline". PCRE does not support \N{name}. However, it does support +quantification such as \N{2,3}. */ + +if (escape == ESC_N && ptr[1] == CHAR_LEFT_CURLY_BRACKET && + !is_counted_repeat(ptr+2)) + *errorcodeptr = ERR37; + +/* If PCRE2_UCP is set, we change the values for \d etc. */ + +if ((options & PCRE2_UCP) != 0 && escape >= ESC_D && escape <= ESC_w) + escape += (ESC_DU - ESC_D); + +/* Set the pointer to the final character before returning. */ + +*ptrptr = ptr; +*chptr = c; +return escape; +} + + + +#ifdef SUPPORT_UTF +/************************************************* +* Handle \P and \p * +*************************************************/ + +/* This function is called after \P or \p has been encountered, provided that +PCRE2 is compiled with support for UTF and Unicode properties. On entry, the +contents of ptrptr are pointing at the P or p. On exit, it is left pointing at +the final code unit of the escape sequence. + +Arguments: + ptrptr the pattern position pointer + negptr a boolean that is set TRUE for negation else FALSE + ptypeptr an unsigned int that is set to the type value + pdataptr an unsigned int that is set to the detailed property value + errorcodeptr the error code variable + cd the compile data + +Returns: TRUE if the type value was found, or FALSE for an invalid type +*/ + +static BOOL +get_ucp(PCRE2_SPTR *ptrptr, BOOL *negptr, unsigned int *ptypeptr, + unsigned int *pdataptr, int *errorcodeptr, compile_data *cd) +{ +register PCRE2_UCHAR c; +int i, bot, top; +PCRE2_SPTR ptr = *ptrptr; +PCRE2_UCHAR name[32]; + +*negptr = FALSE; +c = *(++ptr); + +/* \P or \p can be followed by a name in {}, optionally preceded by ^ for +negation. */ + +if (c == CHAR_LEFT_CURLY_BRACKET) + { + if (ptr[1] == CHAR_CIRCUMFLEX_ACCENT) + { + *negptr = TRUE; + ptr++; + } + for (i = 0; i < (int)(sizeof(name) / sizeof(PCRE2_UCHAR)) - 1; i++) + { + c = *(++ptr); + if (c == CHAR_NULL) goto ERROR_RETURN; + if (c == CHAR_RIGHT_CURLY_BRACKET) break; + name[i] = c; + } + if (c != CHAR_RIGHT_CURLY_BRACKET) goto ERROR_RETURN; + name[i] = 0; + } + +/* Otherwise there is just one following character, which must be an ASCII +letter. */ + +else if (MAX_255(c) && (cd->ctypes[c] & ctype_letter) != 0) + { + name[0] = c; + name[1] = 0; + } +else goto ERROR_RETURN; + +*ptrptr = ptr; + +/* Search for a recognized property name using binary chop. */ + +bot = 0; +top = PRIV(utt_size); + +while (bot < top) + { + int r; + i = (bot + top) >> 1; + r = PRIV(strcmp_c8)(name, PRIV(utt_names) + PRIV(utt)[i].name_offset); + if (r == 0) + { + *ptypeptr = PRIV(utt)[i].type; + *pdataptr = PRIV(utt)[i].value; + return TRUE; + } + if (r > 0) bot = i + 1; else top = i; + } +*errorcodeptr = ERR47; /* Unrecognized name */ +return FALSE; + +ERROR_RETURN: /* Malformed \P or \p */ +*errorcodeptr = ERR46; +*ptrptr = ptr; +return FALSE; +} +#endif + + + +/************************************************* +* Read repeat counts * +*************************************************/ + +/* Read an item of the form {n,m} and return the values. This is called only +after is_counted_repeat() has confirmed that a repeat-count quantifier exists, +so the syntax is guaranteed to be correct, but we need to check the values. + +Arguments: + p pointer to first char after '{' + minp pointer to int for min + maxp pointer to int for max + returned as -1 if no max + errorcodeptr points to error code variable + +Returns: pointer to '}' on success; + current ptr on error, with errorcodeptr set non-zero +*/ + +static PCRE2_SPTR +read_repeat_counts(PCRE2_SPTR p, int *minp, int *maxp, int *errorcodeptr) +{ +int min = 0; +int max = -1; + +while (IS_DIGIT(*p)) + { + min = min * 10 + (int)(*p++ - CHAR_0); + if (min > 65535) + { + *errorcodeptr = ERR5; + return p; + } + } + +if (*p == CHAR_RIGHT_CURLY_BRACKET) max = min; else + { + if (*(++p) != CHAR_RIGHT_CURLY_BRACKET) + { + max = 0; + while(IS_DIGIT(*p)) + { + max = max * 10 + (int)(*p++ - CHAR_0); + if (max > 65535) + { + *errorcodeptr = ERR5; + return p; + } + } + if (max < min) + { + *errorcodeptr = ERR4; + return p; + } + } + } + +*minp = min; +*maxp = max; +return p; +} + + + +/************************************************* +* Scan compiled regex for specific bracket * +*************************************************/ + +/* This function scans through a compiled pattern until it finds a +capturing bracket with the given number, or, if the number is negative, an +instance of OP_REVERSE for a lookbehind. The function is global in the C sense +so that it can be called from pcre2_study() when finding the minimum matching +length. + +Arguments: + code points to start of expression + utf TRUE in UTF mode + number the required bracket number or negative to find a lookbehind + +Returns: pointer to the opcode for the bracket, or NULL if not found +*/ + +PCRE2_SPTR +PRIV(find_bracket)(PCRE2_SPTR code, BOOL utf, int number) +{ +for (;;) + { + register PCRE2_UCHAR c = *code; + + if (c == OP_END) return NULL; + + /* XCLASS is used for classes that cannot be represented just by a bit + map. This includes negated single high-valued characters. The length in + the table is zero; the actual length is stored in the compiled code. */ + + if (c == OP_XCLASS) code += GET(code, 1); + + /* Handle recursion */ + + else if (c == OP_REVERSE) + { + if (number < 0) return (PCRE2_UCHAR *)code; + code += PRIV(OP_lengths)[c]; + } + + /* Handle capturing bracket */ + + else if (c == OP_CBRA || c == OP_SCBRA || + c == OP_CBRAPOS || c == OP_SCBRAPOS) + { + int n = (int)GET2(code, 1+LINK_SIZE); + if (n == number) return (PCRE2_UCHAR *)code; + code += PRIV(OP_lengths)[c]; + } + + /* Otherwise, we can get the item's length from the table, except that for + repeated character types, we have to test for \p and \P, which have an extra + two bytes of parameters, and for MARK/PRUNE/SKIP/THEN with an argument, we + must add in its length. */ + + else + { + switch(c) + { + case OP_TYPESTAR: + case OP_TYPEMINSTAR: + case OP_TYPEPLUS: + case OP_TYPEMINPLUS: + case OP_TYPEQUERY: + case OP_TYPEMINQUERY: + case OP_TYPEPOSSTAR: + case OP_TYPEPOSPLUS: + case OP_TYPEPOSQUERY: + if (code[1] == OP_PROP || code[1] == OP_NOTPROP) code += 2; + break; + + case OP_TYPEUPTO: + case OP_TYPEMINUPTO: + case OP_TYPEEXACT: + case OP_TYPEPOSUPTO: + if (code[1 + IMM2_SIZE] == OP_PROP || code[1 + IMM2_SIZE] == OP_NOTPROP) + code += 2; + break; + + case OP_MARK: + case OP_PRUNE_ARG: + case OP_SKIP_ARG: + case OP_THEN_ARG: + code += code[1]; + break; + } + + /* Add in the fixed length from the table */ + + code += PRIV(OP_lengths)[c]; + + /* In UTF-8 and UTF-16 modes, opcodes that are followed by a character may be + followed by a multi-byte character. The length in the table is a minimum, so + we have to arrange to skip the extra bytes. */ + +#ifdef MAYBE_UTF_MULTI + if (utf) switch(c) + { + case OP_CHAR: + case OP_CHARI: + case OP_EXACT: + case OP_EXACTI: + case OP_UPTO: + case OP_UPTOI: + case OP_MINUPTO: + case OP_MINUPTOI: + case OP_POSUPTO: + case OP_POSUPTOI: + case OP_STAR: + case OP_STARI: + case OP_MINSTAR: + case OP_MINSTARI: + case OP_POSSTAR: + case OP_POSSTARI: + case OP_PLUS: + case OP_PLUSI: + case OP_MINPLUS: + case OP_MINPLUSI: + case OP_POSPLUS: + case OP_POSPLUSI: + case OP_QUERY: + case OP_QUERYI: + case OP_MINQUERY: + case OP_MINQUERYI: + case OP_POSQUERY: + case OP_POSQUERYI: + if (HAS_EXTRALEN(code[-1])) code += GET_EXTRALEN(code[-1]); + break; + } +#else + (void)(utf); /* Keep compiler happy by referencing function argument */ +#endif + } + } +} + + + +/************************************************* +* Scan compiled regex for recursion reference * +*************************************************/ + +/* This function scans through a compiled pattern until it finds an instance of +OP_RECURSE. + +Arguments: + code points to start of expression + utf TRUE in UTF mode + +Returns: pointer to the opcode for OP_RECURSE, or NULL if not found +*/ + +static PCRE2_SPTR +find_recurse(PCRE2_SPTR code, BOOL utf) +{ +for (;;) + { + register PCRE2_UCHAR c = *code; + if (c == OP_END) return NULL; + if (c == OP_RECURSE) return code; + + /* XCLASS is used for classes that cannot be represented just by a bit + map. This includes negated single high-valued characters. The length in + the table is zero; the actual length is stored in the compiled code. */ + + if (c == OP_XCLASS) code += GET(code, 1); + + /* Otherwise, we can get the item's length from the table, except that for + repeated character types, we have to test for \p and \P, which have an extra + two bytes of parameters, and for MARK/PRUNE/SKIP/THEN with an argument, we + must add in its length. */ + + else + { + switch(c) + { + case OP_TYPESTAR: + case OP_TYPEMINSTAR: + case OP_TYPEPLUS: + case OP_TYPEMINPLUS: + case OP_TYPEQUERY: + case OP_TYPEMINQUERY: + case OP_TYPEPOSSTAR: + case OP_TYPEPOSPLUS: + case OP_TYPEPOSQUERY: + if (code[1] == OP_PROP || code[1] == OP_NOTPROP) code += 2; + break; + + case OP_TYPEPOSUPTO: + case OP_TYPEUPTO: + case OP_TYPEMINUPTO: + case OP_TYPEEXACT: + if (code[1 + IMM2_SIZE] == OP_PROP || code[1 + IMM2_SIZE] == OP_NOTPROP) + code += 2; + break; + + case OP_MARK: + case OP_PRUNE_ARG: + case OP_SKIP_ARG: + case OP_THEN_ARG: + code += code[1]; + break; + } + + /* Add in the fixed length from the table */ + + code += PRIV(OP_lengths)[c]; + + /* In UTF-8 and UTF-16 modes, opcodes that are followed by a character may + be followed by a multi-unit character. The length in the table is a + minimum, so we have to arrange to skip the extra units. */ + +#ifdef MAYBE_UTF_MULTI + if (utf) switch(c) + { + case OP_CHAR: + case OP_CHARI: + case OP_NOT: + case OP_NOTI: + case OP_EXACT: + case OP_EXACTI: + case OP_NOTEXACT: + case OP_NOTEXACTI: + case OP_UPTO: + case OP_UPTOI: + case OP_NOTUPTO: + case OP_NOTUPTOI: + case OP_MINUPTO: + case OP_MINUPTOI: + case OP_NOTMINUPTO: + case OP_NOTMINUPTOI: + case OP_POSUPTO: + case OP_POSUPTOI: + case OP_NOTPOSUPTO: + case OP_NOTPOSUPTOI: + case OP_STAR: + case OP_STARI: + case OP_NOTSTAR: + case OP_NOTSTARI: + case OP_MINSTAR: + case OP_MINSTARI: + case OP_NOTMINSTAR: + case OP_NOTMINSTARI: + case OP_POSSTAR: + case OP_POSSTARI: + case OP_NOTPOSSTAR: + case OP_NOTPOSSTARI: + case OP_PLUS: + case OP_PLUSI: + case OP_NOTPLUS: + case OP_NOTPLUSI: + case OP_MINPLUS: + case OP_MINPLUSI: + case OP_NOTMINPLUS: + case OP_NOTMINPLUSI: + case OP_POSPLUS: + case OP_POSPLUSI: + case OP_NOTPOSPLUS: + case OP_NOTPOSPLUSI: + case OP_QUERY: + case OP_QUERYI: + case OP_NOTQUERY: + case OP_NOTQUERYI: + case OP_MINQUERY: + case OP_MINQUERYI: + case OP_NOTMINQUERY: + case OP_NOTMINQUERYI: + case OP_POSQUERY: + case OP_POSQUERYI: + case OP_NOTPOSQUERY: + case OP_NOTPOSQUERYI: + if (HAS_EXTRALEN(code[-1])) code += GET_EXTRALEN(code[-1]); + break; + } +#else + (void)(utf); /* Keep compiler happy by referencing function argument */ +#endif + } + } +} + + + +/************************************************* +* Adjust OP_RECURSE items in repeated group * +*************************************************/ + +/* OP_RECURSE items contain an offset from the start of the regex to the group +that is referenced. This means that groups can be replicated for fixed +repetition simply by copying (because the recursion is allowed to refer to +earlier groups that are outside the current group). However, when a group is +optional (i.e. the minimum quantifier is zero), OP_BRAZERO or OP_SKIPZERO is +inserted before it, after it has been compiled. This means that any OP_RECURSE +items within it that refer to the group itself or any contained groups have to +have their offsets adjusted. That is one of the jobs of this function. Before +it is called, the partially compiled regex must be temporarily terminated with +OP_END. + +This function has been extended with the possibility of forward references for +recursions and subroutine calls. It must also check the list of such references +for the group we are dealing with. If it finds that one of the recursions in +the current group is on this list, it adjusts the offset in the list, not the +value in the reference (which is a group number). + +Arguments: + group points to the start of the group + adjust the amount by which the group is to be moved + utf TRUE in UTF mode + cd compile data + save_hwm the hwm forward reference pointer at the start of the group + +Returns: nothing +*/ + +static void +adjust_recurse(PCRE2_UCHAR *group, int adjust, BOOL utf, compile_data *cd, + PCRE2_UCHAR *save_hwm) +{ +PCRE2_UCHAR *ptr = group; + +while ((ptr = (PCRE2_UCHAR *)find_recurse(ptr, utf)) != NULL) + { + int offset; + PCRE2_UCHAR *hc; + + /* See if this recursion is on the forward reference list. If so, adjust the + reference. */ + + for (hc = save_hwm; hc < cd->hwm; hc += LINK_SIZE) + { + offset = (int)GET(hc, 0); + if (cd->start_code + offset == ptr + 1) + { + PUT(hc, 0, offset + adjust); + break; + } + } + + /* Otherwise, adjust the recursion offset if it's after the start of this + group. */ + + if (hc >= cd->hwm) + { + offset = (int)GET(ptr, 1); + if (cd->start_code + offset >= group) PUT(ptr, 1, offset + adjust); + } + + ptr += 1 + LINK_SIZE; + } +} + + + +/************************************************* +* Check for POSIX class syntax * +*************************************************/ + +/* This function is called when the sequence "[:" or "[." or "[=" is +encountered in a character class. It checks whether this is followed by a +sequence of characters terminated by a matching ":]" or ".]" or "=]". If we +reach an unescaped ']' without the special preceding character, return FALSE. + +Originally, this function only recognized a sequence of letters between the +terminators, but it seems that Perl recognizes any sequence of characters, +though of course unknown POSIX names are subsequently rejected. Perl gives an +"Unknown POSIX class" error for [:f\oo:] for example, where previously PCRE +didn't consider this to be a POSIX class. Likewise for [:1234:]. + +The problem in trying to be exactly like Perl is in the handling of escapes. We +have to be sure that [abc[:x\]pqr] is *not* treated as containing a POSIX +class, but [abc[:x\]pqr:]] is (so that an error can be generated). The code +below handles the special case of \], but does not try to do any other escape +processing. This makes it different from Perl for cases such as [:l\ower:] +where Perl recognizes it as the POSIX class "lower" but PCRE does not recognize +"l\ower". This is a lesser evil than not diagnosing bad classes when Perl does, +I think. + +A user pointed out that PCRE was rejecting [:a[:digit:]] whereas Perl was not. +It seems that the appearance of a nested POSIX class supersedes an apparent +external class. For example, [:a[:digit:]b:] matches "a", "b", ":", or +a digit. + +In Perl, unescaped square brackets may also appear as part of class names. For +example, [:a[:abc]b:] gives unknown POSIX class "[:abc]b:]". However, for +[:a[:abc]b][b:] it gives unknown POSIX class "[:abc]b][b:]", which does not +seem right at all. PCRE does not allow closing square brackets in POSIX class +names. + +Arguments: + ptr pointer to the initial [ + endptr where to return the end pointer + +Returns: TRUE or FALSE +*/ + +static BOOL +check_posix_syntax(PCRE2_SPTR ptr, PCRE2_SPTR *endptr) +{ +PCRE2_UCHAR terminator; /* Don't combine these lines; the Solaris cc */ +terminator = *(++ptr); /* compiler warns about "non-constant" initializer. */ + +for (++ptr; *ptr != CHAR_NULL; ptr++) + { + if (*ptr == CHAR_BACKSLASH && ptr[1] == CHAR_RIGHT_SQUARE_BRACKET) ptr++; + else if (*ptr == CHAR_RIGHT_SQUARE_BRACKET) return FALSE; + else + { + if (*ptr == terminator && ptr[1] == CHAR_RIGHT_SQUARE_BRACKET) + { + *endptr = ptr; + return TRUE; + } + if (*ptr == CHAR_LEFT_SQUARE_BRACKET && + (ptr[1] == CHAR_COLON || ptr[1] == CHAR_DOT || + ptr[1] == CHAR_EQUALS_SIGN) && + check_posix_syntax(ptr, endptr)) + return FALSE; + } + } +return FALSE; +} + + + +/************************************************* +* Check POSIX class name * +*************************************************/ + +/* This function is called to check the name given in a POSIX-style class entry +such as [:alnum:]. + +Arguments: + ptr points to the first letter + len the length of the name + +Returns: a value representing the name, or -1 if unknown +*/ + +static int +check_posix_name(PCRE2_SPTR ptr, int len) +{ +const char *pn = posix_names; +register int yield = 0; +while (posix_name_lengths[yield] != 0) + { + if (len == posix_name_lengths[yield] && + PRIV(strncmp_c8)(ptr, pn, (unsigned int)len) == 0) return yield; + pn += posix_name_lengths[yield] + 1; + yield++; + } +return -1; +} + + + +#ifdef SUPPORT_UTF +/************************************************* +* Get othercase range * +*************************************************/ + +/* This function is passed the start and end of a class range in UCT mode. It +searches up the characters, looking for ranges of characters in the "other" +case. Each call returns the next one, updating the start address. A character +with multiple other cases is returned on its own with a special return value. + +Arguments: + cptr points to starting character value; updated + d end value + ocptr where to put start of othercase range + odptr where to put end of othercase range + +Yield: -1 when no more + 0 when a range is returned + >0 the CASESET offset for char with multiple other cases + in this case, ocptr contains the original +*/ + +static int +get_othercase_range(uint32_t *cptr, uint32_t d, uint32_t *ocptr, + uint32_t *odptr) +{ +uint32_t c, othercase, next; +unsigned int co; + +/* Find the first character that has an other case. If it has multiple other +cases, return its case offset value. */ + +for (c = *cptr; c <= d; c++) + { + if ((co = UCD_CASESET(c)) != 0) + { + *ocptr = c++; /* Character that has the set */ + *cptr = c; /* Rest of input range */ + return (int)co; + } + if ((othercase = UCD_OTHERCASE(c)) != c) break; + } + +if (c > d) return -1; /* Reached end of range */ + +/* Found a character that has a single other case. Search for the end of the +range, which is either the end of the input range, or a character that has zero +or more than one other cases. */ + +*ocptr = othercase; +next = othercase + 1; + +for (++c; c <= d; c++) + { + if ((co = UCD_CASESET(c)) != 0 || UCD_OTHERCASE(c) != next) break; + next++; + } + +*odptr = next - 1; /* End of othercase range */ +*cptr = c; /* Rest of input range */ +return 0; +} +#endif /* SUPPORT_UTF */ + + + +/************************************************* +* Add a character or range to a class * +*************************************************/ + +/* This function packages up the logic of adding a character or range of +characters to a class. The character values in the arguments will be within the +valid values for the current mode (8-bit, 16-bit, UTF, etc). This function is +mutually recursive with the function immediately below. + +Arguments: + classbits the bit map for characters < 256 + uchardptr points to the pointer for extra data + options the options word + cd compile data + start start of range character + end end of range character + +Returns: the number of < 256 characters added + the pointer to extra data is updated +*/ + +static int +add_to_class(uint8_t *classbits, PCRE2_UCHAR **uchardptr, int options, + compile_data *cd, uint32_t start, uint32_t end) +{ +uint32_t c; +uint32_t classbits_end = (end <= 0xff ? end : 0xff); +int n8 = 0; + +/* If caseless matching is required, scan the range and process alternate +cases. In Unicode, there are 8-bit characters that have alternate cases that +are greater than 255 and vice-versa. Sometimes we can just extend the original +range. */ + +if ((options & PCRE2_CASELESS) != 0) + { +#ifdef SUPPORT_UTF + if ((options & PCRE2_UTF) != 0) + { + int rc; + uint32_t oc, od; + + options &= ~PCRE2_CASELESS; /* Remove for recursive calls */ + c = start; + + while ((rc = get_othercase_range(&c, end, &oc, &od)) >= 0) + { + /* Handle a single character that has more than one other case. */ + + if (rc > 0) n8 += add_list_to_class(classbits, uchardptr, options, cd, + PRIV(ucd_caseless_sets) + rc, oc); + + /* Do nothing if the other case range is within the original range. */ + + else if (oc >= start && od <= end) continue; + + /* Extend the original range if there is overlap, noting that if oc < c, we + can't have od > end because a subrange is always shorter than the basic + range. Otherwise, use a recursive call to add the additional range. */ + + else if (oc < start && od >= start - 1) start = oc; /* Extend downwards */ + else if (od > end && oc <= end + 1) end = od; /* Extend upwards */ + else n8 += add_to_class(classbits, uchardptr, options, cd, oc, od); + } + } + else +#endif /* SUPPORT_UTF */ + + /* Not UTF mode */ + + for (c = start; c <= classbits_end; c++) + { + SETBIT(classbits, cd->fcc[c]); + n8++; + } + } + +/* Now handle the original range. Adjust the final value according to the bit +length - this means that the same lists of (e.g.) horizontal spaces can be used +in all cases. */ + +if ((options & PCRE2_UTF) == 0 && end > MAX_NON_UTF_CHAR) + end = MAX_NON_UTF_CHAR; + +/* Use the bitmap for characters < 256. Otherwise use extra data.*/ + +for (c = start; c <= classbits_end; c++) + { + /* Regardless of start, c will always be <= 255. */ + SETBIT(classbits, c); + n8++; + } + +#ifdef SUPPORT_WIDE_CHARS +if (start <= 0xff) start = 0xff + 1; + +if (end >= start) + { + PCRE2_UCHAR *uchardata = *uchardptr; + +#ifdef SUPPORT_UTF + if ((options & PCRE2_UTF) != 0) + { + if (start < end) + { + *uchardata++ = XCL_RANGE; + uchardata += PRIV(ord2utf)(start, uchardata); + uchardata += PRIV(ord2utf)(end, uchardata); + } + else if (start == end) + { + *uchardata++ = XCL_SINGLE; + uchardata += PRIV(ord2utf)(start, uchardata); + } + } + else +#endif /* SUPPORT_UTF */ + + /* Without UTF support, character values are constrained by the bit length, + and can only be > 256 for 16-bit and 32-bit libraries. */ + +#if PCRE2_CODE_UNIT_WIDTH == 8 + {} +#else + if (start < end) + { + *uchardata++ = XCL_RANGE; + *uchardata++ = start; + *uchardata++ = end; + } + else if (start == end) + { + *uchardata++ = XCL_SINGLE; + *uchardata++ = start; + } +#endif + + *uchardptr = uchardata; /* Updata extra data pointer */ + } +#endif /* SUPPORT_WIDE_CHARS */ + +return n8; /* Number of 8-bit characters */ +} + + + +/************************************************* +* Add a list of characters to a class * +*************************************************/ + +/* This function is used for adding a list of case-equivalent characters to a +class, and also for adding a list of horizontal or vertical whitespace. If the +list is in order (which it should be), ranges of characters are detected and +handled appropriately. This function is mutually recursive with the function +above. + +Arguments: + classbits the bit map for characters < 256 + uchardptr points to the pointer for extra data + options the options word + cd contains pointers to tables etc. + p points to row of 32-bit values, terminated by NOTACHAR + except character to omit; this is used when adding lists of + case-equivalent characters to avoid including the one we + already know about + +Returns: the number of < 256 characters added + the pointer to extra data is updated +*/ + +static int +add_list_to_class(uint8_t *classbits, PCRE2_UCHAR **uchardptr, int options, + compile_data *cd, const uint32_t *p, unsigned int except) +{ +int n8 = 0; +while (p[0] < NOTACHAR) + { + int n = 0; + if (p[0] != except) + { + while(p[n+1] == p[0] + n + 1) n++; + n8 += add_to_class(classbits, uchardptr, options, cd, p[0], p[n]); + } + p += n + 1; + } +return n8; +} + + + +/************************************************* +* Add characters not in a list to a class * +*************************************************/ + +/* This function is used for adding the complement of a list of horizontal or +vertical whitespace to a class. The list must be in order. + +Arguments: + classbits the bit map for characters < 256 + uchardptr points to the pointer for extra data + options the options word + cd contains pointers to tables etc. + p points to row of 32-bit values, terminated by NOTACHAR + +Returns: the number of < 256 characters added + the pointer to extra data is updated +*/ + +static int +add_not_list_to_class(uint8_t *classbits, PCRE2_UCHAR **uchardptr, + int options, compile_data *cd, const uint32_t *p) +{ +BOOL utf = (options & PCRE2_UTF) != 0; +int n8 = 0; +if (p[0] > 0) + n8 += add_to_class(classbits, uchardptr, options, cd, 0, p[0] - 1); +while (p[0] < NOTACHAR) + { + while (p[1] == p[0] + 1) p++; + n8 += add_to_class(classbits, uchardptr, options, cd, p[0] + 1, + (p[1] == NOTACHAR) ? (utf ? 0x10ffffu : 0xffffffffu) : p[1] - 1); + p++; + } +return n8; +} + + + +/************************************************* +* Compile one branch * +*************************************************/ + +/* Scan the pattern, compiling it into the a vector. If the options are +changed during the branch, the pointer is used to change the external options +bits. This function is used during the pre-compile phase when we are trying +to find out the amount of memory needed, as well as during the real compile +phase. The value of lengthptr distinguishes the two phases. + +Arguments: + optionsptr pointer to the option bits + codeptr points to the pointer to the current code point + ptrptr points to the current pattern pointer + errorcodeptr points to error code variable + firstcuptr place to put the first required code unit + firstcuflagsptr place to put the first code unit flags, or a negative number + reqcuptr place to put the last required code unit + reqcuflagsptr place to put the last required code unit flags, or a negative number + bcptr points to current branch chain + cond_depth conditional nesting depth + cd contains pointers to tables etc. + lengthptr NULL during the real compile phase + points to length accumulator during pre-compile phase + +Returns: TRUE on success + FALSE, with *errorcodeptr set non-zero on error +*/ + +static BOOL +compile_branch(uint32_t *optionsptr, PCRE2_UCHAR **codeptr, + PCRE2_SPTR *ptrptr, int *errorcodeptr, + uint32_t *firstcuptr, int32_t *firstcuflagsptr, + uint32_t *reqcuptr, int32_t *reqcuflagsptr, + branch_chain *bcptr, int cond_depth, + compile_data *cd, size_t *lengthptr) +{ +int repeat_type, op_type; +int repeat_min = 0, repeat_max = 0; /* To please picky compilers */ +int bravalue = 0; +int greedy_default, greedy_non_default; +uint32_t firstcu, reqcu; +int32_t firstcuflags, reqcuflags; +uint32_t zeroreqcu, zerofirstcu; +int32_t zeroreqcuflags, zerofirstcuflags; +int32_t req_caseopt, reqvary, tempreqvary; +int options = *optionsptr; /* May change dynamically */ +int after_manual_callout = 0; +int escape; +size_t length_prevgroup = 0; +register uint32_t c; +register PCRE2_UCHAR *code = *codeptr; +PCRE2_UCHAR *last_code = code; +PCRE2_UCHAR *orig_code = code; +PCRE2_UCHAR *tempcode; +BOOL inescq = FALSE; +BOOL groupsetfirstcu = FALSE; +PCRE2_SPTR ptr = *ptrptr; +PCRE2_SPTR tempptr; +PCRE2_SPTR nestptr = NULL; +PCRE2_UCHAR *previous = NULL; +PCRE2_UCHAR *previous_callout = NULL; +PCRE2_UCHAR *save_hwm = NULL; +uint8_t classbits[32]; + +/* We can fish out the UTF setting once and for all into a BOOL, but we must +not do this for other options (e.g. PCRE2_EXTENDED) because they may change +dynamically as we process the pattern. */ + +#ifdef SUPPORT_UTF +BOOL utf = (options & PCRE2_UTF) != 0; +#if PCRE2_CODE_UNIT_WIDTH != 32 +#define MAYBE_UTF_MULTI /* UTF chars may use multiple code units */ +PCRE2_UCHAR utf_units[6]; /* For setting up multi-cu chars */ +#endif + +#else /* No UTF support */ +BOOL utf = FALSE; +#endif + +/* Helper variables for OP_XCLASS opcode (for characters > 255). We define +class_uchardata always so that it can be passed to add_to_class() always, +though it will not be used in non-UTF 8-bit cases. This avoids having to supply +alternative calls for the different cases. */ + +PCRE2_UCHAR *class_uchardata; +#ifdef SUPPORT_WIDE_CHARS +BOOL xclass; +PCRE2_UCHAR *class_uchardata_base; +#endif + +/* Set up the default and non-default settings for greediness */ + +greedy_default = ((options & PCRE2_UNGREEDY) != 0); +greedy_non_default = greedy_default ^ 1; + +/* Initialize no first unit, no required unit. REQ_UNSET means "no char +matching encountered yet". It gets changed to REQ_NONE if we hit something that +matches a non-fixed first unit; reqcu just remains unset if we never find one. + +When we hit a repeat whose minimum is zero, we may have to adjust these values +to take the zero repeat into account. This is implemented by setting them to +zerofirstcu and zeroreqcu when such a repeat is encountered. The individual +item types that can be repeated set these backoff variables appropriately. */ + +firstcu = reqcu = zerofirstcu = zeroreqcu = 0; +firstcuflags = reqcuflags = zerofirstcuflags = zeroreqcuflags = REQ_UNSET; + +/* The variable req_caseopt contains either the REQ_CASELESS value or zero, +according to the current setting of the caseless flag. The REQ_CASELESS value +leaves the lower 28 bit empty. It is added into the firstcu or reqcu variables +to record the case status of the value. This is used only for ASCII characters. +*/ + +req_caseopt = ((options & PCRE2_CASELESS) != 0)? REQ_CASELESS:0; + +/* Switch on next character until the end of the branch */ + +for (;; ptr++) + { + BOOL negate_class; + BOOL should_flip_negation; + BOOL possessive_quantifier; + BOOL is_quantifier; + BOOL is_recurse; + BOOL reset_bracount; + int class_has_8bitchar; + int class_one_char; +#ifdef SUPPORT_WIDE_CHARS + BOOL xclass_has_prop; +#endif + int newoptions; + int recno; + int refsign; + int skipbytes; + uint32_t subreqcu, subfirstcu; + int32_t subreqcuflags, subfirstcuflags; + int terminator; + unsigned int mclength; + unsigned int tempbracount; + uint32_t ec; + PCRE2_UCHAR mcbuffer[8]; + + /* Get next character in the pattern */ + + c = *ptr; + + /* If we are at the end of a nested substitution, revert to the outer level + string. Nesting only happens one level deep. */ + + if (c == CHAR_NULL && nestptr != NULL) + { + ptr = nestptr; + nestptr = NULL; + c = *ptr; + } + + /* If we are in the pre-compile phase, accumulate the length used for the + previous cycle of this loop. */ + + if (lengthptr != NULL) + { + if (code > cd->start_workspace + cd->workspace_size - + WORK_SIZE_SAFETY_MARGIN) /* Check for overrun */ + { + *errorcodeptr = ERR52; + goto FAILED; + } + + /* There is at least one situation where code goes backwards: this is the + case of a zero quantifier after a class (e.g. [ab]{0}). At compile time, + the class is simply eliminated. However, it is created first, so we have to + allow memory for it. Therefore, don't ever reduce the length at this point. + */ + + if (code < last_code) code = last_code; + + /* Paranoid check for integer overflow */ + + if (OFLOW_MAX - *lengthptr < (size_t)(code - last_code)) + { + *errorcodeptr = ERR20; + goto FAILED; + } + *lengthptr += code - last_code; + + /* If "previous" is set and it is not at the start of the work space, move + it back to there, in order to avoid filling up the work space. Otherwise, + if "previous" is NULL, reset the current code pointer to the start. */ + + if (previous != NULL) + { + if (previous > orig_code) + { + memmove(orig_code, previous, CU2BYTES(code - previous)); + code -= previous - orig_code; + previous = orig_code; + } + } + else code = orig_code; + + /* Remember where this code item starts so we can pick up the length + next time round. */ + + last_code = code; + } + + /* In the real compile phase, just check the workspace used by the forward + reference list. */ + + else if (cd->hwm > cd->start_workspace + cd->workspace_size - + WORK_SIZE_SAFETY_MARGIN) + { + *errorcodeptr = ERR52; + goto FAILED; + } + + /* If in \Q...\E, check for the end; if not, we have a literal */ + + if (inescq && (c != CHAR_NULL || ptr < cd->end_pattern)) + { + if (c == CHAR_BACKSLASH && ptr[1] == CHAR_E) + { + inescq = FALSE; + ptr++; + continue; + } + else + { + if (previous_callout != NULL) + { + if (lengthptr == NULL) /* Don't attempt in pre-compile phase */ + complete_callout(previous_callout, ptr, cd); + previous_callout = NULL; + } + if ((options & PCRE2_AUTO_CALLOUT) != 0) + { + previous_callout = code; + code = auto_callout(code, ptr, cd); + } + goto NORMAL_CHAR; + } + /* Control does not reach here. */ + } + + /* In extended mode, skip white space and comments. We need a loop in order + to check for more white space and more comments after a comment. */ + + if ((options & PCRE2_EXTENDED) != 0) + { + for (;;) + { + while (MAX_255(c) && (cd->ctypes[c] & ctype_space) != 0) c = *(++ptr); + if (c != CHAR_NUMBER_SIGN) break; + ptr++; + while (*ptr != CHAR_NULL) + { + if (IS_NEWLINE(ptr)) /* For non-fixed-length newline cases, */ + { /* IS_NEWLINE sets cd->nllen. */ + ptr += cd->nllen; + break; + } + ptr++; +#ifdef SUPPORT_UTF + if (utf) FORWARDCHAR(ptr); +#endif + } + c = *ptr; /* Either NULL or the char after a newline */ + } + } + + /* See if the next thing is a quantifier. */ + + is_quantifier = + c == CHAR_ASTERISK || c == CHAR_PLUS || c == CHAR_QUESTION_MARK || + (c == CHAR_LEFT_CURLY_BRACKET && is_counted_repeat(ptr+1)); + + /* Fill in length of a previous callout, except when the next thing is a + quantifier or when processing a property substitution string in UCP mode. */ + + if (!is_quantifier && previous_callout != NULL && nestptr == NULL && + after_manual_callout-- <= 0) + { + if (lengthptr == NULL) /* Don't attempt in pre-compile phase */ + complete_callout(previous_callout, ptr, cd); + previous_callout = NULL; + } + + /* Create auto callout, except for quantifiers, or while processing property + strings that are substituted for \w etc in UCP mode. */ + + if ((options & PCRE2_AUTO_CALLOUT) != 0 && !is_quantifier && nestptr == NULL) + { + previous_callout = code; + code = auto_callout(code, ptr, cd); + } + + /* Process the next pattern item. */ + + switch(c) + { + /* ===================================================================*/ + /* The branch terminates at string end or | or ) */ + + case CHAR_NULL: + if (ptr < cd->end_pattern) goto NORMAL_CHAR; /* Zero data character */ + /* Fall through */ + + case CHAR_VERTICAL_LINE: + case CHAR_RIGHT_PARENTHESIS: + *firstcuptr = firstcu; + *firstcuflagsptr = firstcuflags; + *reqcuptr = reqcu; + *reqcuflagsptr = reqcuflags; + *codeptr = code; + *ptrptr = ptr; + if (lengthptr != NULL) + { + if (OFLOW_MAX - *lengthptr < (size_t)(code - last_code)) + { + *errorcodeptr = ERR20; + goto FAILED; + } + *lengthptr += code - last_code; /* To include callout length */ + } + return TRUE; + + + /* ===================================================================*/ + /* Handle single-character metacharacters. In multiline mode, ^ disables + the setting of any following char as a first character. */ + + case CHAR_CIRCUMFLEX_ACCENT: + previous = NULL; + if ((options & PCRE2_MULTILINE) != 0) + { + if (firstcuflags == REQ_UNSET) firstcuflags = REQ_NONE; + *code++ = OP_CIRCM; + } + else *code++ = OP_CIRC; + break; + + case CHAR_DOLLAR_SIGN: + previous = NULL; + *code++ = ((options & PCRE2_MULTILINE) != 0)? OP_DOLLM : OP_DOLL; + break; + + /* There can never be a first char if '.' is first, whatever happens about + repeats. The value of reqcu doesn't change either. */ + + case CHAR_DOT: + if (firstcuflags == REQ_UNSET) firstcuflags = REQ_NONE; + zerofirstcu = firstcu; + zerofirstcuflags = firstcuflags; + zeroreqcu = reqcu; + zeroreqcuflags = reqcuflags; + previous = code; + *code++ = ((options & PCRE2_DOTALL) != 0)? OP_ALLANY: OP_ANY; + break; + + + /* ===================================================================*/ + /* Character classes. If the included characters are all < 256, we build a + 32-byte bitmap of the permitted characters, except in the special case + where there is only one such character. For negated classes, we build the + map as usual, then invert it at the end. However, we use a different opcode + so that data characters > 255 can be handled correctly. + + If the class contains characters outside the 0-255 range, a different + opcode is compiled. It may optionally have a bit map for characters < 256, + but those above are are explicitly listed afterwards. A flag byte tells + whether the bitmap is present, and whether this is a negated class or not. + + An isolated ']' character is not treated specially, so is just another data + character. In earlier versions of PCRE that used the original API there was + a "JavaScript compatibility mode" in which it gave an error. However, + JavaScript itself has changed in this respect so there is no longer any + need for this special handling. + + In another (POSIX) regex library, the ugly syntax [[:<:]] and [[:>:]] is + used for "start of word" and "end of word". As these are otherwise illegal + sequences, we don't break anything by recognizing them. They are replaced + by \b(?=\w) and \b(?<=\w) respectively. Sequences like [a[:<:]] are + erroneous and are handled by the normal code below. */ + + case CHAR_LEFT_SQUARE_BRACKET: + if (PRIV(strncmp_c8)(ptr+1, STRING_WEIRD_STARTWORD, 6) == 0) + { + nestptr = ptr + 7; + ptr = sub_start_of_word - 1; + continue; + } + + if (PRIV(strncmp_c8)(ptr+1, STRING_WEIRD_ENDWORD, 6) == 0) + { + nestptr = ptr + 7; + ptr = sub_end_of_word - 1; + continue; + } + + /* Handle a real character class. */ + + previous = code; + + /* PCRE supports POSIX class stuff inside a class. Perl gives an error if + they are encountered at the top level, so we'll do that too. */ + + if ((ptr[1] == CHAR_COLON || ptr[1] == CHAR_DOT || + ptr[1] == CHAR_EQUALS_SIGN) && + check_posix_syntax(ptr, &tempptr)) + { + *errorcodeptr = (ptr[1] == CHAR_COLON)? ERR13 : ERR31; + goto FAILED; + } + + /* If the first character is '^', set the negation flag and skip it. Also, + if the first few characters (either before or after ^) are \Q\E or \E we + skip them too. This makes for compatibility with Perl. */ + + negate_class = FALSE; + for (;;) + { + c = *(++ptr); + if (c == CHAR_BACKSLASH) + { + if (ptr[1] == CHAR_E) + ptr++; + else if (PRIV(strncmp_c8)(ptr + 1, STR_Q STR_BACKSLASH STR_E, 3) == 0) + ptr += 3; + else + break; + } + else if (!negate_class && c == CHAR_CIRCUMFLEX_ACCENT) + negate_class = TRUE; + else break; + } + + /* Empty classes are allowed if PCRE2_ALLOW_EMPTY_CLASS is set. Otherwise, + an initial ']' is taken as a data character -- the code below handles + that. When empty classes are allowed, [] must always fail, so generate + OP_FAIL, whereas [^] must match any character, so generate OP_ALLANY. */ + + if (c == CHAR_RIGHT_SQUARE_BRACKET && + (cd->external_options & PCRE2_ALLOW_EMPTY_CLASS) != 0) + { + *code++ = negate_class? OP_ALLANY : OP_FAIL; + if (firstcuflags == REQ_UNSET) firstcuflags = REQ_NONE; + zerofirstcu = firstcu; + zerofirstcuflags = firstcuflags; + break; + } + + /* If a class contains a negative special such as \S, we need to flip the + negation flag at the end, so that support for characters > 255 works + correctly (they are all included in the class). */ + + should_flip_negation = FALSE; + + /* Extended class (xclass) will be used when characters > 255 + might match. */ + +#ifdef SUPPORT_WIDE_CHARS + xclass = FALSE; + class_uchardata = code + LINK_SIZE + 2; /* For XCLASS items */ + class_uchardata_base = class_uchardata; /* Save the start */ +#endif + + /* For optimization purposes, we track some properties of the class: + class_has_8bitchar will be non-zero if the class contains at least one 256 + character with a code point less than 256; class_one_char will be 1 if the + class contains just one character; xclass_has_prop will be TRUE if Unicode + property checks are present in the class. */ + + class_has_8bitchar = 0; + class_one_char = 0; +#ifdef SUPPORT_WIDE_CHARS + xclass_has_prop = FALSE; +#endif + + /* Initialize the 256-bit (32-byte) bit map to all zeros. We build the map + in a temporary bit of memory, in case the class contains fewer than two + 8-bit characters because in that case the compiled code doesn't use the bit + map. */ + + memset(classbits, 0, 32 * sizeof(uint8_t)); + + /* Process characters until ] is reached. As the test is at the end of the + loop, an initial ] is taken as a data character. At the start of the loop, + c contains the first code unit of the character. If it is zero, check for + the end of the pattern, to allow binary zero as data. */ + + for(;;) + { + PCRE2_SPTR oldptr; + + if (c == CHAR_NULL && ptr >= cd->end_pattern) + { + *errorcodeptr = ERR6; /* Missing terminating ']' */ + goto FAILED; + } + +#ifdef SUPPORT_UTF + if (utf && HAS_EXTRALEN(c)) + { /* Braces are required because the */ + GETCHARLEN(c, ptr, ptr); /* macro generates multiple statements */ + } +#endif + +#ifdef SUPPORT_WIDE_CHARS + /* In the pre-compile phase, accumulate the length of any wide characters + and reset the pointer. This is so that very large classes that contain a + zillion wide characters no longer overwrite the work space (which is on + the stack). We have to remember that there was XCLASS data, however. */ + + if (lengthptr != NULL && class_uchardata > class_uchardata_base) + { + xclass = TRUE; + *lengthptr += class_uchardata - class_uchardata_base; + class_uchardata = class_uchardata_base; + } +#endif + + /* Inside \Q...\E everything is literal except \E */ + + if (inescq) + { + if (c == CHAR_BACKSLASH && ptr[1] == CHAR_E) /* If we are at \E */ + { + inescq = FALSE; /* Reset literal state */ + ptr++; /* Skip the 'E' */ + goto CONTINUE_CLASS; /* Carry on with next char */ + } + goto CHECK_RANGE; /* Could be range if \E follows */ + } + + /* Handle POSIX class names. Perl allows a negation extension of the + form [:^name:]. A square bracket that doesn't match the syntax is + treated as a literal. We also recognize the POSIX constructions + [.ch.] and [=ch=] ("collating elements") and fault them, as Perl + 5.6 and 5.8 do. */ + + if (c == CHAR_LEFT_SQUARE_BRACKET && + (ptr[1] == CHAR_COLON || ptr[1] == CHAR_DOT || + ptr[1] == CHAR_EQUALS_SIGN) && check_posix_syntax(ptr, &tempptr)) + { + BOOL local_negate = FALSE; + int posix_class, taboffset, tabopt; + register const uint8_t *cbits = cd->cbits; + uint8_t pbits[32]; + + if (ptr[1] != CHAR_COLON) + { + *errorcodeptr = ERR31; + goto FAILED; + } + + ptr += 2; + if (*ptr == CHAR_CIRCUMFLEX_ACCENT) + { + local_negate = TRUE; + should_flip_negation = TRUE; /* Note negative special */ + ptr++; + } + + posix_class = check_posix_name(ptr, (int)(tempptr - ptr)); + if (posix_class < 0) + { + *errorcodeptr = ERR30; + goto FAILED; + } + + /* If matching is caseless, upper and lower are converted to + alpha. This relies on the fact that the class table starts with + alpha, lower, upper as the first 3 entries. */ + + if ((options & PCRE2_CASELESS) != 0 && posix_class <= 2) + posix_class = 0; + + /* When PCRE2_UCP is set, some of the POSIX classes are converted to + different escape sequences that use Unicode properties \p or \P. Others + that are not available via \p or \P generate XCL_PROP/XCL_NOTPROP + directly. UCP support is not available unless UTF support is.*/ + +#ifdef SUPPORT_UTF + if ((options & PCRE2_UCP) != 0) + { + unsigned int ptype = 0; + int pc = posix_class + ((local_negate)? POSIX_SUBSIZE/2 : 0); + + /* The posix_substitutes table specifies which POSIX classes can be + converted to \p or \P items. */ + + if (posix_substitutes[pc] != NULL) + { + nestptr = tempptr + 1; + ptr = posix_substitutes[pc] - 1; + goto CONTINUE_CLASS; + } + + /* There are three other classes that generate special property calls + that are recognized only in an XCLASS. */ + + else switch(posix_class) + { + case PC_GRAPH: + ptype = PT_PXGRAPH; + /* Fall through */ + case PC_PRINT: + if (ptype == 0) ptype = PT_PXPRINT; + /* Fall through */ + case PC_PUNCT: + if (ptype == 0) ptype = PT_PXPUNCT; + *class_uchardata++ = local_negate? XCL_NOTPROP : XCL_PROP; + *class_uchardata++ = ptype; + *class_uchardata++ = 0; + xclass_has_prop = TRUE; + ptr = tempptr + 1; + goto CONTINUE_CLASS; + + /* For all other POSIX classes, no special action is taken in UCP + mode. Fall through to the non_UCP case. */ + + default: + break; + } + } +#endif /* SUPPORT_UTF */ + + /* In the non-UCP case, or when UCP makes no difference, we build the + bit map for the POSIX class in a chunk of local store because we may be + adding and subtracting from it, and we don't want to subtract bits that + may be in the main map already. At the end we or the result into the + bit map that is being built. */ + + posix_class *= 3; + + /* Copy in the first table (always present) */ + + memcpy(pbits, cbits + posix_class_maps[posix_class], + 32 * sizeof(uint8_t)); + + /* If there is a second table, add or remove it as required. */ + + taboffset = posix_class_maps[posix_class + 1]; + tabopt = posix_class_maps[posix_class + 2]; + + if (taboffset >= 0) + { + if (tabopt >= 0) + for (c = 0; c < 32; c++) pbits[c] |= cbits[c + taboffset]; + else + for (c = 0; c < 32; c++) pbits[c] &= ~cbits[c + taboffset]; + } + + /* Now see if we need to remove any special characters. An option + value of 1 removes vertical space and 2 removes underscore. */ + + if (tabopt < 0) tabopt = -tabopt; + if (tabopt == 1) pbits[1] &= ~0x3c; + else if (tabopt == 2) pbits[11] &= 0x7f; + + /* Add the POSIX table or its complement into the main table that is + being built and we are done. */ + + if (local_negate) + for (c = 0; c < 32; c++) classbits[c] |= ~pbits[c]; + else + for (c = 0; c < 32; c++) classbits[c] |= pbits[c]; + + ptr = tempptr + 1; + /* Every class contains at least one < 256 character. */ + class_has_8bitchar = 1; + /* Every class contains at least two characters. */ + class_one_char = 2; + goto CONTINUE_CLASS; /* End of POSIX syntax handling */ + } + + /* Backslash may introduce a single character, or it may introduce one + of the specials, which just set a flag. The sequence \b is a special + case. Inside a class (and only there) it is treated as backspace. We + assume that other escapes have more than one character in them, so + speculatively set both class_has_8bitchar and class_one_char bigger + than one. Unrecognized escapes fall through and are faulted. */ + + if (c == CHAR_BACKSLASH) + { + escape = check_escape(&ptr, &ec, errorcodeptr, options, TRUE, cd); + if (*errorcodeptr != 0) goto FAILED; + if (escape == 0) c = ec; /* Escaped single char */ + else if (escape == ESC_b) c = CHAR_BS; /* \b is backspace in a class */ + else if (escape == ESC_N) /* \N is not supported in a class */ + { + *errorcodeptr = ERR71; + goto FAILED; + } + else if (escape == ESC_Q) /* Handle start of quoted string */ + { + if (ptr[1] == CHAR_BACKSLASH && ptr[2] == CHAR_E) + { + ptr += 2; /* avoid empty string */ + } + else inescq = TRUE; + goto CONTINUE_CLASS; + } + else if (escape == ESC_E) goto CONTINUE_CLASS; /* Ignore orphan \E */ + + else /* Handle \d-type escapes */ + { + register const uint8_t *cbits = cd->cbits; + /* Every class contains at least two < 256 characters. */ + class_has_8bitchar++; + /* Every class contains at least two characters. */ + class_one_char += 2; + + switch (escape) + { +#ifdef SUPPORT_UTF + case ESC_du: /* These are the values given for \d etc */ + case ESC_DU: /* when PCRE2_UCP is set. We replace the */ + case ESC_wu: /* escape sequence with an appropriate \p */ + case ESC_WU: /* or \P to test Unicode properties instead */ + case ESC_su: /* of the default ASCII testing. */ + case ESC_SU: + nestptr = ptr; + ptr = substitutes[escape - ESC_DU] - 1; /* Just before substitute */ + class_has_8bitchar--; /* Undo! */ + break; +#endif + case ESC_d: + for (c = 0; c < 32; c++) classbits[c] |= cbits[c+cbit_digit]; + break; + + case ESC_D: + should_flip_negation = TRUE; + for (c = 0; c < 32; c++) classbits[c] |= ~cbits[c+cbit_digit]; + break; + + case ESC_w: + for (c = 0; c < 32; c++) classbits[c] |= cbits[c+cbit_word]; + break; + + case ESC_W: + should_flip_negation = TRUE; + for (c = 0; c < 32; c++) classbits[c] |= ~cbits[c+cbit_word]; + break; + + /* Perl 5.004 onwards omitted VT from \s, but restored it at Perl + 5.18. Before PCRE 8.34, we had to preserve the VT bit if it was + previously set by something earlier in the character class. + Luckily, the value of CHAR_VT is 0x0b in both ASCII and EBCDIC, so + we could just adjust the appropriate bit. From PCRE 8.34 we no + longer treat \s and \S specially. */ + + case ESC_s: + for (c = 0; c < 32; c++) classbits[c] |= cbits[c+cbit_space]; + break; + + case ESC_S: + should_flip_negation = TRUE; + for (c = 0; c < 32; c++) classbits[c] |= ~cbits[c+cbit_space]; + break; + + /* The rest apply in both UCP and non-UCP cases. */ + + case ESC_h: + (void)add_list_to_class(classbits, &class_uchardata, options, cd, + PRIV(hspace_list), NOTACHAR); + break; + + case ESC_H: + (void)add_not_list_to_class(classbits, &class_uchardata, options, + cd, PRIV(hspace_list)); + break; + + case ESC_v: + (void)add_list_to_class(classbits, &class_uchardata, options, cd, + PRIV(vspace_list), NOTACHAR); + break; + + case ESC_V: + (void)add_not_list_to_class(classbits, &class_uchardata, options, + cd, PRIV(vspace_list)); + break; + +#ifdef SUPPORT_UTF + case ESC_p: + case ESC_P: + { + BOOL negated; + unsigned int ptype = 0, pdata = 0; + if (!get_ucp(&ptr, &negated, &ptype, &pdata, errorcodeptr, cd)) + goto FAILED; + *class_uchardata++ = ((escape == ESC_p) != negated)? + XCL_PROP : XCL_NOTPROP; + *class_uchardata++ = ptype; + *class_uchardata++ = pdata; + xclass_has_prop = TRUE; + class_has_8bitchar--; /* Undo! */ + } + break; +#endif + /* Unrecognized escapes are faulted. */ + + default: + *errorcodeptr = ERR7; + goto FAILED; + } + + /* Handled \d-type escape */ + + goto CONTINUE_CLASS; + } + + /* Control gets here if the escape just defined a single character. + This is in c and may be greater than 256. */ + + escape = 0; + } /* End of backslash handling */ + + /* A character may be followed by '-' to form a range. However, Perl does + not permit ']' to be the end of the range. A '-' character at the end is + treated as a literal. Perl ignores orphaned \E sequences entirely. The + code for handling \Q and \E is messy. */ + + CHECK_RANGE: + while (ptr[1] == CHAR_BACKSLASH && ptr[2] == CHAR_E) + { + inescq = FALSE; + ptr += 2; + } + oldptr = ptr; + + /* Remember if \r or \n were explicitly used */ + + if (c == CHAR_CR || c == CHAR_NL) cd->external_flags |= PCRE2_HASCRORLF; + + /* Check for range */ + + if (!inescq && ptr[1] == CHAR_MINUS) + { + uint32_t d; + ptr += 2; + while (*ptr == CHAR_BACKSLASH && ptr[1] == CHAR_E) ptr += 2; + + /* If we hit \Q (not followed by \E) at this point, go into escaped + mode. */ + + while (*ptr == CHAR_BACKSLASH && ptr[1] == CHAR_Q) + { + ptr += 2; + if (*ptr == CHAR_BACKSLASH && ptr[1] == CHAR_E) + { ptr += 2; goto CONTINUE_CLASS; } + inescq = TRUE; + break; + } + + /* Minus (hyphen) at the end of a class is treated as a literal, so put + back the pointer and jump to handle the character that preceded it. */ + + if (*ptr == CHAR_NULL || (!inescq && *ptr == CHAR_RIGHT_SQUARE_BRACKET)) + { + ptr = oldptr; + goto CLASS_SINGLE_CHARACTER; + } + + /* Otherwise, we have a potential range; pick up the next character */ + +#ifdef SUPPORT_UTF + if (utf) + { /* Braces are required because the */ + GETCHARLEN(d, ptr, ptr); /* macro generates multiple statements */ + } + else +#endif + d = *ptr; /* Not UTF mode */ + + /* The second part of a range can be a single-character escape + sequence, but not any of the other escapes. Perl treats a hyphen as a + literal in such circumstances. However, in Perl's warning mode, a + warning is given, so PCRE now faults it as it is almost certainly a + mistake on the user's part. */ + + if (!inescq) + { + if (d == CHAR_BACKSLASH) + { + int descape; + descape = check_escape(&ptr, &d, errorcodeptr, options, TRUE, cd); + if (*errorcodeptr != 0) goto FAILED; + + /* 0 means a character was put into d; \b is backspace; any other + special causes an error. */ + + if (descape != 0) + { + if (descape == ESC_b) d = CHAR_BS; else + { + *errorcodeptr = ERR83; + goto FAILED; + } + } + } + + /* A hyphen followed by a POSIX class is treated in the same way. */ + + else if (d == CHAR_LEFT_SQUARE_BRACKET && + (ptr[1] == CHAR_COLON || ptr[1] == CHAR_DOT || + ptr[1] == CHAR_EQUALS_SIGN) && + check_posix_syntax(ptr, &tempptr)) + { + *errorcodeptr = ERR83; + goto FAILED; + } + } + + /* Check that the two values are in the correct order. Optimize + one-character ranges. */ + + if (d < c) + { + *errorcodeptr = ERR8; + goto FAILED; + } + if (d == c) goto CLASS_SINGLE_CHARACTER; /* A few lines below */ + + /* We have found a character range, so single character optimizations + cannot be done anymore. Any value greater than 1 indicates that there + is more than one character. */ + + class_one_char = 2; + + /* Remember an explicit \r or \n, and add the range to the class. */ + + if (d == CHAR_CR || d == CHAR_NL) cd->external_flags |= PCRE2_HASCRORLF; + + class_has_8bitchar += + add_to_class(classbits, &class_uchardata, options, cd, c, d); + + goto CONTINUE_CLASS; /* Go get the next char in the class */ + } + + /* Handle a single character - we can get here for a normal non-escape + char, or after \ that introduces a single character or for an apparent + range that isn't. Only the value 1 matters for class_one_char, so don't + increase it if it is already 2 or more ... just in case there's a class + with a zillion characters in it. */ + + CLASS_SINGLE_CHARACTER: + if (class_one_char < 2) class_one_char++; + + /* If class_one_char is 1, we have the first single character in the + class, and there have been no prior ranges, or XCLASS items generated by + escapes. If this is the final character in the class, we can optimize by + turning the item into a 1-character OP_CHAR[I] if it's positive, or + OP_NOT[I] if it's negative. In the positive case, it can cause firstcu + to be set. Otherwise, there can be no first char if this item is first, + whatever repeat count may follow. In the case of reqcu, save the + previous value for reinstating. */ + + if (class_one_char == 1 && ptr[1] == CHAR_RIGHT_SQUARE_BRACKET) + { + ptr++; + zeroreqcu = reqcu; + zeroreqcuflags = reqcuflags; + + if (negate_class) + { +#ifdef SUPPORT_UTF + int d; +#endif + if (firstcuflags == REQ_UNSET) firstcuflags = REQ_NONE; + zerofirstcu = firstcu; + zerofirstcuflags = firstcuflags; + + /* For caseless UTF mode, check whether this character has more than + one other case. If so, generate a special OP_NOTPROP item instead of + OP_NOTI. */ + +#ifdef SUPPORT_UTF + if (utf && (options & PCRE2_CASELESS) != 0 && + (d = UCD_CASESET(c)) != 0) + { + *code++ = OP_NOTPROP; + *code++ = PT_CLIST; + *code++ = d; + } + else +#endif + /* Char has only one other case, or UCP not available */ + + { + *code++ = ((options & PCRE2_CASELESS) != 0)? OP_NOTI: OP_NOT; + code += PUTCHAR(c, code); + } + + /* We are finished with this character class */ + + goto END_CLASS; + } + + /* For a single, positive character, get the value into mcbuffer, and + then we can handle this with the normal one-character code. */ + + mclength = PUTCHAR(c, mcbuffer); + goto ONE_CHAR; + } /* End of 1-char optimization */ + + /* There is more than one character in the class, or an XCLASS item + has been generated. Add this character to the class. */ + + class_has_8bitchar += + add_to_class(classbits, &class_uchardata, options, cd, c, c); + + /* Continue to the next character in the class. Closing square bracket + not within \Q..\E ends the class. A NULL character terminates a + nested substitution string, but may be a data character in the main + pattern (tested at the start of this loop). */ + + CONTINUE_CLASS: + c = *(++ptr); + if (c == CHAR_RIGHT_SQUARE_BRACKET && !inescq) break; + if (c == 0 && nestptr != NULL) + { + ptr = nestptr; + nestptr = NULL; + c = *(++ptr); + } + } /* End of main class-processing loop */ + + /* We will need an XCLASS if data has been placed in class_uchardata. In + the second phase this is a sufficient test. However, in the pre-compile + phase, class_uchardata gets emptied to prevent workspace overflow, so it + only if the very last character in the class needs XCLASS will it contain + anything at this point. For this reason, xclass gets set TRUE above when + class_uchardata is emptied, and that's why this code is the way it is here + instead of just doing a test on class_uchardata below. */ + +#ifdef SUPPORT_WIDE_CHARS + if (class_uchardata > class_uchardata_base) xclass = TRUE; +#endif + + /* If this is the first thing in the branch, there can be no first char + setting, whatever the repeat count. Any reqcu setting must remain + unchanged after any kind of repeat. */ + + if (firstcuflags == REQ_UNSET) firstcuflags = REQ_NONE; + zerofirstcu = firstcu; + zerofirstcuflags = firstcuflags; + zeroreqcu = reqcu; + zeroreqcuflags = reqcuflags; + + /* If there are characters with values > 255, we have to compile an + extended class, with its own opcode, unless there was a negated special + such as \S in the class, and PCRE2_UCP is not set, because in that case all + characters > 255 are in the class, so any that were explicitly given as + well can be ignored. If (when there are explicit characters > 255 that must + be listed) there are no characters < 256, we can omit the bitmap in the + actual compiled code. */ + +#ifdef SUPPORT_UTF + if (xclass && (!should_flip_negation || (options & PCRE2_UCP) != 0)) +#elif PCRE2_CODE_UNIT_WIDTH != 8 + if (xclass && !should_flip_negation) +#endif +#ifdef SUPPORT_WIDE_CHARS + { + *class_uchardata++ = XCL_END; /* Marks the end of extra data */ + *code++ = OP_XCLASS; + code += LINK_SIZE; + *code = negate_class? XCL_NOT:0; + if (xclass_has_prop) *code |= XCL_HASPROP; + + /* If the map is required, move up the extra data to make room for it; + otherwise just move the code pointer to the end of the extra data. */ + + if (class_has_8bitchar > 0) + { + *code++ |= XCL_MAP; + memmove(code + (32 / sizeof(PCRE2_UCHAR)), code, + CU2BYTES(class_uchardata - code)); + if (negate_class && !xclass_has_prop) + for (c = 0; c < 32; c++) classbits[c] = ~classbits[c]; + memcpy(code, classbits, 32); + code = class_uchardata + (32 / sizeof(PCRE2_UCHAR)); + } + else code = class_uchardata; + + /* Now fill in the complete length of the item */ + + PUT(previous, 1, (int)(code - previous)); + break; /* End of class handling */ + } +#endif + + /* If there are no characters > 255, or they are all to be included or + excluded, set the opcode to OP_CLASS or OP_NCLASS, depending on whether the + whole class was negated and whether there were negative specials such as \S + (non-UCP) in the class. Then copy the 32-byte map into the code vector, + negating it if necessary. */ + + *code++ = (negate_class == should_flip_negation) ? OP_CLASS : OP_NCLASS; + if (lengthptr == NULL) /* Save time in the pre-compile phase */ + { + if (negate_class) + for (c = 0; c < 32; c++) classbits[c] = ~classbits[c]; + memcpy(code, classbits, 32); + } + code += 32 / sizeof(PCRE2_UCHAR); + + END_CLASS: + break; + + + /* ===================================================================*/ + /* Various kinds of repeat; '{' is not necessarily a quantifier, but this + has been tested above. */ + + case CHAR_LEFT_CURLY_BRACKET: + if (!is_quantifier) goto NORMAL_CHAR; + ptr = read_repeat_counts(ptr+1, &repeat_min, &repeat_max, errorcodeptr); + if (*errorcodeptr != 0) goto FAILED; + goto REPEAT; + + case CHAR_ASTERISK: + repeat_min = 0; + repeat_max = -1; + goto REPEAT; + + case CHAR_PLUS: + repeat_min = 1; + repeat_max = -1; + goto REPEAT; + + case CHAR_QUESTION_MARK: + repeat_min = 0; + repeat_max = 1; + + REPEAT: + if (previous == NULL) + { + *errorcodeptr = ERR9; + goto FAILED; + } + + if (repeat_min == 0) + { + firstcu = zerofirstcu; /* Adjust for zero repeat */ + firstcuflags = zerofirstcuflags; + reqcu = zeroreqcu; /* Ditto */ + reqcuflags = zeroreqcuflags; + } + + /* Remember whether this is a variable length repeat */ + + reqvary = (repeat_min == repeat_max)? 0 : REQ_VARY; + + op_type = 0; /* Default single-char op codes */ + possessive_quantifier = FALSE; /* Default not possessive quantifier */ + + /* Save start of previous item, in case we have to move it up in order to + insert something before it. */ + + tempcode = previous; + + /* Before checking for a possessive quantifier, we must skip over + whitespace and comments in extended mode because Perl allows white space at + this point. */ + + if ((options & PCRE2_EXTENDED) != 0) + { + PCRE2_SPTR p = ptr + 1; + for (;;) + { + while (MAX_255(*p) && (cd->ctypes[*p] & ctype_space) != 0) p++; + if (*p != CHAR_NUMBER_SIGN) break; + p++; + while (*p != CHAR_NULL) + { + if (IS_NEWLINE(p)) /* For non-fixed-length newline cases, */ + { /* IS_NEWLINE sets cd->nllen. */ + p += cd->nllen; + break; + } + p++; +#ifdef SUPPORT_UTF + if (utf) FORWARDCHAR(p); +#endif + } /* Loop for comment characters */ + } /* Loop for multiple comments */ + ptr = p - 1; /* Character before the next significant one. */ + } + + /* If the next character is '+', we have a possessive quantifier. This + implies greediness, whatever the setting of the PCRE2_UNGREEDY option. + If the next character is '?' this is a minimizing repeat, by default, + but if PCRE2_UNGREEDY is set, it works the other way round. We change the + repeat type to the non-default. */ + + if (ptr[1] == CHAR_PLUS) + { + repeat_type = 0; /* Force greedy */ + possessive_quantifier = TRUE; + ptr++; + } + else if (ptr[1] == CHAR_QUESTION_MARK) + { + repeat_type = greedy_non_default; + ptr++; + } + else repeat_type = greedy_default; + + /* If previous was a recursion call, wrap it in atomic brackets so that + previous becomes the atomic group. All recursions were so wrapped in the + past, but it no longer happens for non-repeated recursions. In fact, the + repeated ones could be re-implemented independently so as not to need this, + but for the moment we rely on the code for repeating groups. */ + + if (*previous == OP_RECURSE) + { + memmove(previous + 1 + LINK_SIZE, previous, CU2BYTES(1 + LINK_SIZE)); + *previous = OP_ONCE; + PUT(previous, 1, 2 + 2*LINK_SIZE); + previous[2 + 2*LINK_SIZE] = OP_KET; + PUT(previous, 3 + 2*LINK_SIZE, 2 + 2*LINK_SIZE); + code += 2 + 2 * LINK_SIZE; + length_prevgroup = 3 + 3*LINK_SIZE; + + /* When actually compiling, we need to check whether this was a forward + reference, and if so, adjust the offset. */ + + if (lengthptr == NULL && cd->hwm >= cd->start_workspace + LINK_SIZE) + { + int offset = GET(cd->hwm, -LINK_SIZE); + if (offset == previous + 1 - cd->start_code) + PUT(cd->hwm, -LINK_SIZE, offset + 1 + LINK_SIZE); + } + } + + /* Now handle repetition for the different types of item. */ + + /* If previous was a character or negated character match, abolish the item + and generate a repeat item instead. If a char item has a minimum of more + than one, ensure that it is set in reqcu - it might not be if a sequence + such as x{3} is the first thing in a branch because the x will have gone + into firstcu instead. */ + + if (*previous == OP_CHAR || *previous == OP_CHARI + || *previous == OP_NOT || *previous == OP_NOTI) + { + switch (*previous) + { + default: /* Make compiler happy. */ + case OP_CHAR: op_type = OP_STAR - OP_STAR; break; + case OP_CHARI: op_type = OP_STARI - OP_STAR; break; + case OP_NOT: op_type = OP_NOTSTAR - OP_STAR; break; + case OP_NOTI: op_type = OP_NOTSTARI - OP_STAR; break; + } + + /* Deal with UTF characters that take up more than one code unit. It's + easier to write this out separately than try to macrify it. Use c to + hold the length of the character in code units, plus UTF_LENGTH to flag + that it's a length rather than a small character. */ + +#ifdef MAYBE_UTF_MULTI + if (utf && NOT_FIRSTCHAR(code[-1])) + { + PCRE2_UCHAR *lastchar = code - 1; + BACKCHAR(lastchar); + c = (int)(code - lastchar); /* Length of UTF character */ + memcpy(utf_units, lastchar, CU2BYTES(c)); /* Save the char */ + c |= UTF_LENGTH; /* Flag c as a length */ + } + else +#endif /* SUPPORT_UTF */ + + /* Handle the case of a single charater - either with no UTF support, or + with UTF disabled, or for a single-code-unit UTF character. */ + { + c = code[-1]; + if (*previous <= OP_CHARI && repeat_min > 1) + { + reqcu = c; + reqcuflags = req_caseopt | cd->req_varyopt; + } + } + + goto OUTPUT_SINGLE_REPEAT; /* Code shared with single character types */ + } + + /* If previous was a character type match (\d or similar), abolish it and + create a suitable repeat item. The code is shared with single-character + repeats by setting op_type to add a suitable offset into repeat_type. Note + the the Unicode property types will be present only when SUPPORT_UTF is + defined, but we don't wrap the little bits of code here because it just + makes it horribly messy. */ + + else if (*previous < OP_EODN) + { + PCRE2_UCHAR *oldcode; + int prop_type, prop_value; + op_type = OP_TYPESTAR - OP_STAR; /* Use type opcodes */ + c = *previous; /* Save previous opcode */ + if (c == OP_PROP || c == OP_NOTPROP) + { + prop_type = previous[1]; + prop_value = previous[2]; + } + else + { + /* Come here from just above with a character in c */ + OUTPUT_SINGLE_REPEAT: + prop_type = prop_value = -1; + } + + /* At this point we either have prop_type == prop_value == -1 and either + a code point or a character type that is not OP_[NOT]PROP in c, or we + have OP_[NOT]PROP in c and prop_type/prop_value not negative. */ + + oldcode = code; /* Save where we were */ + code = previous; /* Usually overwrite previous item */ + + /* If the maximum is zero then the minimum must also be zero; Perl allows + this case, so we do too - by simply omitting the item altogether. */ + + if (repeat_max == 0) goto END_REPEAT; + + /* Combine the op_type with the repeat_type */ + + repeat_type += op_type; + + /* A minimum of zero is handled either as the special case * or ?, or as + an UPTO, with the maximum given. */ + + if (repeat_min == 0) + { + if (repeat_max == -1) *code++ = OP_STAR + repeat_type; + else if (repeat_max == 1) *code++ = OP_QUERY + repeat_type; + else + { + *code++ = OP_UPTO + repeat_type; + PUT2INC(code, 0, repeat_max); + } + } + + /* A repeat minimum of 1 is optimized into some special cases. If the + maximum is unlimited, we use OP_PLUS. Otherwise, the original item is + left in place and, if the maximum is greater than 1, we use OP_UPTO with + one less than the maximum. */ + + else if (repeat_min == 1) + { + if (repeat_max == -1) + *code++ = OP_PLUS + repeat_type; + else + { + code = oldcode; /* Leave previous item in place */ + if (repeat_max == 1) goto END_REPEAT; + *code++ = OP_UPTO + repeat_type; + PUT2INC(code, 0, repeat_max - 1); + } + } + + /* The case {n,n} is just an EXACT, while the general case {n,m} is + handled as an EXACT followed by an UPTO or STAR or QUERY. */ + + else + { + *code++ = OP_EXACT + op_type; /* NB EXACT doesn't have repeat_type */ + PUT2INC(code, 0, repeat_min); + + /* Unless repeat_max equals repeat_min, fill in the data for EXACT, and + then generate the second opcode. In UTF mode, multi-code-unit + characters have their length in c, with the UTF_LENGTH bit as a flag, + and the code units in utf_units. For a repeated Unicode property match, + there are two extra values that define the required property, and c + never has the UTF_LENGTH bit set. */ + + if (repeat_max != repeat_min) + { +#ifdef MAYBE_UTF_MULTI + if (utf && (c & UTF_LENGTH) != 0) + { + memcpy(code, utf_units, CU2BYTES(c & 7)); + code += c & 7; + } + else +#endif + { + *code++ = c; + if (prop_type >= 0) + { + *code++ = prop_type; + *code++ = prop_value; + } + } + + /* Now set up the following opcode */ + + if (repeat_max < 0) *code++ = OP_STAR + repeat_type; else + { + repeat_max -= repeat_min; + if (repeat_max == 1) + { + *code++ = OP_QUERY + repeat_type; + } + else + { + *code++ = OP_UPTO + repeat_type; + PUT2INC(code, 0, repeat_max); + } + } + } + } + + /* Fill in the character or character type for the final opcode. */ + +#ifdef MAYBE_UTF_MULTI + if (utf && (c & UTF_LENGTH) != 0) + { + memcpy(code, utf_units, CU2BYTES(c & 7)); + code += c & 7; + } + else +#endif + { + *code++ = c; + if (prop_type >= 0) + { + *code++ = prop_type; + *code++ = prop_value; + } + } + } + + /* If previous was a character class or a back reference, we put the repeat + stuff after it, but just skip the item if the repeat was {0,0}. */ + + else if (*previous == OP_CLASS || *previous == OP_NCLASS || +#ifdef SUPPORT_WIDE_CHARS + *previous == OP_XCLASS || +#endif + *previous == OP_REF || *previous == OP_REFI || + *previous == OP_DNREF || *previous == OP_DNREFI) + { + if (repeat_max == 0) + { + code = previous; + goto END_REPEAT; + } + + if (repeat_min == 0 && repeat_max == -1) + *code++ = OP_CRSTAR + repeat_type; + else if (repeat_min == 1 && repeat_max == -1) + *code++ = OP_CRPLUS + repeat_type; + else if (repeat_min == 0 && repeat_max == 1) + *code++ = OP_CRQUERY + repeat_type; + else + { + *code++ = OP_CRRANGE + repeat_type; + PUT2INC(code, 0, repeat_min); + if (repeat_max == -1) repeat_max = 0; /* 2-byte encoding for max */ + PUT2INC(code, 0, repeat_max); + } + } + + /* If previous was a bracket group, we may have to replicate it in certain + cases. Note that at this point we can encounter only the "basic" bracket + opcodes such as BRA and CBRA, as this is the place where they get converted + into the more special varieties such as BRAPOS and SBRA. A test for >= + OP_ASSERT and <= OP_COND includes ASSERT, ASSERT_NOT, ASSERTBACK, + ASSERTBACK_NOT, ONCE, ONCE_NC, BRA, BRAPOS, CBRA, CBRAPOS, and COND. + Originally, PCRE did not allow repetition of assertions, but now it does, + for Perl compatibility. */ + + else if (*previous >= OP_ASSERT && *previous <= OP_COND) + { + register int i; + int len = (int)(code - previous); + PCRE2_UCHAR *bralink = NULL; + PCRE2_UCHAR *brazeroptr = NULL; + + /* Repeating a DEFINE group is pointless, but Perl allows the syntax, so + we just ignore the repeat. */ + + if (*previous == OP_COND && previous[LINK_SIZE+1] == OP_DEF) + goto END_REPEAT; + + /* There is no sense in actually repeating assertions. The only potential + use of repetition is in cases when the assertion is optional. Therefore, + if the minimum is greater than zero, just ignore the repeat. If the + maximum is not zero or one, set it to 1. */ + + if (*previous < OP_ONCE) /* Assertion */ + { + if (repeat_min > 0) goto END_REPEAT; + if (repeat_max < 0 || repeat_max > 1) repeat_max = 1; + } + + /* The case of a zero minimum is special because of the need to stick + OP_BRAZERO in front of it, and because the group appears once in the + data, whereas in other cases it appears the minimum number of times. For + this reason, it is simplest to treat this case separately, as otherwise + the code gets far too messy. There are several special subcases when the + minimum is zero. */ + + if (repeat_min == 0) + { + /* If the maximum is also zero, we used to just omit the group from the + output altogether, like this: + + ** if (repeat_max == 0) + ** { + ** code = previous; + ** goto END_REPEAT; + ** } + + However, that fails when a group or a subgroup within it is referenced + as a subroutine from elsewhere in the pattern, so now we stick in + OP_SKIPZERO in front of it so that it is skipped on execution. As we + don't have a list of which groups are referenced, we cannot do this + selectively. + + If the maximum is 1 or unlimited, we just have to stick in the BRAZERO + and do no more at this point. However, we do need to adjust any + OP_RECURSE calls inside the group that refer to the group itself or any + internal or forward referenced group, because the offset is from the + start of the whole regex. Temporarily terminate the pattern while doing + this. */ + + if (repeat_max <= 1) /* Covers 0, 1, and unlimited */ + { + *code = OP_END; + adjust_recurse(previous, 1, utf, cd, save_hwm); + memmove(previous + 1, previous, CU2BYTES(len)); + code++; + if (repeat_max == 0) + { + *previous++ = OP_SKIPZERO; + goto END_REPEAT; + } + brazeroptr = previous; /* Save for possessive optimizing */ + *previous++ = OP_BRAZERO + repeat_type; + } + + /* If the maximum is greater than 1 and limited, we have to replicate + in a nested fashion, sticking OP_BRAZERO before each set of brackets. + The first one has to be handled carefully because it's the original + copy, which has to be moved up. The remainder can be handled by code + that is common with the non-zero minimum case below. We have to + adjust the value or repeat_max, since one less copy is required. Once + again, we may have to adjust any OP_RECURSE calls inside the group. */ + + else + { + int offset; + *code = OP_END; + adjust_recurse(previous, 2 + LINK_SIZE, utf, cd, save_hwm); + memmove(previous + 2 + LINK_SIZE, previous, CU2BYTES(len)); + code += 2 + LINK_SIZE; + *previous++ = OP_BRAZERO + repeat_type; + *previous++ = OP_BRA; + + /* We chain together the bracket offset fields that have to be + filled in later when the ends of the brackets are reached. */ + + offset = (bralink == NULL)? 0 : (int)(previous - bralink); + bralink = previous; + PUTINC(previous, 0, offset); + } + + repeat_max--; + } + + /* If the minimum is greater than zero, replicate the group as many + times as necessary, and adjust the maximum to the number of subsequent + copies that we need. If we set a first char from the group, and didn't + set a required char, copy the latter from the former. If there are any + forward reference subroutine calls in the group, there will be entries on + the workspace list; replicate these with an appropriate increment. */ + + else + { + if (repeat_min > 1) + { + /* In the pre-compile phase, we don't actually do the replication. We + just adjust the length as if we had. Do some paranoid checks for + potential integer overflow. The INT64_OR_DOUBLE type is a 64-bit + integer type when available, otherwise double. */ + + if (lengthptr != NULL) + { + size_t delta = (repeat_min - 1)*length_prevgroup; + if ((INT64_OR_DOUBLE)(repeat_min - 1)* + (INT64_OR_DOUBLE)length_prevgroup > + (INT64_OR_DOUBLE)INT_MAX || + OFLOW_MAX - *lengthptr < delta) + { + *errorcodeptr = ERR20; + goto FAILED; + } + *lengthptr += delta; + } + + /* This is compiling for real. If there is a set first byte for + the group, and we have not yet set a "required byte", set it. Make + sure there is enough workspace for copying forward references before + doing the copy. */ + + else + { + if (groupsetfirstcu && reqcuflags < 0) + { + reqcu = firstcu; + reqcuflags = firstcuflags; + } + + for (i = 1; i < repeat_min; i++) + { + PCRE2_UCHAR *hc; + PCRE2_UCHAR *this_hwm = cd->hwm; + memcpy(code, previous, CU2BYTES(len)); + + while (cd->hwm > cd->start_workspace + cd->workspace_size - + WORK_SIZE_SAFETY_MARGIN - (this_hwm - save_hwm)) + { + size_t save_offset = save_hwm - cd->start_workspace; + size_t this_offset = this_hwm - cd->start_workspace; + *errorcodeptr = expand_workspace(cd); + if (*errorcodeptr != 0) goto FAILED; + save_hwm = (PCRE2_UCHAR *)cd->start_workspace + save_offset; + this_hwm = (PCRE2_UCHAR *)cd->start_workspace + this_offset; + } + + for (hc = save_hwm; hc < this_hwm; hc += LINK_SIZE) + { + PUT(cd->hwm, 0, GET(hc, 0) + len); + cd->hwm += LINK_SIZE; + } + save_hwm = this_hwm; + code += len; + } + } + } + + if (repeat_max > 0) repeat_max -= repeat_min; + } + + /* This code is common to both the zero and non-zero minimum cases. If + the maximum is limited, it replicates the group in a nested fashion, + remembering the bracket starts on a stack. In the case of a zero minimum, + the first one was set up above. In all cases the repeat_max now specifies + the number of additional copies needed. Again, we must remember to + replicate entries on the forward reference list. */ + + if (repeat_max >= 0) + { + /* In the pre-compile phase, we don't actually do the replication. We + just adjust the length as if we had. For each repetition we must add 1 + to the length for BRAZERO and for all but the last repetition we must + add 2 + 2*LINKSIZE to allow for the nesting that occurs. Do some + paranoid checks to avoid integer overflow. The INT64_OR_DOUBLE type is + a 64-bit integer type when available, otherwise double. */ + + if (lengthptr != NULL && repeat_max > 0) + { + size_t delta = repeat_max*(length_prevgroup + 1 + 2 + 2*LINK_SIZE) - + 2 - 2*LINK_SIZE; /* Last one doesn't nest */ + if ((INT64_OR_DOUBLE)repeat_max * + (INT64_OR_DOUBLE)(length_prevgroup + 1 + 2 + 2*LINK_SIZE) + > (INT64_OR_DOUBLE)INT_MAX || + OFLOW_MAX - *lengthptr < delta) + { + *errorcodeptr = ERR20; + goto FAILED; + } + *lengthptr += delta; + } + + /* This is compiling for real */ + + else for (i = repeat_max - 1; i >= 0; i--) + { + PCRE2_UCHAR *hc; + PCRE2_UCHAR *this_hwm = cd->hwm; + + *code++ = OP_BRAZERO + repeat_type; + + /* All but the final copy start a new nesting, maintaining the + chain of brackets outstanding. */ + + if (i != 0) + { + int offset; + *code++ = OP_BRA; + offset = (bralink == NULL)? 0 : (int)(code - bralink); + bralink = code; + PUTINC(code, 0, offset); + } + + memcpy(code, previous, CU2BYTES(len)); + + /* Ensure there is enough workspace for forward references before + copying them. */ + + while (cd->hwm > cd->start_workspace + cd->workspace_size - + WORK_SIZE_SAFETY_MARGIN - (this_hwm - save_hwm)) + { + size_t save_offset = save_hwm - cd->start_workspace; + size_t this_offset = this_hwm - cd->start_workspace; + *errorcodeptr = expand_workspace(cd); + if (*errorcodeptr != 0) goto FAILED; + save_hwm = (PCRE2_UCHAR *)cd->start_workspace + save_offset; + this_hwm = (PCRE2_UCHAR *)cd->start_workspace + this_offset; + } + + for (hc = save_hwm; hc < this_hwm; hc += LINK_SIZE) + { + PUT(cd->hwm, 0, GET(hc, 0) + len + ((i != 0)? 2+LINK_SIZE : 1)); + cd->hwm += LINK_SIZE; + } + save_hwm = this_hwm; + code += len; + } + + /* Now chain through the pending brackets, and fill in their length + fields (which are holding the chain links pro tem). */ + + while (bralink != NULL) + { + int oldlinkoffset; + int offset = (int)(code - bralink + 1); + PCRE2_UCHAR *bra = code - offset; + oldlinkoffset = GET(bra, 1); + bralink = (oldlinkoffset == 0)? NULL : bralink - oldlinkoffset; + *code++ = OP_KET; + PUTINC(code, 0, offset); + PUT(bra, 1, offset); + } + } + + /* If the maximum is unlimited, set a repeater in the final copy. For + ONCE brackets, that's all we need to do. However, possessively repeated + ONCE brackets can be converted into non-capturing brackets, as the + behaviour of (?:xx)++ is the same as (?>xx)++ and this saves having to + deal with possessive ONCEs specially. + + Otherwise, when we are doing the actual compile phase, check to see + whether this group is one that could match an empty string. If so, + convert the initial operator to the S form (e.g. OP_BRA -> OP_SBRA) so + that runtime checking can be done. [This check is also applied to ONCE + groups at runtime, but in a different way.] + + Then, if the quantifier was possessive and the bracket is not a + conditional, we convert the BRA code to the POS form, and the KET code to + KETRPOS. (It turns out to be convenient at runtime to detect this kind of + subpattern at both the start and at the end.) The use of special opcodes + makes it possible to reduce greatly the stack usage in pcre_exec(). If + the group is preceded by OP_BRAZERO, convert this to OP_BRAPOSZERO. + + Then, if the minimum number of matches is 1 or 0, cancel the possessive + flag so that the default action below, of wrapping everything inside + atomic brackets, does not happen. When the minimum is greater than 1, + there will be earlier copies of the group, and so we still have to wrap + the whole thing. */ + + else + { + PCRE2_UCHAR *ketcode = code - 1 - LINK_SIZE; + PCRE2_UCHAR *bracode = ketcode - GET(ketcode, 1); + + /* Convert possessive ONCE brackets to non-capturing */ + + if ((*bracode == OP_ONCE || *bracode == OP_ONCE_NC) && + possessive_quantifier) *bracode = OP_BRA; + + /* For non-possessive ONCE brackets, all we need to do is to + set the KET. */ + + if (*bracode == OP_ONCE || *bracode == OP_ONCE_NC) + *ketcode = OP_KETRMAX + repeat_type; + + /* Handle non-ONCE brackets and possessive ONCEs (which have been + converted to non-capturing above). */ + + else + { + /* In the compile phase, check for empty string matching. */ + + if (lengthptr == NULL) + { + PCRE2_UCHAR *scode = bracode; + do + { + if (could_be_empty_branch(scode, ketcode, utf, cd, NULL)) + { + *bracode += OP_SBRA - OP_BRA; + break; + } + scode += GET(scode, 1); + } + while (*scode == OP_ALT); + } + + /* Handle possessive quantifiers. */ + + if (possessive_quantifier) + { + /* For COND brackets, we wrap the whole thing in a possessively + repeated non-capturing bracket, because we have not invented POS + versions of the COND opcodes. Because we are moving code along, we + must ensure that any pending recursive references are updated. */ + + if (*bracode == OP_COND || *bracode == OP_SCOND) + { + int nlen = (int)(code - bracode); + *code = OP_END; + adjust_recurse(bracode, 1 + LINK_SIZE, utf, cd, save_hwm); + memmove(bracode + 1 + LINK_SIZE, bracode, CU2BYTES(nlen)); + code += 1 + LINK_SIZE; + nlen += 1 + LINK_SIZE; + *bracode = OP_BRAPOS; + *code++ = OP_KETRPOS; + PUTINC(code, 0, nlen); + PUT(bracode, 1, nlen); + } + + /* For non-COND brackets, we modify the BRA code and use KETRPOS. */ + + else + { + *bracode += 1; /* Switch to xxxPOS opcodes */ + *ketcode = OP_KETRPOS; + } + + /* If the minimum is zero, mark it as possessive, then unset the + possessive flag when the minimum is 0 or 1. */ + + if (brazeroptr != NULL) *brazeroptr = OP_BRAPOSZERO; + if (repeat_min < 2) possessive_quantifier = FALSE; + } + + /* Non-possessive quantifier */ + + else *ketcode = OP_KETRMAX + repeat_type; + } + } + } + + /* If previous is OP_FAIL, it was generated by an empty class [] + (PCRE2_ALLOW_EMPTY_CLASS is set). The other ways in which OP_FAIL can be + generated, that is by (*FAIL) or (?!), set previous to NULL, which gives a + "nothing to repeat" error above. We can just ignore the repeat in empty + class case. */ + + else if (*previous == OP_FAIL) goto END_REPEAT; + + /* Else there's some kind of shambles */ + + else + { + *errorcodeptr = ERR11; + goto FAILED; + } + + /* If the character following a repeat is '+', possessive_quantifier is + TRUE. For some opcodes, there are special alternative opcodes for this + case. For anything else, we wrap the entire repeated item inside OP_ONCE + brackets. Logically, the '+' notation is just syntactic sugar, taken from + Sun's Java package, but the special opcodes can optimize it. + + Some (but not all) possessively repeated subpatterns have already been + completely handled in the code just above. For them, possessive_quantifier + is always FALSE at this stage. Note that the repeated item starts at + tempcode, not at previous, which might be the first part of a string whose + (former) last char we repeated. */ + + if (possessive_quantifier) + { + int len; + + /* Possessifying an EXACT quantifier has no effect, so we can ignore it. + However, QUERY, STAR, or UPTO may follow (for quantifiers such as {5,6}, + {5,}, or {5,10}). We skip over an EXACT item; if the length of what + remains is greater than zero, there's a further opcode that can be + handled. If not, do nothing, leaving the EXACT alone. */ + + switch(*tempcode) + { + case OP_TYPEEXACT: + tempcode += PRIV(OP_lengths)[*tempcode] + + ((tempcode[1 + IMM2_SIZE] == OP_PROP + || tempcode[1 + IMM2_SIZE] == OP_NOTPROP)? 2 : 0); + break; + + /* CHAR opcodes are used for exacts whose count is 1. */ + + case OP_CHAR: + case OP_CHARI: + case OP_NOT: + case OP_NOTI: + case OP_EXACT: + case OP_EXACTI: + case OP_NOTEXACT: + case OP_NOTEXACTI: + tempcode += PRIV(OP_lengths)[*tempcode]; +#ifdef SUPPORT_UTF + if (utf && HAS_EXTRALEN(tempcode[-1])) + tempcode += GET_EXTRALEN(tempcode[-1]); +#endif + break; + + /* For the class opcodes, the repeat operator appears at the end; + adjust tempcode to point to it. */ + + case OP_CLASS: + case OP_NCLASS: + tempcode += 1 + 32/sizeof(PCRE2_UCHAR); + break; + +#ifdef SUPPORT_WIDE_CHARS + case OP_XCLASS: + tempcode += GET(tempcode, 1); + break; +#endif + } + + /* If tempcode is equal to code (which points to the end of the repeated + item), it means we have skipped an EXACT item but there is no following + QUERY, STAR, or UPTO; the value of len will be 0, and we do nothing. In + all other cases, tempcode will be pointing to the repeat opcode, and will + be less than code, so the value of len will be greater than 0. */ + + len = (int)(code - tempcode); + if (len > 0) + { + unsigned int repcode = *tempcode; + + /* There is a table for possessifying opcodes, all of which are less + than OP_CALLOUT. A zero entry means there is no possessified version. + */ + + if (repcode < OP_CALLOUT && opcode_possessify[repcode] > 0) + *tempcode = opcode_possessify[repcode]; + + /* For opcode without a special possessified version, wrap the item in + ONCE brackets. Because we are moving code along, we must ensure that + any pending recursive references are updated. */ + + else + { + *code = OP_END; + adjust_recurse(tempcode, 1 + LINK_SIZE, utf, cd, save_hwm); + memmove(tempcode + 1 + LINK_SIZE, tempcode, CU2BYTES(len)); + code += 1 + LINK_SIZE; + len += 1 + LINK_SIZE; + tempcode[0] = OP_ONCE; + *code++ = OP_KET; + PUTINC(code, 0, len); + PUT(tempcode, 1, len); + } + } + } + + /* In all case we no longer have a previous item. We also set the + "follows varying string" flag for subsequently encountered reqcus if + it isn't already set and we have just passed a varying length item. */ + + END_REPEAT: + previous = NULL; + cd->req_varyopt |= reqvary; + break; + + + /* ===================================================================*/ + /* Start of nested parenthesized sub-expression, or comment or lookahead or + lookbehind or option setting or condition or all the other extended + parenthesis forms. */ + + case CHAR_LEFT_PARENTHESIS: + newoptions = options; + skipbytes = 0; + bravalue = OP_CBRA; + save_hwm = cd->hwm; + reset_bracount = FALSE; + + /* First deal with various "verbs" that can be introduced by '*'. */ + + ptr++; + if (ptr[0] == CHAR_ASTERISK && (ptr[1] == ':' + || (MAX_255(ptr[1]) && ((cd->ctypes[ptr[1]] & ctype_letter) != 0)))) + { + int i, namelen; + int arglen = 0; + const char *vn = verbnames; + PCRE2_SPTR name = ptr + 1; + PCRE2_SPTR arg = NULL; + previous = NULL; + ptr++; + while (MAX_255(*ptr) && (cd->ctypes[*ptr] & ctype_letter) != 0) ptr++; + namelen = (int)(ptr - name); + + /* It appears that Perl allows any characters whatsoever, other than + a closing parenthesis, to appear in arguments, so we no longer insist on + letters, digits, and underscores. */ + + if (*ptr == CHAR_COLON) + { + arg = ++ptr; + while (*ptr != CHAR_NULL && *ptr != CHAR_RIGHT_PARENTHESIS) ptr++; + arglen = (int)(ptr - arg); + if ((unsigned int)arglen > MAX_MARK) + { + *errorcodeptr = ERR75; + goto FAILED; + } + } + + if (*ptr != CHAR_RIGHT_PARENTHESIS) + { + *errorcodeptr = ERR60; + goto FAILED; + } + + /* Scan the table of verb names */ + + for (i = 0; i < verbcount; i++) + { + if (namelen == verbs[i].len && + PRIV(strncmp_c8)(name, vn, namelen) == 0) + { + int setverb; + + /* Check for open captures before ACCEPT and convert it to + ASSERT_ACCEPT if in an assertion. */ + + if (verbs[i].op == OP_ACCEPT) + { + open_capitem *oc; + if (arglen != 0) + { + *errorcodeptr = ERR59; + goto FAILED; + } + cd->had_accept = TRUE; + for (oc = cd->open_caps; oc != NULL; oc = oc->next) + { + *code++ = OP_CLOSE; + PUT2INC(code, 0, oc->number); + } + setverb = *code++ = + (cd->assert_depth > 0)? OP_ASSERT_ACCEPT : OP_ACCEPT; + + /* Do not set firstcu after *ACCEPT */ + if (firstcuflags == REQ_UNSET) firstcuflags = REQ_NONE; + } + + /* Handle other cases with/without an argument */ + + else if (arglen == 0) + { + if (verbs[i].op < 0) /* Argument is mandatory */ + { + *errorcodeptr = ERR66; + goto FAILED; + } + setverb = *code++ = verbs[i].op; + } + + else + { + if (verbs[i].op_arg < 0) /* Argument is forbidden */ + { + *errorcodeptr = ERR59; + goto FAILED; + } + setverb = *code++ = verbs[i].op_arg; + *code++ = arglen; + memcpy(code, arg, CU2BYTES(arglen)); + code += arglen; + *code++ = 0; + } + + switch (setverb) + { + case OP_THEN: + case OP_THEN_ARG: + cd->external_flags |= PCRE2_HASTHEN; + break; + + case OP_PRUNE: + case OP_PRUNE_ARG: + case OP_SKIP: + case OP_SKIP_ARG: + cd->had_pruneorskip = TRUE; + break; + } + + break; /* Found verb, exit loop */ + } + + vn += verbs[i].len + 1; + } + + if (i < verbcount) continue; /* Successfully handled a verb */ + *errorcodeptr = ERR60; /* Verb not recognized */ + goto FAILED; + } + + /* Deal with the extended parentheses; all are introduced by '?', and the + appearance of any of them means that this is not a capturing group. */ + + else if (*ptr == CHAR_QUESTION_MARK) + { + int i, set, unset, namelen; + int *optset; + PCRE2_SPTR name; + PCRE2_UCHAR *slot; + + switch (*(++ptr)) + { + case CHAR_NUMBER_SIGN: /* Comment; skip to ket */ + ptr++; + while (*ptr != CHAR_NULL && *ptr != CHAR_RIGHT_PARENTHESIS) ptr++; + if (*ptr == CHAR_NULL) + { + *errorcodeptr = ERR18; + goto FAILED; + } + continue; + + + /* ------------------------------------------------------------ */ + case CHAR_VERTICAL_LINE: /* Reset capture count for each branch */ + reset_bracount = TRUE; + /* Fall through */ + + /* ------------------------------------------------------------ */ + case CHAR_COLON: /* Non-capturing bracket */ + bravalue = OP_BRA; + ptr++; + break; + + + /* ------------------------------------------------------------ */ + case CHAR_LEFT_PARENTHESIS: + bravalue = OP_COND; /* Conditional group */ + tempptr = ptr; + + /* A condition can be an assertion, a number (referring to a numbered + group's having been set), a name (referring to a named group), or 'R', + referring to recursion. R and R&name are also permitted for + recursion tests. + + There are ways of testing a named group: (?(name)) is used by Python; + Perl 5.10 onwards uses (?() or (?('name')). + + There is one unfortunate ambiguity, caused by history. 'R' can be the + recursive thing or the name 'R' (and similarly for 'R' followed by + digits). We look for a name first; if not found, we try the other case. + + For compatibility with auto-callouts, we allow a callout to be + specified before a condition that is an assertion. First, check for the + syntax of a callout; if found, adjust the temporary pointer that is + used to check for an assertion condition. That's all that is needed! */ + + if (ptr[1] == CHAR_QUESTION_MARK && ptr[2] == CHAR_C) + { + for (i = 3;; i++) if (!IS_DIGIT(ptr[i])) break; + if (ptr[i] == CHAR_RIGHT_PARENTHESIS) + tempptr += i + 1; + } + + /* For conditions that are assertions, check the syntax, and then exit + the switch. This will take control down to where bracketed groups, + including assertions, are processed. */ + + if (tempptr[1] == CHAR_QUESTION_MARK && + (tempptr[2] == CHAR_EQUALS_SIGN || + tempptr[2] == CHAR_EXCLAMATION_MARK || + tempptr[2] == CHAR_LESS_THAN_SIGN)) + break; + + /* Other conditions use OP_CREF/OP_DNCREF/OP_RREF/OP_DNRREF, and all + need to skip at least 1+IMM2_SIZE bytes at the start of the group. */ + + code[1+LINK_SIZE] = OP_CREF; + skipbytes = 1+IMM2_SIZE; + refsign = -1; /* => not a number */ + namelen = -1; /* => not a name; must set to avoid warning */ + name = NULL; /* Always set to avoid warning */ + recno = 0; /* Always set to avoid warning */ + + /* Check for a test for recursion in a named group. */ + + ptr++; + if (*ptr == CHAR_R && ptr[1] == CHAR_AMPERSAND) + { + terminator = -1; + ptr += 2; + code[1+LINK_SIZE] = OP_RREF; /* Change the type of test */ + } + + /* Check for a test for a named group's having been set, using the Perl + syntax (?() or (?('name'), and also allow for the original PCRE + syntax of (?(name) or for (?(+n), (?(-n), and just (?(n). */ + + else if (*ptr == CHAR_LESS_THAN_SIGN) + { + terminator = CHAR_GREATER_THAN_SIGN; + ptr++; + } + else if (*ptr == CHAR_APOSTROPHE) + { + terminator = CHAR_APOSTROPHE; + ptr++; + } + else + { + terminator = CHAR_NULL; + if (*ptr == CHAR_MINUS || *ptr == CHAR_PLUS) refsign = *ptr++; + else if (IS_DIGIT(*ptr)) refsign = 0; + } + + /* Handle a number */ + + if (refsign >= 0) + { + while (IS_DIGIT(*ptr)) + { + recno = recno * 10 + (int)(*ptr - CHAR_0); + ptr++; + } + } + + /* Otherwise we expect to read a name; anything else is an error. When + a name is one of a number of duplicates, a different opcode is used and + it needs more memory. Unfortunately we cannot tell whether a name is a + duplicate in the first pass, so we have to allow for more memory. */ + + else + { + if (IS_DIGIT(*ptr)) + { + *errorcodeptr = ERR84; + goto FAILED; + } + if (!MAX_255(*ptr) || (cd->ctypes[*ptr] & ctype_word) == 0) + { + *errorcodeptr = ERR28; /* Assertion expected */ + goto FAILED; + } + name = ptr++; + while (MAX_255(*ptr) && (cd->ctypes[*ptr] & ctype_word) != 0) + { + ptr++; + } + namelen = (int)(ptr - name); + if (lengthptr != NULL && (options & PCRE2_DUPNAMES) != 0) + *lengthptr += IMM2_SIZE; + } + + /* Check the terminator */ + + if ((terminator > 0 && *ptr++ != (PCRE2_UCHAR)terminator) || + *ptr++ != CHAR_RIGHT_PARENTHESIS) + { + ptr--; /* Error offset */ + *errorcodeptr = ERR26; /* Malformed number or name */ + goto FAILED; + } + + /* Do no further checking in the pre-compile phase. */ + + if (lengthptr != NULL) break; + + /* In the real compile we do the work of looking for the actual + reference. If refsign is not negative, it means we have a number in + recno. */ + + if (refsign >= 0) + { + if (recno <= 0) + { + *errorcodeptr = ERR35; + goto FAILED; + } + if (refsign != 0) recno = (refsign == CHAR_MINUS)? + cd->bracount - recno + 1 : recno + cd->bracount; + if (recno <= 0 || recno > cd->final_bracount) + { + *errorcodeptr = ERR15; + goto FAILED; + } + PUT2(code, 2+LINK_SIZE, recno); + break; + } + + /* Otherwise look for the name. */ + + slot = cd->name_table; + for (i = 0; i < cd->names_found; i++) + { + if (PRIV(strncmp)(name, slot+IMM2_SIZE, namelen) == 0) break; + slot += cd->name_entry_size; + } + + /* Found the named subpattern. If the name is duplicated, add one to + the opcode to change CREF/RREF into DNCREF/DNRREF and insert + appropriate data values. Otherwise, just insert the unique subpattern + number. */ + + if (i < cd->names_found) + { + int offset = i++; + int count = 1; + recno = GET2(slot, 0); /* Number from first found */ + for (; i < cd->names_found; i++) + { + slot += cd->name_entry_size; + if (PRIV(strncmp)(name, slot+IMM2_SIZE, namelen) != 0 || + (slot+IMM2_SIZE)[namelen] != 0) break; + count++; + } + + if (count > 1) + { + PUT2(code, 2+LINK_SIZE, offset); + PUT2(code, 2+LINK_SIZE+IMM2_SIZE, count); + skipbytes += IMM2_SIZE; + code[1+LINK_SIZE]++; + } + else /* Not a duplicated name */ + { + PUT2(code, 2+LINK_SIZE, recno); + } + } + + /* If terminator == CHAR_NULL it means that the name followed directly + after the opening parenthesis [e.g. (?(abc)...] and in this case there + are some further alternatives to try. For the cases where terminator != + CHAR_NULL [things like (?(... or (?('name')... or (?(R&name)... ] + we have now checked all the possibilities, so give an error. */ + + else if (terminator != CHAR_NULL) + { + *errorcodeptr = ERR15; + goto FAILED; + } + + /* Check for (?(R) for recursion. Allow digits after R to specify a + specific group number. */ + + else if (*name == CHAR_R) + { + recno = 0; + for (i = 1; i < namelen; i++) + { + if (!IS_DIGIT(name[i])) + { + *errorcodeptr = ERR15; + goto FAILED; + } + recno = recno * 10 + name[i] - CHAR_0; + } + if (recno == 0) recno = RREF_ANY; + code[1+LINK_SIZE] = OP_RREF; /* Change test type */ + PUT2(code, 2+LINK_SIZE, recno); + } + + /* Similarly, check for the (?(DEFINE) "condition", which is always + false. */ + + else if (namelen == 6 && PRIV(strncmp_c8)(name, STRING_DEFINE, 6) == 0) + { + code[1+LINK_SIZE] = OP_DEF; + skipbytes = 1; + } + + /* Reference to an unidentified subpattern. */ + + else + { + *errorcodeptr = ERR15; + goto FAILED; + } + break; + + + /* ------------------------------------------------------------ */ + case CHAR_EQUALS_SIGN: /* Positive lookahead */ + bravalue = OP_ASSERT; + cd->assert_depth += 1; + ptr++; + break; + + /* Optimize (?!) to (*FAIL) unless it is quantified - which is a weird + thing to do, but Perl allows all assertions to be quantified, and when + they contain capturing parentheses there may be a potential use for + this feature. Not that that applies to a quantified (?!) but we allow + it for uniformity. */ + + /* ------------------------------------------------------------ */ + case CHAR_EXCLAMATION_MARK: /* Negative lookahead */ + ptr++; + if (*ptr == CHAR_RIGHT_PARENTHESIS && ptr[1] != CHAR_ASTERISK && + ptr[1] != CHAR_PLUS && ptr[1] != CHAR_QUESTION_MARK && + (ptr[1] != CHAR_LEFT_CURLY_BRACKET || !is_counted_repeat(ptr+2))) + { + *code++ = OP_FAIL; + previous = NULL; + continue; + } + bravalue = OP_ASSERT_NOT; + cd->assert_depth += 1; + break; + + + /* ------------------------------------------------------------ */ + case CHAR_LESS_THAN_SIGN: /* Lookbehind or named define */ + switch (ptr[1]) + { + case CHAR_EQUALS_SIGN: /* Positive lookbehind */ + bravalue = OP_ASSERTBACK; + cd->assert_depth += 1; + ptr += 2; + break; + + case CHAR_EXCLAMATION_MARK: /* Negative lookbehind */ + bravalue = OP_ASSERTBACK_NOT; + cd->assert_depth += 1; + ptr += 2; + break; + + default: /* Could be name define, else bad */ + if (MAX_255(ptr[1]) && (cd->ctypes[ptr[1]] & ctype_word) != 0) + goto DEFINE_NAME; + ptr++; /* Correct offset for error */ + *errorcodeptr = ERR24; + goto FAILED; + } + break; + + + /* ------------------------------------------------------------ */ + case CHAR_GREATER_THAN_SIGN: /* One-time brackets */ + bravalue = OP_ONCE; + ptr++; + break; + + + /* ------------------------------------------------------------ */ + case CHAR_C: /* Callout - may be followed by digits; */ + previous_callout = code; /* Save for later completion */ + after_manual_callout = 1; /* Skip one item before completing */ + *code++ = OP_CALLOUT; + { + int n = 0; + ptr++; + while(IS_DIGIT(*ptr)) + n = n * 10 + *ptr++ - CHAR_0; + if (*ptr != CHAR_RIGHT_PARENTHESIS) + { + *errorcodeptr = ERR39; + goto FAILED; + } + if (n > 255) + { + *errorcodeptr = ERR38; + goto FAILED; + } + *code++ = n; + PUT(code, 0, (int)(ptr - cd->start_pattern + 1)); /* Pattern offset */ + PUT(code, LINK_SIZE, 0); /* Default length */ + code += 2 * LINK_SIZE; + } + previous = NULL; + continue; + + + /* ------------------------------------------------------------ */ + case CHAR_P: /* Python-style named subpattern handling */ + if (*(++ptr) == CHAR_EQUALS_SIGN || + *ptr == CHAR_GREATER_THAN_SIGN) /* Reference or recursion */ + { + is_recurse = *ptr == CHAR_GREATER_THAN_SIGN; + terminator = CHAR_RIGHT_PARENTHESIS; + goto NAMED_REF_OR_RECURSE; + } + else if (*ptr != CHAR_LESS_THAN_SIGN) /* Test for Python-style defn */ + { + *errorcodeptr = ERR41; + goto FAILED; + } + /* Fall through to handle (?P< as (?< is handled */ + + + /* ------------------------------------------------------------ */ + DEFINE_NAME: /* Come here from (?< handling */ + case CHAR_APOSTROPHE: + terminator = (*ptr == CHAR_LESS_THAN_SIGN)? + CHAR_GREATER_THAN_SIGN : CHAR_APOSTROPHE; + name = ++ptr; + if (IS_DIGIT(*ptr)) + { + *errorcodeptr = ERR84; /* Group name must start with non-digit */ + goto FAILED; + } + while (MAX_255(*ptr) && (cd->ctypes[*ptr] & ctype_word) != 0) ptr++; + namelen = (int)(ptr - name); + + /* In the pre-compile phase, do a syntax check, remember the longest + name, and then remember the group in a vector, expanding it if + necessary. Duplicates for the same number are skipped; other duplicates + are checked for validity. In the actual compile, there is nothing to + do. */ + + if (lengthptr != NULL) + { + named_group *ng; + uint32_t number = cd->bracount + 1; + + if (*ptr != (PCRE2_UCHAR)terminator) + { + *errorcodeptr = ERR42; + goto FAILED; + } + + if (cd->names_found >= MAX_NAME_COUNT) + { + *errorcodeptr = ERR49; + goto FAILED; + } + + if (namelen + IMM2_SIZE + 1 > cd->name_entry_size) + { + cd->name_entry_size = namelen + IMM2_SIZE + 1; + if (namelen > MAX_NAME_SIZE) + { + *errorcodeptr = ERR48; + goto FAILED; + } + } + + /* Scan the list to check for duplicates. For duplicate names, if the + number is the same, break the loop, which causes the name to be + discarded; otherwise, if DUPNAMES is not set, give an error. + If it is set, allow the name with a different number, but continue + scanning in case this is a duplicate with the same number. For + non-duplicate names, give an error if the number is duplicated. */ + + ng = cd->named_groups; + for (i = 0; i < cd->names_found; i++, ng++) + { + if (namelen == ng->length && + PRIV(strncmp)(name, ng->name, namelen) == 0) + { + if (ng->number == number) break; + if ((options & PCRE2_DUPNAMES) == 0) + { + *errorcodeptr = ERR43; + goto FAILED; + } + cd->dupnames = TRUE; /* Duplicate names exist */ + } + else if (ng->number == number) + { + *errorcodeptr = ERR65; + goto FAILED; + } + } + + if (i >= cd->names_found) /* Not a duplicate with same number */ + { + /* Increase the list size if necessary */ + + if (cd->names_found >= cd->named_group_list_size) + { + int newsize = cd->named_group_list_size * 2; + named_group *newspace = + cd->cx->memctl.malloc(newsize * sizeof(named_group), + cd->cx->memctl.memory_data); + + if (newspace == NULL) + { + *errorcodeptr = ERR21; + goto FAILED; + } + + memcpy(newspace, cd->named_groups, + cd->named_group_list_size * sizeof(named_group)); + if (cd->named_group_list_size > NAMED_GROUP_LIST_SIZE) + cd->cx->memctl.free((void *)cd->named_groups, + cd->cx->memctl.memory_data); + cd->named_groups = newspace; + cd->named_group_list_size = newsize; + } + + cd->named_groups[cd->names_found].name = name; + cd->named_groups[cd->names_found].length = namelen; + cd->named_groups[cd->names_found].number = number; + cd->names_found++; + } + } + + ptr++; /* Move past > or ' in both passes. */ + goto NUMBERED_GROUP; + + + /* ------------------------------------------------------------ */ + case CHAR_AMPERSAND: /* Perl recursion/subroutine syntax */ + terminator = CHAR_RIGHT_PARENTHESIS; + is_recurse = TRUE; + /* Fall through */ + + /* We come here from the Python syntax above that handles both + references (?P=name) and recursion (?P>name), as well as falling + through from the Perl recursion syntax (?&name). We also come here from + the Perl \k or \k'name' back reference syntax and the \k{name} + .NET syntax, and the Oniguruma \g<...> and \g'...' subroutine syntax. */ + + NAMED_REF_OR_RECURSE: + name = ++ptr; + if (IS_DIGIT(*ptr)) + { + *errorcodeptr = ERR84; /* Group name must start with non-digit */ + goto FAILED; + } + while (MAX_255(*ptr) && (cd->ctypes[*ptr] & ctype_word) != 0) ptr++; + namelen = (int)(ptr - name); + + /* In the pre-compile phase, do a syntax check. We used to just set + a dummy reference number, because it was not used in the first pass. + However, with the change of recursive back references to be atomic, + we have to look for the number so that this state can be identified, as + otherwise the incorrect length is computed. If it's not a backwards + reference, the dummy number will do. */ + + if (lengthptr != NULL) + { + named_group *ng; + + if (namelen == 0) + { + *errorcodeptr = ERR62; + goto FAILED; + } + if (*ptr != (PCRE2_UCHAR)terminator) + { + *errorcodeptr = ERR42; + goto FAILED; + } + if (namelen > MAX_NAME_SIZE) + { + *errorcodeptr = ERR48; + goto FAILED; + } + + /* The name table does not exist in the first pass; instead we must + scan the list of names encountered so far in order to get the + number. If the name is not found, set the value to 0 for a forward + reference. */ + + ng = cd->named_groups; + for (i = 0; i < cd->names_found; i++, ng++) + { + if (namelen == ng->length && + PRIV(strncmp)(name, ng->name, namelen) == 0) + break; + } + recno = (i < cd->names_found)? ng->number : 0; + + /* Count named back references. */ + + if (!is_recurse) cd->namedrefcount++; + + /* If duplicate names are permitted, we have to allow for a named + reference to a duplicated name (this cannot be determined until the + second pass). This needs an extra 16-bit data item. */ + + if ((options & PCRE2_DUPNAMES) != 0) *lengthptr += IMM2_SIZE; + } + + /* In the real compile, search the name table. We check the name + first, and then check that we have reached the end of the name in the + table. That way, if the name is longer than any in the table, the + comparison will fail without reading beyond the table entry. */ + + else + { + slot = cd->name_table; + for (i = 0; i < cd->names_found; i++) + { + if (PRIV(strncmp)(name, slot+IMM2_SIZE, namelen) == 0 && + slot[IMM2_SIZE+namelen] == 0) + break; + slot += cd->name_entry_size; + } + + if (i < cd->names_found) + { + recno = GET2(slot, 0); + } + else + { + *errorcodeptr = ERR15; + goto FAILED; + } + } + + /* In both phases, for recursions, we can now go to the code than + handles numerical recursion. */ + + if (is_recurse) goto HANDLE_RECURSION; + + /* In the second pass we must see if the name is duplicated. If so, we + generate a different opcode. */ + + if (lengthptr == NULL && cd->dupnames) + { + int count = 1; + unsigned int index = i; + PCRE2_UCHAR *cslot = slot + cd->name_entry_size; + + for (i++; i < cd->names_found; i++) + { + if (PRIV(strcmp)(slot + IMM2_SIZE, cslot + IMM2_SIZE) != 0) break; + count++; + cslot += cd->name_entry_size; + } + + if (count > 1) + { + if (firstcuflags == REQ_UNSET) firstcuflags = REQ_NONE; + previous = code; + *code++ = ((options & PCRE2_CASELESS) != 0)? OP_DNREFI : OP_DNREF; + PUT2INC(code, 0, index); + PUT2INC(code, 0, count); + + /* Process each potentially referenced group. */ + + for (; slot < cslot; slot += cd->name_entry_size) + { + open_capitem *oc; + recno = GET2(slot, 0); + cd->backref_map |= (recno < 32)? (1 << recno) : 1; + if (recno > cd->top_backref) cd->top_backref = recno; + + /* Check to see if this back reference is recursive, that it, it + is inside the group that it references. A flag is set so that the + group can be made atomic. */ + + for (oc = cd->open_caps; oc != NULL; oc = oc->next) + { + if (oc->number == recno) + { + oc->flag = TRUE; + break; + } + } + } + + continue; /* End of back ref handling */ + } + } + + /* First pass, or a non-duplicated name. */ + + goto HANDLE_REFERENCE; + + + /* ------------------------------------------------------------ */ + case CHAR_R: /* Recursion */ + ptr++; /* Same as (?0) */ + /* Fall through */ + + + /* ------------------------------------------------------------ */ + case CHAR_MINUS: case CHAR_PLUS: /* Recursion or subroutine */ + case CHAR_0: case CHAR_1: case CHAR_2: case CHAR_3: case CHAR_4: + case CHAR_5: case CHAR_6: case CHAR_7: case CHAR_8: case CHAR_9: + { + PCRE2_SPTR called; + terminator = CHAR_RIGHT_PARENTHESIS; + + /* Come here from the \g<...> and \g'...' code (Oniguruma + compatibility). However, the syntax has been checked to ensure that + the ... are a (signed) number, so that neither ERR63 nor ERR29 will + be called on this path, nor with the jump to OTHER_CHAR_AFTER_QUERY + ever be taken. */ + + HANDLE_NUMERICAL_RECURSION: + + if ((refsign = *ptr) == CHAR_PLUS) + { + ptr++; + if (!IS_DIGIT(*ptr)) + { + *errorcodeptr = ERR63; + goto FAILED; + } + } + else if (refsign == CHAR_MINUS) + { + if (!IS_DIGIT(ptr[1])) + goto OTHER_CHAR_AFTER_QUERY; + ptr++; + } + + recno = 0; + while(IS_DIGIT(*ptr)) + recno = recno * 10 + *ptr++ - CHAR_0; + + if (*ptr != (PCRE2_UCHAR)terminator) + { + *errorcodeptr = ERR29; + goto FAILED; + } + + if (refsign == CHAR_MINUS) + { + if (recno == 0) + { + *errorcodeptr = ERR58; + goto FAILED; + } + recno = cd->bracount - recno + 1; + if (recno <= 0) + { + *errorcodeptr = ERR15; + goto FAILED; + } + } + else if (refsign == CHAR_PLUS) + { + if (recno == 0) + { + *errorcodeptr = ERR58; + goto FAILED; + } + recno += cd->bracount; + } + + /* Come here from code above that handles a named recursion */ + + HANDLE_RECURSION: + + previous = code; + called = cd->start_code; + + /* When we are actually compiling, find the bracket that is being + referenced. Temporarily end the regex in case it doesn't exist before + this point. If we end up with a forward reference, first check that + the bracket does occur later so we can give the error (and position) + now. Then remember this forward reference in the workspace so it can + be filled in at the end. */ + + if (lengthptr == NULL) + { + *code = OP_END; + if (recno != 0) + called = PRIV(find_bracket)(cd->start_code, utf, recno); + + /* Forward reference */ + + if (called == NULL) + { + if (recno > cd->final_bracount) + { + *errorcodeptr = ERR15; + goto FAILED; + } + + /* Fudge the value of "called" so that when it is inserted as an + offset below, what it actually inserted is the reference number + of the group. Then remember the forward reference. */ + + called = cd->start_code + recno; + if (cd->hwm >= cd->start_workspace + cd->workspace_size - + WORK_SIZE_SAFETY_MARGIN) + { + *errorcodeptr = expand_workspace(cd); + if (*errorcodeptr != 0) goto FAILED; + } + PUTINC(cd->hwm, 0, (int)(code + 1 - cd->start_code)); + } + + /* If not a forward reference, and the subpattern is still open, + this is a recursive call. We check to see if this is a left + recursion that could loop for ever, and diagnose that case. We + must not, however, do this check if we are in a conditional + subpattern because the condition might be testing for recursion in + a pattern such as /(?(R)a+|(?R)b)/, which is perfectly valid. + Forever loops are also detected at runtime, so those that occur in + conditional subpatterns will be picked up then. */ + + else if (GET(called, 1) == 0 && cond_depth <= 0 && + could_be_empty(called, code, bcptr, utf, cd)) + { + *errorcodeptr = ERR40; + goto FAILED; + } + } + + /* Insert the recursion/subroutine item. It does not have a set first + character (relevant if it is repeated, because it will then be + wrapped with ONCE brackets). */ + + *code = OP_RECURSE; + PUT(code, 1, (int)(called - cd->start_code)); + code += 1 + LINK_SIZE; + groupsetfirstcu = FALSE; + } + + /* Can't determine a first byte now */ + + if (firstcuflags == REQ_UNSET) firstcuflags = REQ_NONE; + continue; + + + /* ------------------------------------------------------------ */ + default: /* Other characters: check option setting */ + OTHER_CHAR_AFTER_QUERY: + set = unset = 0; + optset = &set; + + while (*ptr != CHAR_RIGHT_PARENTHESIS && *ptr != CHAR_COLON) + { + switch (*ptr++) + { + case CHAR_MINUS: optset = &unset; break; + + case CHAR_J: /* Record that it changed in the external options */ + *optset |= PCRE2_DUPNAMES; + cd->external_flags |= PCRE2_JCHANGED; + break; + + case CHAR_i: *optset |= PCRE2_CASELESS; break; + case CHAR_m: *optset |= PCRE2_MULTILINE; break; + case CHAR_s: *optset |= PCRE2_DOTALL; break; + case CHAR_x: *optset |= PCRE2_EXTENDED; break; + case CHAR_U: *optset |= PCRE2_UNGREEDY; break; + + default: *errorcodeptr = ERR12; + ptr--; /* Correct the offset */ + goto FAILED; + } + } + + /* Set up the changed option bits, but don't change anything yet. */ + + newoptions = (options | set) & (~unset); + + /* If the options ended with ')' this is not the start of a nested + group with option changes, so the options change at this level. If this + item is right at the start of the pattern, the options can be + abstracted and made external in the pre-compile phase, and ignored in + the compile phase. This can be helpful when matching -- for instance in + caseless checking of required bytes. + + If the code pointer is not (cd->start_code + 1 + LINK_SIZE), we are + definitely *not* at the start of the pattern because something has been + compiled. In the pre-compile phase, however, the code pointer can have + that value after the start, because it gets reset as code is discarded + during the pre-compile. However, this can happen only at top level - if + we are within parentheses, the starting BRA will still be present. At + any parenthesis level, the length value can be used to test if anything + has been compiled at that level. Thus, a test for both these conditions + is necessary to ensure we correctly detect the start of the pattern in + both phases. + + If we are not at the pattern start, reset the greedy defaults and the + case value for firstcu and reqcu. */ + + if (*ptr == CHAR_RIGHT_PARENTHESIS) + { + if (code == cd->start_code + 1 + LINK_SIZE && + (lengthptr == NULL || *lengthptr == 2 + 2*LINK_SIZE)) + { + cd->external_options = newoptions; + } + else + { + greedy_default = ((newoptions & PCRE2_UNGREEDY) != 0); + greedy_non_default = greedy_default ^ 1; + req_caseopt = ((newoptions & PCRE2_CASELESS) != 0)? REQ_CASELESS:0; + } + + /* Change options at this level, and pass them back for use + in subsequent branches. */ + + *optionsptr = options = newoptions; + previous = NULL; /* This item can't be repeated */ + continue; /* It is complete */ + } + + /* If the options ended with ':' we are heading into a nested group + with possible change of options. Such groups are non-capturing and are + not assertions of any kind. All we need to do is skip over the ':'; + the newoptions value is handled below. */ + + bravalue = OP_BRA; + ptr++; + } /* End of switch for character following (? */ + } /* End of (? handling */ + + /* Opening parenthesis not followed by '*' or '?'. If PCRE2_NO_AUTO_CAPTURE + is set, all unadorned brackets become non-capturing and behave like (?:...) + brackets. */ + + else if ((options & PCRE2_NO_AUTO_CAPTURE) != 0) + { + bravalue = OP_BRA; + } + + /* Else we have a capturing group. */ + + else + { + NUMBERED_GROUP: + cd->bracount += 1; + PUT2(code, 1+LINK_SIZE, cd->bracount); + skipbytes = IMM2_SIZE; + } + + /* Process nested bracketed regex. First check for parentheses nested too + deeply. */ + + if ((cd->parens_depth += 1) > PARENS_NEST_LIMIT) + { + *errorcodeptr = ERR82; + goto FAILED; + } + + /* Assertions used not to be repeatable, but this was changed for Perl + compatibility, so all kinds can now be repeated. We copy code into a + non-register variable (tempcode) in order to be able to pass its address + because some compilers complain otherwise. */ + + previous = code; /* For handling repetition */ + *code = bravalue; + tempcode = code; + tempreqvary = cd->req_varyopt; /* Save value before bracket */ + tempbracount = cd->bracount; /* Save value before bracket */ + length_prevgroup = 0; /* Initialize for pre-compile phase */ + + if (!compile_regex( + newoptions, /* The complete new option state */ + &tempcode, /* Where to put code (updated) */ + &ptr, /* Input pointer (updated) */ + errorcodeptr, /* Where to put an error message */ + (bravalue == OP_ASSERTBACK || + bravalue == OP_ASSERTBACK_NOT), /* TRUE if back assert */ + reset_bracount, /* True if (?| group */ + skipbytes, /* Skip over bracket number */ + cond_depth + + ((bravalue == OP_COND)?1:0), /* Depth of condition subpatterns */ + &subfirstcu, /* For possible first char */ + &subfirstcuflags, + &subreqcu, /* For possible last char */ + &subreqcuflags, + bcptr, /* Current branch chain */ + cd, /* Compile data block */ + (lengthptr == NULL)? NULL : /* Actual compile phase */ + &length_prevgroup /* Pre-compile phase */ + )) + goto FAILED; + + cd->parens_depth -= 1; + + /* If this was an atomic group and there are no capturing groups within it, + generate OP_ONCE_NC instead of OP_ONCE. */ + + if (bravalue == OP_ONCE && cd->bracount <= tempbracount) + *code = OP_ONCE_NC; + + if (bravalue >= OP_ASSERT && bravalue <= OP_ASSERTBACK_NOT) + cd->assert_depth -= 1; + + /* At the end of compiling, code is still pointing to the start of the + group, while tempcode has been updated to point past the end of the group. + The pattern pointer (ptr) is on the bracket. + + If this is a conditional bracket, check that there are no more than + two branches in the group, or just one if it's a DEFINE group. We do this + in the real compile phase, not in the pre-pass, where the whole group may + not be available. */ + + if (bravalue == OP_COND && lengthptr == NULL) + { + PCRE2_UCHAR *tc = code; + int condcount = 0; + + do { + condcount++; + tc += GET(tc,1); + } + while (*tc != OP_KET); + + /* A DEFINE group is never obeyed inline (the "condition" is always + false). It must have only one branch. */ + + if (code[LINK_SIZE+1] == OP_DEF) + { + if (condcount > 1) + { + *errorcodeptr = ERR54; + goto FAILED; + } + bravalue = OP_DEF; /* Just a flag to suppress char handling below */ + } + + /* A "normal" conditional group. If there is just one branch, we must not + make use of its firstcu or reqcu, because this is equivalent to an + empty second branch. */ + + else + { + if (condcount > 2) + { + *errorcodeptr = ERR27; + goto FAILED; + } + if (condcount == 1) subfirstcuflags = subreqcuflags = REQ_NONE; + } + } + + /* Error if hit end of pattern */ + + if (*ptr != CHAR_RIGHT_PARENTHESIS) + { + *errorcodeptr = ERR14; + goto FAILED; + } + + /* In the pre-compile phase, update the length by the length of the group, + less the brackets at either end. Then reduce the compiled code to just a + set of non-capturing brackets so that it doesn't use much memory if it is + duplicated by a quantifier.*/ + + if (lengthptr != NULL) + { + if (OFLOW_MAX - *lengthptr < length_prevgroup - 2 - 2*LINK_SIZE) + { + *errorcodeptr = ERR20; + goto FAILED; + } + *lengthptr += length_prevgroup - 2 - 2*LINK_SIZE; + code++; /* This already contains bravalue */ + PUTINC(code, 0, 1 + LINK_SIZE); + *code++ = OP_KET; + PUTINC(code, 0, 1 + LINK_SIZE); + break; /* No need to waste time with special character handling */ + } + + /* Otherwise update the main code pointer to the end of the group. */ + + code = tempcode; + + /* For a DEFINE group, required and first character settings are not + relevant. */ + + if (bravalue == OP_DEF) break; + + /* Handle updating of the required and first characters for other types of + group. Update for normal brackets of all kinds, and conditions with two + branches (see code above). If the bracket is followed by a quantifier with + zero repeat, we have to back off. Hence the definition of zeroreqcu and + zerofirstcu outside the main loop so that they can be accessed for the + back off. */ + + zeroreqcu = reqcu; + zeroreqcuflags = reqcuflags; + zerofirstcu = firstcu; + zerofirstcuflags = firstcuflags; + groupsetfirstcu = FALSE; + + if (bravalue >= OP_ONCE) + { + /* If we have not yet set a firstcu in this branch, take it from the + subpattern, remembering that it was set here so that a repeat of more + than one can replicate it as reqcu if necessary. If the subpattern has + no firstcu, set "none" for the whole branch. In both cases, a zero + repeat forces firstcu to "none". */ + + if (firstcuflags == REQ_UNSET) + { + if (subfirstcuflags >= 0) + { + firstcu = subfirstcu; + firstcuflags = subfirstcuflags; + groupsetfirstcu = TRUE; + } + else firstcuflags = REQ_NONE; + zerofirstcuflags = REQ_NONE; + } + + /* If firstcu was previously set, convert the subpattern's firstcu + into reqcu if there wasn't one, using the vary flag that was in + existence beforehand. */ + + else if (subfirstcuflags >= 0 && subreqcuflags < 0) + { + subreqcu = subfirstcu; + subreqcuflags = subfirstcuflags | tempreqvary; + } + + /* If the subpattern set a required byte (or set a first byte that isn't + really the first byte - see above), set it. */ + + if (subreqcuflags >= 0) + { + reqcu = subreqcu; + reqcuflags = subreqcuflags; + } + } + + /* For a forward assertion, we take the reqcu, if set. This can be + helpful if the pattern that follows the assertion doesn't set a different + char. For example, it's useful for /(?=abcde).+/. We can't set firstcu + for an assertion, however because it leads to incorrect effect for patterns + such as /(?=a)a.+/ when the "real" "a" would then become a reqcu instead + of a firstcu. This is overcome by a scan at the end if there's no + firstcu, looking for an asserted first char. */ + + else if (bravalue == OP_ASSERT && subreqcuflags >= 0) + { + reqcu = subreqcu; + reqcuflags = subreqcuflags; + } + break; /* End of processing '(' */ + + + /* ===================================================================*/ + /* Handle metasequences introduced by \. For ones like \d, the ESC_ values + are arranged to be the negation of the corresponding OP_values in the + default case when PCRE2_UCP is not set. For the back references, the values + are negative the reference number. Only back references and those types + that consume a character may be repeated. We can test for values between + ESC_b and ESC_Z for the latter; this may have to change if any new ones are + ever created. */ + + case CHAR_BACKSLASH: + tempptr = ptr; + escape = check_escape(&ptr, &ec, errorcodeptr, options, FALSE, cd); + if (*errorcodeptr != 0) goto FAILED; + + if (escape == 0) /* The escape coded a single character */ + c = ec; + else + { + if (escape == ESC_Q) /* Handle start of quoted string */ + { + if (ptr[1] == CHAR_BACKSLASH && ptr[2] == CHAR_E) + ptr += 2; /* avoid empty string */ + else inescq = TRUE; + continue; + } + + if (escape == ESC_E) continue; /* Perl ignores an orphan \E */ + + /* For metasequences that actually match a character, we disable the + setting of a first character if it hasn't already been set. */ + + if (firstcuflags == REQ_UNSET && escape > ESC_b && escape < ESC_Z) + firstcuflags = REQ_NONE; + + /* Set values to reset to if this is followed by a zero repeat. */ + + zerofirstcu = firstcu; + zerofirstcuflags = firstcuflags; + zeroreqcu = reqcu; + zeroreqcuflags = reqcuflags; + + /* \g or \g'name' is a subroutine call by name and \g or \g'n' + is a subroutine call by number (Oniguruma syntax). In fact, the value + ESC_g is returned only for these cases. So we don't need to check for < + or ' if the value is ESC_g. For the Perl syntax \g{n} the value is + -n, and for the Perl syntax \g{name} the result is ESC_k (as + that is a synonym for a named back reference). */ + + if (escape == ESC_g) + { + PCRE2_SPTR p; + uint32_t cf; + + save_hwm = cd->hwm; /* Normally this is set when '(' is read */ + terminator = (*(++ptr) == CHAR_LESS_THAN_SIGN)? + CHAR_GREATER_THAN_SIGN : CHAR_APOSTROPHE; + + /* These two statements stop the compiler for warning about possibly + unset variables caused by the jump to HANDLE_NUMERICAL_RECURSION. In + fact, because we do the check for a number below, the paths that + would actually be in error are never taken. */ + + skipbytes = 0; + reset_bracount = FALSE; + + /* If it's not a signed or unsigned number, treat it as a name. */ + + cf = ptr[1]; + if (cf != CHAR_PLUS && cf != CHAR_MINUS && !IS_DIGIT(cf)) + { + is_recurse = TRUE; + goto NAMED_REF_OR_RECURSE; + } + + /* Signed or unsigned number (cf = ptr[1]) is known to be plus or minus + or a digit. */ + + p = ptr + 2; + while (IS_DIGIT(*p)) p++; + if (*p != (PCRE2_UCHAR)terminator) + { + *errorcodeptr = ERR57; + break; + } + ptr++; + goto HANDLE_NUMERICAL_RECURSION; + } + + /* \k or \k'name' is a back reference by name (Perl syntax). + We also support \k{name} (.NET syntax). */ + + if (escape == ESC_k) + { + if ((ptr[1] != CHAR_LESS_THAN_SIGN && + ptr[1] != CHAR_APOSTROPHE && ptr[1] != CHAR_LEFT_CURLY_BRACKET)) + { + *errorcodeptr = ERR69; + break; + } + is_recurse = FALSE; + terminator = (*(++ptr) == CHAR_LESS_THAN_SIGN)? + CHAR_GREATER_THAN_SIGN : (*ptr == CHAR_APOSTROPHE)? + CHAR_APOSTROPHE : CHAR_RIGHT_CURLY_BRACKET; + goto NAMED_REF_OR_RECURSE; + } + + /* Back references are handled specially; must disable firstcu if + not set to cope with cases like (?=(\w+))\1: which would otherwise set + ':' later. */ + + if (escape < 0) + { + open_capitem *oc; + recno = -escape; + + /* Come here from named backref handling when the reference is to a + single group (i.e. not to a duplicated name. */ + + HANDLE_REFERENCE: + if (firstcuflags == REQ_UNSET) firstcuflags = REQ_NONE; + previous = code; + *code++ = ((options & PCRE2_CASELESS) != 0)? OP_REFI : OP_REF; + PUT2INC(code, 0, recno); + cd->backref_map |= (recno < 32)? (1 << recno) : 1; + if (recno > cd->top_backref) cd->top_backref = recno; + + /* Check to see if this back reference is recursive, that it, it + is inside the group that it references. A flag is set so that the + group can be made atomic. */ + + for (oc = cd->open_caps; oc != NULL; oc = oc->next) + { + if (oc->number == recno) + { + oc->flag = TRUE; + break; + } + } + } + + /* So are Unicode property matches, if supported. */ + +#ifdef SUPPORT_UTF + else if (escape == ESC_P || escape == ESC_p) + { + BOOL negated; + unsigned int ptype = 0, pdata = 0; + if (!get_ucp(&ptr, &negated, &ptype, &pdata, errorcodeptr, cd)) + goto FAILED; + previous = code; + *code++ = ((escape == ESC_p) != negated)? OP_PROP : OP_NOTPROP; + *code++ = ptype; + *code++ = pdata; + } +#else + + /* If Unicode properties are not supported, \X, \P, and \p are not + allowed. */ + + else if (escape == ESC_X || escape == ESC_P || escape == ESC_p) + { + *errorcodeptr = ERR45; + goto FAILED; + } +#endif + + /* For the rest (including \X when Unicode properties are supported), we + can obtain the OP value by negating the escape value in the default + situation when PCRE2_UCP is not set. When it *is* set, we substitute + Unicode property tests. Note that \b and \B do a one-character + lookbehind, and \A also behaves as if it does. */ + + else + { + if ((escape == ESC_b || escape == ESC_B || escape == ESC_A) && + cd->max_lookbehind == 0) + cd->max_lookbehind = 1; +#ifdef SUPPORT_UTF + if (escape >= ESC_DU && escape <= ESC_wu) + { + nestptr = ptr + 1; /* Where to resume */ + ptr = substitutes[escape - ESC_DU] - 1; /* Just before substitute */ + } + else +#endif + /* In non-UTF mode, we turn \C into OP_ALLANY instead of OP_ANYBYTE + so that it works in DFA mode and in lookbehinds. */ + + { + previous = (escape > ESC_b && escape < ESC_Z)? code : NULL; + *code++ = (!utf && escape == ESC_C)? OP_ALLANY : escape; + } + } + continue; + } + + /* We have a data character whose value is in c. In UTF-8 mode it may have + a value > 127. We set its representation in the length/buffer, and then + handle it as a data character. */ + + mclength = PUTCHAR(c, mcbuffer); + goto ONE_CHAR; + + + /* ===================================================================*/ + /* Handle a literal character. It is guaranteed not to be whitespace or # + when the extended flag is set. If we are in a UTF mode, it may be a + multi-unit literal character. */ + + default: + NORMAL_CHAR: + mclength = 1; + mcbuffer[0] = c; + +#ifdef SUPPORT_UTF + if (utf && HAS_EXTRALEN(c)) + ACROSSCHAR(TRUE, ptr[1], mcbuffer[mclength++] = *(++ptr)); +#endif + + /* At this point we have the character's bytes in mcbuffer, and the length + in mclength. When not in UTF mode, the length is always 1. */ + + ONE_CHAR: + previous = code; + + /* For caseless UTF mode, check whether this character has more than one + other case. If so, generate a special OP_PROP item instead of OP_CHARI. */ + +#ifdef SUPPORT_UTF + if (utf && (options & PCRE2_CASELESS) != 0) + { + GETCHAR(c, mcbuffer); + if ((c = UCD_CASESET(c)) != 0) + { + *code++ = OP_PROP; + *code++ = PT_CLIST; + *code++ = c; + if (firstcuflags == REQ_UNSET) + firstcuflags = zerofirstcuflags = REQ_NONE; + break; + } + } +#endif + + /* Caseful matches, or not one of the multicase characters. */ + + *code++ = ((options & PCRE2_CASELESS) != 0)? OP_CHARI : OP_CHAR; + for (c = 0; c < mclength; c++) *code++ = mcbuffer[c]; + + /* Remember if \r or \n were seen */ + + if (mcbuffer[0] == CHAR_CR || mcbuffer[0] == CHAR_NL) + cd->external_flags |= PCRE2_HASCRORLF; + + /* Set the first and required bytes appropriately. If no previous first + byte, set it from this character, but revert to none on a zero repeat. + Otherwise, leave the firstcu value alone, and don't change it on a zero + repeat. */ + + if (firstcuflags == REQ_UNSET) + { + zerofirstcuflags = REQ_NONE; + zeroreqcu = reqcu; + zeroreqcuflags = reqcuflags; + + /* If the character is more than one byte long, we can set firstcu + only if it is not to be matched caselessly. */ + + if (mclength == 1 || req_caseopt == 0) + { + firstcu = mcbuffer[0] | req_caseopt; + firstcu = mcbuffer[0]; + firstcuflags = req_caseopt; + + if (mclength != 1) + { + reqcu = code[-1]; + reqcuflags = cd->req_varyopt; + } + } + else firstcuflags = reqcuflags = REQ_NONE; + } + + /* firstcu was previously set; we can set reqcu only if the length is + 1 or the matching is caseful. */ + + else + { + zerofirstcu = firstcu; + zerofirstcuflags = firstcuflags; + zeroreqcu = reqcu; + zeroreqcuflags = reqcuflags; + if (mclength == 1 || req_caseopt == 0) + { + reqcu = code[-1]; + reqcuflags = req_caseopt | cd->req_varyopt; + } + } + + break; /* End of literal character handling */ + } + } /* end of big loop */ + +/* Control never reaches here by falling through, only by a goto for all the +error states. Pass back the position in the pattern so that it can be displayed +to the user for diagnosing the error. */ + +FAILED: +*ptrptr = ptr; +return FALSE; +} + + + +/************************************************* +* Compile regex: a sequence of alternatives * +*************************************************/ + +/* On entry, ptr is pointing past the bracket character, but on return it +points to the closing bracket, or vertical bar, or end of string. The code +variable is pointing at the byte into which the BRA operator has been stored. +This function is used during the pre-compile phase when we are trying to find +out the amount of memory needed, as well as during the real compile phase. The +value of lengthptr distinguishes the two phases. + +Arguments: + options option bits, including any changes for this subpattern + codeptr -> the address of the current code pointer + ptrptr -> the address of the current pattern pointer + errorcodeptr -> pointer to error code variable + lookbehind TRUE if this is a lookbehind assertion + reset_bracount TRUE to reset the count for each branch + skipunits skip this many code units at start (for brackets and OP_COND) + cond_depth depth of nesting for conditional subpatterns + firstcuptr place to put the first required code unit + firstcuflagsptr place to put the first code unit flags, or a negative number + reqcuptr place to put the last required code unit + reqcuflagsptr place to put the last required code unit flags, or a negative number + bcptr pointer to the chain of currently open branches + cd points to the data block with tables pointers etc. + lengthptr NULL during the real compile phase + points to length accumulator during pre-compile phase + +Returns: TRUE on success +*/ + +static BOOL +compile_regex(uint32_t options, PCRE2_UCHAR **codeptr, PCRE2_SPTR *ptrptr, + int *errorcodeptr, BOOL lookbehind, BOOL reset_bracount, int skipunits, + int cond_depth, + uint32_t *firstcuptr, int32_t *firstcuflagsptr, + uint32_t *reqcuptr, int32_t *reqcuflagsptr, + branch_chain *bcptr, compile_data *cd, size_t *lengthptr) +{ +PCRE2_SPTR ptr = *ptrptr; +PCRE2_UCHAR *code = *codeptr; +PCRE2_UCHAR *last_branch = code; +PCRE2_UCHAR *start_bracket = code; +PCRE2_UCHAR *reverse_count = NULL; +open_capitem capitem; +int capnumber = 0; +uint32_t firstcu, reqcu; +int32_t firstcuflags, reqcuflags; +uint32_t branchfirstcu, branchreqcu; +int32_t branchfirstcuflags, branchreqcuflags; +size_t length; +unsigned int orig_bracount; +unsigned int max_bracount; +branch_chain bc; + +#ifdef FIXME +/* If set, call the external function that checks for stack availability. */ + +if (ccontext->stack_guard != NULL && ccontext->stack_guard(0)) + { + *errorcodeptr= ERR85; + return FALSE; + } +#endif + +/* Miscellaneous initialization */ + +bc.outer = bcptr; +bc.current_branch = code; + +firstcu = reqcu = 0; +firstcuflags = reqcuflags = REQ_UNSET; + +/* Accumulate the length for use in the pre-compile phase. Start with the +length of the BRA and KET and any extra code units that are required at the +beginning. We accumulate in a local variable to save frequent testing of +lengthptr for NULL. We cannot do this by looking at the value of 'code' at the +start and end of each alternative, because compiled items are discarded during +the pre-compile phase so that the work space is not exceeded. */ + +length = 2 + 2*LINK_SIZE + skipunits; + +/* WARNING: If the above line is changed for any reason, you must also change +the code that abstracts option settings at the start of the pattern and makes +them global. It tests the value of length for (2 + 2*LINK_SIZE) in the +pre-compile phase to find out whether or not anything has yet been compiled. + +If this is a capturing subpattern, add to the chain of open capturing items +so that we can detect them if (*ACCEPT) is encountered. This is also used to +detect groups that contain recursive back references to themselves. Note that +only OP_CBRA need be tested here; changing this opcode to one of its variants, +e.g. OP_SCBRAPOS, happens later, after the group has been compiled. */ + +if (*code == OP_CBRA) + { + capnumber = GET2(code, 1 + LINK_SIZE); + capitem.number = capnumber; + capitem.next = cd->open_caps; + capitem.flag = FALSE; + cd->open_caps = &capitem; + } + +/* Offset is set zero to mark that this bracket is still open */ + +PUT(code, 1, 0); +code += 1 + LINK_SIZE + skipunits; + +/* Loop for each alternative branch */ + +orig_bracount = max_bracount = cd->bracount; + +for (;;) + { + /* For a (?| group, reset the capturing bracket count so that each branch + uses the same numbers. */ + + if (reset_bracount) cd->bracount = orig_bracount; + + /* Set up dummy OP_REVERSE if lookbehind assertion */ + + if (lookbehind) + { + *code++ = OP_REVERSE; + reverse_count = code; + PUTINC(code, 0, 0); + length += 1 + LINK_SIZE; + } + + /* Now compile the branch; in the pre-compile phase its length gets added + into the length. */ + + if (!compile_branch(&options, &code, &ptr, errorcodeptr, &branchfirstcu, + &branchfirstcuflags, &branchreqcu, &branchreqcuflags, &bc, + cond_depth, cd, (lengthptr == NULL)? NULL : &length)) + { + *ptrptr = ptr; + return FALSE; + } + + /* Keep the highest bracket count in case (?| was used and some branch + has fewer than the rest. */ + + if (cd->bracount > max_bracount) max_bracount = cd->bracount; + + /* In the real compile phase, there is some post-processing to be done. */ + + if (lengthptr == NULL) + { + /* If this is the first branch, the firstcu and reqcu values for the + branch become the values for the regex. */ + + if (*last_branch != OP_ALT) + { + firstcu = branchfirstcu; + firstcuflags = branchfirstcuflags; + reqcu = branchreqcu; + reqcuflags = branchreqcuflags; + } + + /* If this is not the first branch, the first char and reqcu have to + match the values from all the previous branches, except that if the + previous value for reqcu didn't have REQ_VARY set, it can still match, + and we set REQ_VARY for the regex. */ + + else + { + /* If we previously had a firstcu, but it doesn't match the new branch, + we have to abandon the firstcu for the regex, but if there was + previously no reqcu, it takes on the value of the old firstcu. */ + + if (firstcuflags >= 0 && + (firstcuflags != branchfirstcuflags || + firstcu != branchfirstcu)) + { + if (reqcuflags < 0) + { + reqcu = firstcu; + reqcuflags = firstcuflags; + } + firstcuflags = REQ_NONE; + } + + /* If we (now or from before) have no firstcu, a firstcu from the + branch becomes a reqcu if there isn't a branch reqcu. */ + + if (firstcuflags < 0 && branchfirstcuflags >= 0 && + branchreqcuflags < 0) + { + branchreqcu = branchfirstcu; + branchreqcuflags = branchfirstcuflags; + } + + /* Now ensure that the reqcus match */ + + if (((reqcuflags & ~REQ_VARY) != (branchreqcuflags & ~REQ_VARY)) || + reqcu != branchreqcu) + reqcuflags = REQ_NONE; + else + { + reqcu = branchreqcu; + reqcuflags |= branchreqcuflags; /* To "or" REQ_VARY */ + } + } + + /* If lookbehind, check that this branch matches a fixed-length string, and + put the length into the OP_REVERSE item. Temporarily mark the end of the + branch with OP_END. If the branch contains OP_RECURSE, the result is -3 + because there may be forward references that we can't check here. Set a + flag to cause another lookbehind check at the end. Why not do it all at the + end? Because common, erroneous checks are picked up here and the offset of + the problem can be shown. */ + + if (lookbehind) + { + int fixed_length; + *code = OP_END; + fixed_length = find_fixedlength(last_branch, (options & PCRE2_UTF) != 0, + FALSE, cd); + if (fixed_length == -3) + { + cd->check_lookbehind = TRUE; + } + else if (fixed_length < 0) + { + *errorcodeptr = (fixed_length == -2)? ERR36 : + (fixed_length == -4)? ERR70: ERR25; + *ptrptr = ptr; + return FALSE; + } + else + { + if (fixed_length > cd->max_lookbehind) + cd->max_lookbehind = fixed_length; + PUT(reverse_count, 0, fixed_length); + } + } + } + + /* Reached end of expression, either ')' or end of pattern. In the real + compile phase, go back through the alternative branches and reverse the chain + of offsets, with the field in the BRA item now becoming an offset to the + first alternative. If there are no alternatives, it points to the end of the + group. The length in the terminating ket is always the length of the whole + bracketed item. Return leaving the pointer at the terminating char. */ + + if (*ptr != CHAR_VERTICAL_LINE) + { + if (lengthptr == NULL) + { + size_t branch_length = code - last_branch; + do + { + size_t prev_length = GET(last_branch, 1); + PUT(last_branch, 1, branch_length); + branch_length = prev_length; + last_branch -= branch_length; + } + while (branch_length > 0); + } + + /* Fill in the ket */ + + *code = OP_KET; + PUT(code, 1, (int)(code - start_bracket)); + code += 1 + LINK_SIZE; + + /* If it was a capturing subpattern, check to see if it contained any + recursive back references. If so, we must wrap it in atomic brackets. + In any event, remove the block from the chain. */ + + if (capnumber > 0) + { + if (cd->open_caps->flag) + { + memmove(start_bracket + 1 + LINK_SIZE, start_bracket, + CU2BYTES(code - start_bracket)); + *start_bracket = OP_ONCE; + code += 1 + LINK_SIZE; + PUT(start_bracket, 1, (int)(code - start_bracket)); + *code = OP_KET; + PUT(code, 1, (int)(code - start_bracket)); + code += 1 + LINK_SIZE; + length += 2 + 2*LINK_SIZE; + } + cd->open_caps = cd->open_caps->next; + } + + /* Retain the highest bracket number, in case resetting was used. */ + + cd->bracount = max_bracount; + + /* Set values to pass back */ + + *codeptr = code; + *ptrptr = ptr; + *firstcuptr = firstcu; + *firstcuflagsptr = firstcuflags; + *reqcuptr = reqcu; + *reqcuflagsptr = reqcuflags; + if (lengthptr != NULL) + { + if (OFLOW_MAX - *lengthptr < length) + { + *errorcodeptr = ERR20; + return FALSE; + } + *lengthptr += length; + } + return TRUE; + } + + /* Another branch follows. In the pre-compile phase, we can move the code + pointer back to where it was for the start of the first branch. (That is, + pretend that each branch is the only one.) + + In the real compile phase, insert an ALT node. Its length field points back + to the previous branch while the bracket remains open. At the end the chain + is reversed. It's done like this so that the start of the bracket has a + zero offset until it is closed, making it possible to detect recursion. */ + + if (lengthptr != NULL) + { + code = *codeptr + 1 + LINK_SIZE + skipunits; + length += 1 + LINK_SIZE; + } + else + { + *code = OP_ALT; + PUT(code, 1, (int)(code - last_branch)); + bc.current_branch = last_branch = code; + code += 1 + LINK_SIZE; + } + + /* Advance past the vertical bar */ + + ptr++; + } +/* Control never reaches here */ +} + + + +/************************************************* +* Check for anchored pattern * +*************************************************/ + +/* Try to find out if this is an anchored regular expression. Consider each +alternative branch. If they all start with OP_SOD or OP_CIRC, or with a bracket +all of whose alternatives start with OP_SOD or OP_CIRC (recurse ad lib), then +it's anchored. However, if this is a multiline pattern, then only OP_SOD will +be found, because ^ generates OP_CIRCM in that mode. + +We can also consider a regex to be anchored if OP_SOM starts all its branches. +This is the code for \G, which means "match at start of match position, taking +into account the match offset". + +A branch is also implicitly anchored if it starts with .* and DOTALL is set, +because that will try the rest of the pattern at all possible matching points, +so there is no point trying again.... er .... + +.... except when the .* appears inside capturing parentheses, and there is a +subsequent back reference to those parentheses. We haven't enough information +to catch that case precisely. + +At first, the best we could do was to detect when .* was in capturing brackets +and the highest back reference was greater than or equal to that level. +However, by keeping a bitmap of the first 31 back references, we can catch some +of the more common cases more precisely. + +... A second exception is when the .* appears inside an atomic group, because +this prevents the number of characters it matches from being adjusted. + +Arguments: + code points to start of the compiled pattern + bracket_map a bitmap of which brackets we are inside while testing; this + handles up to substring 31; after that we just have to take + the less precise approach + cd points to the compile data block + atomcount atomic group level + +Returns: TRUE or FALSE +*/ + +static BOOL +is_anchored(register PCRE2_SPTR code, unsigned int bracket_map, + compile_data *cd, int atomcount) +{ +do { + PCRE2_SPTR scode = first_significant_code( + code + PRIV(OP_lengths)[*code], FALSE); + register int op = *scode; + + /* Non-capturing brackets */ + + if (op == OP_BRA || op == OP_BRAPOS || + op == OP_SBRA || op == OP_SBRAPOS) + { + if (!is_anchored(scode, bracket_map, cd, atomcount)) return FALSE; + } + + /* Capturing brackets */ + + else if (op == OP_CBRA || op == OP_CBRAPOS || + op == OP_SCBRA || op == OP_SCBRAPOS) + { + int n = GET2(scode, 1+LINK_SIZE); + int new_map = bracket_map | ((n < 32)? (1 << n) : 1); + if (!is_anchored(scode, new_map, cd, atomcount)) return FALSE; + } + + /* Positive forward assertions and conditions */ + + else if (op == OP_ASSERT || op == OP_COND) + { + if (!is_anchored(scode, bracket_map, cd, atomcount)) return FALSE; + } + + /* Atomic groups */ + + else if (op == OP_ONCE || op == OP_ONCE_NC) + { + if (!is_anchored(scode, bracket_map, cd, atomcount + 1)) + return FALSE; + } + + /* .* is not anchored unless DOTALL is set (which generates OP_ALLANY) and + it isn't in brackets that are or may be referenced or inside an atomic + group. */ + + else if ((op == OP_TYPESTAR || op == OP_TYPEMINSTAR || + op == OP_TYPEPOSSTAR)) + { + if (scode[1] != OP_ALLANY || (bracket_map & cd->backref_map) != 0 || + atomcount > 0 || cd->had_pruneorskip) + return FALSE; + } + + /* Check for explicit anchoring */ + + else if (op != OP_SOD && op != OP_SOM && op != OP_CIRC) return FALSE; + + code += GET(code, 1); + } +while (*code == OP_ALT); /* Loop for each alternative */ +return TRUE; +} + + + +/************************************************* +* Check for starting with ^ or .* * +*************************************************/ + +/* This is called to find out if every branch starts with ^ or .* so that +"first char" processing can be done to speed things up in multiline +matching and for non-DOTALL patterns that start with .* (which must start at +the beginning or after \n). As in the case of is_anchored() (see above), we +have to take account of back references to capturing brackets that contain .* +because in that case we can't make the assumption. Also, the appearance of .* +inside atomic brackets or in a pattern that contains *PRUNE or *SKIP does not +count, because once again the assumption no longer holds. + +Arguments: + code points to start of the compiled pattern + bracket_map a bitmap of which brackets we are inside while testing; this + handles up to substring 31; after that we just have to take + the less precise approach + cd points to the compile data + atomcount atomic group level + +Returns: TRUE or FALSE +*/ + +static BOOL +is_startline(PCRE2_SPTR code, unsigned int bracket_map, compile_data *cd, + int atomcount) +{ +do { + PCRE2_SPTR scode = first_significant_code( + code + PRIV(OP_lengths)[*code], FALSE); + register int op = *scode; + + /* If we are at the start of a conditional assertion group, *both* the + conditional assertion *and* what follows the condition must satisfy the test + for start of line. Other kinds of condition fail. Note that there may be an + auto-callout at the start of a condition. */ + + if (op == OP_COND) + { + scode += 1 + LINK_SIZE; + if (*scode == OP_CALLOUT) scode += PRIV(OP_lengths)[OP_CALLOUT]; + switch (*scode) + { + case OP_CREF: + case OP_DNCREF: + case OP_RREF: + case OP_DNRREF: + case OP_DEF: + return FALSE; + + default: /* Assertion */ + if (!is_startline(scode, bracket_map, cd, atomcount)) return FALSE; + do scode += GET(scode, 1); while (*scode == OP_ALT); + scode += 1 + LINK_SIZE; + break; + } + scode = first_significant_code(scode, FALSE); + op = *scode; + } + + /* Non-capturing brackets */ + + if (op == OP_BRA || op == OP_BRAPOS || + op == OP_SBRA || op == OP_SBRAPOS) + { + if (!is_startline(scode, bracket_map, cd, atomcount)) return FALSE; + } + + /* Capturing brackets */ + + else if (op == OP_CBRA || op == OP_CBRAPOS || + op == OP_SCBRA || op == OP_SCBRAPOS) + { + int n = GET2(scode, 1+LINK_SIZE); + int new_map = bracket_map | ((n < 32)? (1 << n) : 1); + if (!is_startline(scode, new_map, cd, atomcount)) return FALSE; + } + + /* Positive forward assertions */ + + else if (op == OP_ASSERT) + { + if (!is_startline(scode, bracket_map, cd, atomcount)) return FALSE; + } + + /* Atomic brackets */ + + else if (op == OP_ONCE || op == OP_ONCE_NC) + { + if (!is_startline(scode, bracket_map, cd, atomcount + 1)) return FALSE; + } + + /* .* means "start at start or after \n" if it isn't in atomic brackets or + brackets that may be referenced, as long as the pattern does not contain + *PRUNE or *SKIP, because these break the feature. Consider, for example, + /.*?a(*PRUNE)b/ with the subject "aab", which matches "ab", i.e. not at the + start of a line. */ + + else if (op == OP_TYPESTAR || op == OP_TYPEMINSTAR || op == OP_TYPEPOSSTAR) + { + if (scode[1] != OP_ANY || (bracket_map & cd->backref_map) != 0 || + atomcount > 0 || cd->had_pruneorskip) + return FALSE; + } + + /* Check for explicit circumflex; anything else gives a FALSE result. Note + in particular that this includes atomic brackets OP_ONCE and OP_ONCE_NC + because the number of characters matched by .* cannot be adjusted inside + them. */ + + else if (op != OP_CIRC && op != OP_CIRCM) return FALSE; + + /* Move on to the next alternative */ + + code += GET(code, 1); + } +while (*code == OP_ALT); /* Loop for each alternative */ +return TRUE; +} + + + +/************************************************* +* Check for asserted fixed first code unit * +*************************************************/ + +/* During compilation, the "first code unit" settings from forward assertions +are discarded, because they can cause conflicts with actual literals that +follow. However, if we end up without a first code unit setting for an +unanchored pattern, it is worth scanning the regex to see if there is an +initial asserted first code unit. If all branches start with the same asserted +code unit, or with a non-conditional bracket all of whose alternatives start +with the same asserted code unit (recurse ad lib), then we return that code +unit, with the flags set to zero or REQ_CASELESS; otherwise return zero with +REQ_NONE in the flags. + +Arguments: + code points to start of compiled pattern + flags points to the first code unit flags + inassert TRUE if in an assertion + +Returns: the fixed first code unit, or 0 with REQ_NONE in flags +*/ + +static uint32_t +find_firstassertedcu(PCRE2_SPTR code, int32_t *flags, BOOL inassert) +{ +register uint32_t c = 0; +int cflags = REQ_NONE; + +*flags = REQ_NONE; +do { + uint32_t d; + int dflags; + int xl = (*code == OP_CBRA || *code == OP_SCBRA || + *code == OP_CBRAPOS || *code == OP_SCBRAPOS)? IMM2_SIZE:0; + PCRE2_SPTR scode = first_significant_code(code + 1+LINK_SIZE + xl, TRUE); + register PCRE2_UCHAR op = *scode; + + switch(op) + { + default: + return 0; + + case OP_BRA: + case OP_BRAPOS: + case OP_CBRA: + case OP_SCBRA: + case OP_CBRAPOS: + case OP_SCBRAPOS: + case OP_ASSERT: + case OP_ONCE: + case OP_ONCE_NC: + d = find_firstassertedcu(scode, &dflags, op == OP_ASSERT); + if (dflags < 0) + return 0; + if (cflags < 0) { c = d; cflags = dflags; } + else if (c != d || cflags != dflags) return 0; + break; + + case OP_EXACT: + scode += IMM2_SIZE; + /* Fall through */ + + case OP_CHAR: + case OP_PLUS: + case OP_MINPLUS: + case OP_POSPLUS: + if (!inassert) return 0; + if (cflags < 0) { c = scode[1]; cflags = 0; } + else if (c != scode[1]) return 0; + break; + + case OP_EXACTI: + scode += IMM2_SIZE; + /* Fall through */ + + case OP_CHARI: + case OP_PLUSI: + case OP_MINPLUSI: + case OP_POSPLUSI: + if (!inassert) return 0; + if (cflags < 0) { c = scode[1]; cflags = REQ_CASELESS; } + else if (c != scode[1]) return 0; + break; + } + + code += GET(code, 1); + } +while (*code == OP_ALT); + +*flags = cflags; +return c; +} + + + +/************************************************* +* Add an entry to the name/number table * +*************************************************/ + +/* This function is called between compiling passes to add an entry to the +name/number table, maintaining alphabetical order. Checking for permitted +and forbidden duplicates has already been done. + +Arguments: + cd the compile data block + name the name to add + length the length of the name + groupno the group number + +Returns: nothing +*/ + +static void +add_name_to_table(compile_data *cd, PCRE2_SPTR name, int length, + unsigned int groupno) +{ +int i; +PCRE2_UCHAR *slot = cd->name_table; + +for (i = 0; i < cd->names_found; i++) + { + int crc = memcmp(name, slot+IMM2_SIZE, CU2BYTES(length)); + if (crc == 0 && slot[IMM2_SIZE+length] != 0) + crc = -1; /* Current name is a substring */ + + /* Make space in the table and break the loop for an earlier name. For a + duplicate or later name, carry on. We do this for duplicates so that in the + simple case (when ?(| is not used) they are in order of their numbers. In all + cases they are in the order in which they appear in the pattern. */ + + if (crc < 0) + { + memmove(slot + cd->name_entry_size, slot, + CU2BYTES((cd->names_found - i) * cd->name_entry_size)); + break; + } + + /* Continue the loop for a later or duplicate name */ + + slot += cd->name_entry_size; + } + +PUT2(slot, 0, groupno); +memcpy(slot + IMM2_SIZE, name, CU2BYTES(length)); +slot[IMM2_SIZE + length] = 0; +cd->names_found++; +} + + + +/************************************************* +* External function to compile a pattern * *************************************************/ /* This function reads a regular expression in the form of a string and returns @@ -71,100 +7253,661 @@ Arguments: pattern the regular expression patlen the length of the pattern, or < 0 for zero-terminated options option bits - errorcode pointer to error code variable (positive error code) - erroroffset pointer for offset in pattern where error was detected + errorptr pointer to errorcode + erroroffset pointer to error offset ccontext points to a compile context or is NULL Returns: pointer to compiled data block, or NULL on error, with errorcode and erroroffset set */ -/* FIXME: this is currently a placeholder function */ - PCRE2_EXP_DEFN pcre2_code * PCRE2_CALL_CONVENTION -pcre2_compile(PCRE2_SPTR pattern, int patlen, uint32_t options, int *errorcode, - size_t *erroroffset, pcre2_compile_context *ccontext) +pcre2_compile(PCRE2_SPTR pattern, int patlen, uint32_t options, + int *errorptr, size_t *erroroffset, pcre2_compile_context *ccontext) { -pcre2_compile_context default_context; -pcre2_code *c = NULL; +pcre2_real_code *re = NULL; /* What we will return */ +pcre2_compile_context default_context; /* For use if no context given */ -patlen = patlen; +size_t length = 1; /* Allow or final END opcode */ + +int32_t firstcuflags, reqcuflags; /* Type of first/req code unit */ +uint32_t firstcu, reqcu; /* Value of first/req code unit */ + +uint32_t limit_match = MATCH_LIMIT; /* Default match limits */ +uint32_t limit_recursion = MATCH_LIMIT_RECURSION; + +compile_data cd; /* "Static" compile-time data */ + +PCRE2_SPTR codestart; /* Start of compiled code */ +PCRE2_SPTR ptr; /* Current pointer in pattern */ +PCRE2_UCHAR *code; /* Current pointer in compiled code */ + +uint32_t skipatstart; /* When checking (*UTF) etc */ + +const uint8_t *tables; /* Char tables base pointer */ + + +BOOL utf; + +int newline = 0; /* Unset; can be set by the pattern */ +int bsr = 0; /* Unset; can be set by the pattern */ + +int errorcode = 0; /* Initialize */ + +size_t re_blocksize; + + + +/* Comments at the head of this file explain about these variables. */ + +PCRE2_UCHAR *copied_pattern = NULL; +PCRE2_UCHAR stack_copied_pattern[COPIED_PATTERN_SIZE]; +PCRE2_UCHAR cworkspace[COMPILE_WORK_SIZE]; +named_group named_groups[NAMED_GROUP_LIST_SIZE]; + + +/* -------------- Check arguments and set up the pattern ----------------- */ + +/* There must be error code and offset pointers. */ + +if (errorptr == NULL || erroroffset == NULL) return NULL; +*errorptr = ERR0; +*erroroffset = 0; + +/* There must be a pattern! */ + +if (pattern == NULL) + { + *errorptr = ERR16; + return NULL; + } + +/* Check that all undefined public option bits are zero. */ + +if ((options & ~PUBLIC_COMPILE_OPTIONS) != 0) + { + *errorptr = ERR17; + return NULL; + } + +/* A NULL compile context means "use a default context" */ if (ccontext == NULL) { PRIV(compile_context_init)(&default_context, TRUE); ccontext = &default_context; + } + +/* A negative pattern length means "zero-terminated". Otherwise, we make +a copy of the pattern and add a zero. */ + +if (patlen < 0) patlen = PRIV(strlen)(pattern); else + { + if (patlen < COPIED_PATTERN_SIZE) + copied_pattern = stack_copied_pattern; + else + { + copied_pattern = ccontext->memctl.malloc(CU2BYTES(patlen + 1), + ccontext->memctl.memory_data); + if (copied_pattern == NULL) + { + *errorptr = ERR21; + return NULL; + } + } + memcpy(copied_pattern, pattern, CU2BYTES(patlen)); + copied_pattern[patlen] = 0; + pattern = copied_pattern; + } + +/* ------------ Initialize the "static" compile data -------------- */ + + +tables = (ccontext->tables != NULL)? ccontext->tables : PRIV(default_tables); + +cd.lcc = tables + lcc_offset; /* Individual */ +cd.fcc = tables + fcc_offset; /* character */ +cd.cbits = tables + cbits_offset; /* tables */ +cd.ctypes = tables + ctypes_offset; + +cd.assert_depth = 0; +cd.bracount = cd.final_bracount = 0; +cd.cx = ccontext; +cd.dupnames = FALSE; +cd.end_pattern = pattern + patlen; +cd.external_flags = 0; +cd.external_options = options; +cd.hwm = cworkspace; +cd.max_lookbehind = 0; +cd.name_entry_size = 0; +cd.name_table = NULL; +cd.named_groups = named_groups; +cd.named_group_list_size = NAMED_GROUP_LIST_SIZE; +cd.namedrefcount = 0; +cd.names_found = 0; +cd.open_caps = NULL; +cd.parens_depth = 0; +cd.req_varyopt = 0; +cd.start_code = cworkspace; +cd.start_pattern = pattern; +cd.start_workspace = cworkspace; +cd.workspace_size = COMPILE_WORK_SIZE; + +/* Maximum back reference and backref bitmap. The bitmap records up to 31 back +references to help in deciding whether (.*) can be treated as anchored or not. +*/ + +cd.top_backref = 0; +cd.backref_map = 0; + + +/* --------------- Start looking at the pattern --------------- */ + +/* Check for global one-time option settings at the start of the pattern, and +remember the offset to the actual regex. */ + +ptr = pattern; +skipatstart = 0; + +while (ptr[skipatstart] == CHAR_LEFT_PARENTHESIS && + ptr[skipatstart+1] == CHAR_ASTERISK) + { + unsigned int i; + for (i = 0; i < sizeof(pso_list)/sizeof(pso); i++) + { + pso *p = pso_list + i; + + if (PRIV(strncmp_c8)(ptr+skipatstart+2, (char *)(p->name), p->length) == 0) + { + uint32_t c, pp; + + skipatstart += p->length + 2; + switch(p->type) + { + case PSO_OPT: + cd.external_options |= p->value; + break; + + case PSO_NL: + newline = p->value; + break; + + case PSO_BSR: + bsr = p->value; + break; + + case PSO_LIMM: + case PSO_LIMR: + c = 0; + pp = skipatstart; + while (IS_DIGIT(ptr[pp])) + { + if (c > UINT32_MAX / 10 - 1) break; /* Integer overflow */ + c = c*10 + ptr[pp++] - CHAR_0; + } + if (ptr[pp++] != CHAR_RIGHT_PARENTHESIS) goto END_PSO; + if (p->type == PSO_LIMM) + { + if (c < limit_match) + { + limit_match = c; + cd.external_flags |= PCRE2_MLSET; + } + } + else + { + if (c < limit_recursion) + { + limit_recursion = c; + cd.external_flags |= PCRE2_RLSET; + } + } + skipatstart += pp - skipatstart; + break; + } + break; /* Out of the table scan loop */ + } + } + if (i > sizeof(pso_list)/sizeof(pso)) break; /* Out of pso loop */ + } + +/* End of pattern-start options; advance to start of real regex. */ + +END_PSO: +ptr += skipatstart; + +/* Can't support UTF or UCP unless PCRE2 has been compiled with UTF support. */ + +#ifndef SUPPORT_UTF +if ((cd->external_options & (PCRE2_UTF|PCRE2_UCP)) != 0) + { + errorcode = ERR32; + goto HAD_ERROR; + } +#endif + +/* Check UTF. We have the original options in 'options', with that value as +modified by (*UTF) etc in cd->external_options. */ + +utf = (cd.external_options & PCRE2_UTF) != 0; +if (utf) + { + if ((options & PCRE2_NEVER_UTF) != 0) + { + errorcode = ERR78; + goto HAD_ERROR; + } + if ((options & PCRE2_NO_UTF_CHECK) == 0 && + (errorcode = PRIV(valid_utf)(pattern, -1, erroroffset)) != 0) + goto HAD_ERROR; } -/* Fudge while testing pcre2test. */ +/* Process the BSR setting. */ -*erroroffset = 0; +if (bsr == 0) bsr = ccontext->bsr_convention; -if (pattern[0] == 'Y') +/* Process the newline setting. */ + +if (newline == 0) newline = ccontext->newline_convention; +switch(newline) { - PCRE2_UCHAR *n; - int lennumber = (PCRE2_CODE_UNIT_WIDTH == 8)? 2 : 1; - size_t size = sizeof(pcre2_real_code) + - (12 + 3*lennumber)*(PCRE2_CODE_UNIT_WIDTH/8) + CU2BYTES(20); - c = ccontext->memctl.malloc(size, NULL); - c->memctl = ccontext->memctl; - c->magic_number = MAGIC_NUMBER; - c->size = size; - c->compile_options = options; - c->pattern_options = 0; - c->flags = PCRE2_CODE_UNIT_WIDTH/8; - c->limit_match = 0; - c->limit_recursion = 0; - c->max_lookbehind = 0; - c->minlength = 3; - c->top_bracket = 5; - c->top_backref = 1; - c->bsr_convention = ccontext->bsr_convention; - c->newline_convention = ccontext->newline_convention; - c->name_count = 3; - c->name_entry_size = 4 + lennumber; - - n = (PCRE2_UCHAR *)((char *)c + sizeof(pcre2_real_code)); - - if (lennumber == 2) *n++ = 0 ; - *n++ = 1; - *n++ = 'x'; *n++ = 'x'; *n++ = 'x'; *n++ = 0; + case PCRE2_NEWLINE_CR: + cd.nllen = 1; + cd.nl[0] = CHAR_CR; + break; - if (lennumber == 2) *n++ = 0 ; - *n++ = 2; - *n++ = 'y'; *n++ = 'y'; *n++ = 'y'; *n++ = 0; + case PCRE2_NEWLINE_LF: + cd.nllen = 1; + cd.nl[0] = CHAR_NL; + break; - if (lennumber == 2) *n++ = 0 ; - *n++ = 3; - *n++ = 'y'; *n++ = 'y'; *n++ = 'y'; *n++ = 0; + case PCRE2_NEWLINE_CRLF: + cd.nltype = NLTYPE_FIXED; + cd.nl[0] = CHAR_CR; + cd.nl[1] = CHAR_NL; + break; + case PCRE2_NEWLINE_ANY: + cd.nltype = NLTYPE_ANY; + break; - *n++ = OP_CHAR; - *n++ = 'x'; - *n++ = OP_CHARI; - *n++ = 'Y'; - - *n++ = OP_PROP; - *n++ = PT_SC; - *n++ = 0; - - *n++ = OP_DNRREF; - *n++ = 0; - *n++ = 0; - *n++ = 0; - *n++ = 0; - - *n++ = OP_END; + case PCRE2_NEWLINE_ANYCRLF: + cd.nltype = NLTYPE_ANYCRLF; + break; - - } - -else - { - *errorcode = 1; + default: + errorcode = ERR56; + goto HAD_ERROR; } + +/* Pretend to compile the pattern while actually just accumulating the amount +of memory required in the 'length' variable. This behaviour is triggered by +passing a non-NULL final argument to compile_regex(). We pass a block of +workspace (cworkspace) for it to compile parts of the pattern into; the +compiled code is discarded when it is no longer needed, so hopefully this +workspace will never overflow, though there is a test for its doing so. + +On error, errorcode will be set non-zero, so we don't need to look at the +result of the function. The initial options have been put into the cd block so +that they can be changed if an option setting is found within the regex right +at the beginning. Bringing initial option settings outside can help speed up +starting point checks. We still have to pass a separate options variable (the +first argument) because that may change as the pattern is processed. */ + +code = cworkspace; +*code = OP_BRA; + +(void)compile_regex(cd.external_options, &code, &ptr, &errorcode, FALSE, + FALSE, 0, 0, &firstcu, &firstcuflags, &reqcu, &reqcuflags, NULL, + &cd, &length); + +if (errorcode != 0) goto HAD_ERROR; +if (length > MAX_PATTERN_SIZE) + { + errorcode = ERR20; + goto HAD_ERROR; + } + +/* If there are groups with duplicate names and there are also references by +name, we must allow for the possibility of named references to duplicated +groups. These require an extra data item each. */ + +if (cd.dupnames && cd.namedrefcount > 0) + length += cd.namedrefcount * IMM2_SIZE * sizeof(PCRE2_UCHAR); + +/* Compute the size of, and then get and initialize, the data block for storing +the compiled pattern and names table. Integer overflow should no longer be +possible because nowadays we limit the maximum value of cd.names_found and +cd.name_entry_size. */ + +re_blocksize = sizeof(pcre2_real_code) + + CU2BYTES(length + cd.names_found * cd.name_entry_size); +re = (pcre2_real_code *) + ccontext->memctl.malloc(re_blocksize, ccontext->memctl.memory_data); +if (re == NULL) + { + errorcode = ERR21; + goto HAD_ERROR; + } + +re->memctl = ccontext->memctl; +re->tables = tables; +re->executable_jit = NULL; +memset(re->start_bitmap, 32, 0); +re->blocksize = re_blocksize; +re->magic_number = MAGIC_NUMBER; +re->compile_options = options; +re->pattern_options = cd.external_options; +re->flags = PCRE2_CODE_UNIT_WIDTH/8 | cd.external_flags; +re->limit_match = limit_match; +re->limit_recursion = limit_recursion; +re->first_codeunit = 0; +re->last_codeunit = 0; +re->bsr_convention = bsr; +re->newline_convention = newline; +re->max_lookbehind = +re->minlength = 0; +re->top_bracket = 0; +re->top_backref = 0; +re->name_entry_size = cd.name_entry_size; +re->name_count = cd.names_found; + +/* The basic block is immediately followed by the name table, and the compiled +code follows after that. */ + +codestart = (PCRE2_SPTR)((uint8_t *)re + sizeof(pcre2_real_code)) + + re->name_entry_size * re->name_count; + + +/* FIXME: Is this necessary? The nullpad field is at the end; it's there to +help in the case when a regex compiled on a system with 4-byte pointers is run +on another with 8-byte pointers. */ + +#ifdef FIXME +#ifdef COMPILE_PCRE32 +re->dummy = 0; +#else +re->dummy1 = re->dummy2 = re->dummy3 = 0; +#endif +#endif + + + + +/* Update the compile data block for the actual compile. The starting points of +the name/number translation table and of the code are passed around in the +compile data block. The start/end pattern and initial options are already set +from the pre-compile phase, as is the name_entry_size field. Reset the bracket +count and the names_found field. Also reset the hwm field; this time it's used +for remembering forward references to subpatterns. */ + +cd.final_bracount = cd.bracount; /* Save for checking forward references */ +cd.parens_depth = 0; +cd.assert_depth = 0; +cd.bracount = 0; +cd.max_lookbehind = 0; +cd.name_table = (PCRE2_UCHAR *)((uint8_t *)re + sizeof(pcre2_real_code)); +cd.start_code = codestart; +cd.hwm = (PCRE2_UCHAR *)(cd.start_workspace); +cd.req_varyopt = 0; +cd.had_accept = FALSE; +cd.had_pruneorskip = FALSE; +cd.check_lookbehind = FALSE; +cd.open_caps = NULL; + +/* If any named groups were found, create the name/number table from the list +created in the first pass. If the list was longer than the in-stack list, free +the heap memory. */ + +if (cd.names_found > 0) + { + int i = cd.names_found; + named_group *ng = cd.named_groups; + cd.names_found = 0; + for (; i > 0; i--, ng++) + add_name_to_table(&cd, ng->name, ng->length, ng->number); + if (cd.named_group_list_size > NAMED_GROUP_LIST_SIZE) + ccontext->memctl.free((void *)cd.named_groups, ccontext->memctl.memory_data); + } + +/* Set up a starting, non-extracting bracket, then compile the expression. On +error, errorcode will be set non-zero, so we don't need to look at the result +of the function here. */ + +ptr = pattern + skipatstart; +code = (PCRE2_UCHAR *)codestart; +*code = OP_BRA; +(void)compile_regex(re->pattern_options, &code, &ptr, &errorcode, FALSE, FALSE, + 0, 0, &firstcu, &firstcuflags, &reqcu, &reqcuflags, NULL, &cd, NULL); + +re->top_bracket = cd.bracount; +re->top_backref = cd.top_backref; +re->max_lookbehind = cd.max_lookbehind; + +#ifdef FIXME /* Is this necessary? Set above */ +re->flags = cd.external_flags | PCRE2_MODE; +#endif + +if (cd.had_accept) + { + reqcu = 0; /* Must disable after (*ACCEPT) */ + reqcuflags = REQ_NONE; + } + +/* If we have not reached end of pattern after a successful compile, there's an +excess bracket. Otherwise, fill in the final opcode and check for disastrous +overflow. */ + +if (errorcode == 0 && ptr < cd.end_pattern) errorcode = ERR22; +*code++ = OP_END; +if ((size_t)(code - codestart) > length) errorcode = ERR23; + +#ifdef SUPPORT_VALGRIND +/* If the estimated length exceeds the really used length, mark the extra +allocated memory as unaddressable, so that any out-of-bound reads can be +detected. */ +VALGRIND_MAKE_MEM_NOACCESS(code, (length - (code - codestart)) * sizeof(PCRE2_UCHAR)); +#endif + +/* Fill in any forward references that are required. There may be repeated +references; optimize for them, as searching a large regex takes time. */ + +if (cd.hwm > cd.start_workspace) + { + int prev_recno = -1; + PCRE2_SPTR groupptr = NULL; + while (errorcode == 0 && cd.hwm > cd.start_workspace) + { + int offset, recno; + cd.hwm -= LINK_SIZE; + offset = GET(cd.hwm, 0); + recno = GET(codestart, offset); + if (recno != prev_recno) + { + groupptr = PRIV(find_bracket)(codestart, utf, recno); + prev_recno = recno; + } + if (groupptr == NULL) errorcode = ERR53; + else PUT(((PCRE2_UCHAR *)codestart), offset, (int)(groupptr - codestart)); + } + } + +/* If the workspace had to be expanded, free the new memory. Set the pointer to +NULL to indicate that forward references have been filled in. */ + +if (cd.workspace_size > COMPILE_WORK_SIZE) + ccontext->memctl.free((void *)cd.start_workspace, + ccontext->memctl.memory_data); +cd.start_workspace = NULL; + +/* Give an error if there's back reference to a non-existent capturing +subpattern. */ + +if (errorcode == 0 && re->top_backref > re->top_bracket) errorcode = ERR15; + +/* Unless disabled, check whether single character iterators can be +auto-possessified. The function overwrites the appropriate opcode values. */ + +#ifdef FIXME +if ((options & PCRE2_NO_AUTO_POSSESS) == 0) + auto_possessify((PCRE2_UCHAR *)codestart, utf, cd); +#endif + +/* If there were any lookbehind assertions that contained OP_RECURSE +(recursions or subroutine calls), a flag is set for them to be checked here, +because they may contain forward references. Actual recursions cannot be fixed +length, but subroutine calls can. It is done like this so that those without +OP_RECURSE that are not fixed length get a diagnosic with a useful offset. The +exceptional ones forgo this. We scan the pattern to check that they are fixed +length, and set their lengths. */ + +if (cd.check_lookbehind) + { + PCRE2_UCHAR *cc = (PCRE2_UCHAR *)codestart; + + /* Loop, searching for OP_REVERSE items, and process those that do not have + their length set. (Actually, it will also re-process any that have a length + of zero, but that is a pathological case, and it does no harm.) When we find + one, we temporarily terminate the branch it is in while we scan it. */ + + for (cc = (PCRE2_UCHAR *)PRIV(find_bracket)(codestart, utf, -1); + cc != NULL; + cc = (PCRE2_UCHAR *)PRIV(find_bracket)(cc, utf, -1)) + { + if (GET(cc, 1) == 0) + { + int fixed_length; + PCRE2_UCHAR *be = cc - 1 - LINK_SIZE + GET(cc, -LINK_SIZE); + int end_op = *be; + *be = OP_END; + fixed_length = find_fixedlength(cc, utf, TRUE, &cd); + *be = end_op; + if (fixed_length < 0) + { + errorcode = (fixed_length == -2)? ERR36 : + (fixed_length == -4)? ERR70 : ERR25; + break; + } + if (fixed_length > cd.max_lookbehind) cd.max_lookbehind = fixed_length; + PUT(cc, 1, fixed_length); + } + cc += 1 + LINK_SIZE; + } + } + +/* Failed to compile, or error while post-processing. Earlier errors get here +via the dreaded goto. */ + +if (errorcode != 0) + { + HAD_ERROR: + pcre2_code_free(re); + re = NULL; + *errorptr = errorcode; + *erroroffset = (int)(ptr - pattern); + goto EXIT; + } + +/* Successful compile. If the anchored option was not passed, set it if +we can determine that the pattern is anchored by virtue of ^ characters or \A +or anything else, such as starting with non-atomic .* when DOTALL is set and +there are no occurrences of *PRUNE or *SKIP. */ + +if ((re->pattern_options & PCRE2_ANCHORED) == 0 && + is_anchored(codestart, 0, &cd, 0)) + re->pattern_options |= PCRE2_ANCHORED; + +/* If the pattern is still not anchored and we do not have a first code unit, +see if there is one that is asserted (these are not saved during the compile +because they can cause conflicts with actual literals that follow). */ + +if ((re->pattern_options & PCRE2_ANCHORED) == 0) + { + if (firstcuflags < 0) + firstcu = find_firstassertedcu(codestart, &firstcuflags, FALSE); -return c; + /* Save the data for a first code unit. */ + + if (firstcuflags >= 0) + { + re->first_codeunit = firstcu; + re->flags |= PCRE2_FIRSTSET; + + /* Handle caseless first code units. */ + + if ((firstcuflags & REQ_CASELESS) != 0) + { + if (firstcu < 128 || (!utf && firstcu < 255)) + { + if (cd.fcc[firstcu] != firstcu) re->flags |= PCRE2_FIRSTCASELESS; + } + + /* The first code unit is > 128 in UTF mode, or > 255 otherwise. In + 8-bit UTF mode, codepoints in the range 128-255 are introductory code + points and cannot have another case. In 16-bit and 32-bit mode, we can + check wide characters when UTF (and therefore UCP) is supported. */ + +#if defined SUPPORT_UTF && PCRE2_CODE_UNIT_WIDTH != 8 + else if (UCD_OTHERCASE(firstcu) != firstcu) + re->flags |= PCRE2_FIRSTCASELESS; +#endif + } + } + + /* When there is no first code unit, see if we can set the PCRE2_STARTLINE + flag. This is helpful for multiline matches when all branches start with ^ + and also when all branches start with non-atomic .* for non-DOTALL matches + when *PRUNE and SKIP are not present. */ + + else if (is_startline(codestart, 0, &cd, 0)) re->flags |= PCRE2_STARTLINE; + } + +/* Handle the "required code unit", if one is set. In the case of an anchored +pattern, do this only if it follows a variable length item in the pattern. */ + +if (reqcuflags >= 0 && + ((re->pattern_options & PCRE2_ANCHORED) == 0 || + (reqcuflags & REQ_VARY) != 0)) + { + re->last_codeunit = reqcu; + re->flags |= PCRE2_LASTSET; + + /* Handle caseless required code units as for first code units (above). */ + + if ((reqcuflags & REQ_CASELESS) != 0) + { + if (reqcu < 128 || (!utf && reqcu < 255)) + { + if (cd.fcc[reqcu] != reqcu) re->flags |= PCRE2_LASTCASELESS; + } +#if defined SUPPORT_UTF && PCRE2_CODE_UNIT_WIDTH != 8 + else if (UCD_OTHERCASE(reqcu) != reqcu) + re->flags |= PCRE2_LASTCASELESS; +#endif + } + } + +/* Check for a pattern than can match an empty string, so that this information +can be provided to applications. */ + +do + { + if (could_be_empty_branch(codestart, code, utf, &cd, NULL)) + { + re->flags |= PCRE2_MATCH_EMPTY; + break; + } + codestart += GET(codestart, 1); + } +while (*codestart == OP_ALT); + +/* Control ends up here in all cases. If memory was obtained for a +zero-terminated copy of the pattern, remember to free it before returning. */ + +EXIT: +if (copied_pattern != stack_copied_pattern) + ccontext->memctl.free(copied_pattern, ccontext->memctl.memory_data); +return re; /* Will be NULL after an error */ } /* End of pcre2_compile.c */ diff --git a/src/pcre2_config.c b/src/pcre2_config.c index 21c14f3..b5abc5b 100644 --- a/src/pcre2_config.c +++ b/src/pcre2_config.c @@ -102,7 +102,7 @@ switch (what) break; case PCRE2_CONFIG_NEWLINE: - *((int *)where) = NEWLINE; + *((int *)where) = NEWLINE_DEFAULT; break; case PCRE2_CONFIG_PARENS_LIMIT: diff --git a/src/pcre2_context.c b/src/pcre2_context.c index 541bba3..376a807 100644 --- a/src/pcre2_context.c +++ b/src/pcre2_context.c @@ -139,9 +139,9 @@ if (defmemctl) } ccontext->stack_guard = NULL; ccontext->tables = PRIV(default_tables); -ccontext->bsr_convention = PCRE2_BSR_DEFAULT; -ccontext->newline_convention = PCRE2_NEWLINE_DEFAULT; ccontext->parens_nest_limit = PARENS_NEST_LIMIT; +ccontext->newline_convention = NEWLINE_DEFAULT; +ccontext->bsr_convention = BSR_DEFAULT; } diff --git a/src/pcre2_error.c b/src/pcre2_error.c index 079ef83..7e8aa6d 100644 --- a/src/pcre2_error.c +++ b/src/pcre2_error.c @@ -51,7 +51,10 @@ POSSIBILITY OF SUCH DAMAGE. #define STRING(a) # a #define XSTRING(s) STRING(s) -/* The texts of compile-time error messages. Do not ever re-use any error +/* The texts of compile-time error messages. Compile-time error numbers start +at COMPILE_ERROR_BASE (100). + +Do not ever re-use any error number, because they are documented. Always add a new error instead. Messages marked DEAD below are no longer used. This used to be a table of strings, but in order to reduce the number of relocations needed when a shared library is @@ -85,7 +88,7 @@ static const char compile_error_texts[] = "missing )\0" /* 15 */ "reference to non-existent subpattern\0" - "erroffset passed as NULL\0" + "pattern or erroffset passed as NULL\0" "unknown option bit(s) set\0" "missing ) after comment\0" "parentheses nested too deeply\0" /** DEAD **/ @@ -104,7 +107,7 @@ static const char compile_error_texts[] = /* 30 */ "unknown POSIX class name\0" "POSIX collating elements are not supported\0" - "this version of PCRE is compiled without UTF support\0" + "this version of PCRE does not have UTF or Unicode property support\0" "spare error\0" /** DEAD **/ "character value in \\x{} or \\o{} is too large\0" /* 35 */ @@ -133,7 +136,7 @@ static const char compile_error_texts[] = "DEFINE group contains more than one branch\0" /* 55 */ "repeating a DEFINE group is not allowed\0" /** DEAD **/ - "inconsistent NEWLINE options\0" + "internal error: unknown newline setting\0" "\\g is not followed by a braced, angle-bracketed, or quoted name/number or by a plain number\0" "a numbered reference must not be zero\0" "an argument is not allowed for (*ACCEPT), (*FAIL), or (*COMMIT)\0" @@ -171,58 +174,74 @@ static const char compile_error_texts[] = "parentheses are too deeply nested (stack check)\0" ; -/* Match-time error texts are in the same format. */ +/* Match-time and UTF error texts are in the same format. */ static const char match_error_texts[] = "no error\0" "no match\0" + "partial match\0" + "UTF-8 error: 1 byte missing at end\0" + "UTF-8 error: 2 bytes missing at end\0" + /* 5 */ + "UTF-8 error: 3 bytes missing at end\0" + "UTF-8 error: 4 bytes missing at end\0" + "UTF-8 error: 5 bytes missing at end\0" + "UTF-8 error: byte 2 top bits not 0x80\0" + "UTF-8 error: byte 3 top bits not 0x80\0" + /* 10 */ + "UTF-8 error: byte 4 top bits not 0x80\0" + "UTF-8 error: byte 5 top bits not 0x80\0" + "UTF-8 error: byte 6 top bits not 0x80\0" + "UTF-8 error: 5-byte character is not allowed (RFC 3629)\0" + "UTF-8 error: 6-byte character is not allowed (RFC 3629)\0" + /* 15 */ + "UTF-8 error: code point > 0x10ffff is not defined\0" + "UTF-8 error: code points 0xd000-0xdfff are not defined\0" + "UTF-8 error: overlong 2-byte sequence\0" + "UTF-8 error: overlong 3-byte sequence\0" + "UTF-8 error: overlong 4-byte sequence\0" + /* 20 */ + "UTF-8 error: overlong 5-byte sequence\0" + "UTF-8 error: overlong 6-byte sequence\0" + "UTF-8 error: isolated 0x80 byte\0" + "UTF-8 error: illegal byte (0xfe or 0xff)\0" + "UTF-16 error: missing low surrogate at end\0" + /* 25 */ + "UTF-16 error: invalid low surrogate\0" + "UTF-16 error: isolated low surrogate\0" + "UTF-32 error: surrogate character not allowed\0" + "UTF-32 error: code point > 0x10ffff is not defined\0" "bad count value\0" + /* 30 */ "pattern compiled with other endianness\0" "bad length\0" - /* -5 */ "magic number missing\0" "pattern compiled in wrong mode: 8/16/32-bit error\0" "bad offset value\0" + /* 35 */ "bad option value\0" - "bad UTF string\0" - /* -10 */ "bad offset into UTF string\0" "callout error code\0" /* Never returned by PCRE2 itself */ + "invalid data in workspace for DFA restart\0" + "too much recursion for DFA matching\0" + /* 40 */ + "backreference condition or recursion test not supported for DFA matching\0" + "item unsupported for DFA matching\0" + "match limit not supported for DFA matching\0" + "workspace size exceeded in DFA matching\0" "internal error - pattern overwritten?\0" + /* 45 */ "bad JIT option\0" "JIT stack limit reached\0" - /* -15 */ "match limit exceeded\0" "no more memory\0" "unknown substring\0" + /* 50 */ "NULL argument passed\0" - "partial match\0" - /* -20 */ "nested recursion at the same subject position\0" "recursion limit exceeded\0" "unknown opcode - pattern overwritten?\0" "value unset\0" /* Used by pcre2_pattern_info() */ - "spare -24\0" - /* -25 */ - "spare -25\0" - "spare -26\0" - "spare -27\0" - "spare -28\0" - "spare -29\0" - /* -30 */ - "invalid data in workspace for DFA restart\0" - "too much recursion for DFA matching\0" - "backreference condition or recursion test not supported for DFA matching\0" - "item unsupported for DFA matching\0" - "match limit not supported for DFA matching\0" - /* -35 */ - "workspace size exceeded in DFA matching\0" - "spare -36\0" - "spare -37\0" - "spare -38\0" - "spare -39\0" - /* -40 */ - "spare -39\0" ; @@ -232,7 +251,8 @@ static const char match_error_texts[] = /* This function copies an error message into a buffer whose units are of an appropriate width. Error numbers are positive for compile-time errors, and -negative for exec-time errors. +negative for match-time errors (except for UTF errors), but the numbers are all +distinct. Arguments: enumber error number @@ -253,16 +273,15 @@ uint32_t n; if (size == 0) return PCRE2_ERROR_NOMEMORY; -if (enumber > 0) /* Compile-time error */ +if (enumber > COMPILE_ERROR_BASE) /* Compile error */ { message = compile_error_texts; - n = enumber; + n = enumber - COMPILE_ERROR_BASE; } - -else /* Match-time error */ +else /* Match or UTF error */ { message = match_error_texts; - n = -enumber; + n = -enumber; } for (; n > 0; n--) diff --git a/src/pcre2_internal.h b/src/pcre2_internal.h index 862b19a..f70da5a 100644 --- a/src/pcre2_internal.h +++ b/src/pcre2_internal.h @@ -40,78 +40,235 @@ POSSIBILITY OF SUCH DAMAGE. /* FIXME: this file is incomplete, being gradually built. */ +/* We do not support both EBCDIC and UTF at the same time. The "configure" +script prevents both being selected, but not everybody uses "configure". */ + +#if defined EBCDIC && defined SUPPORT_UTF +#error The use of both EBCDIC and SUPPORT_UTF is not supported. +#endif + +/* Standard C headers */ + #include #include #include -#include #include #include #include -#include "pcre2.h" -#include "pcre2_ucp.h" - -#define PUBL(name) pcre2_##name - -#ifndef PRIV -#define PRIV(name) _pcre2_##name -#endif - -#define PCRE2_CALL_CONVENTION - -extern const uint8_t PRIV(default_tables)[]; - - - +/* Macros to make boolean values more obvious. The #ifndef is to pacify +compiler warnings in environments where these macros are defined elsewhere. +Unfortunately, there is no way to do the same for the typedef. */ typedef int BOOL; - #ifndef FALSE #define FALSE 0 #define TRUE 1 #endif - /* Valgrind (memcheck) support */ #ifdef SUPPORT_VALGRIND #include #endif -/* This is an unsigned int value that no character can ever have, as +/* When compiling a DLL for Windows, the exported symbols have to be declared +using some MS magic. I found some useful information on this web page: +http://msdn2.microsoft.com/en-us/library/y4h7bcy6(VS.80).aspx. According to the +information there, using __declspec(dllexport) without "extern" we have a +definition; with "extern" we have a declaration. The settings here override the +setting in pcre.h (which is included below); it defines only PCRE2_EXP_DECL, +which is all that is needed for applications (they just import the symbols). We +use: + + PCRE2_EXP_DECL for declarations + PCRE2_EXP_DEFN for definitions of exported functions + PCRE2_EXP_DATA_DEFN for definitions of exported variables + +The reason for the two DEFN macros is that in non-Windows environments, one +does not want to have "extern" before variable definitions because it leads to +compiler warnings. So we distinguish between functions and variables. In +Windows, the two should always be the same. + +The reason for wrapping this in #ifndef PCRE2_EXP_DECL is so that pcretest, +which is an application, but needs to import this file in order to "peek" at +internals, can #include pcre.h first to get an application's-eye view. + +In principle, people compiling for non-Windows, non-Unix-like (i.e. uncommon, +special-purpose environments) might want to stick other stuff in front of +exported symbols. That's why, in the non-Windows case, we set PCRE2_EXP_DEFN and +PCRE2_EXP_DATA_DEFN only if they are not already set. */ + +#ifndef PCRE2_EXP_DECL +# ifdef _WIN32 +# ifndef PCRE2_STATIC +# define PCRE2_EXP_DECL extern __declspec(dllexport) +# define PCRE2_EXP_DEFN __declspec(dllexport) +# define PCRE2_EXP_DATA_DEFN __declspec(dllexport) +# else +# define PCRE2_EXP_DECL extern +# define PCRE2_EXP_DEFN +# define PCRE2_EXP_DATA_DEFN +# endif +# else +# ifdef __cplusplus +# define PCRE2_EXP_DECL extern "C" +# else +# define PCRE2_EXP_DECL extern +# endif +# ifndef PCRE2_EXP_DEFN +# define PCRE2_EXP_DEFN PCRE2_EXP_DECL +# endif +# ifndef PCRE2_EXP_DATA_DEFN +# define PCRE2_EXP_DATA_DEFN +# endif +# endif +#endif + +/* Include the public PCRE2 header and the definitions of UCP character +property values. This must follow the setting of PCRE2_EXP_DECL above. */ + +#include "pcre2.h" +#include "pcre2_ucp.h" + +/* When PCRE is compiled as a C++ library, the subject pointer can be replaced +with a custom type. This makes it possible, for example, to allow pcre2_match() +to process subject strings that are discontinuous by using a smart pointer +class. It must always be possible to inspect all of the subject string in +pcre2_match() because of the way it backtracks. */ + +/* WARNING: This is as yet untested for PCRE2. */ + +#ifdef CUSTOM_SUBJECT_PTR +#undef PCRE2_SPTR +#define PCRE2_SPTR CUSTOM_SUBJECT_PTR +#endif + +/* When compiling with the MSVC compiler, it is sometimes necessary to include +a "calling convention" before exported function names. (This is secondhand +information; I know nothing about MSVC myself). For example, something like + + void __cdecl function(....) + +might be needed. In order so make this easy, all the exported functions have +PCRE2_CALL_CONVENTION just before their names. It is rarely needed; if not +set, we ensure here that it has no effect. */ + +#ifndef PCRE2_CALL_CONVENTION +#define PCRE2_CALL_CONVENTION +#endif + +/* When checking for integer overflow in pcre2_compile(), we need to handle +large integers. If a 64-bit integer type is available, we can use that. +Otherwise we have to cast to double, which of course requires floating point +arithmetic. Handle this by defining a macro for the appropriate type. If +stdint.h is available, include it; it may define INT64_MAX. Systems that do not +have stdint.h (e.g. Solaris) may have inttypes.h. The macro int64_t may be set +by "configure". */ + +#if defined HAVE_STDINT_H +#include +#elif defined HAVE_INTTYPES_H +#include +#endif + +#if defined INT64_MAX || defined int64_t +#define INT64_OR_DOUBLE int64_t +#else +#define INT64_OR_DOUBLE double +#endif + +/* When compiling for use with the Virtual Pascal compiler, these functions +need to have their names changed. PCRE must be compiled with the -DVPCOMPAT +option on the command line. */ + +#ifdef VPCOMPAT +#define strlen(s) _strlen(s) +#define strncmp(s1,s2,m) _strncmp(s1,s2,m) +#define memcmp(s,c,n) _memcmp(s,c,n) +#define memcpy(d,s,n) _memcpy(d,s,n) +#define memmove(d,s,n) _memmove(d,s,n) +#define memset(s,c,n) _memset(s,c,n) +#else /* VPCOMPAT */ + +/* To cope with SunOS4 and other systems that lack memmove() but have bcopy(), +define a macro for memmove() if HAVE_MEMMOVE is false, provided that HAVE_BCOPY +is set. Otherwise, include an emulating function for those systems that have +neither (there some non-Unix environments where this is the case). */ + +#ifndef HAVE_MEMMOVE +#undef memmove /* some systems may have a macro */ +#ifdef HAVE_BCOPY +#define memmove(a, b, c) bcopy(b, a, c) +#else /* HAVE_BCOPY */ +static void * +pcre_memmove(void *d, const void *s, size_t n) +{ +size_t i; +unsigned char *dest = (unsigned char *)d; +const unsigned char *src = (const unsigned char *)s; +if (dest > src) + { + dest += n; + src += n; + for (i = 0; i < n; ++i) *(--dest) = *(--src); + return (void *)dest; + } +else + { + for (i = 0; i < n; ++i) *dest++ = *src++; + return (void *)(dest - n); + } +} +#define memmove(a, b, c) pcre_memmove(a, b, c) +#endif /* not HAVE_BCOPY */ +#endif /* not HAVE_MEMMOVE */ +#endif /* not VPCOMPAT */ + +/* External (in the C sense) functions and macros that are private to the +libraries are always referenced using the PRIV macro. This makes it possible +for pcre2test.c to include some of the source files from the libraries using a +different PRIV definition to avoid name clashes. */ + +#ifndef PRIV +#define PRIV(name) _pcre2_##name +#endif + +/* This is an unsigned int value that no UTF character can ever have, as Unicode doesn't go beyond 0x0010ffff. */ #define NOTACHAR 0xffffffff -/* When UTF encoding is being used, a character is no longer just a single -byte in 8-bit mode or a single short in 16-bit mode. The macros for character -handling generate simple sequences when used in the basic mode, and more -complicated ones for UTF characters. GETCHARLENTEST and other macros are not -used when UTF is not supported. To make sure they can never even appear when -UTF support is omitted, we don't even define them. */ +/* Compile-time errors are added to this value. As they are documented, it +should probably never be changed. */ -#ifndef SUPPORT_UTF +#define COMPILE_ERROR_BASE 100 -/* #define MAX_VALUE_FOR_SINGLE_CHAR */ -/* #define HAS_EXTRALEN(c) */ -/* #define GET_EXTRALEN(c) */ -/* #define NOT_FIRSTCHAR(c) */ -#define GETCHAR(c, eptr) c = *eptr; -#define GETCHARTEST(c, eptr) c = *eptr; -#define GETCHARINC(c, eptr) c = *eptr++; -#define GETCHARINCTEST(c, eptr) c = *eptr++; -#define GETCHARLEN(c, eptr, len) c = *eptr; -/* #define GETCHARLENTEST(c, eptr, len) */ -/* #define BACKCHAR(eptr) */ -/* #define FORWARDCHAR(eptr) */ -/* #define ACROSSCHAR(condition, eptr, action) */ +/* Define the default BSR convention. */ -#else /* SUPPORT_UTF */ +#ifdef BSR_ANYCRLF +#define BSR_DEFAULT PCRE2_BSR_ANYCRLF +#else +#define BSR_DEFAULT PCRE2_BSR_UNICODE +#endif + + +/* ---------------- Basic UTF-8 macros ---------------- */ + +/* These UTF-8 macros are always defined because they are used in pcre2test for +handling wide characters in 16-bit and 32-bit modes, even if an 8-bit library +is not supported. */ /* Tests whether a UTF-8 code point needs extra bytes to decode. */ #define HASUTF8EXTRALEN(c) ((c) >= 0xc0) +/* The following macros were originally written in the form of loops that used +data from the tables whose names start with PRIV(utf8_table). They were +rewritten by a user so as not to use loops, because in some environments this +gives a significant performance advantage, and it seems never to do any harm. +*/ + /* Base macro to pick up the remaining bytes of a UTF-8 character, not advancing the pointer. */ @@ -168,8 +325,44 @@ the pointer. */ } \ } -#endif /* SUPPORT_UTF */ +/* Base macro to pick up the remaining bytes of a UTF-8 character, not +advancing the pointer, incrementing the length. */ +#define GETUTF8LEN(c, eptr, len) \ + { \ + if ((c & 0x20) == 0) \ + { \ + c = ((c & 0x1f) << 6) | (eptr[1] & 0x3f); \ + len++; \ + } \ + else if ((c & 0x10) == 0) \ + { \ + c = ((c & 0x0f) << 12) | ((eptr[1] & 0x3f) << 6) | (eptr[2] & 0x3f); \ + len += 2; \ + } \ + else if ((c & 0x08) == 0) \ + {\ + c = ((c & 0x07) << 18) | ((eptr[1] & 0x3f) << 12) | \ + ((eptr[2] & 0x3f) << 6) | (eptr[3] & 0x3f); \ + len += 3; \ + } \ + else if ((c & 0x04) == 0) \ + { \ + c = ((c & 0x03) << 24) | ((eptr[1] & 0x3f) << 18) | \ + ((eptr[2] & 0x3f) << 12) | ((eptr[3] & 0x3f) << 6) | \ + (eptr[4] & 0x3f); \ + len += 4; \ + } \ + else \ + {\ + c = ((c & 0x01) << 30) | ((eptr[1] & 0x3f) << 24) | \ + ((eptr[2] & 0x3f) << 18) | ((eptr[3] & 0x3f) << 12) | \ + ((eptr[4] & 0x3f) << 6) | (eptr[5] & 0x3f); \ + len += 5; \ + } \ + } + +/* --------------- Whitespace macros ---------------- */ /* Tests for Unicode horizontal and vertical whitespace characters must check a number of different values. Using a switch statement for this generates the @@ -187,7 +380,7 @@ NOTACHAR (which is 0xffffffff). Any changes should ensure that the various macros are kept in step with each other. NOTE: The values also appear in pcre2_jit_compile.c. */ -/* ------ ASCII/Unicode environments ------ */ +/* -------------- ASCII/Unicode environments -------------- */ #ifndef EBCDIC @@ -242,7 +435,7 @@ other. NOTE: The values also appear in pcre2_jit_compile.c. */ VSPACE_BYTE_CASES: \ VSPACE_MULTIBYTE_CASES -/* ------ EBCDIC environments ------ */ +/* -------------- EBCDIC environments -------------- */ #else #define HSPACE_LIST CHAR_HT, CHAR_SPACE @@ -271,9 +464,47 @@ other. NOTE: The values also appear in pcre2_jit_compile.c. */ #define VSPACE_CASES VSPACE_BYTE_CASES #endif /* EBCDIC */ -/* ------ End of whitespace macros ------ */ +/* -------------- End of whitespace macros -------------- */ +/* PCRE2 is able to support several different kinds of newline (CR, LF, CRLF, +"any" and "anycrlf" at present). The following macros are used to package up +testing for newlines. NLBLOCK, PSSTART, and PSEND are defined in the various +modules to indicate in which datablock the parameters exist, and what the +start/end of string field names are. */ + +#define NLTYPE_FIXED 0 /* Newline is a fixed length string */ +#define NLTYPE_ANY 1 /* Newline is any Unicode line ending */ +#define NLTYPE_ANYCRLF 2 /* Newline is CR, LF, or CRLF */ + +/* This macro checks for a newline at the given position */ + +#define IS_NEWLINE(p) \ + ((NLBLOCK->nltype != NLTYPE_FIXED)? \ + ((p) < NLBLOCK->PSEND && \ + PRIV(is_newline)((p), NLBLOCK->nltype, NLBLOCK->PSEND, \ + &(NLBLOCK->nllen), utf)) \ + : \ + ((p) <= NLBLOCK->PSEND - NLBLOCK->nllen && \ + UCHAR21TEST(p) == NLBLOCK->nl[0] && \ + (NLBLOCK->nllen == 1 || UCHAR21TEST(p+1) == NLBLOCK->nl[1]) \ + ) \ + ) + +/* This macro checks for a newline immediately preceding the given position */ + +#define WAS_NEWLINE(p) \ + ((NLBLOCK->nltype != NLTYPE_FIXED)? \ + ((p) > NLBLOCK->PSSTART && \ + PRIV(was_newline)((p), NLBLOCK->nltype, NLBLOCK->PSSTART, \ + &(NLBLOCK->nllen), utf)) \ + : \ + ((p) >= NLBLOCK->PSSTART + NLBLOCK->nllen && \ + UCHAR21TEST(p - NLBLOCK->nllen) == NLBLOCK->nl[0] && \ + (NLBLOCK->nllen == 1 || UCHAR21TEST(p - NLBLOCK->nllen + 1) == NLBLOCK->nl[1]) \ + ) \ + ) + /* Private flags containing information about the compiled pattern. The first three must not be changed, because whichever is set is actually the number of bytes in a code unit in that mode. */ @@ -296,16 +527,55 @@ bytes in a code unit in that mode. */ #define PCRE2_MODE_MASK (PCRE2_MODE8 | PCRE2_MODE16 | PCRE2_MODE32) - /* Magic number to provide a small check against being handed junk. */ #define MAGIC_NUMBER 0x50435245UL /* 'PCRE' */ -/* This variable is used to detect a loaded regular expression -in different endianness. */ +/* This value is used to detect a loaded regular expression in different +endianness. */ #define REVERSED_MAGIC_NUMBER 0x45524350UL /* 'ERCP' */ +/* The maximum remaining length of subject we are prepared to search for a +req_unit match. */ + +#define REQ_UNIT_MAX 1000 + +/* Bit definitions for entries in the pcre_ctypes table. */ + +#define ctype_space 0x01 +#define ctype_letter 0x02 +#define ctype_digit 0x04 +#define ctype_xdigit 0x08 +#define ctype_word 0x10 /* alphanumeric or '_' */ +#define ctype_meta 0x80 /* regexp meta char or zero (end pattern) */ + +/* Offsets for the bitmap tables in pcre_cbits. Each table contains a set +of bits for a class map. Some classes are built by combining these tables. */ + +#define cbit_space 0 /* [:space:] or \s */ +#define cbit_xdigit 32 /* [:xdigit:] */ +#define cbit_digit 64 /* [:digit:] or \d */ +#define cbit_upper 96 /* [:upper:] */ +#define cbit_lower 128 /* [:lower:] */ +#define cbit_word 160 /* [:word:] or \w */ +#define cbit_graph 192 /* [:graph:] */ +#define cbit_print 224 /* [:print:] */ +#define cbit_punct 256 /* [:punct:] */ +#define cbit_cntrl 288 /* [:cntrl:] */ +#define cbit_length 320 /* Length of the cbits table */ + +/* Offsets of the various tables from the base tables pointer, and +total length. */ + +#define lcc_offset 0 +#define fcc_offset 256 +#define cbits_offset 512 +#define ctypes_offset (cbits_offset + cbit_length) +#define tables_length (ctypes_offset + 256) + + + /* -------------------- Character and string names ------------------------ */ @@ -1432,6 +1702,17 @@ typedef struct pcre2_memctl { void *memory_data; } pcre2_memctl; +/* Structure for building a chain of open capturing subpatterns during +compiling, so that instructions to close them can be compiled when (*ACCEPT) is +encountered. This is also used to identify subpatterns that contain recursive +back references to themselves, so that they can be made atomic. */ + +typedef struct open_capitem { + struct open_capitem *next; /* Chain link */ + uint16_t number; /* Capture number */ + uint16_t flag; /* Set TRUE if recursive back ref */ +} open_capitem; + /* Layout of the UCP type table that translates property names into types and codes. Each entry used to point directly to a name, but to reduce the number of relocations in shared libraries, it now has an offset into a single string @@ -1481,13 +1762,52 @@ extern const int PRIV(ucp_typerange)[]; /* ----------------- Items that need PCRE2_CODE_UNIT_WIDTH ----------------- */ +/* When this file is included by pcre2test, PCRE2_CODE_UNIT_WIDTH is not +defined, so the following items are omitted. */ + #ifdef PCRE2_CODE_UNIT_WIDTH -/* Mode-dependent macros and private structures are defined in a separate file. -When compiling the library, PCRE2_CODE_UNIT_WIDTH will be defined, and we -include them at the appropriate width. When compiling pcre2test, however, that -macro is not set at this point because pcre2test needs to include them at all -supported widths. */ +/* This is the largest non-UTF code point. */ + +#define MAX_NON_UTF_CHAR (0xffffffffU >> (32 - PCRE2_CODE_UNIT_WIDTH)) + + +/* Internal shared data tables. These are tables that are used by more than one +of the exported public functions. They have to be "external" in the C sense, +but are not part of the PCRE2 public API. The data for these tables is in the +pcre2_tables.c module. Even though some of them are identical in each library, +they must have different names so that more than one library can be linked with +an application. UTF-8 tables are needed only when compiling the 8-bit library. +*/ + +#if PCRE2_CODE_UNIT_WIDTH == 8 +extern const int PRIV(utf8_table1)[]; +extern const int PRIV(utf8_table1_size); +extern const int PRIV(utf8_table2)[]; +extern const int PRIV(utf8_table3)[]; +extern const uint8_t PRIV(utf8_table4)[]; +#endif + +extern const uint8_t PRIV(default_tables)[]; +extern const uint8_t PRIV(OP_lengths)[]; + +extern const uint32_t PRIV(hspace_list)[]; +extern const uint32_t PRIV(vspace_list)[]; + +extern const ucp_type_table PRIV(utt)[]; +extern const char PRIV(utt_names)[]; +extern const size_t PRIV(utt_size); + + +/* Mode-dependent macros and hidden and private structures are defined in a +separate file so that pcre2test can include them at all supported widths. When +compiling the library, PCRE2_CODE_UNIT_WIDTH will be defined, and we can +include them at the appropriate width, after setting up suffix macros for the +private structures. */ + +#define compile_data PCRE2_SUFFIX(compile_data_) +#define branch_chain PCRE2_SUFFIX(branch_chain_) +#define named_group PCRE2_SUFFIX(named_group_) #include "pcre2_intmodedep.h" @@ -1498,14 +1818,32 @@ from pcre2test, and must not be defined when no code unit width is available. */ #define _pcre2_compile_context_init PCRE2_SUFFIX(_pcre2_compile_context_init_) +#define _pcre2_find_bracket PCRE2_SUFFIX(_pcre2_find_bracket_) +#define _pcre2_is_newline PCRE2_SUFFIX(_pcre2_is_newline_) #define _pcre2_match_context_init PCRE2_SUFFIX(_pcre2_match_context_init_) #define _pcre2_memctl_malloc PCRE2_SUFFIX(_pcre2_memctl_malloc_) +#define _pcre2_ord2utf PCRE2_SUFFIX(_pcre2_ord2utf_) #define _pcre2_strcmp PCRE2_SUFFIX(_pcre_strcmp_) +#define _pcre2_strcmp_c8 PCRE2_SUFFIX(_pcre_strcmp_c8_) +#define _pcre2_strlen PCRE2_SUFFIX(_pcre_strlen_) +#define _pcre2_strncmp PCRE2_SUFFIX(_pcre_strncmp_) +#define _pcre2_strncmp_c8 PCRE2_SUFFIX(_pcre_strncmp_c8_) +#define _pcre2_valid_utf PCRE2_SUFFIX(_pcre_valid_utf_) +#define _pcre2_was_newline PCRE2_SUFFIX(_pcre2_was_newline_) -extern void _pcre2_compile_context_init(pcre2_compile_context *, BOOL); -extern void _pcre2_match_context_init(pcre2_match_context *, BOOL); -extern void *_pcre2_memctl_malloc(size_t, size_t, pcre2_memctl *); -extern int _pcre2_strcmp(PCRE2_SPTR, PCRE2_SPTR); -#endif +extern void _pcre2_compile_context_init(pcre2_compile_context *, BOOL); +extern PCRE2_SPTR _pcre2_find_bracket(PCRE2_SPTR, BOOL, int); +extern BOOL _pcre2_is_newline(PCRE2_SPTR, int, PCRE2_SPTR, int *, BOOL); +extern void _pcre2_match_context_init(pcre2_match_context *, BOOL); +extern void *_pcre2_memctl_malloc(size_t, size_t, pcre2_memctl *); +extern unsigned int _pcre2_ord2utf(uint32_t, PCRE2_UCHAR *); +extern int _pcre2_strcmp(PCRE2_SPTR, PCRE2_SPTR); +extern int _pcre2_strcmp_c8(PCRE2_SPTR, const char *); +extern int _pcre2_strlen(PCRE2_SPTR); +extern int _pcre2_strncmp(PCRE2_SPTR, PCRE2_SPTR, size_t); +extern int _pcre2_strncmp_c8(PCRE2_SPTR, const char *, size_t); +extern int _pcre2_valid_utf(PCRE2_SPTR, int, size_t *); +extern BOOL _pcre2_was_newline(PCRE2_SPTR, int, PCRE2_SPTR, int *, BOOL); +#endif /* PCRE2_CODE_UNIT_WIDTH */ /* End of pcre2_internal.h */ diff --git a/src/pcre2_intmodedep.h b/src/pcre2_intmodedep.h index b86e91b..47e094a 100644 --- a/src/pcre2_intmodedep.h +++ b/src/pcre2_intmodedep.h @@ -42,21 +42,44 @@ POSSIBILITY OF SUCH DAMAGE. /* This module contains mode-dependent macro and structure definitions. The file is #included by pcre2_internal.h if PCRE2_CODE_UNIT_WIDTH is defined. These mode-dependent items are kept in a separate file so that they can also be -#included multiple times for different code unit widths by pcre2test. Start by -undefining all the new macros defined herein so that they can be redefined for -multiple inclusions. */ +#included multiple times for different code unit widths by pcre2test in order +to have access to the hidden structures at all supported widths. +Some of the mode-dependent macros are required at different widths for +different parts of the pcre2test code (in particular, the included +pcre_printint.c file). We undefine them here so that they can be re-defined for +multiple inclusions. Not all of these are used in pcretest, but it's easier +just to undefine them all. */ + +#undef ACROSSCHAR +#undef BACKCHAR #undef CU2BYTES +#undef FORWARDCHAR #undef GET #undef GET2 +#undef GETCHAR +#undef GETCHARINC +#undef GETCHARINCTEST +#undef GETCHARLEN +#undef GETCHARLENTEST +#undef GETCHARTEST +#undef GET_EXTRALEN +#undef HAS_EXTRALEN #undef IMM2_SIZE +#undef MAX_255 +#undef MAX_MARK #undef MAX_PATTERN_SIZE +#undef MAX_UTF_SINGLE_CU +#undef NOT_FIRSTCHAR #undef PUT #undef PUT2 +#undef PUT2INC +#undef PUTCHAR #undef PUTINC -/* ---------------------------MACROS ----------------------------- */ + +/* -------------------------- MACROS ----------------------------- */ /* PCRE keeps offsets in its compiled code as at least 16-bit quantities (always stored in big-endian order in 8-bit mode) by default. These are used, @@ -70,7 +93,6 @@ unit string is now handled by the macros that are defined here. The macros are controlled by the value of LINK_SIZE. This defaults to 2, but values of 2 or 4 are also supported. */ - /* ------------------- 8-bit support ------------------ */ #if PCRE2_CODE_UNIT_WIDTH == 8 @@ -150,8 +172,8 @@ values of 2 or 4 are also supported. */ #error Unsupported compiling mode #endif -/* -------------------------------------------------------*/ +/* --------------- Other mode-specific macros ----------------- */ /* PCRE uses some other (at least) 16-bit quantities that do not change when the size of offsets changes. There are used for repeat counts and for other @@ -166,7 +188,7 @@ arithmetic results in a signed value. Hence the cast. */ #if PCRE2_CODE_UNIT_WIDTH == 8 #define IMM2_SIZE 2 #define GET2(a,n) (unsigned int)(((a)[n] << 8) | (a)[(n)+1]) -#define PUT2(a,n,d) { a[n] = (d) >> 8; a[(n)+1] = (d) & 255; } +#define PUT2(a,n,d) a[n] = (d) >> 8, a[(n)+1] = (d) & 255 #else /* Code units are 16 or 32 bits */ #define IMM2_SIZE 1 @@ -174,14 +196,338 @@ arithmetic results in a signed value. Hence the cast. */ #define PUT2(a,n,d) a[n] = d #endif +/* Other macros that are different for 8-bit mode. The maximum length of a MARK +name must fit in one code unit; currently it is set to 255 or 65535. */ + +#if PCRE2_CODE_UNIT_WIDTH == 8 +#define MAX_255(c) TRUE +#define MAX_MARK ((1u << 8) - 1) +#ifdef SUPPORT_UTF +#define SUPPORT_WIDE_CHARS +#endif /* SUPPORT_UTF */ + +#else /* Code units are 16 or 32 bits */ +#define MAX_255(c) ((c) <= 255u) +#define MAX_MARK ((1u << 16) - 1) +#define SUPPORT_WIDE_CHARS +#endif + + + +/* ----------------- Character-handling macros ----------------- */ + +/* There is a proposed future special "UTF-21" mode, in which only the lowest +21 bits of a 32-bit character are interpreted as UTF, with the remaining 11 +high-order bits available to the application for other uses. In preparation for +the future implementation of this mode, there are macros that load a data item +and, if in this special mode, mask it to 21 bits. These macros all have names +starting with UCHAR21. In all other modes, including the normal 32-bit +library, the macros all have the same simple definitions. When the new mode is +implemented, it is expected that these definitions will be varied appropriately +using #ifdef when compiling the library that supports the special mode. */ + +#define UCHAR21(eptr) (*(eptr)) +#define UCHAR21TEST(eptr) (*(eptr)) +#define UCHAR21INC(eptr) (*(eptr)++) +#define UCHAR21INCTEST(eptr) (*(eptr)++) + +/* When UTF encoding is being used, a character is no longer just a single +byte in 8-bit mode or a single short in 16-bit mode. The macros for character +handling generate simple sequences when used in the basic mode, and more +complicated ones for UTF characters. GETCHARLENTEST and other macros are not +used when UTF is not supported. To make sure they can never even appear when +UTF support is omitted, we don't even define them. */ + +#ifndef SUPPORT_UTF + +/* #define MAX_UTF_SINGLE_CU */ +/* #define HAS_EXTRALEN(c) */ +/* #define GET_EXTRALEN(c) */ +/* #define NOT_FIRSTCHAR(c) */ +#define GETCHAR(c, eptr) c = *eptr; +#define GETCHARTEST(c, eptr) c = *eptr; +#define GETCHARINC(c, eptr) c = *eptr++; +#define GETCHARINCTEST(c, eptr) c = *eptr++; +#define GETCHARLEN(c, eptr, len) c = *eptr; +#define PUTCHAR(c, p) (*p = c, 1) +/* #define GETCHARLENTEST(c, eptr, len) */ +/* #define BACKCHAR(eptr) */ +/* #define FORWARDCHAR(eptr) */ +/* #define ACROSSCHAR(condition, eptr, action) */ + +#else /* SUPPORT_UTF */ + +/* ------------------- 8-bit support ------------------ */ + +#if PCRE2_CODE_UNIT_WIDTH == 8 + +/* The largest UTF code point that can be encoded as a single code unit. */ + +#define MAX_UTF_SINGLE_CU 127 + +/* Tests whether the code point needs extra characters to decode. */ + +#define HAS_EXTRALEN(c) HASUTF8EXTRALEN(c) + +/* Returns with the additional number of characters if IS_MULTICHAR(c) is TRUE. +Otherwise it has an undefined behaviour. */ + +#define GET_EXTRALEN(c) (PRIV(utf8_table4)[(c) & 0x3f]) + +/* Returns TRUE, if the given character is not the first character +of a UTF sequence. */ + +#define NOT_FIRSTCHAR(c) (((c) & 0xc0) == 0x80) + +/* Get the next UTF-8 character, not advancing the pointer. This is called when +we know we are in UTF-8 mode. */ + +#define GETCHAR(c, eptr) \ + c = *eptr; \ + if (c >= 0xc0) GETUTF8(c, eptr); + +/* Get the next UTF-8 character, testing for UTF-8 mode, and not advancing the +pointer. */ + +#define GETCHARTEST(c, eptr) \ + c = *eptr; \ + if (utf && c >= 0xc0) GETUTF8(c, eptr); + +/* Get the next UTF-8 character, advancing the pointer. This is called when we +know we are in UTF-8 mode. */ + +#define GETCHARINC(c, eptr) \ + c = *eptr++; \ + if (c >= 0xc0) GETUTF8INC(c, eptr); + +/* Get the next character, testing for UTF-8 mode, and advancing the pointer. +This is called when we don't know if we are in UTF-8 mode. */ + +#define GETCHARINCTEST(c, eptr) \ + c = *eptr++; \ + if (utf && c >= 0xc0) GETUTF8INC(c, eptr); + +/* Get the next UTF-8 character, not advancing the pointer, incrementing length +if there are extra bytes. This is called when we know we are in UTF-8 mode. */ + +#define GETCHARLEN(c, eptr, len) \ + c = *eptr; \ + if (c >= 0xc0) GETUTF8LEN(c, eptr, len); + +/* Get the next UTF-8 character, testing for UTF-8 mode, not advancing the +pointer, incrementing length if there are extra bytes. This is called when we +do not know if we are in UTF-8 mode. */ + +#define GETCHARLENTEST(c, eptr, len) \ + c = *eptr; \ + if (utf && c >= 0xc0) GETUTF8LEN(c, eptr, len); + +/* If the pointer is not at the start of a character, move it back until +it is. This is called only in UTF-8 mode - we don't put a test within the macro +because almost all calls are already within a block of UTF-8 only code. */ + +#define BACKCHAR(eptr) while((*eptr & 0xc0) == 0x80) eptr-- + +/* Same as above, just in the other direction. */ +#define FORWARDCHAR(eptr) while((*eptr & 0xc0) == 0x80) eptr++ + +/* Same as above, but it allows a fully customizable form. */ +#define ACROSSCHAR(condition, eptr, action) \ + while((condition) && ((eptr) & 0xc0) == 0x80) action + +/* Deposit a character into memory, returning the number of code units. */ + +#define PUTCHAR(c, p) ((utf && c > MAX_UTF_SINGLE_CU)? \ + PRIV(ord2utf)(c,p) : (*p = c, 1)) + + +/* ------------------- 16-bit support ------------------ */ + +#elif PCRE2_CODE_UNIT_WIDTH == 16 + +/* The largest UTF code point that can be encoded as a single code unit. */ + +#define MAX_UTF_SINGLE_CU 65535 + +/* Tests whether the code point needs extra characters to decode. */ + +#define HAS_EXTRALEN(c) (((c) & 0xfc00) == 0xd800) + +/* Returns with the additional number of characters if IS_MULTICHAR(c) is TRUE. +Otherwise it has an undefined behaviour. */ + +#define GET_EXTRALEN(c) 1 + +/* Returns TRUE, if the given character is not the first character +of a UTF sequence. */ + +#define NOT_FIRSTCHAR(c) (((c) & 0xfc00) == 0xdc00) + +/* Base macro to pick up the low surrogate of a UTF-16 character, not +advancing the pointer. */ + +#define GETUTF16(c, eptr) \ + { c = (((c & 0x3ff) << 10) | (eptr[1] & 0x3ff)) + 0x10000; } + +/* Get the next UTF-16 character, not advancing the pointer. This is called when +we know we are in UTF-16 mode. */ + +#define GETCHAR(c, eptr) \ + c = *eptr; \ + if ((c & 0xfc00) == 0xd800) GETUTF16(c, eptr); + +/* Get the next UTF-16 character, testing for UTF-16 mode, and not advancing the +pointer. */ + +#define GETCHARTEST(c, eptr) \ + c = *eptr; \ + if (utf && (c & 0xfc00) == 0xd800) GETUTF16(c, eptr); + +/* Base macro to pick up the low surrogate of a UTF-16 character, advancing +the pointer. */ + +#define GETUTF16INC(c, eptr) \ + { c = (((c & 0x3ff) << 10) | (*eptr++ & 0x3ff)) + 0x10000; } + +/* Get the next UTF-16 character, advancing the pointer. This is called when we +know we are in UTF-16 mode. */ + +#define GETCHARINC(c, eptr) \ + c = *eptr++; \ + if ((c & 0xfc00) == 0xd800) GETUTF16INC(c, eptr); + +/* Get the next character, testing for UTF-16 mode, and advancing the pointer. +This is called when we don't know if we are in UTF-16 mode. */ + +#define GETCHARINCTEST(c, eptr) \ + c = *eptr++; \ + if (utf && (c & 0xfc00) == 0xd800) GETUTF16INC(c, eptr); + +/* Base macro to pick up the low surrogate of a UTF-16 character, not +advancing the pointer, incrementing the length. */ + +#define GETUTF16LEN(c, eptr, len) \ + { c = (((c & 0x3ff) << 10) | (eptr[1] & 0x3ff)) + 0x10000; len++; } + +/* Get the next UTF-16 character, not advancing the pointer, incrementing +length if there is a low surrogate. This is called when we know we are in +UTF-16 mode. */ + +#define GETCHARLEN(c, eptr, len) \ + c = *eptr; \ + if ((c & 0xfc00) == 0xd800) GETUTF16LEN(c, eptr, len); + +/* Get the next UTF-816character, testing for UTF-16 mode, not advancing the +pointer, incrementing length if there is a low surrogate. This is called when +we do not know if we are in UTF-16 mode. */ + +#define GETCHARLENTEST(c, eptr, len) \ + c = *eptr; \ + if (utf && (c & 0xfc00) == 0xd800) GETUTF16LEN(c, eptr, len); + +/* If the pointer is not at the start of a character, move it back until +it is. This is called only in UTF-16 mode - we don't put a test within the +macro because almost all calls are already within a block of UTF-16 only +code. */ + +#define BACKCHAR(eptr) if ((*eptr & 0xfc00) == 0xdc00) eptr-- + +/* Same as above, just in the other direction. */ +#define FORWARDCHAR(eptr) if ((*eptr & 0xfc00) == 0xdc00) eptr++ + +/* Same as above, but it allows a fully customizable form. */ +#define ACROSSCHAR(condition, eptr, action) \ + if ((condition) && ((eptr) & 0xfc00) == 0xdc00) action + +/* Deposit a character into memory, returning the number of code units. */ + +#define PUTCHAR(c, p) ((utf && c > MAX_UTF_SINGLE_CU)? \ + PRIV(ord2utf)(c,p) : (*p = c, 1)) + + +/* ------------------- 32-bit support ------------------ */ + +#else + +/* These are trivial for the 32-bit library, since all UTF-32 characters fit +into one PCRE_UCHAR unit. */ + +#define MAX_UTF_SINGLE_CU (0x10ffffu) +#define HAS_EXTRALEN(c) (0) +#define GET_EXTRALEN(c) (0) +#define NOT_FIRSTCHAR(c) (0) + +/* Get the next UTF-32 character, not advancing the pointer. This is called when +we know we are in UTF-32 mode. */ + +#define GETCHAR(c, eptr) \ + c = *(eptr); + +/* Get the next UTF-32 character, testing for UTF-32 mode, and not advancing the +pointer. */ + +#define GETCHARTEST(c, eptr) \ + c = *(eptr); + +/* Get the next UTF-32 character, advancing the pointer. This is called when we +know we are in UTF-32 mode. */ + +#define GETCHARINC(c, eptr) \ + c = *((eptr)++); + +/* Get the next character, testing for UTF-32 mode, and advancing the pointer. +This is called when we don't know if we are in UTF-32 mode. */ + +#define GETCHARINCTEST(c, eptr) \ + c = *((eptr)++); + +/* Get the next UTF-32 character, not advancing the pointer, not incrementing +length (since all UTF-32 is of length 1). This is called when we know we are in +UTF-32 mode. */ + +#define GETCHARLEN(c, eptr, len) \ + GETCHAR(c, eptr) + +/* Get the next UTF-32character, testing for UTF-32 mode, not advancing the +pointer, not incrementing the length (since all UTF-32 is of length 1). +This is called when we do not know if we are in UTF-32 mode. */ + +#define GETCHARLENTEST(c, eptr, len) \ + GETCHARTEST(c, eptr) + +/* If the pointer is not at the start of a character, move it back until +it is. This is called only in UTF-32 mode - we don't put a test within the +macro because almost all calls are already within a block of UTF-32 only +code. + +These are all no-ops since all UTF-32 characters fit into one pcre_uchar. */ + +#define BACKCHAR(eptr) do { } while (0) + +/* Same as above, just in the other direction. */ + +#define FORWARDCHAR(eptr) do { } while (0) + +/* Same as above, but it allows a fully customizable form. */ + +#define ACROSSCHAR(condition, eptr, action) do { } while (0) + +/* Deposit a character into memory, returning the number of code units. */ + +#define PUTCHAR(c, p) (*p = c, 1) + +#endif /* UTF-32 character handling */ +#endif /* SUPPORT_UTF */ + /* Mode-dependent macros that have the same definition in all modes. */ #define CU2BYTES(x) (x)*((PCRE2_CODE_UNIT_WIDTH/8)) #define PUTINC(a,n,d) PUT(a,n,d), a += LINK_SIZE +#define PUT2INC(a,n,d) PUT2(a,n,d), a += IMM2_SIZE -/* --------------------------- STRUCTURES ----------------------------- */ +/* ----------------------- HIDDEN STRUCTURES ----------------------------- */ /* The real general context structure. At present it hold only data for custom memory control. */ @@ -195,7 +541,7 @@ typedef struct pcre2_real_general_context { typedef struct pcre2_real_compile_context { pcre2_memctl memctl; int (*stack_guard)(uint32_t); - const unsigned char *tables; + const uint8_t *tables; uint16_t bsr_convention; uint16_t newline_convention; uint32_t parens_nest_limit; @@ -217,11 +563,12 @@ typedef struct pcre2_real_match_context { /* The real compiled code structure */ typedef struct pcre2_real_code { - pcre2_memctl memctl; + pcre2_memctl memctl; /* Memory control fields */ + const uint8_t *tables; /* The character tables */ void *executable_jit; /* Pointer to JIT code */ uint8_t start_bitmap[32]; /* Bitmap for starting code unit < 256 */ + size_t blocksize; /* Total (bytes) that was malloc-ed */ uint32_t magic_number; /* Paranoid and endianness check */ - uint32_t size; /* Total (bytes) that was malloc-ed */ uint32_t compile_options; /* Options passed to pcre2_compile() */ uint32_t pattern_options; /* Options taken from the pattern */ uint32_t flags; /* Various state flags */ @@ -239,14 +586,13 @@ typedef struct pcre2_real_code { uint16_t name_count; /* Number of name entries in the table */ } pcre2_real_code; -/* The reat match data structure. */ +/* The real match data structure. */ typedef struct pcre2_real_match_data { pcre2_memctl memctl; const pcre2_real_code *code; /* The pattern used for the match */ PCRE2_SPTR subject; /* The subject that was matched */ int rc; /* The return code from the match */ - int utf_reason; /* Reason code for bad UTF */ size_t leftchar; /* Offset to leftmost code unit */ size_t rightchar; /* Offset to rightmost code unit */ size_t startchar; /* Offset to starting code unit */ @@ -255,4 +601,71 @@ typedef struct pcre2_real_match_data { size_t ovector[1]; /* The first field */ } pcre2_real_match_data; + +/* ----------------------- PRIVATE STRUCTURES ----------------------------- */ + +/* These structures are not needed for pcre2test. */ + +#ifndef PCRE2_PCRE2TEST + +/* Structure for maintaining a chain of pointers to the currently incomplete +branches, for testing for left recursion while compiling. */ + +typedef struct branch_chain { + struct branch_chain *outer; + PCRE2_UCHAR *current_branch; +} branch_chain; + +/* Structure for building a list of named groups during the first pass of +compiling. */ + +typedef struct named_group { + PCRE2_SPTR name; /* Points to the name in the pattern */ + int length; /* Length of the name */ + uint32_t number; /* Group number */ +} named_group; + +/* Structure for passing "static" information around between the functions +doing the compiling, so that they are thread-safe. */ + +typedef struct compile_data { + pcre2_real_compile_context *cx; /* Points to the compile context */ + const uint8_t *lcc; /* Points to lower casing table */ + const uint8_t *fcc; /* Points to case-flipping table */ + const uint8_t *cbits; /* Points to character type table */ + const uint8_t *ctypes; /* Points to table of type maps */ + PCRE2_SPTR start_workspace; /* The start of working space */ + PCRE2_SPTR start_code; /* The start of the compiled code */ + PCRE2_SPTR start_pattern; /* The start of the pattern */ + PCRE2_SPTR end_pattern; /* The end of the pattern */ + PCRE2_UCHAR *hwm; /* High watermark of workspace */ + open_capitem *open_caps; /* Chain of open capture items */ + named_group *named_groups; /* Points to vector in pre-compile */ + PCRE2_UCHAR *name_table; /* The name/number table */ + int names_found; /* Number of entries so far */ + int name_entry_size; /* Size of each entry */ + int named_group_list_size; /* Number of entries in the list */ + int workspace_size; /* Size of workspace */ + unsigned int bracount; /* Count of capturing parens as we compile */ + int final_bracount; /* Saved value after first pass */ + int max_lookbehind; /* Maximum lookbehind (characters) */ + int top_backref; /* Maximum back reference */ + unsigned int backref_map; /* Bitmap of low back refs */ + unsigned int namedrefcount; /* Number of backreferences by name */ + int parens_depth; /* Depth of nested parentheses */ + int assert_depth; /* Depth of nested assertions */ + uint32_t external_options; /* External (initial) options */ + uint32_t external_flags; /* External flag bits to be set */ + int req_varyopt; /* "After variable item" flag for reqbyte */ + BOOL had_accept; /* (*ACCEPT) encountered */ + BOOL had_pruneorskip; /* (*PRUNE) or (*SKIP) encountered */ + BOOL check_lookbehind; /* Lookbehinds need later checking */ + BOOL dupnames; /* Duplicate names exist */ + int nltype; /* Newline type */ + int nllen; /* Newline string length */ + PCRE2_UCHAR nl[4]; /* Newline string when fixed length */ +} compile_data; + +#endif /* PCRE2_PCRE2TEST */ + /* End of pcre2_intmodedep.h */ diff --git a/src/pcre2_newline.c b/src/pcre2_newline.c new file mode 100644 index 0000000..1110db7 --- /dev/null +++ b/src/pcre2_newline.c @@ -0,0 +1,213 @@ +/************************************************* +* Perl-Compatible Regular Expressions * +*************************************************/ + +/* PCRE is a library of functions to support regular expressions whose syntax +and semantics are as close as possible to those of the Perl 5 language. + + Written by Philip Hazel + Original API code Copyright (c) 1997-2012 University of Cambridge + New API code Copyright (c) 2014 University of Cambridge + +----------------------------------------------------------------------------- +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + + * Redistributions of source code must retain the above copyright notice, + this list of conditions and the following disclaimer. + + * Redistributions in binary form must reproduce the above copyright + notice, this list of conditions and the following disclaimer in the + documentation and/or other materials provided with the distribution. + + * Neither the name of the University of Cambridge nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE +ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE +LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR +CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF +SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS +INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN +CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) +ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE +POSSIBILITY OF SUCH DAMAGE. +----------------------------------------------------------------------------- +*/ + + +/* This module contains internal functions for testing newlines when more than +one kind of newline is to be recognized. When a newline is found, its length is +returned. In principle, we could implement several newline "types", each +referring to a different set of newline characters. At present, PCRE2 supports +only NLTYPE_FIXED, which gets handled without these functions, NLTYPE_ANYCRLF, +and NLTYPE_ANY. The full list of Unicode newline characters is taken from +http://unicode.org/unicode/reports/tr18/. */ + + +#ifdef HAVE_CONFIG_H +#include "config.h" +#endif + +#include "pcre2_internal.h" + + + +/************************************************* +* Check for newline at given position * +*************************************************/ + +/* It is guaranteed that the initial value of ptr is less than the end of the +string that is being processed. + +Arguments: + ptr pointer to possible newline + type the newline type + endptr pointer to the end of the string + lenptr where to return the length + utf TRUE if in utf mode + +Returns: TRUE or FALSE +*/ + +BOOL +PRIV(is_newline)(PCRE2_SPTR ptr, int type, PCRE2_SPTR endptr, int *lenptr, + BOOL utf) +{ +uint32_t c; + +#ifdef SUPPORT_UTF +if (utf) { GETCHAR(c, ptr); } else +#else +(void)utf; +#endif /* SUPPORT_UTF */ + + c = *ptr; + +/* Note that this function is called only for ANY or ANYCRLF. */ + +if (type == NLTYPE_ANYCRLF) switch(c) + { + case CHAR_LF: *lenptr = 1; return TRUE; + case CHAR_CR: *lenptr = (ptr < endptr - 1 && ptr[1] == CHAR_LF)? 2 : 1; + return TRUE; + default: return FALSE; + } + +/* NLTYPE_ANY */ + +else switch(c) + { +#ifdef EBCDIC + case CHAR_NEL: +#endif + case CHAR_LF: + case CHAR_VT: + case CHAR_FF: *lenptr = 1; return TRUE; + + case CHAR_CR: + *lenptr = (ptr < endptr - 1 && ptr[1] == CHAR_LF)? 2 : 1; + return TRUE; + +#ifndef EBCDIC +#if PCRE2_CODE_UNIT_WIDTH == 8 + case CHAR_NEL: *lenptr = utf? 2 : 1; return TRUE; + case 0x2028: /* LS */ + case 0x2029: *lenptr = 3; return TRUE; /* PS */ +#else /* 16-bit or 32-bit code units */ + case CHAR_NEL: + case 0x2028: /* LS */ + case 0x2029: *lenptr = 1; return TRUE; /* PS */ +#endif +#endif /* Not EBCDIC */ + + default: return FALSE; + } +} + + + +/************************************************* +* Check for newline at previous position * +*************************************************/ + +/* It is guaranteed that the initial value of ptr is greater than the start of +the string that is being processed. + +Arguments: + ptr pointer to possible newline + type the newline type + startptr pointer to the start of the string + lenptr where to return the length + utf TRUE if in utf mode + +Returns: TRUE or FALSE +*/ + +BOOL +PRIV(was_newline)(PCRE2_SPTR ptr, int type, PCRE2_SPTR startptr, int *lenptr, + BOOL utf) +{ +uint32_t c; +ptr--; + +#ifdef SUPPORT_UTF +if (utf) + { + BACKCHAR(ptr); + GETCHAR(c, ptr); + } +else +#else +(void)utf; +#endif /* SUPPORT_UTF */ + + c = *ptr; + +/* Note that this function is called only for ANY or ANYCRLF. */ + +if (type == NLTYPE_ANYCRLF) switch(c) + { + case CHAR_LF: + *lenptr = (ptr > startptr && ptr[-1] == CHAR_CR)? 2 : 1; + return TRUE; + + case CHAR_CR: *lenptr = 1; return TRUE; + default: return FALSE; + } + +/* NLTYPE_ANY */ + +else switch(c) + { + case CHAR_LF: + *lenptr = (ptr > startptr && ptr[-1] == CHAR_CR)? 2 : 1; + return TRUE; + +#ifdef EBCDIC + case CHAR_NEL: +#endif + case CHAR_VT: + case CHAR_FF: + case CHAR_CR: *lenptr = 1; return TRUE; + +#ifndef EBCDIC +#if PCRE2_CODE_UNIT_WIDTH == 8 + case CHAR_NEL: *lenptr = utf? 2 : 1; return TRUE; + case 0x2028: /* LS */ + case 0x2029: *lenptr = 3; return TRUE; /* PS */ +#else /* 16-bit or 32-bit code units */ + case CHAR_NEL: + case 0x2028: /* LS */ + case 0x2029: *lenptr = 1; return TRUE; /* PS */ +#endif +#endif /* Not EBCDIC */ + + default: return FALSE; + } +} + +/* End of pcre2_newline.c */ diff --git a/src/pcre2_ord2utf.c b/src/pcre2_ord2utf.c new file mode 100644 index 0000000..1dc5b14 --- /dev/null +++ b/src/pcre2_ord2utf.c @@ -0,0 +1,119 @@ +/************************************************* +* Perl-Compatible Regular Expressions * +*************************************************/ + +/* PCRE is a library of functions to support regular expressions whose syntax +and semantics are as close as possible to those of the Perl 5 language. + + Written by Philip Hazel + Original API code Copyright (c) 1997-2012 University of Cambridge + New API code Copyright (c) 2014 University of Cambridge + +----------------------------------------------------------------------------- +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + + * Redistributions of source code must retain the above copyright notice, + this list of conditions and the following disclaimer. + + * Redistributions in binary form must reproduce the above copyright + notice, this list of conditions and the following disclaimer in the + documentation and/or other materials provided with the distribution. + + * Neither the name of the University of Cambridge nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE +ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE +LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR +CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF +SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS +INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN +CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) +ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE +POSSIBILITY OF SUCH DAMAGE. +----------------------------------------------------------------------------- +*/ + + +/* This file contains a function that converts a Unicode character code point +into a UTF string. The behaviour is different for each code unit width. */ + + +#ifdef HAVE_CONFIG_H +#include "config.h" +#endif + +#include "pcre2_internal.h" + + +/* If SUPPORT_UTF is not defined, this function will never be called. Supply a +dummy function because some compilers do not like empty source modules. */ + +#ifndef SUPPORT_UTF +unsigned int +PRIV(ord2utf)(uint32_t cvalue, PCRE2_UCHAR *buffer) +{ +(void)(cvalue); +(void)(buffer); +return 0; +} +#else /* SUPPORT_UTF */ + + +/************************************************* +* Convert code point to UTF * +*************************************************/ + +/* +Arguments: + cvalue the character value + buffer pointer to buffer for result + +Returns: number of code units placed in the buffer +*/ + +unsigned int +PRIV(ord2utf)(uint32_t cvalue, PCRE2_UCHAR *buffer) +{ +/* Convert to UTF-8 */ + +#if PCRE2_CODE_UNIT_WIDTH == 8 +register int i, j; +for (i = 0; i < PRIV(utf8_table1_size); i++) + if ((int)cvalue <= PRIV(utf8_table1)[i]) break; +buffer += i; +for (j = i; j > 0; j--) + { + *buffer-- = 0x80 | (cvalue & 0x3f); + cvalue >>= 6; + } +*buffer = PRIV(utf8_table2)[i] | cvalue; +return i + 1; + +/* Convert to UTF-16 */ + +#elif PCRE2_CODE_UNIT_WIDTH == 16 +if (cvalue <= 0xffff) + { + *buffer = (PCRE2_UCHAR)cvalue; + return 1; + } +cvalue -= 0x10000; +*buffer++ = 0xd800 | (cvalue >> 10); +*buffer = 0xdc00 | (cvalue & 0x3ff); +return 2; + +/* Convert to UTF-32 */ + +#else +*buffer = (PCRE2_UCHAR)cvalue; +return 1; +#endif +} +#endif /* SUPPORT_UTF */ + +/* End of pcre_ord2utf.c */ diff --git a/src/pcre2_pattern_info.c b/src/pcre2_pattern_info.c index 555da9b..158c982 100644 --- a/src/pcre2_pattern_info.c +++ b/src/pcre2_pattern_info.c @@ -184,7 +184,7 @@ switch(what) break; case PCRE2_INFO_SIZE: - *((size_t *)where) = re->size; + *((size_t *)where) = re->blocksize; break; default: return PCRE2_ERROR_BADOPTION; diff --git a/src/pcre2_string_utils.c b/src/pcre2_string_utils.c index f4f4de0..2b585d0 100644 --- a/src/pcre2_string_utils.c +++ b/src/pcre2_string_utils.c @@ -53,7 +53,7 @@ functions work only on 8-bit data. */ /************************************************* -* Compare two strings * +* Compare two zero-terminated PCRE2 strings * *************************************************/ /* @@ -77,4 +77,105 @@ while (*str1 != '\0' || *str2 != '\0') return 0; } + +/************************************************* +* Compare zero-terminated PCRE2 & 8-bit strings * +*************************************************/ + +/* +Arguments: + str1 first string + str2 second string + +Returns: 0, 1, or -1 +*/ + +int +PRIV(strcmp_c8)(PCRE2_SPTR str1, const char *str2) +{ +PCRE2_UCHAR c1, c2; +while (*str1 != '\0' || *str2 != '\0') + { + c1 = *str1++; + c2 = *str2++; + if (c1 != c2) return ((c1 > c2) << 1) - 1; + } +return 0; +} + + +/************************************************* +* Compare two PCRE2 strings, given a length * +*************************************************/ + +/* +Arguments: + str1 first string + str2 second string + len the length + +Returns: 0, 1, or -1 +*/ + +int +PRIV(strncmp)(PCRE2_SPTR str1, PCRE2_SPTR str2, size_t len) +{ +PCRE2_UCHAR c1, c2; +while (len-- > 0) + { + c1 = *str1++; + c2 = *str2++; + if (c1 != c2) return ((c1 > c2) << 1) - 1; + } +return 0; +} + + +/************************************************* +* Compare PCRE2 string to 8-bit string by length * +*************************************************/ + +/* As the 8-bit string is almost always a literal, its type is specified as +'const char *'. + +Arguments: + str1 first string + str2 second string + len the length + +Returns: 0, 1, or -1 +*/ + +int +PRIV(strncmp_c8)(PCRE2_SPTR str1, const char *str2, size_t len) +{ +PCRE2_UCHAR c1, c2; +while (len-- > 0) + { + c1 = *str1++; + c2 = *str2++; + if (c1 != c2) return ((c1 > c2) << 1) - 1; + } +return 0; +} + + + +/************************************************* +* Find the length of a string * +*************************************************/ + +/* +Argument: the string +Returns: the length +*/ + +int +PRIV(strlen)(PCRE2_SPTR str) +{ +int c = 0; +while (*str++ != 0) c++; +return c; +} + /* End of pcre2_string_utils.c */ diff --git a/src/pcre2_tables.c b/src/pcre2_tables.c index a8843a8..9d60276 100644 --- a/src/pcre2_tables.c +++ b/src/pcre2_tables.c @@ -41,22 +41,22 @@ POSSIBILITY OF SUCH DAMAGE. /* This module contains some fixed tables that are used by more than one of the PCRE code modules. The tables are also #included by the pcre2test program, which uses macros to change their names from _pcre2_xxx to xxxx, thereby -avoiding name clashes with the library. In this case, PCRE2_INCLUDED is +avoiding name clashes with the library. In this case, PCRE2_PCRE2TEST is defined. */ -#ifndef PCRE2_INCLUDED /* We're compiling the library */ +#ifndef PCRE2_PCRE2TEST /* We're compiling the library */ #ifdef HAVE_CONFIG_H #include "config.h" #endif #include "pcre2_internal.h" -#endif /* PCRE2_INCLUDED */ +#endif /* PCRE2_PCRE2TEST */ /* Table of sizes for the fixed-length opcodes. It's defined in a macro so that -the definition is next to the definition of the opcodes in pcre2_internal.h. +the definition is next to the definition of the opcodes in pcre2_internal.h. This is mode-dependent, so is skipped when this file is included by pcre2test. */ -#ifndef PCRE2_INCLUDED +#ifndef PCRE2_PCRE2TEST const uint8_t PRIV(OP_lengths)[] = { OP_LENGTHS }; #endif @@ -71,14 +71,18 @@ const uint32_t PRIV(vspace_list)[] = { VSPACE_LIST }; * Tables for UTF-8 support * *************************************************/ +/* These tables are required by pcre2test in 16- or 32-bit mode, as well +as for the library in 8-bit mode, because pcre2test uses UTF-8 internally for +handling wide characters. */ + +#if defined PCRE2_PCRE2TEST || \ + (defined SUPPORT_UTF && \ + defined PCRE2_CODE_UNIT_WIDTH && \ + PCRE2_CODE_UNIT_WIDTH == 8) + /* These are the breakpoints for different numbers of bytes in a UTF-8 character. */ -#if (defined SUPPORT_UTF && defined COMPILE_PCRE8) \ - || (defined PCRE2_INCLUDED && (defined SUPPORT_PCRE16 || defined SUPPORT_PCRE32)) - -/* These tables are also required by pcretest in 16- or 32-bit mode. */ - const int PRIV(utf8_table1)[] = { 0x7f, 0x7ff, 0xffff, 0x1fffff, 0x3ffffff, 0x7fffffff}; @@ -99,7 +103,7 @@ const uint8_t PRIV(utf8_table4)[] = { 2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2, 3,3,3,3,3,3,3,3,4,4,4,4,5,5,5,5 }; -#endif /* (SUPPORT_UTF && COMPILE_PCRE8) || (PCRE2_INCLUDED && SUPPORT_PCRE[16|32])*/ +#endif /* UTF-8 support needed */ #ifdef SUPPORT_UTF @@ -653,7 +657,7 @@ const ucp_type_table PRIV(utt)[] = { { 1042, PT_PC, ucp_Zs } }; -const int PRIV(utt_size) = sizeof(PRIV(utt)) / sizeof(ucp_type_table); +const size_t PRIV(utt_size) = sizeof(PRIV(utt)) / sizeof(ucp_type_table); #endif /* SUPPORT_UTF */ diff --git a/src/pcre2_ucd.c b/src/pcre2_ucd.c index 671acd7..eecf892 100644 --- a/src/pcre2_ucd.c +++ b/src/pcre2_ucd.c @@ -8,7 +8,7 @@ table names from _pcre2_xxx to xxxx, thereby avoiding name clashes with the library. At present, just one of these tables is actually needed. */ -#ifndef PCRE2_INCLUDED +#ifndef PCRE2_PCRE2TEST #ifdef HAVE_CONFIG_H #include "config.h" @@ -16,7 +16,7 @@ needed. */ #include "pcre2_internal.h" -#endif /* PCRE2_INCLUDED */ +#endif /* PCRE2_PCRE2TEST */ /* Unicode character database. */ /* This file was autogenerated by the MultiStage2.py script. */ @@ -78,7 +78,7 @@ const uint32_t PRIV(ucd_caseless_sets)[] = { /* When #included in pcre2test, we don't need this large table. */ -#ifndef PCRE2_INCLUDED +#ifndef PCRE2_PCRE2TEST const ucd_record PRIV(ucd_records)[] = { /* 5016 bytes, record size 8 */ { 9, 0, 2, 0, 0, }, /* 0 */ @@ -3295,4 +3295,4 @@ const uint16_t PRIV(ucd_stage2)[] = { /* 51968 bytes, block = 128 */ #endif #endif /* SUPPORT_UTF */ -#endif /* PCRE2_INCLUDED */ +#endif /* PCRE2_PCRE2TEST */ diff --git a/src/pcre2_valid_utf.c b/src/pcre2_valid_utf.c new file mode 100644 index 0000000..ccda391 --- /dev/null +++ b/src/pcre2_valid_utf.c @@ -0,0 +1,399 @@ +/************************************************* +* Perl-Compatible Regular Expressions * +*************************************************/ + +/* PCRE is a library of functions to support regular expressions whose syntax +and semantics are as close as possible to those of the Perl 5 language. + + Written by Philip Hazel + Original API code Copyright (c) 1997-2012 University of Cambridge + New API code Copyright (c) 2014 University of Cambridge + +----------------------------------------------------------------------------- +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + + * Redistributions of source code must retain the above copyright notice, + this list of conditions and the following disclaimer. + + * Redistributions in binary form must reproduce the above copyright + notice, this list of conditions and the following disclaimer in the + documentation and/or other materials provided with the distribution. + + * Neither the name of the University of Cambridge nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE +ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE +LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR +CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF +SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS +INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN +CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) +ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE +POSSIBILITY OF SUCH DAMAGE. +----------------------------------------------------------------------------- +*/ + + +/* This module contains an internal function for validating UTF character +strings. */ + + +#ifdef HAVE_CONFIG_H +#include "config.h" +#endif + +#include "pcre2_internal.h" + + +#ifndef SUPPORT_UTF +/************************************************* +* Dummy function when UTF not supported * +*************************************************/ + +/* This function should never be called when UTF is not supported. */ + +int +PRIV(valid_utf)(PCRE2_SPTR string, int length, size_t *erroroffset) +{ +(void)string; +(void)length; +(void)erroroffset); +return 0; +} +#else + + + +/************************************************* +* Validate a UTF string * +*************************************************/ + +/* This function is called (optionally) at the start of compile or match, to +check that a supposed UTF string is actually valid. The early check means +that subsequent code can assume it is dealing with a valid string. The check +can be turned off for maximum performance, but the consequences of supplying an +invalid string are then undefined. + +Arguments: + string points to the string + length length of string, or -1 if the string is zero-terminated + errp pointer to an error position offset variable + +Returns: == 0 if the string is a valid UTF string + != 0 otherwise, setting the offset of the bad character +*/ + +int +PRIV(valid_utf)(PCRE2_SPTR string, int length, size_t *erroroffset) +{ +register PCRE2_SPTR p; +register uint32_t c; + +if (length < 0) + { + for (p = string; *p != 0; p++); + length = (int)(p - string); + } + +/* ----------------- Check a UTF-8 string ----------------- */ + +#if PCRE2_CODE_UNIT_WIDTH == 8 + +/* Originally, this function checked according to RFC 2279, allowing for values +in the range 0 to 0x7fffffff, up to 6 bytes long, but ensuring that they were +in the canonical format. Once somebody had pointed out RFC 3629 to me (it +obsoletes 2279), additional restrictions were applied. The values are now +limited to be between 0 and 0x0010ffff, no more than 4 bytes long, and the +subrange 0xd000 to 0xdfff is excluded. However, the format of 5-byte and 6-byte +characters is still checked. Error returns are as follows: + +PCRE2_ERROR_UTF8_ERR1 Missing 1 byte at the end of the string +PCRE2_ERROR_UTF8_ERR2 Missing 2 bytes at the end of the string +PCRE2_ERROR_UTF8_ERR3 Missing 3 bytes at the end of the string +PCRE2_ERROR_UTF8_ERR4 Missing 4 bytes at the end of the string +PCRE2_ERROR_UTF8_ERR5 Missing 5 bytes at the end of the string +PCRE2_ERROR_UTF8_ERR6 2nd-byte's two top bits are not 0x80 +PCRE2_ERROR_UTF8_ERR7 3rd-byte's two top bits are not 0x80 +PCRE2_ERROR_UTF8_ERR8 4th-byte's two top bits are not 0x80 +PCRE2_ERROR_UTF8_ERR9 5th-byte's two top bits are not 0x80 +PCRE2_ERROR_UTF8_ERR10 6th-byte's two top bits are not 0x80 +PCRE2_ERROR_UTF8_ERR11 5-byte character is not permitted by RFC 3629 +PCRE2_ERROR_UTF8_ERR12 6-byte character is not permitted by RFC 3629 +PCRE2_ERROR_UTF8_ERR13 4-byte character with value > 0x10ffff is not permitted +PCRE2_ERROR_UTF8_ERR14 3-byte character with value 0xd000-0xdfff is not permitted +PCRE2_ERROR_UTF8_ERR15 Overlong 2-byte sequence +PCRE2_ERROR_UTF8_ERR16 Overlong 3-byte sequence +PCRE2_ERROR_UTF8_ERR17 Overlong 4-byte sequence +PCRE2_ERROR_UTF8_ERR18 Overlong 5-byte sequence (won't ever occur) +PCRE2_ERROR_UTF8_ERR19 Overlong 6-byte sequence (won't ever occur) +PCRE2_ERROR_UTF8_ERR20 Isolated 0x80 byte (not within UTF-8 character) +PCRE2_ERROR_UTF8_ERR21 Byte with the illegal value 0xfe or 0xff +*/ + +for (p = string; length-- > 0; p++) + { + register uint32_t ab, d; + + c = *p; + if (c < 128) continue; /* ASCII character */ + + if (c < 0xc0) /* Isolated 10xx xxxx byte */ + { + *erroroffset = (int)(p - string); + return PCRE2_ERROR_UTF8_ERR20; + } + + if (c >= 0xfe) /* Invalid 0xfe or 0xff bytes */ + { + *erroroffset = (int)(p - string); + return PCRE2_ERROR_UTF8_ERR21; + } + + ab = PRIV(utf8_table4)[c & 0x3f]; /* Number of additional bytes (1-5) */ + if (length < (int)ab) /* Missing bytes */ + { + *erroroffset = (int)(p - string); + switch(ab - length) + { + case 1: return PCRE2_ERROR_UTF8_ERR1; + case 2: return PCRE2_ERROR_UTF8_ERR2; + case 3: return PCRE2_ERROR_UTF8_ERR3; + case 4: return PCRE2_ERROR_UTF8_ERR4; + case 5: return PCRE2_ERROR_UTF8_ERR5; + } + } + length -= ab; /* Length remaining */ + + /* Check top bits in the second byte */ + + if (((d = *(++p)) & 0xc0) != 0x80) + { + *erroroffset = (int)(p - string) - 1; + return PCRE2_ERROR_UTF8_ERR6; + } + + /* For each length, check that the remaining bytes start with the 0x80 bit + set and not the 0x40 bit. Then check for an overlong sequence, and for the + excluded range 0xd800 to 0xdfff. */ + + switch (ab) + { + /* 2-byte character. No further bytes to check for 0x80. Check first byte + for for xx00 000x (overlong sequence). */ + + case 1: if ((c & 0x3e) == 0) + { + *erroroffset = (int)(p - string) - 1; + return PCRE2_ERROR_UTF8_ERR15; + } + break; + + /* 3-byte character. Check third byte for 0x80. Then check first 2 bytes + for 1110 0000, xx0x xxxx (overlong sequence) or + 1110 1101, 1010 xxxx (0xd800 - 0xdfff) */ + + case 2: + if ((*(++p) & 0xc0) != 0x80) /* Third byte */ + { + *erroroffset = (int)(p - string) - 2; + return PCRE2_ERROR_UTF8_ERR7; + } + if (c == 0xe0 && (d & 0x20) == 0) + { + *erroroffset = (int)(p - string) - 2; + return PCRE2_ERROR_UTF8_ERR16; + } + if (c == 0xed && d >= 0xa0) + { + *erroroffset = (int)(p - string) - 2; + return PCRE2_ERROR_UTF8_ERR14; + } + break; + + /* 4-byte character. Check 3rd and 4th bytes for 0x80. Then check first 2 + bytes for for 1111 0000, xx00 xxxx (overlong sequence), then check for a + character greater than 0x0010ffff (f4 8f bf bf) */ + + case 3: + if ((*(++p) & 0xc0) != 0x80) /* Third byte */ + { + *erroroffset = (int)(p - string) - 2; + return PCRE2_ERROR_UTF8_ERR7; + } + if ((*(++p) & 0xc0) != 0x80) /* Fourth byte */ + { + *erroroffset = (int)(p - string) - 3; + return PCRE2_ERROR_UTF8_ERR8; + } + if (c == 0xf0 && (d & 0x30) == 0) + { + *erroroffset = (int)(p - string) - 3; + return PCRE2_ERROR_UTF8_ERR17; + } + if (c > 0xf4 || (c == 0xf4 && d > 0x8f)) + { + *erroroffset = (int)(p - string) - 3; + return PCRE2_ERROR_UTF8_ERR13; + } + break; + + /* 5-byte and 6-byte characters are not allowed by RFC 3629, and will be + rejected by the length test below. However, we do the appropriate tests + here so that overlong sequences get diagnosed, and also in case there is + ever an option for handling these larger code points. */ + + /* 5-byte character. Check 3rd, 4th, and 5th bytes for 0x80. Then check for + 1111 1000, xx00 0xxx */ + + case 4: + if ((*(++p) & 0xc0) != 0x80) /* Third byte */ + { + *erroroffset = (int)(p - string) - 2; + return PCRE2_ERROR_UTF8_ERR7; + } + if ((*(++p) & 0xc0) != 0x80) /* Fourth byte */ + { + *erroroffset = (int)(p - string) - 3; + return PCRE2_ERROR_UTF8_ERR8; + } + if ((*(++p) & 0xc0) != 0x80) /* Fifth byte */ + { + *erroroffset = (int)(p - string) - 4; + return PCRE2_ERROR_UTF8_ERR9; + } + if (c == 0xf8 && (d & 0x38) == 0) + { + *erroroffset = (int)(p - string) - 4; + return PCRE2_ERROR_UTF8_ERR18; + } + break; + + /* 6-byte character. Check 3rd-6th bytes for 0x80. Then check for + 1111 1100, xx00 00xx. */ + + case 5: + if ((*(++p) & 0xc0) != 0x80) /* Third byte */ + { + *erroroffset = (int)(p - string) - 2; + return PCRE2_ERROR_UTF8_ERR7; + } + if ((*(++p) & 0xc0) != 0x80) /* Fourth byte */ + { + *erroroffset = (int)(p - string) - 3; + return PCRE2_ERROR_UTF8_ERR8; + } + if ((*(++p) & 0xc0) != 0x80) /* Fifth byte */ + { + *erroroffset = (int)(p - string) - 4; + return PCRE2_ERROR_UTF8_ERR9; + } + if ((*(++p) & 0xc0) != 0x80) /* Sixth byte */ + { + *erroroffset = (int)(p - string) - 5; + return PCRE2_ERROR_UTF8_ERR10; + } + if (c == 0xfc && (d & 0x3c) == 0) + { + *erroroffset = (int)(p - string) - 5; + return PCRE2_ERROR_UTF8_ERR19; + } + break; + } + + /* Character is valid under RFC 2279, but 4-byte and 5-byte characters are + excluded by RFC 3629. The pointer p is currently at the last byte of the + character. */ + + if (ab > 3) + { + *erroroffset = (int)(p - string) - ab; + return (ab == 4)? PCRE2_ERROR_UTF8_ERR11 : PCRE2_ERROR_UTF8_ERR12; + } + } +return 0; + + +/* ----------------- Check a UTF-16 string ----------------- */ + +#elif PCRE2_CODE_UNIT_WIDTH == 16 + +/* There's not so much work, nor so many errors, for UTF-16. +PCRE2_ERROR_UTF16_ERR1 Missing low surrogate at the end of the string +PCRE2_ERROR_UTF16_ERR2 Invalid low surrogate +PCRE2_ERROR_UTF16_ERR3 Isolated low surrogate +*/ + +for (p = string; length-- > 0; p++) + { + c = *p; + + if ((c & 0xf800) != 0xd800) + { + /* Normal UTF-16 code point. Neither high nor low surrogate. */ + } + else if ((c & 0x0400) == 0) + { + /* High surrogate. Must be a followed by a low surrogate. */ + if (length == 0) + { + *erroroffset = p - string; + return PCRE2_ERROR_UTF16_ERR1; + } + p++; + length--; + if ((*p & 0xfc00) != 0xdc00) + { + *erroroffset = p - string; + return PCRE2_ERROR_UTF16_ERR2; + } + } + else + { + /* Isolated low surrogate. Always an error. */ + *erroroffset = p - string; + return PCRE2_ERROR_UTF16_ERR3; + } + } +return 0; + + + +/* ----------------- Check a UTF-32 string ----------------- */ + +#else + +/* There is very little to do for a UTF-32 string. +PCRE2_ERROR_UTF32_ERR1 Surrogate character +PCRE2_ERROR_UTF32_ERR2 Character > 0x10ffff +*/ + +for (p = string; length-- > 0; p++) + { + c = *p; + if ((c & 0xfffff800u) != 0xd800u) + { + /* Normal UTF-32 code point. Neither high nor low surrogate. */ + if (c > 0x10ffffu) + { + *erroroffset = p - string; + return PCRE2_ERROR_UTF32_ERR2; + } + } + else + { + /* A surrogate */ + *erroroffset = p - string; + return PCRE2_ERROR_UTF32_ERR1; + } + } +return 0; +#endif /* CODE_UNIT_WIDTH */ +#endif /* SUPPORT_UTF */ +} + +/* End of pcre2_valid_utf.c */ diff --git a/src/pcre2posix.c b/src/pcre2posix.c index aabc0a4..bf1ea6d 100644 --- a/src/pcre2posix.c +++ b/src/pcre2posix.c @@ -108,7 +108,7 @@ static const int eint2[] = { 30, REG_ECTYPE, /* unknown POSIX class name */ 32, REG_INVARG, /* this version of PCRE2 is not compiled with PCRE2_UTF8 support */ 37, REG_EESCAPE, /* PCRE2 does not support \L, \l, \N, \U, or \u */ - 56, REG_INVARG, /* inconsistent NEWLINE options */ + 56, REG_INVARG, /* internal error: unknown newline setting */ 67, REG_INVARG, /* this version of PCRE2 is not compiled with PCRE2_UCP support */ }; @@ -148,6 +148,8 @@ regerror(int errcode, const regex_t *preg, char *errbuf, size_t errbuf_size) const char *message, *addmessage; size_t length, addlength; +errcode -= COMPILE_ERROR_BASE; + message = (errcode >= (int)(sizeof(pstring)/sizeof(char *)))? "unknown error code" : pstring[errcode]; length = strlen(message) + 1; @@ -224,6 +226,8 @@ preg->re_erroffset = erroffset; if (preg->re_pcre2_code == NULL) { unsigned int i; + if (errorcode < 0) return REG_BADPAT; /* UTF error */ + errorcode -= COMPILE_ERROR_BASE; if (errorcode < (int)(sizeof(eint1)/sizeof(const int))) return eint1[errorcode]; for (i = 0; i < sizeof(eint2)/(2*sizeof(const int)); i += 2) @@ -307,13 +311,15 @@ if (rc >= 0) /* Unsuccessful match */ +if (rc <= PCRE2_ERROR_UTF8_ERR1 && rc >= PCRE2_ERROR_UTF8_ERR21) + return REG_INVARG; + switch(rc) { default: return REG_ASSERT; case PCRE2_ERROR_BADMODE: return REG_INVARG; case PCRE2_ERROR_BADMAGIC: return REG_INVARG; case PCRE2_ERROR_BADOPTION: return REG_INVARG; - case PCRE2_ERROR_BADUTF: return REG_INVARG; case PCRE2_ERROR_BADUTF_OFFSET: return REG_INVARG; case PCRE2_ERROR_MATCHLIMIT: return REG_ESPACE; case PCRE2_ERROR_NOMATCH: return REG_NOMATCH; diff --git a/src/pcre2test.c b/src/pcre2test.c index 99db0e7..0c219e8 100644 --- a/src/pcre2test.c +++ b/src/pcre2test.c @@ -211,10 +211,10 @@ for building the library. */ #include "pcre2_internal.h" /* We need access to some of the data tables that PCRE uses. Defining -PCRE2_INCLUDED makes some minor changes in the files. The previous definition +PCRE2_PCRETEST makes some minor changes in the files. The previous definition of PRIV avoids name clashes. */ -#define PCRE2_INCLUDED +#define PCRE2_PCRE2TEST #include "pcre2_tables.c" #include "pcre2_ucd.c" @@ -340,12 +340,14 @@ either on a pattern or a data line, so they must all be distinct. */ #define CTL_FULLBINCODE 0x00000200 #define CTL_GETALL 0x00000400 #define CTL_GLOBAL 0x00000800 -#define CTL_INFO 0x00001000 -#define CTL_JITVERIFY 0x00002000 -#define CTL_LIMITS 0x00004000 -#define CTL_MARK 0x00008000 -#define CTL_MEMORY 0x00010000 -#define CTL_POSIX 0x00020000 +#define CTL_HEXPAT 0x00001000 +#define CTL_INFO 0x00002000 +#define CTL_JITVERIFY 0x00004000 +#define CTL_LIMITS 0x00008000 +#define CTL_MARK 0x00010000 +#define CTL_MEMORY 0x00020000 +#define CTL_PATLEN 0x00040000 +#define CTL_POSIX 0x00080000 #define CTL_DEBUG (CTL_FULLBINCODE|CTL_INFO) /* For setting */ #define CTL_ANYINFO (CTL_DEBUG|CTL_BINCODE) /* For testing */ @@ -441,6 +443,7 @@ static modstruct modlist[] = { { "get", MOD_DAT, MOD_NN, DO(get_numbers), DO(get_names) }, { "getall", MOD_DAT, MOD_CTL, CTL_GETALL, DO(control) }, { "global", MOD_PNDP, MOD_CTL, CTL_GLOBAL, PO(control) }, + { "hex", MOD_PAT, MOD_CTL, CTL_HEXPAT, PO(control) }, { "info", MOD_PAT, MOD_CTL, CTL_INFO, PO(control) }, { "jit", MOD_PAT, MOD_IND, 7, PO(jit) }, { "jitstack", MOD_DAT, MOD_INT, 0, DO(jitstack) }, @@ -475,6 +478,7 @@ static modstruct modlist[] = { { "tables", MOD_PAT, MOD_INT, 0, PO(tables_id) }, { "ucp", MOD_PATP, MOD_OPT, PCRE2_UCP, PO(options) }, { "ungreedy", MOD_PAT, MOD_OPT, PCRE2_UNGREEDY, PO(options) }, + { "use_length", MOD_PAT, MOD_CTL, CTL_PATLEN, PO(control) }, { "utf", MOD_PATP, MOD_OPT, PCRE2_UTF, PO(options) } }; @@ -625,14 +629,9 @@ buffer is where all input lines are read. Its size is the same as pbuffer8. Pattern lines are always copied to pbuffer8 for use in callouts, even if they are actually compiled from pbuffer16 or pbuffer32. */ -static int pbuffer8_size = 50000; /* Initial size, bytes */ -static int pbuffer16_size = 0; /* Only set once needed */ -static int pbuffer32_size = 0; /* Only set once needed */ - -static uint8_t *buffer = NULL; +static int pbuffer8_size = 50000; /* Initial size, bytes */ static uint8_t *pbuffer8 = NULL; -static uint16_t *pbuffer16 = NULL; -static uint32_t *pbuffer32 = NULL; +static uint8_t *buffer = NULL; /* The dbuffer is where all processed data lines are put. In non-8-bit modes it is cast as needed. For long data lines it grows as necessary. */ @@ -655,6 +654,8 @@ pcre2_code_16 *compiled_code16; pcre2_compile_context_16 *pat_context16, *default_pat_context16; pcre2_match_context_16 *dat_context16, *default_dat_context16; pcre2_match_data_16 *match_data16; +static int pbuffer16_size = 0; /* Only set once needed */ +static uint16_t *pbuffer16 = NULL; #endif #ifdef SUPPORT_PCRE32 @@ -662,6 +663,8 @@ pcre2_code_32 *compiled_code32; pcre2_compile_context_32 *pat_context32, *default_pat_context32; pcre2_match_context_32 *dat_context32, *default_dat_context32; pcre2_match_data_32 *match_data32; +static int pbuffer32_size = 0; /* Only set once needed */ +static uint32_t *pbuffer32 = NULL; #endif @@ -997,10 +1000,10 @@ the three different cases. */ #define PCRE2_DFA_MATCH(a,b,c,d,e,f,g,h,i,j) \ if (test_mode == G(G(PCRE,BITONE),_MODE)) \ - a = G(pcre2_dfa-match_,BITONE)(G(b,BITONE),(G(PCRE2_SPTR,BITONE))c,d,e,f, \ + a = G(pcre2_dfa_match_,BITONE)(G(b,BITONE),(G(PCRE2_SPTR,BITONE))c,d,e,f, \ G(g,BITONE),G(h,BITONE),i,j); \ else \ - a = G(pcre2_dfa-match_,BITTWO)(G(b,BITTWO),(G(PCRE2_SPTR,BITTWO))c,d,e,f, \ + a = G(pcre2_dfa_match_,BITTWO)(G(b,BITTWO),(G(PCRE2_SPTR,BITTWO))c,d,e,f, \ G(g,BITTWO),G(h,BITTWO),i,j) #define PCRE2_GET_ERROR_MESSAGE(r,a,b) \ @@ -1178,20 +1181,20 @@ the three different cases. */ #define PCRE2_COMPILE(a,b,c,d,e,f,g) \ G(a,8) = pcre2_compile_8(G(b,8),c,d,e,f,G(g,8)) #define PCRE2_DFA_MATCH(a,b,c,d,e,f,g,h,i,j) \ - G(a,8) = pcre2_dfa-match_8(G(b,8),(PCRE2_SPTR8)c,d,e,f,G(g,8),G(h,8),i,j) + a = pcre2_dfa_match_8(G(b,8),(PCRE2_SPTR8)c,d,e,f,G(g,8),G(h,8),i,j) #define PCRE2_GET_ERROR_MESSAGE(r,a,b) \ r = pcre2_get_error_message_8(a,G(b,8),G(G(b,8),_size)) #define PCRE2_JIT_COMPILE(a,b) pcre2_jit_compile_8(G(a,8),b) #define PCRE2_MATCH(a,b,c,d,e,f,g,h) \ - G(a,8) = pcre2_match_8(G(b,8),(PCRE2_SPTR8)c,d,e,f,G(g,8),G(h,8)) + a = pcre2_match_8(G(b,8),(PCRE2_SPTR8)c,d,e,f,G(g,8),G(h,8)) #define PCRE2_MAKETABLES(a) a = pcre2_maketables_8(NULL) #define PCRE2_MATCH_DATA_CREATE(a,b,c) G(a,8) = pcre2_match_data_create_8(b,c) -#define PCRE2_MATCH_DATA_FREE(a) pcre2_match_data_free_8(a) -#define PCRE2_PATTERN_INFO(a,b,c,d) G(a,8) = pcre2_pattern_info_8(G(b,8),c,d) -#define PCRE2_PRINTINT(a,b) pcre2_printint_8(compiled_code8,outfile,a) -#define PCRE2_SET_CHARACTER_TABLES(a,b) pcre2_set_character_tables_8(G(a,8),b) \ +#define PCRE2_MATCH_DATA_FREE(a) pcre2_match_data_free_8(G(a,8)) +#define PCRE2_PATTERN_INFO(a,b,c,d) a = pcre2_pattern_info_8(G(b,8),c,d) +#define PCRE2_PRINTINT(a) pcre2_printint_8(compiled_code8,outfile,a) +#define PCRE2_SET_CHARACTER_TABLES(a,b) pcre2_set_character_tables_8(G(a,8),b) #define PCRE2_SUBSTRING_COPY_BYNAME(a,b,c,d,e) \ - a = pcre2_substring_copy_bynumber_8(G(b,8),G(c,8),(PCRE2_UCHAR8 *)d,e) + a = pcre2_substring_copy_byname_8(G(b,8),G(c,8),(PCRE2_UCHAR8 *)d,e) #define PCRE2_SUBSTRING_COPY_BYNUMBER(a,b,c,d,e) \ a = pcre2_substring_copy_bynumber_8(G(b,8),c,(PCRE2_UCHAR8 *)d,e) #define PCRE2_SUBSTRING_FREE(a) pcre2_substring_free_8((PCRE2_UCHAR8 *)a) @@ -1207,12 +1210,12 @@ the three different cases. */ #define SETFLD(x,y,z) G(x,8)->y = z #define SETFLDVEC(x,y,v,z) G(x,8)->y[v] = z #define SETOP(x,y,z) G(x,8) z y -#define SETCASTPTR(x,y) G(x,8) = (uint8_t) *)y -#define STRLEN(p) (int)strlen(p) +#define SETCASTPTR(x,y) G(x,8) = (uint8_t *)y +#define STRLEN(p) (int)strlen((char *)p) #define SUB1(a,b) G(a,8)(G(b,8)) #define SUB2(a,b,c) G(a,8)(G(b,8),G(c,8)) -#define TEST(x,r,y) (G(a,8) r (y)) -#define TESTFLD(x,f,r,y) (G(a,8)->f r (y)) +#define TEST(x,r,y) (G(x,8) r (y)) +#define TESTFLD(x,f,r,y) (G(x,8)->f r (y)) /* ----- Only 16-bit mode is supported ----- */ @@ -1231,20 +1234,20 @@ the three different cases. */ #define PCRE2_COMPILE(a,b,c,d,e,f,g) \ G(a,16) = pcre2_compile_16(G(b,16),c,d,e,f,G(g,16)) #define PCRE2_DFA_MATCH(a,b,c,d,e,f,g,h,i,j) \ - G(a,16) = pcre2_dfa_match_16(G(b,16),(PCRE2_SPTR16)c,d,e,f,G(g,16),G(h,16),i,j) + a = pcre2_dfa_match_16(G(b,16),(PCRE2_SPTR16)c,d,e,f,G(g,16),G(h,16),i,j) #define PCRE2_GET_ERROR_MESSAGE(r,a,b) \ r = pcre2_get_error_message_16(a,G(b,16),G(G(b,16),_size)) #define PCRE2_JIT_COMPILE(a,b) pcre2_jit_compile_16(G(a,16),b) #define PCRE2_MAKETABLES(a) a = pcre2_maketables_16(NULL) #define PCRE2_MATCH(a,b,c,d,e,f,g,h) \ - G(a,16) = pcre2_match_16(G(b,16),(PCRE2_SPTR16)c,d,e,f,G(g,16),G(h,16)) + a = pcre2_match_16(G(b,16),(PCRE2_SPTR16)c,d,e,f,G(g,16),G(h,16)) #define PCRE2_MATCH_DATA_CREATE(a,b,c) G(a,16) = pcre2_match_data_create_16(b,c) -#define PCRE2_MATCH_DATA_FREE(a) pcre2_match_data_free_16(a) +#define PCRE2_MATCH_DATA_FREE(a) pcre2_match_data_free_16(G(a,16)) #define PCRE2_PATTERN_INFO(a,b,c,d) G(a,16) = pcre2_pattern_info_16(G(b,16),c,d) -#define PCRE2_PRINTINT(a,b) pcre2_printint_16(compiled_code16,outfile,a) +#define PCRE2_PRINTINT(a) pcre2_printint_16(compiled_code16,outfile,a) #define PCRE2_SET_CHARACTER_TABLES(a,b) pcre2_set_character_tables_16(G(a,16),b) #define PCRE2_SUBSTRING_COPY_BYNAME(a,b,c,d,e) \ - a = pcre2_substring_copy_bynumber_16(G(b,16),G(c,16),(PCRE2_UCHAR16 *)d,e); + a = pcre2_substring_copy_byname_16(G(b,16),G(c,16),(PCRE2_UCHAR16 *)d,e); #define PCRE2_SUBSTRING_COPY_BYNUMBER(a,b,c,d,e) \ a = pcre2_substring_copy_bynumber_16(G(b,16),c,(PCRE2_UCHAR16 *)d,e); #define PCRE2_SUBSTRING_FREE(a) pcre2_substring_free_16((PCRE2_UCHAR16 *)a) @@ -1260,12 +1263,12 @@ the three different cases. */ #define SETFLD(x,y,z) G(x,16)->y = z #define SETFLDVEC(x,y,v,z) G(x,16)->y[v] = z #define SETOP(x,y,z) G(x,16) z y -#define SETCASTPTR(x,y) G(x,16) = (uint16_t) *)y +#define SETCASTPTR(x,y) G(x,16) = (uint16_t *)y #define STRLEN(p) (int)strlen16(p) #define SUB1(a,b) G(a,16)(G(b,16)) #define SUB2(a,b,c) G(a,16)(G(b,16),G(c,16)) -#define TEST(x,r,y) (G(a,16) r (y)) -#define TESTFLD(x,f,r,y) (G(a,16)->f r (y)) +#define TEST(x,r,y) (G(x,16) r (y)) +#define TESTFLD(x,f,r,y) (G(x,16)->f r (y)) /* ----- Only 32-bit mode is supported ----- */ @@ -1284,20 +1287,20 @@ the three different cases. */ #define PCRE2_COMPILE(a,b,c,d,e,f,g) \ G(a,32) = pcre2_compile_32(G(b,32),c,d,e,f,G(g,32)) #define PCRE2_DFA_MATCH(a,b,c,d,e,f,g,h,i,j) \ - G(a,32) = pcre2_dfa_match_32(G(b,32),(PCRE2_SPTR32)c,d,e,f,G(g,32),G(h,32),i,j) + a = pcre2_dfa_match_32(G(b,32),(PCRE2_SPTR32)c,d,e,f,G(g,32),G(h,32),i,j) #define PCRE2_GET_ERROR_MESSAGE(r,a,b) \ r = pcre2_get_error_message_32(a,G(b,32),G(G(b,32),_size)) #define PCRE2_JIT_COMPILE(a,b) pcre2_jit_compile_32(G(a,32),b) #define PCRE2_MATCH(a,b,c,d,e,f,g,h) \ - G(a,32) = pcre2_match_32(G(b,32),(PCRE2_SPTR32)c,d,e,f,G(g,32),g(h,32)) + a = pcre2_match_32(G(b,32),(PCRE2_SPTR32)c,d,e,f,G(g,32),g(h,32)) #define PCRE2_MAKETABLES(a) a = pcre2_maketables_32(NULL) #define PCRE2_MATCH_DATA_CREATE(a,b,c) G(a,32) = pcre2_match_data_create_32(b,c) -#define PCRE2_MATCH_DATA_FREE(a) pcre2_match_data_free_32(a) +#define PCRE2_MATCH_DATA_FREE(a) pcre2_match_data_free_32(G(a,32)) #define PCRE2_PATTERN_INFO(a,b,c,d) G(a,32) = pcre2_pattern_info_32(G(b,32),c,d) -#define PCRE2_PRINTINT(a,b) pcre2_printint_32(compiled_code32,outfile,a) +#define PCRE2_PRINTINT(a) pcre2_printint_32(compiled_code32,outfile,a) #define PCRE2_SET_CHARACTER_TABLES(a,b) pcre2_set_character_tables_32(G(a,32),b) #define PCRE2_SUBSTRING_COPY_BYNAME(a,b,c,d,e) \ - a = pcre2_substring_copy_bynumber_32(G(b,32),G(c,32),(PCRE2_UCHAR32 *)d,e); + a = pcre2_substring_copy_byname_32(G(b,32),G(c,32),(PCRE2_UCHAR32 *)d,e); #define PCRE2_SUBSTRING_COPY_BYNUMBER(a,b,c,d,e) \ a = pcre2_substring_copy_bynumber_32(G(b,32),c,(PCRE2_UCHAR32 *)d,e); #define PCRE2_SUBSTRING_FREE(a) pcre2_substring_free_32((PCRE2_UCHAR32 *)a) @@ -1313,12 +1316,12 @@ define PCRE2_SUBSTRING_GET_BYNUMBER(a,b,c,d) \ #define SETFLD(x,y,z) G(x,32)->y = z #define SETFLDVEC(x,y,v,z) G(x,32)->y[v] = z #define SETOP(x,y,z) G(x,32) z y -#define SETCASTPTR(x,y) G(x,32) = (uint32_t) *)y +#define SETCASTPTR(x,y) G(x,32) = (uint32_t *)y #define STRLEN(p) (int)strle32(p) #define SUB1(a,b) G(a,32)(G(b,32)) #define SUB2(a,b,c) G(a,32)(G(b,32),G(c,32)) -#define TEST(x,r,y) (G(a,32) r (y)) -#define TESTFLD(x,f,r,y) (G(a,32)->f r (y)) +#define TEST(x,r,y) (G(x,32) r (y)) +#define TESTFLD(x,f,r,y) (G(x,32)->f r (y)) #endif @@ -2669,7 +2672,7 @@ Returns: nothing static void show_compile_controls(uint32_t controls, const char *before, const char *after) { -fprintf(outfile, "%s%s%s%s%s%s%s%s%s%s%s%s%s%s", +fprintf(outfile, "%s%s%s%s%s%s%s%s%s%s%s%s%s%s%s%s", before, ((controls & CTL_AFTERTEXT) != 0)? " aftertext" : "", ((controls & CTL_ALLAFTERTEXT) != 0)? " allaftertext" : "", @@ -2679,9 +2682,11 @@ fprintf(outfile, "%s%s%s%s%s%s%s%s%s%s%s%s%s%s", ((controls & CTL_FLIPBYTES) != 0)? " flipbytes" : "", ((controls & CTL_FULLBINCODE) != 0)? " fullbincode" : "", ((controls & CTL_GLOBAL) != 0)? " global" : "", + ((controls & CTL_HEXPAT) != 0)? " hex" : "", ((controls & CTL_INFO) != 0)? " info" : "", ((controls & CTL_JITVERIFY) != 0)? " jitverify" : "", ((controls & CTL_MARK) != 0)? " mark" : "", + ((controls & CTL_PATLEN) != 0)? " use_length" : "", ((controls & CTL_POSIX) != 0)? " posix" : "", after); } @@ -2705,7 +2710,8 @@ Returns: nothing static void show_compile_options(uint32_t options, const char *before, const char *after) { -fprintf(outfile, "%s%s%s%s%s%s%s%s%s%s%s%s%s%s%s%s%s%s%s%s%s%s%s", +if (options == 0) fprintf(outfile, "%s %s", before, after); +else fprintf(outfile, "%s%s%s%s%s%s%s%s%s%s%s%s%s%s%s%s%s%s%s%s%s%s%s", before, ((options & PCRE2_ANCHORED) != 0)? " anchored" : "", ((options & PCRE2_CASELESS) != 0)? " caseless" : "", @@ -2896,15 +2902,13 @@ if ((pat_patctl.control & CTL_INFO) != 0) fprintf(outfile, "No options\n"); else { - if (compile_options != 0) - show_compile_options(compile_options, "Compile options:", "\n"); - if (pattern_options != 0) - show_compile_options(pattern_options, "Pattern options:", "\n"); + show_compile_options(compile_options, "Compile options:", "\n"); + show_compile_options(pattern_options, "Pattern options:", "\n"); } if (jchanged) fprintf(outfile, "Duplicate name status changes\n"); - if (bsr_convention != PCRE2_BSR_DEFAULT) + if (bsr_convention != BSR_DEFAULT) fprintf(outfile, "\\R matches %s\n", (bsr_convention == PCRE2_BSR_UNICODE)? "any Unicode newline" : "CR, LF, or CRLF"); @@ -3272,25 +3276,63 @@ for(;;) if (infile != stdin) fprintf(outfile, "%s", (char *)p); } -/* If the first character after the delimiter is backslash, make -the pattern end with backslash. This is purely to provide a way -of testing for the error message when a pattern ends with backslash. */ +/* If the first character after the delimiter is backslash, make the pattern +end with backslash. This is purely to provide a way of testing for the error +message when a pattern ends with backslash. */ if (p[1] == '\\') *p++ = '\\'; -/* Terminate the pattern at the delimiter, and save a copy of the pattern -for callouts. */ +/* Terminate the pattern at the delimiter, and compute the length. */ *p++ = 0; -patlen = p - buffer - 1; -strncpy((char *)pbuffer8, (char *)(buffer+1), patlen); +patlen = p - buffer - 2; -/* Look for modifiers and options after the final delimiter. If successful, -compile the pattern. */ +/* Look for modifiers and options after the final delimiter. */ if (!decode_modifiers(p, CTX_PAT, &pat_patctl, NULL)) return PR_SKIP; utf = (pat_patctl.options & PCRE2_UTF) != 0; +/* Now copy the pattern to pbuffer8 for use in 8-bit testing and for reflecting +in callouts. Convert to binary if required. */ + +if ((pat_patctl.control & CTL_HEXPAT) != 0) + { + uint8_t *pp, *pt; + uint32_t c, d; + + if ((pat_patctl.control & CTL_POSIX) != 0) + { + fprintf(outfile, "** Hex patterns are not supported for the POSIX API\n"); + return PR_SKIP; + } + + pt = pbuffer8; + for (pp = buffer + 1; *pp != 0; pp++) + { + if (isspace(*pp)) continue; + c = toupper(*pp++); + if (*pp == 0) + { + fprintf(outfile, "** Odd number of digits in hex pattern.\n"); + return PR_SKIP; + } + d = toupper(*pp); + if (!isxdigit(c) || !isxdigit(d)) + { + fprintf(outfile, "** Non-hex-digit in hex pattern.\n"); + return PR_SKIP; + } + *pt++ = ((isdigit(c)? (c - '0') : (c - 'A' + 10)) << 4) + + (isdigit(d)? (d - '0') : (d - 'A' + 10)); + } + *pt = 0; + patlen = pt - pbuffer8; + } +else + { + strncpy((char *)pbuffer8, (char *)(buffer+1), patlen + 1); + } + /* Sort out character tables */ if (pat_patctl.locale[0] != 0) @@ -3394,12 +3436,12 @@ modes. */ #ifdef SUPPORT_PCRE16 if (test_mode == PCRE16_MODE) - patlen = to16(pbuffer8, utf, (int)strlen((char *)pbuffer8)); + patlen = to16(pbuffer8, utf, patlen); #endif #ifdef SUPPORT_PCRE32 if (test_mode == PCRE32_MODE) - patlen = to32(pbuffer8, utf, (int)strlen((char *)pbuffer8)); + patlen = to32(pbuffer8, utf, patlen); #endif switch(patlen) @@ -3423,8 +3465,13 @@ switch(patlen) break; } -/* The pattern in now in pbuffer[8|16|32], with the length in patlen. Compile -many times when timing. */ +/* The pattern in now in pbuffer[8|16|32], with the length in patlen. By +default, however, we pass a zero-terminated pattern. The length is passed only +if we had a hex pattern or if use_length was set. */ + +if ((pat_patctl.control & (CTL_PATLEN|CTL_HEXPAT)) == 0) patlen = -1; + +/* Compile many times when timing. */ if (timeit > 0) { @@ -3960,9 +4007,9 @@ if ((pat_patctl.control & CTL_POSIX) != 0) if (dat_datctl.cfail[0] != 0 || dat_datctl.cfail[1] != 0) prmsg(&msg, "callout_fail"); - if (dat_datctl.copy_numbers[0] != 0 || dat_datctl.copy_names[0] != 0) + if (dat_datctl.copy_numbers[0] >= 0 || dat_datctl.copy_names[0] != 0) prmsg(&msg, "copy"); - if (dat_datctl.get_numbers[0] != 0 || dat_datctl.get_names[0] != 0) + if (dat_datctl.get_numbers[0] >= 0 || dat_datctl.get_names[0] != 0) prmsg(&msg, "get"); if (dat_datctl.jitstack != 0) prmsg(&msg, "jitstack"); @@ -4059,6 +4106,9 @@ for (gmatched = 0;; gmatched++) #ifdef FIXME jit_was_used = FALSE; + +Need to set newline and bsr in match context and allow them to be +set in the datctl block. #endif /* Adjust match_data according to size of offsets required. */ @@ -4502,12 +4552,6 @@ if ((dat_datctl.control & CTL_DFA) != 0) } break; - case PCRE2_ERROR_BADUTF: - fprintf(outfile, "Error %d (bad UTF-%d string) offset=%d reason=%d\n", - capcount, test_mode, CASTFLD(int, match_data, startchar), - CASTFLD(int, match_data, utf_reason)); - break; - case PCRE2_ERROR_BADUTF_OFFSET: fprintf(outfile, "Error %d (bad UTF-%d offset)\n", capcount, test_mode); break;