Add serialization functions and tests with updated pcre2test. Fix

PCRE2_INFO_SIZE issues.

ChangeLog

@@ -1,8 +1,8 @@
 Change Log for PCRE2
 --------------------
 
-Version 10.10 13-January-2015
+Version 10.10 xx-xxx-2015
------------------------------
+-------------------------
 
 1. When a pattern is compiled, it remembers the highest back reference so that
 when matching, if the ovector is too small, extra memory can be obtained to
@@ -16,6 +16,19 @@ bug was that the condition was always treated as FALSE when the capture could
 not be consulted, leading to a incorrect behaviour by pcre2_match(). This bug
 has been fixed.
 
+2. Functions for serialization and deserialization of sets of compiled patterns
+have been added.
+
+3. The value that is returned by PCRE2_INFO_SIZE has been corrected to remove
+excess code units at the end of the data block that may occasionally occur if
+the code for calculating the size over-estimates. This change stops the
+serialization code copying uninitialized data, to which valgrind objects. The
+documentation of PCRE2_INFO_SIZE was incorrect in stating that the size did not
+include the general overhead. This has been corrected.
+
+4. All code units in every slot in the table of group names are now set, again
+in order to avoid accessing uninitialized data when serializing.
+
 
 Version 10.00 05-January-2015
 -----------------------------
@@ -30,8 +43,9 @@ logged. In addition to the API changes, the following changes were made. They
 are either new functionality, or bug fixes and other noticeable changes of
 behaviour that were implemented after the code had been forked.
 
-1. Unicode support is now enabled by default, but it can optionally be
+1. Including Unicode support at build time is now enabled by default, but it
-disabled.
+can optionally be disabled. It is not enabled by default at run time (no
+change).
 
 2. The test program, now called pcre2test, was re-specified and almost
 completely re-written. Its input is not compatible with input for pcretest.

Makefile.am

@@ -54,6 +54,10 @@ dist_html_DATA = \
   doc/html/pcre2_match_data_create_from_pattern.html \
   doc/html/pcre2_match_data_free.html \
   doc/html/pcre2_pattern_info.html \
+  doc/html/pcre2_serialize_decode.html \
+  doc/html/pcre2_serialize_encode.html \
+  doc/html/pcre2_serialize_free.html \
+  doc/html/pcre2_serialize_get_number_of_codes.html \
   doc/html/pcre2_set_bsr.html \
   doc/html/pcre2_set_callout.html \
   doc/html/pcre2_set_character_tables.html \
@@ -89,6 +93,7 @@ dist_html_DATA = \
   doc/html/pcre2perform.html \
   doc/html/pcre2posix.html \
   doc/html/pcre2sample.html \
+  doc/html/pcre2serialize.html \
   doc/html/pcre2stack.html \
   doc/html/pcre2syntax.html \
   doc/html/pcre2test.html \
@@ -127,6 +132,10 @@ dist_man_MANS = \
   doc/pcre2_match_data_create_from_pattern.3 \
   doc/pcre2_match_data_free.3 \
   doc/pcre2_pattern_info.3 \
+  doc/pcre2_serialize_decode.3 \
+  doc/pcre2_serialize_encode.3 \
+  doc/pcre2_serialize_free.3 \
+  doc/pcre2_serialize_get_number_of_codes.3 \
   doc/pcre2_set_bsr.3 \
   doc/pcre2_set_callout.3 \
   doc/pcre2_set_character_tables.3 \
@@ -162,6 +171,7 @@ dist_man_MANS = \
   doc/pcre2perform.3 \
   doc/pcre2posix.3 \
   doc/pcre2sample.3 \
+  doc/pcre2serialize.3 \
   doc/pcre2stack.3 \
   doc/pcre2syntax.3 \
   doc/pcre2test.1 \
@@ -316,6 +326,7 @@ COMMON_SOURCES = \
   src/pcre2_newline.c \
   src/pcre2_ord2utf.c \
   src/pcre2_pattern_info.c \
+  src/pcre2_serialize.c \
   src/pcre2_string_utils.c \
   src/pcre2_study.c \
   src/pcre2_substitute.c \
@@ -573,6 +584,7 @@ EXTRA_DIST += \
   testdata/testinput16 \
   testdata/testinput17 \
   testdata/testinput18 \
+  testdata/testinput19 \
   testdata/testinputEBC \
   testdata/testoutput1 \
   testdata/testoutput2 \
@@ -598,6 +610,7 @@ EXTRA_DIST += \
   testdata/testoutput16 \
   testdata/testoutput17 \
   testdata/testoutput18 \
+  testdata/testoutput19 \
   testdata/testoutputEBC \
   perltest.sh
 

NON-AUTOTOOLS-BUILD

@@ -108,6 +108,7 @@ can skip ahead to the CMake section.
        pcre2_newline.c
        pcre2_ord2utf.c
        pcre2_pattern_info.c
+       pcre2_serialize.c
        pcre2_string_utils.c
        pcre2_study.c
        pcre2_substitute.c
@@ -391,4 +392,4 @@ The site currently has ports for PCRE1 releases, but PCRE2 should follow in due
 course.
 
 =============================
-Last Updated: 05 January 2015
+Last Updated: 19 January 2015

README

@@ -527,11 +527,10 @@ Testing PCRE2
 ------------
 
 To test the basic PCRE2 library on a Unix-like system, run the RunTest script.
-There is another script called RunGrepTest that tests the options of the
+There is another script called RunGrepTest that tests the pcre2grep command.
-pcre2grep command. When JIT support is enabled, a third test program called
+When JIT support is enabled, a third test program called pcre2_jit_test is
-pcre2_jit_test is built. Both the scripts and all the program tests are run if
+built. Both the scripts and all the program tests are run if you obey "make
-you obey "make check". For other environments, see the instructions in
+check". For other environments, see the instructions in NON-AUTOTOOLS-BUILD.
-NON-AUTOTOOLS-BUILD.
 
 The RunTest script runs the pcre2test test program (which is documented in its
 own man page) on each of the relevant testinput files in the testdata
@@ -544,9 +543,9 @@ Some tests are relevant only when certain build-time options were selected. For
 example, the tests for UTF-8/16/32 features are run only when Unicode support
 is available. RunTest outputs a comment when it skips a test.
 
-Many of the tests that are not skipped are run twice if JIT support is
+Many (but not all) of the tests that are not skipped are run twice if JIT
-available. On the second run, JIT compilation is forced. This testing can be
+support is available. On the second run, JIT compilation is forced. This
-suppressed by putting "nojit" on the RunTest command line.
+testing can be suppressed by putting "nojit" on the RunTest command line.
 
 The entire set of tests is run once for each of the 8-bit, 16-bit and 32-bit
 libraries that are enabled. If you want to run just one set of tests, call
@@ -570,14 +569,20 @@ in numerical order.
 You can also call RunTest with the single argument "list" to cause it to output
 a list of tests.
 
-The first two tests can always be run, as they expect only plain text strings
+The test sequence starts with "test 0", which is a special test that has no
-(not UTF) and make no use of Unicode properties. The first test file can be fed
+input file, and whose output is not checked. This is because it will be
+different on different hardware and with different configurations. The test
+exists in order to exercise some of pcre2test's code that would not otherwise
+be run.
+
+Tests 1 and 2 can always be run, as they expect only plain text strings (not
+UTF) and make no use of Unicode properties. The first test file can be fed
 directly into the perltest.sh script to check that Perl gives the same results.
 The only difference you should see is in the first few lines, where the Perl
 version is given instead of the PCRE2 version. The second set of tests check
 auxiliary functions, error detection, and run-time flags that are specific to
-PCRE2, as well as the POSIX wrapper API. It also uses the debugging flags to
+PCRE2. It also uses the debugging flags to check some of the internals of
-check some of the internals of pcre2_compile().
+pcre2_compile().
 
 If you build PCRE2 with a locale setting that is not the standard C locale, the
 character tables may be different (see next paragraph). In some cases, this may
@@ -585,18 +590,17 @@ cause failures in the second set of tests. For example, in a locale where the
 isprint() function yields TRUE for characters in the range 128-255, the use of
 [:isascii:] inside a character class defines a different set of characters, and
 this shows up in this test as a difference in the compiled code, which is being
-listed for checking. Where the comparison test output contains [\x00-\x7f] the
+listed for checking. For example, where the comparison test output contains
-test will contain [\x00-\xff], and similarly in some other cases. This is not a
+[\x00-\x7f] the test might contain [\x00-\xff], and similarly in some other
-bug in PCRE2.
+cases. This is not a bug in PCRE2.
 
-The third set of tests checks pcre2_maketables(), the facility for building a
+Test 3 checks pcre2_maketables(), the facility for building a set of character
-set of character tables for a specific locale and using them instead of the
+tables for a specific locale and using them instead of the default tables. The
-default tables. The script uses the "locale" command to check for the
+script uses the "locale" command to check for the availability of the "fr_FR",
-availability of the "fr_FR", "french", or "fr" locale, and uses the first one
+"french", or "fr" locale, and uses the first one that it finds. If the "locale"
-that it finds. If the "locale" command fails, or if its output doesn't include
+command fails, or if its output doesn't include "fr_FR", "french", or "fr" in
-"fr_FR", "french", or "fr" in the list of available locales, the third test
+the list of available locales, the third test cannot be run, and a comment is
-cannot be run, and a comment is output to say why. If running this test
+output to say why. If running this test produces an error like this:
-produces an error like this
 
   ** Failed to set locale "fr_FR"
 
@@ -606,33 +610,37 @@ alternative output files for the third test, because three different versions
 of the French locale have been encountered. The test passes if its output
 matches any one of them.
 
-The fourth and fifth tests check UTF and Unicode property support, the fourth
+Tests 4 and 5 check UTF and Unicode property support, test 4 being compatible
-being compatible with the perltest.sh script, and the fifth checking
+with the perltest.sh script, and test 5 checking PCRE2-specific things.
-PCRE2-specific things.
 
-The sixth and seventh tests check the pcre2_dfa_match() alternative matching
+Tests 6 and 7 check the pcre2_dfa_match() alternative matching function, in
-function, in non-UTF mode and UTF-mode with Unicode property support,
+non-UTF mode and UTF-mode with Unicode property support, respectively.
-respectively.
 
-The eighth test checks some internal offsets and code size features; it is
+Test 8 checks some internal offsets and code size features; it is run only when
-run only when the default "link size" of 2 is set (in other cases the sizes
+the default "link size" of 2 is set (in other cases the sizes change) and when
-change) and when Unicode support is enabled.
+Unicode support is enabled.
 
-The ninth and tenth tests are run only in 8-bit mode, and the eleventh and
+Tests 9 and 10 are run only in 8-bit mode, and tests 11 and 12 are run only in
-twelfth tests are run only in 16-bit and 32-bit modes. These are tests that
+16-bit and 32-bit modes. These are tests that generate different output in
-generate different output in 8-bit mode. Each pair are for general cases and
+8-bit mode. Each pair are for general cases and Unicode support, respectively.
-Unicode support, respectively. The thirteenth test checks the handling of
+Test 13 checks the handling of non-UTF characters greater than 255 by
-non-UTF characters greater than 255 by pcre2_dfa_match() in 16-bit and 32-bit
+pcre2_dfa_match() in 16-bit and 32-bit modes.
-modes.
 
-The fourteenth test is run only when JIT support is not available, and the
+Test 14 contains a number of tests that must not be run with JIT. They check,
-fifteenth test is run only when JIT support is available. They test some
+among other non-JIT things, the match-limiting features of the intepretive
-JIT-specific features such as information output from pcre2test about JIT
+matcher.
-compilation.
 
-The sixteenth and seventeenth tests are run only in 8-bit mode. They check the
+Test 15 is run only when JIT support is not available. It checks that an
-POSIX interface to the 8-bit library, without and with Unicode support,
+attempt to use JIT has the expected behaviour.
-respectively.
+
+Test 16 is run only when JIT support is available. It checks JIT complete and
+partial modes, match-limiting under JIT, and other JIT-specific features.
+
+Tests 17 and 18 are run only in 8-bit mode. They check the POSIX interface to
+the 8-bit library, without and with Unicode support, respectively.
+
+Test 19 checks the serialization functions by writing a set of compiled
+patterns to a file, and then reloading and checking them.
 
 
 Character tables
@@ -718,6 +726,7 @@ The distribution should contain the files listed below.
   src/pcre2_newline.c      )
   src/pcre2_ord2utf.c      )
   src/pcre2_pattern_info.c )
+  src/pcre2_serialize.c    )
   src/pcre2_string_utils.c )
   src/pcre2_study.c        )
   src/pcre2_substitute.c   )
@@ -816,4 +825,4 @@ The distribution should contain the files listed below.
 Philip Hazel
 Email local part: ph10
 Email domain: cam.ac.uk
-Last updated: 05 January 2015
+Last updated: 20 January 2015

RunTest

@@ -65,6 +65,7 @@ title15="Test 15: JIT-specific features when JIT is not available"
 title16="Test 16: JIT-specific features when JIT is available"
 title17="Test 17: Tests of the POSIX interface, excluding UTF/UCP"
 title18="Test 18: Tests of the POSIX interface with UTF/UCP"
+title19="Test 19: Serialization tests"
 maxtest=18
 
 if [ $# -eq 1 -a "$1" = "list" ]; then
@@ -87,6 +88,7 @@ if [ $# -eq 1 -a "$1" = "list" ]; then
   echo $title16
   echo $title17
   echo $title18
+  echo $title19
   exit 0
 fi
 
@@ -207,6 +209,7 @@ do15=no
 do16=no
 do17=no
 do18=no
+do19=no
 
 while [ $# -gt 0 ] ; do
   case $1 in
@@ -229,6 +232,7 @@ while [ $# -gt 0 ] ; do
    16) do16=yes;;
    17) do17=yes;;
    18) do18=yes;;
+   19) do19=yes;;
    -8) arg8=yes;;
   -16) arg16=yes;;
   -32) arg32=yes;;
@@ -364,7 +368,7 @@ if [ $do0  = no -a $do1  = no -a $do2  = no -a $do3  = no -a \
      $do4  = no -a $do5  = no -a $do6  = no -a $do7  = no -a \
      $do8  = no -a $do9  = no -a $do10 = no -a $do11 = no -a \
      $do12 = no -a $do13 = no -a $do14 = no -a $do15 = no -a \
-     $do16 = no -a $do17 = no -a $do18 = no \
+     $do16 = no -a $do17 = no -a $do18 = no -a $do19 = no \
    ]; then
   do0=yes
   do1=yes
@@ -385,6 +389,7 @@ if [ $do0  = no -a $do1  = no -a $do2  = no -a $do3  = no -a \
   do16=yes
   do17=yes
   do18=yes
+  do19=yes
 fi
 
 # Handle any explicit skips at this stage, so that an argument list may consist
@@ -721,10 +726,18 @@ for bmode in "$test8" "$test16" "$test32"; do
     fi
   fi
 
+  # Serialization tests
+
+  if [ $do19 = yes ] ; then
+    echo $title19
+    $sim $valgrind ./pcre2test -q $bmode $testdata/testinput19 testtry
+    checkresult $? 19 ""
+  fi
+
 # End of loop for 8/16/32-bit tests
 done
 
 # Clean up local working files
-rm -f testSinput test3input test3output test3outputA test3outputB teststdout testtry
+rm -f testSinput test3input testsaved1 testsaved2 test3output test3outputA test3outputB teststdout testtry
 
 # End

doc/html/NON-AUTOTOOLS-BUILD.txt

@@ -108,6 +108,7 @@ can skip ahead to the CMake section.
        pcre2_newline.c
        pcre2_ord2utf.c
        pcre2_pattern_info.c
+       pcre2_serialize.c
        pcre2_string_utils.c
        pcre2_study.c
        pcre2_substitute.c
@@ -391,4 +392,4 @@ The site currently has ports for PCRE1 releases, but PCRE2 should follow in due
 course.
 
 =============================
-Last Updated: 05 January 2015
+Last Updated: 19 January 2015

doc/html/README.txt

@@ -527,11 +527,10 @@ Testing PCRE2
 ------------
 
 To test the basic PCRE2 library on a Unix-like system, run the RunTest script.
-There is another script called RunGrepTest that tests the options of the
+There is another script called RunGrepTest that tests the pcre2grep command.
-pcre2grep command. When JIT support is enabled, a third test program called
+When JIT support is enabled, a third test program called pcre2_jit_test is
-pcre2_jit_test is built. Both the scripts and all the program tests are run if
+built. Both the scripts and all the program tests are run if you obey "make
-you obey "make check". For other environments, see the instructions in
+check". For other environments, see the instructions in NON-AUTOTOOLS-BUILD.
-NON-AUTOTOOLS-BUILD.
 
 The RunTest script runs the pcre2test test program (which is documented in its
 own man page) on each of the relevant testinput files in the testdata
@@ -544,9 +543,9 @@ Some tests are relevant only when certain build-time options were selected. For
 example, the tests for UTF-8/16/32 features are run only when Unicode support
 is available. RunTest outputs a comment when it skips a test.
 
-Many of the tests that are not skipped are run twice if JIT support is
+Many (but not all) of the tests that are not skipped are run twice if JIT
-available. On the second run, JIT compilation is forced. This testing can be
+support is available. On the second run, JIT compilation is forced. This
-suppressed by putting "nojit" on the RunTest command line.
+testing can be suppressed by putting "nojit" on the RunTest command line.
 
 The entire set of tests is run once for each of the 8-bit, 16-bit and 32-bit
 libraries that are enabled. If you want to run just one set of tests, call
@@ -570,14 +569,20 @@ in numerical order.
 You can also call RunTest with the single argument "list" to cause it to output
 a list of tests.
 
-The first two tests can always be run, as they expect only plain text strings
+The test sequence starts with "test 0", which is a special test that has no
-(not UTF) and make no use of Unicode properties. The first test file can be fed
+input file, and whose output is not checked. This is because it will be
+different on different hardware and with different configurations. The test
+exists in order to exercise some of pcre2test's code that would not otherwise
+be run.
+
+Tests 1 and 2 can always be run, as they expect only plain text strings (not
+UTF) and make no use of Unicode properties. The first test file can be fed
 directly into the perltest.sh script to check that Perl gives the same results.
 The only difference you should see is in the first few lines, where the Perl
 version is given instead of the PCRE2 version. The second set of tests check
 auxiliary functions, error detection, and run-time flags that are specific to
-PCRE2, as well as the POSIX wrapper API. It also uses the debugging flags to
+PCRE2. It also uses the debugging flags to check some of the internals of
-check some of the internals of pcre2_compile().
+pcre2_compile().
 
 If you build PCRE2 with a locale setting that is not the standard C locale, the
 character tables may be different (see next paragraph). In some cases, this may
@@ -585,18 +590,17 @@ cause failures in the second set of tests. For example, in a locale where the
 isprint() function yields TRUE for characters in the range 128-255, the use of
 [:isascii:] inside a character class defines a different set of characters, and
 this shows up in this test as a difference in the compiled code, which is being
-listed for checking. Where the comparison test output contains [\x00-\x7f] the
+listed for checking. For example, where the comparison test output contains
-test will contain [\x00-\xff], and similarly in some other cases. This is not a
+[\x00-\x7f] the test might contain [\x00-\xff], and similarly in some other
-bug in PCRE2.
+cases. This is not a bug in PCRE2.
 
-The third set of tests checks pcre2_maketables(), the facility for building a
+Test 3 checks pcre2_maketables(), the facility for building a set of character
-set of character tables for a specific locale and using them instead of the
+tables for a specific locale and using them instead of the default tables. The
-default tables. The script uses the "locale" command to check for the
+script uses the "locale" command to check for the availability of the "fr_FR",
-availability of the "fr_FR", "french", or "fr" locale, and uses the first one
+"french", or "fr" locale, and uses the first one that it finds. If the "locale"
-that it finds. If the "locale" command fails, or if its output doesn't include
+command fails, or if its output doesn't include "fr_FR", "french", or "fr" in
-"fr_FR", "french", or "fr" in the list of available locales, the third test
+the list of available locales, the third test cannot be run, and a comment is
-cannot be run, and a comment is output to say why. If running this test
+output to say why. If running this test produces an error like this:
-produces an error like this
 
   ** Failed to set locale "fr_FR"
 
@@ -606,33 +610,37 @@ alternative output files for the third test, because three different versions
 of the French locale have been encountered. The test passes if its output
 matches any one of them.
 
-The fourth and fifth tests check UTF and Unicode property support, the fourth
+Tests 4 and 5 check UTF and Unicode property support, test 4 being compatible
-being compatible with the perltest.sh script, and the fifth checking
+with the perltest.sh script, and test 5 checking PCRE2-specific things.
-PCRE2-specific things.
 
-The sixth and seventh tests check the pcre2_dfa_match() alternative matching
+Tests 6 and 7 check the pcre2_dfa_match() alternative matching function, in
-function, in non-UTF mode and UTF-mode with Unicode property support,
+non-UTF mode and UTF-mode with Unicode property support, respectively.
-respectively.
 
-The eighth test checks some internal offsets and code size features; it is
+Test 8 checks some internal offsets and code size features; it is run only when
-run only when the default "link size" of 2 is set (in other cases the sizes
+the default "link size" of 2 is set (in other cases the sizes change) and when
-change) and when Unicode support is enabled.
+Unicode support is enabled.
 
-The ninth and tenth tests are run only in 8-bit mode, and the eleventh and
+Tests 9 and 10 are run only in 8-bit mode, and tests 11 and 12 are run only in
-twelfth tests are run only in 16-bit and 32-bit modes. These are tests that
+16-bit and 32-bit modes. These are tests that generate different output in
-generate different output in 8-bit mode. Each pair are for general cases and
+8-bit mode. Each pair are for general cases and Unicode support, respectively.
-Unicode support, respectively. The thirteenth test checks the handling of
+Test 13 checks the handling of non-UTF characters greater than 255 by
-non-UTF characters greater than 255 by pcre2_dfa_match() in 16-bit and 32-bit
+pcre2_dfa_match() in 16-bit and 32-bit modes.
-modes.
 
-The fourteenth test is run only when JIT support is not available, and the
+Test 14 contains a number of tests that must not be run with JIT. They check,
-fifteenth test is run only when JIT support is available. They test some
+among other non-JIT things, the match-limiting features of the intepretive
-JIT-specific features such as information output from pcre2test about JIT
+matcher.
-compilation.
 
-The sixteenth and seventeenth tests are run only in 8-bit mode. They check the
+Test 15 is run only when JIT support is not available. It checks that an
-POSIX interface to the 8-bit library, without and with Unicode support,
+attempt to use JIT has the expected behaviour.
-respectively.
+
+Test 16 is run only when JIT support is available. It checks JIT complete and
+partial modes, match-limiting under JIT, and other JIT-specific features.
+
+Tests 17 and 18 are run only in 8-bit mode. They check the POSIX interface to
+the 8-bit library, without and with Unicode support, respectively.
+
+Test 19 checks the serialization functions by writing a set of compiled
+patterns to a file, and then reloading and checking them.
 
 
 Character tables
@@ -718,6 +726,7 @@ The distribution should contain the files listed below.
   src/pcre2_newline.c      )
   src/pcre2_ord2utf.c      )
   src/pcre2_pattern_info.c )
+  src/pcre2_serialize.c    )
   src/pcre2_string_utils.c )
   src/pcre2_study.c        )
   src/pcre2_substitute.c   )
@@ -816,4 +825,4 @@ The distribution should contain the files listed below.
 Philip Hazel
 Email local part: ph10
 Email domain: cam.ac.uk
-Last updated: 05 January 2015
+Last updated: 20 January 2015

doc/html/index.html

@@ -65,6 +65,9 @@ first.
 <tr><td><a href="pcre2sample.html">pcre2sample</a></td>
     <td>&nbsp;&nbsp;Discussion of the pcre2demo program</td></tr>
 
+<tr><td><a href="pcre2serialize.html">pcre2serialize</a></td>
+    <td>&nbsp;&nbsp;Serializing functions for saving precompiled patterns</td></tr>
+
 <tr><td><a href="pcre2stack.html">pcre2stack</a></td>
     <td>&nbsp;&nbsp;Discussion of PCRE2's stack usage</td></tr>
 
@@ -177,6 +180,18 @@ in the library.
 <tr><td><a href="pcre2_pattern_info.html">pcre2_pattern_info</a></td>
     <td>&nbsp;&nbsp;Extract information about a pattern</td></tr>
 
+<tr><td><a href="pcre2_serialize_decode.html">pcre2_serialize_decode</a></td>
+    <td>&nbsp;&nbsp;Decode serialized compiled patterns</td></tr>
+
+<tr><td><a href="pcre2_serialize_encode.html">pcre2_serialize_encode</a></td>
+    <td>&nbsp;&nbsp;Serialize compiled patterns for save/restore</td></tr>
+
+<tr><td><a href="pcre2_serialize_free.html">pcre2_serialize_free</a></td>
+    <td>&nbsp;&nbsp;Free serialized compiled patterns</td></tr>
+
+<tr><td><a href="pcre2_serialize_get_number_of_codes.html">pcre2_serialize_get_number_of_codes</a></td>
+    <td>&nbsp;&nbsp;Get number of serialized compiled patterns</td></tr>
+
 <tr><td><a href="pcre2_set_bsr.html">pcre2_set_bsr</a></td>
     <td>&nbsp;&nbsp;Set \R convention</td></tr>
 

doc/html/pcre2_serialize_decode.html

@@ -0,0 +1,62 @@
+<html>
+<head>
+<title>pcre2_serialize_decode specification</title>
+</head>
+<body bgcolor="#FFFFFF" text="#00005A" link="#0066FF" alink="#3399FF" vlink="#2222BB">
+<h1>pcre2_serialize_decode man page</h1>
+<p>
+Return to the <a href="index.html">PCRE2 index page</a>.
+</p>
+<p>
+This page is part of the PCRE2 HTML documentation. It was generated
+automatically from the original man page. If there is any nonsense in it,
+please consult the man page, in case the conversion went wrong.
+<br>
+<br><b>
+SYNOPSIS
+</b><br>
+<P>
+<b>#include &#60;pcre2.h&#62;</b>
+</P>
+<P>
+<b>int32_t pcre2_serialize_decode(pcre2_code **<i>codes</i>,</b>
+<b>  int32_t <i>number_of_codes</i>, const uint32_t *<i>bytes</i>,</b>
+<b>  pcre2_general_context *<i>gcontext</i>);</b>
+</P>
+<br><b>
+DESCRIPTION
+</b><br>
+<P>
+This function decodes a serialized set of compiled patterns back into a list of
+individual patterns. Its arguments are:
+<pre>
+  <i>codes</i>            pointer to a vector in which to build the list
+  <i>number_of_codes</i>  number of slots in the vector
+  <i>bytes</i>            the serialized byte stream
+  <i>gcontext</i>         pointer to a general context or NULL
+</pre>
+The <i>bytes</i> argument must point to a block of data that was originally
+created by <b>pcre2_serialize_encode()</b>, though it may have been saved on
+disc or elsewhere in the meantime. If there are more codes in the serialized
+data than slots in the list, only those compiled patterns that will fit are
+decoded. The yield of the function is the number of decoded patterns, or one of
+the following negative error codes:
+<pre>
+  PCRE2_ERROR_BADDATA   <i>number_of_codes</i> is zero or less
+  PCRE2_ERROR_BADMAGIC  mismatch of id bytes in <i>bytes</i>
+  PCRE2_ERROR_BADMODE   mismatch of variable unit size or PCRE version
+  PCRE2_ERROR_MEMORY    memory allocation failed
+  PCRE2_ERROR_NULL      <i>codes</i> or <i>bytes</i> is NULL
+</pre>
+PCRE2_ERROR_BADMAGIC may mean that the data is corrupt, or that it was compiled
+on a system with different endianness.
+</P>
+<P>
+There is a complete description of the PCRE2 native API in the
+<a href="pcre2api.html"><b>pcre2api</b></a>
+page and a description of the POSIX API in the
+<a href="pcre2posix.html"><b>pcre2posix</b></a>
+page.
+<p>
+Return to the <a href="index.html">PCRE2 index page</a>.
+</p>

doc/html/pcre2_serialize_encode.html

@@ -0,0 +1,61 @@
+<html>
+<head>
+<title>pcre2_serialize_encode specification</title>
+</head>
+<body bgcolor="#FFFFFF" text="#00005A" link="#0066FF" alink="#3399FF" vlink="#2222BB">
+<h1>pcre2_serialize_encode man page</h1>
+<p>
+Return to the <a href="index.html">PCRE2 index page</a>.
+</p>
+<p>
+This page is part of the PCRE2 HTML documentation. It was generated
+automatically from the original man page. If there is any nonsense in it,
+please consult the man page, in case the conversion went wrong.
+<br>
+<br><b>
+SYNOPSIS
+</b><br>
+<P>
+<b>#include &#60;pcre2.h&#62;</b>
+</P>
+<P>
+<b>int32_t pcre2_serialize_encode(pcre2_code **<i>codes</i>,</b>
+<b>  int32_t <i>number_of_codes</i>, uint32_t **<i>serialized_bytes</i>,</b>
+<b>  PCRE2_SIZE *<i>serialized_size</i>, pcre2_general_context *<i>gcontext</i>);</b>
+</P>
+<br><b>
+DESCRIPTION
+</b><br>
+<P>
+This function encodes a list of compiled patterns into a byte stream that can
+be saved on disc or elsewhere. Its arguments are:
+<pre>
+  <i>codes</i>             pointer to a vector containing the list
+  <i>number_of_codes</i>   number of slots in the vector
+  <i>serialized_bytes</i>  set to point to the serialized byte stream
+  <i>serialized_size</i>   set to the number of bytes in the byte stream
+  <i>gcontext</i>          pointer to a general context or NULL
+</pre>
+The context argument is used to obtain memory for the byte stream. When the
+serialized data is no longer needed, it must be freed by calling
+<b>pcre2_serialize_free()</b>. The yield of the function is the number of
+serialized patterns, or one of the following negative error codes:
+<pre>
+  PCRE2_ERROR_BADDATA      <i>number_of_codes</i> is zero or less
+  PCRE2_ERROR_BADMAGIC     mismatch of id bytes in one of the patterns
+  PCRE2_ERROR_MEMORY       memory allocation failed
+  PCRE2_ERROR_MIXEDTABLES  the patterns do not all use the same tables
+  PCRE2_ERROR_NULL         an argument other than <i>gcontext</i> is NULL
+</pre>
+PCRE2_ERROR_BADMAGIC means either that a pattern's code has been corrupted, or
+that a slot in the vector does not point to a compiled pattern.
+</P>
+<P>
+There is a complete description of the PCRE2 native API in the
+<a href="pcre2api.html"><b>pcre2api</b></a>
+page and a description of the POSIX API in the
+<a href="pcre2posix.html"><b>pcre2posix</b></a>
+page.
+<p>
+Return to the <a href="index.html">PCRE2 index page</a>.
+</p>

doc/html/pcre2_serialize_free.html

@@ -0,0 +1,40 @@
+<html>
+<head>
+<title>pcre2_serialize_free specification</title>
+</head>
+<body bgcolor="#FFFFFF" text="#00005A" link="#0066FF" alink="#3399FF" vlink="#2222BB">
+<h1>pcre2_serialize_free man page</h1>
+<p>
+Return to the <a href="index.html">PCRE2 index page</a>.
+</p>
+<p>
+This page is part of the PCRE2 HTML documentation. It was generated
+automatically from the original man page. If there is any nonsense in it,
+please consult the man page, in case the conversion went wrong.
+<br>
+<br><b>
+SYNOPSIS
+</b><br>
+<P>
+<b>#include &#60;pcre2.h&#62;</b>
+</P>
+<P>
+<b>void pcre2_serialize_free(uint8_t *<i>bytes</i>);</b>
+</P>
+<br><b>
+DESCRIPTION
+</b><br>
+<P>
+This function frees the memory that was obtained by
+<b>pcre2_serialize_encode()</b> to hold a serialized byte stream. The argument
+must point to such a byte stream.
+</P>
+<P>
+There is a complete description of the PCRE2 native API in the
+<a href="pcre2api.html"><b>pcre2api</b></a>
+page and a description of the POSIX API in the
+<a href="pcre2posix.html"><b>pcre2posix</b></a>
+page.
+<p>
+Return to the <a href="index.html">PCRE2 index page</a>.
+</p>

doc/html/pcre2_serialize_get_number_of_codes.html

@@ -0,0 +1,49 @@
+<html>
+<head>
+<title>pcre2_serialize_get_number_of_codes specification</title>
+</head>
+<body bgcolor="#FFFFFF" text="#00005A" link="#0066FF" alink="#3399FF" vlink="#2222BB">
+<h1>pcre2_serialize_get_number_of_codes man page</h1>
+<p>
+Return to the <a href="index.html">PCRE2 index page</a>.
+</p>
+<p>
+This page is part of the PCRE2 HTML documentation. It was generated
+automatically from the original man page. If there is any nonsense in it,
+please consult the man page, in case the conversion went wrong.
+<br>
+<br><b>
+SYNOPSIS
+</b><br>
+<P>
+<b>#include &#60;pcre2.h&#62;</b>
+</P>
+<P>
+<b>int32_t pcre2_serialize_get_number_of_codes(const uint8_t *<i>bytes</i>);</b>
+</P>
+<br><b>
+DESCRIPTION
+</b><br>
+<P>
+The <i>bytes</i> argument must point to a serialized byte stream that was
+originally created by <b>pcre2_serialize_encode()</b> (though it may have been
+saved on disc or elsewhere in the meantime). The function returns the number of
+serialized patterns in the byte stream, or one of the following negative error
+codes:
+<pre>
+  PCRE2_ERROR_BADMAGIC  mismatch of id bytes in <i>bytes</i>
+  PCRE2_ERROR_BADMODE   mismatch of variable unit size or PCRE version
+  PCRE2_ERROR_NULL      the argument is NULL
+</pre>
+PCRE2_ERROR_BADMAGIC may mean that the data is corrupt, or that it was compiled
+on a system with different endianness.
+</P>
+<P>
+There is a complete description of the PCRE2 native API in the
+<a href="pcre2api.html"><b>pcre2api</b></a>
+page and a description of the POSIX API in the
+<a href="pcre2posix.html"><b>pcre2posix</b></a>
+page.
+<p>
+Return to the <a href="index.html">PCRE2 index page</a>.
+</p>

doc/html/pcre2api.html

@@ -21,35 +21,37 @@ please consult the man page, in case the conversion went wrong.
 <li><a name="TOC6" href="#SEC6">PCRE2 NATIVE API STRING EXTRACTION FUNCTIONS</a>
 <li><a name="TOC7" href="#SEC7">PCRE2 NATIVE API STRING SUBSTITUTION FUNCTION</a>
 <li><a name="TOC8" href="#SEC8">PCRE2 NATIVE API JIT FUNCTIONS</a>
-<li><a name="TOC9" href="#SEC9">PCRE2 NATIVE API AUXILIARY FUNCTIONS</a>
+<li><a name="TOC9" href="#SEC9">PCRE2 NATIVE API SERIALIZATION FUNCTIONS</a>
-<li><a name="TOC10" href="#SEC10">PCRE2 8-BIT, 16-BIT, AND 32-BIT LIBRARIES</a>
+<li><a name="TOC10" href="#SEC10">PCRE2 NATIVE API AUXILIARY FUNCTIONS</a>
-<li><a name="TOC11" href="#SEC11">PCRE2 API OVERVIEW</a>
+<li><a name="TOC11" href="#SEC11">PCRE2 8-BIT, 16-BIT, AND 32-BIT LIBRARIES</a>
-<li><a name="TOC12" href="#SEC12">STRING LENGTHS AND OFFSETS</a>
+<li><a name="TOC12" href="#SEC12">PCRE2 API OVERVIEW</a>
-<li><a name="TOC13" href="#SEC13">NEWLINES</a>
+<li><a name="TOC13" href="#SEC13">STRING LENGTHS AND OFFSETS</a>
-<li><a name="TOC14" href="#SEC14">MULTITHREADING</a>
+<li><a name="TOC14" href="#SEC14">NEWLINES</a>
-<li><a name="TOC15" href="#SEC15">PCRE2 CONTEXTS</a>
+<li><a name="TOC15" href="#SEC15">MULTITHREADING</a>
-<li><a name="TOC16" href="#SEC16">CHECKING BUILD-TIME OPTIONS</a>
+<li><a name="TOC16" href="#SEC16">PCRE2 CONTEXTS</a>
-<li><a name="TOC17" href="#SEC17">COMPILING A PATTERN</a>
+<li><a name="TOC17" href="#SEC17">CHECKING BUILD-TIME OPTIONS</a>
-<li><a name="TOC18" href="#SEC18">COMPILATION ERROR CODES</a>
+<li><a name="TOC18" href="#SEC18">COMPILING A PATTERN</a>
-<li><a name="TOC19" href="#SEC19">JUST-IN-TIME (JIT) COMPILATION</a>
+<li><a name="TOC19" href="#SEC19">COMPILATION ERROR CODES</a>
-<li><a name="TOC20" href="#SEC20">LOCALE SUPPORT</a>
+<li><a name="TOC20" href="#SEC20">JUST-IN-TIME (JIT) COMPILATION</a>
-<li><a name="TOC21" href="#SEC21">INFORMATION ABOUT A COMPILED PATTERN</a>
+<li><a name="TOC21" href="#SEC21">LOCALE SUPPORT</a>
-<li><a name="TOC22" href="#SEC22">THE MATCH DATA BLOCK</a>
+<li><a name="TOC22" href="#SEC22">INFORMATION ABOUT A COMPILED PATTERN</a>
-<li><a name="TOC23" href="#SEC23">MATCHING A PATTERN: THE TRADITIONAL FUNCTION</a>
+<li><a name="TOC23" href="#SEC23">SERIALIZATION AND PRECOMPILING</a>
-<li><a name="TOC24" href="#SEC24">NEWLINE HANDLING WHEN MATCHING</a>
+<li><a name="TOC24" href="#SEC24">THE MATCH DATA BLOCK</a>
-<li><a name="TOC25" href="#SEC25">HOW PCRE2_MATCH() RETURNS A STRING AND CAPTURED SUBSTRINGS</a>
+<li><a name="TOC25" href="#SEC25">MATCHING A PATTERN: THE TRADITIONAL FUNCTION</a>
-<li><a name="TOC26" href="#SEC26">OTHER INFORMATION ABOUT A MATCH</a>
+<li><a name="TOC26" href="#SEC26">NEWLINE HANDLING WHEN MATCHING</a>
-<li><a name="TOC27" href="#SEC27">ERROR RETURNS FROM <b>pcre2_match()</b></a>
+<li><a name="TOC27" href="#SEC27">HOW PCRE2_MATCH() RETURNS A STRING AND CAPTURED SUBSTRINGS</a>
-<li><a name="TOC28" href="#SEC28">EXTRACTING CAPTURED SUBSTRINGS BY NUMBER</a>
+<li><a name="TOC28" href="#SEC28">OTHER INFORMATION ABOUT A MATCH</a>
-<li><a name="TOC29" href="#SEC29">EXTRACTING A LIST OF ALL CAPTURED SUBSTRINGS</a>
+<li><a name="TOC29" href="#SEC29">ERROR RETURNS FROM <b>pcre2_match()</b></a>
-<li><a name="TOC30" href="#SEC30">EXTRACTING CAPTURED SUBSTRINGS BY NAME</a>
+<li><a name="TOC30" href="#SEC30">EXTRACTING CAPTURED SUBSTRINGS BY NUMBER</a>
-<li><a name="TOC31" href="#SEC31">CREATING A NEW STRING WITH SUBSTITUTIONS</a>
+<li><a name="TOC31" href="#SEC31">EXTRACTING A LIST OF ALL CAPTURED SUBSTRINGS</a>
-<li><a name="TOC32" href="#SEC32">DUPLICATE SUBPATTERN NAMES</a>
+<li><a name="TOC32" href="#SEC32">EXTRACTING CAPTURED SUBSTRINGS BY NAME</a>
-<li><a name="TOC33" href="#SEC33">FINDING ALL POSSIBLE MATCHES AT ONE POSITION</a>
+<li><a name="TOC33" href="#SEC33">CREATING A NEW STRING WITH SUBSTITUTIONS</a>
-<li><a name="TOC34" href="#SEC34">MATCHING A PATTERN: THE ALTERNATIVE FUNCTION</a>
+<li><a name="TOC34" href="#SEC34">DUPLICATE SUBPATTERN NAMES</a>
-<li><a name="TOC35" href="#SEC35">SEE ALSO</a>
+<li><a name="TOC35" href="#SEC35">FINDING ALL POSSIBLE MATCHES AT ONE POSITION</a>
-<li><a name="TOC36" href="#SEC36">AUTHOR</a>
+<li><a name="TOC36" href="#SEC36">MATCHING A PATTERN: THE ALTERNATIVE FUNCTION</a>
-<li><a name="TOC37" href="#SEC37">REVISION</a>
+<li><a name="TOC37" href="#SEC37">SEE ALSO</a>
+<li><a name="TOC38" href="#SEC38">AUTHOR</a>
+<li><a name="TOC39" href="#SEC39">REVISION</a>
 </ul>
 <P>
 <b>#include &#60;pcre2.h&#62;</b>
@@ -260,7 +262,24 @@ document for an overview of all the PCRE2 documentation.
 <br>
 <b>void pcre2_jit_stack_free(pcre2_jit_stack *<i>jit_stack</i>);</b>
 </P>
-<br><a name="SEC9" href="#TOC1">PCRE2 NATIVE API AUXILIARY FUNCTIONS</a><br>
+<br><a name="SEC9" href="#TOC1">PCRE2 NATIVE API SERIALIZATION FUNCTIONS</a><br>
+<P>
+<b>int32_t pcre2_serialize_decode(pcre2_code **<i>codes</i>,</b>
+<b>  int32_t <i>number_of_codes</i>, const uint32_t *<i>bytes</i>,</b>
+<b>  pcre2_general_context *<i>gcontext</i>);</b>
+<br>
+<br>
+<b>int32_t pcre2_serialize_encode(pcre2_code **<i>codes</i>,</b>
+<b>  int32_t <i>number_of_codes</i>, uint32_t **<i>serialized_bytes</i>,</b>
+<b>  PCRE2_SIZE *<i>serialized_size</i>, pcre2_general_context *<i>gcontext</i>);</b>
+<br>
+<br>
+<b>void pcre2_serialize_free(uint8_t *<i>bytes</i>);</b>
+<br>
+<br>
+<b>int32_t pcre2_serialize_get_number_of_codes(const uint8_t *<i>bytes</i>);</b>
+</P>
+<br><a name="SEC10" href="#TOC1">PCRE2 NATIVE API AUXILIARY FUNCTIONS</a><br>
 <P>
 <b>int pcre2_get_error_message(int <i>errorcode</i>, PCRE2_UCHAR *<i>buffer</i>,</b>
 <b>  PCRE2_SIZE <i>bufflen</i>);</b>
@@ -274,7 +293,7 @@ document for an overview of all the PCRE2 documentation.
 <br>
 <b>int pcre2_config(uint32_t <i>what</i>, void *<i>where</i>);</b>
 </P>
-<br><a name="SEC10" href="#TOC1">PCRE2 8-BIT, 16-BIT, AND 32-BIT LIBRARIES</a><br>
+<br><a name="SEC11" href="#TOC1">PCRE2 8-BIT, 16-BIT, AND 32-BIT LIBRARIES</a><br>
 <P>
 There are three PCRE2 libraries, supporting 8-bit, 16-bit, and 32-bit code
 units, respectively. However, there is just one header file, <b>pcre2.h</b>.
@@ -335,7 +354,7 @@ In the function summaries above, and in the rest of this document and other
 PCRE2 documents, functions and data types are described using their generic
 names, without the 8, 16, or 32 suffix.
 </P>
-<br><a name="SEC11" href="#TOC1">PCRE2 API OVERVIEW</a><br>
+<br><a name="SEC12" href="#TOC1">PCRE2 API OVERVIEW</a><br>
 <P>
 PCRE2 has its own native API, which is described in this document. There are
 also some wrapper functions for the 8-bit library that correspond to the
@@ -426,7 +445,7 @@ Finally, there are functions for finding out information about a compiled
 pattern (<b>pcre2_pattern_info()</b>) and about the configuration with which
 PCRE2 was built (<b>pcre2_config()</b>).
 </P>
-<br><a name="SEC12" href="#TOC1">STRING LENGTHS AND OFFSETS</a><br>
+<br><a name="SEC13" href="#TOC1">STRING LENGTHS AND OFFSETS</a><br>
 <P>
 The PCRE2 API uses string lengths and offsets into strings of code units in
 several places. These values are always of type PCRE2_SIZE, which is an
@@ -436,7 +455,7 @@ as a special indicator for zero-terminated strings and unset offsets.
 Therefore, the longest string that can be handled is one less than this
 maximum.
 <a name="newlines"></a></P>
-<br><a name="SEC13" href="#TOC1">NEWLINES</a><br>
+<br><a name="SEC14" href="#TOC1">NEWLINES</a><br>
 <P>
 PCRE2 supports five different conventions for indicating line breaks in
 strings: a single CR (carriage return) character, a single LF (linefeed)
@@ -471,7 +490,7 @@ The choice of newline convention does not affect the interpretation of
 the \n or \r escape sequences, nor does it affect what \R matches; this has
 its own separate convention.
 </P>
-<br><a name="SEC14" href="#TOC1">MULTITHREADING</a><br>
+<br><a name="SEC15" href="#TOC1">MULTITHREADING</a><br>
 <P>
 In a multithreaded application it is important to keep thread-specific data
 separate from data that can be shared between threads. The PCRE2 library code
@@ -516,7 +535,7 @@ storing the results of a match. This includes details of what was matched, as
 well as additional information such as the name of a (*MARK) setting. Each
 thread must provide its own version of this memory.
 </P>
-<br><a name="SEC15" href="#TOC1">PCRE2 CONTEXTS</a><br>
+<br><a name="SEC16" href="#TOC1">PCRE2 CONTEXTS</a><br>
 <P>
 Some PCRE2 functions have a lot of parameters, many of which are used only by
 specialist applications, for example, those that use custom memory management
@@ -797,7 +816,7 @@ exit so that they can be re-used when possible during the match. In the absence
 of these functions, the normal custom memory management functions are used, if
 supplied, otherwise the system functions.
 </P>
-<br><a name="SEC16" href="#TOC1">CHECKING BUILD-TIME OPTIONS</a><br>
+<br><a name="SEC17" href="#TOC1">CHECKING BUILD-TIME OPTIONS</a><br>
 <P>
 <b>int pcre2_config(uint32_t <i>what</i>, void *<i>where</i>);</b>
 </P>
@@ -929,7 +948,7 @@ the PCRE2 version string, zero-terminated. The number of code units used is
 returned. This is the length of the string plus one unit for the terminating
 zero.
 <a name="compiling"></a></P>
-<br><a name="SEC17" href="#TOC1">COMPILING A PATTERN</a><br>
+<br><a name="SEC18" href="#TOC1">COMPILING A PATTERN</a><br>
 <P>
 <b>pcre2_code *pcre2_compile(PCRE2_SPTR <i>pattern</i>, PCRE2_SIZE <i>length</i>,</b>
 <b>  uint32_t <i>options</i>, int *<i>errorcode</i>, PCRE2_SIZE *<i>erroroffset,</i></b>
@@ -1305,7 +1324,7 @@ the behaviour of PCRE2 are given in the
 <a href="pcre2unicode.html"><b>pcre2unicode</b></a>
 page.
 </P>
-<br><a name="SEC18" href="#TOC1">COMPILATION ERROR CODES</a><br>
+<br><a name="SEC19" href="#TOC1">COMPILATION ERROR CODES</a><br>
 <P>
 There are over 80 positive error codes that <b>pcre2_compile()</b> may return if
 it finds an error in the pattern. There are also some negative error codes that
@@ -1315,7 +1334,7 @@ are used for invalid UTF strings. These are the same as given by
 page. The <b>pcre2_get_error_message()</b> function can be called to obtain a
 textual error message from any error code.
 </P>
-<br><a name="SEC19" href="#TOC1">JUST-IN-TIME (JIT) COMPILATION</a><br>
+<br><a name="SEC20" href="#TOC1">JUST-IN-TIME (JIT) COMPILATION</a><br>
 <P>
 <b>int pcre2_jit_compile(pcre2_code *<i>code</i>, uint32_t <i>options</i>);</b>
 <br>
@@ -1353,7 +1372,7 @@ patterns to be analyzed, and for one-off matches and simple patterns the
 benefit of faster execution might be offset by a much slower compilation time.
 Most, but not all patterns can be optimized by the JIT compiler.
 <a name="localesupport"></a></P>
-<br><a name="SEC20" href="#TOC1">LOCALE SUPPORT</a><br>
+<br><a name="SEC21" href="#TOC1">LOCALE SUPPORT</a><br>
 <P>
 PCRE2 handles caseless matching, and determines whether characters are letters,
 digits, or whatever, by reference to a set of tables, indexed by character code
@@ -1409,7 +1428,7 @@ is saved with the compiled pattern, and the same tables are used by
 compilation, and matching all happen in the same locale, but different patterns
 can be processed in different locales.
 <a name="infoaboutpattern"></a></P>
-<br><a name="SEC21" href="#TOC1">INFORMATION ABOUT A COMPILED PATTERN</a><br>
+<br><a name="SEC22" href="#TOC1">INFORMATION ABOUT A COMPILED PATTERN</a><br>
 <P>
 <b>int pcre2_pattern_info(const pcre2 *<i>code</i>, uint32_t <i>what</i>, void *<i>where</i>);</b>
 </P>
@@ -1478,8 +1497,12 @@ options returned for PCRE2_INFO_ALLOPTIONS.
   PCRE2_INFO_BACKREFMAX
 </pre>
 Return the number of the highest back reference in the pattern. The third
-argument should point to an <b>uint32_t</b> variable. Zero is returned if there
+argument should point to an <b>uint32_t</b> variable. Named subpatterns acquire
-are no back references.
+numbers as well as names, and these count towards the highest back reference.
+Back references such as \4 or \g{12} match the captured characters of the
+given group, but in addition, the check that a capturing group is set in a
+conditional subpattern such as (?(3)a|b) is also a back reference. Zero is
+returned if there are no back references.
 <pre>
   PCRE2_INFO_BSR
 </pre>
@@ -1689,14 +1712,24 @@ set, the call to <b>pcre2_pattern_info()</b> returns the error PCRE2_ERROR_UNSET
   PCRE2_INFO_SIZE
 </pre>
 Return the size of the compiled pattern in bytes (for all three libraries). The
-third argument should point to a <b>size_t</b> variable. This value does not
+third argument should point to a <b>size_t</b> variable. This value includes the
-include the size of the <b>pcre2_code</b> structure that is returned by
+size of the general data block that precedes the code units of the compiled
-<b>pcre_compile()</b>. The value that is used when <b>pcre2_compile()</b> is
+pattern itself. The value that is used when <b>pcre2_compile()</b> is getting
-getting memory in which to place the compiled data is the value returned by
+memory in which to place the compiled pattern may be slightly larger than the
-this option plus the size of the <b>pcre2_code</b> structure. Processing a
+value returned by this option, because there are cases where the code that
-pattern with the JIT compiler does not alter the value returned by this option.
+calculates the size has to over-estimate. Processing a pattern with the JIT
+compiler does not alter the value returned by this option.
+</P>
+<br><a name="SEC23" href="#TOC1">SERIALIZATION AND PRECOMPILING</a><br>
+<P>
+It is possible to save compiled patterns on disc or elsewhere, and reload them
+later, subject to a number of restrictions. The functions whose names begin
+with <b>pcre2_serialize_</b> are used for this purpose. They are described in
+the
+<a href="pcre2serialize.html"><b>pcre2serialize</b></a>
+documentation.
 <a name="matchdatablock"></a></P>
-<br><a name="SEC22" href="#TOC1">THE MATCH DATA BLOCK</a><br>
+<br><a name="SEC24" href="#TOC1">THE MATCH DATA BLOCK</a><br>
 <P>
 <b>pcre2_match_data_create(uint32_t <i>ovecsize</i>,</b>
 <b>  pcre2_general_context *<i>gcontext</i>);</b>
@@ -1767,7 +1800,7 @@ match data block (for that match) have taken place.
 When a match data block itself is no longer needed, it should be freed by
 calling <b>pcre2_match_data_free()</b>.
 </P>
-<br><a name="SEC23" href="#TOC1">MATCHING A PATTERN: THE TRADITIONAL FUNCTION</a><br>
+<br><a name="SEC25" href="#TOC1">MATCHING A PATTERN: THE TRADITIONAL FUNCTION</a><br>
 <P>
 <b>int pcre2_match(const pcre2_code *<i>code</i>, PCRE2_SPTR <i>subject</i>,</b>
 <b>  PCRE2_SIZE <i>length</i>, PCRE2_SIZE <i>startoffset</i>,</b>
@@ -1981,7 +2014,7 @@ examples, in the
 <a href="pcre2partial.html"><b>pcre2partial</b></a>
 documentation.
 </P>
-<br><a name="SEC24" href="#TOC1">NEWLINE HANDLING WHEN MATCHING</a><br>
+<br><a name="SEC26" href="#TOC1">NEWLINE HANDLING WHEN MATCHING</a><br>
 <P>
 When PCRE2 is built, a default newline convention is set; this is usually the
 standard convention for the operating system. The default can be overridden in
@@ -2016,7 +2049,7 @@ LF in the characters that it matches.
 Notwithstanding the above, anomalous effects may still occur when CRLF is a
 valid newline sequence and explicit \r or \n escapes appear in the pattern.
 <a name="matchedstrings"></a></P>
-<br><a name="SEC25" href="#TOC1">HOW PCRE2_MATCH() RETURNS A STRING AND CAPTURED SUBSTRINGS</a><br>
+<br><a name="SEC27" href="#TOC1">HOW PCRE2_MATCH() RETURNS A STRING AND CAPTURED SUBSTRINGS</a><br>
 <P>
 <b>uint32_t pcre2_get_ovector_count(pcre2_match_data *<i>match_data</i>);</b>
 <br>
@@ -2118,7 +2151,7 @@ parentheses, no more than <i>ovector[0]</i> to <i>ovector[2n+1]</i> are set by
 <b>pcre2_match()</b>. The other elements retain whatever values they previously
 had.
 <a name="matchotherdata"></a></P>
-<br><a name="SEC26" href="#TOC1">OTHER INFORMATION ABOUT A MATCH</a><br>
+<br><a name="SEC28" href="#TOC1">OTHER INFORMATION ABOUT A MATCH</a><br>
 <P>
 <b>PCRE2_SPTR pcre2_get_mark(pcre2_match_data *<i>match_data</i>);</b>
 <br>
@@ -2162,7 +2195,7 @@ the code unit offset of the invalid UTF character. Details are given in the
 <a href="pcre2unicode.html"><b>pcre2unicode</b></a>
 page.
 <a name="errorlist"></a></P>
-<br><a name="SEC27" href="#TOC1">ERROR RETURNS FROM <b>pcre2_match()</b></a><br>
+<br><a name="SEC29" href="#TOC1">ERROR RETURNS FROM <b>pcre2_match()</b></a><br>
 <P>
 If <b>pcre2_match()</b> fails, it returns a negative number. This can be
 converted to a text string by calling <b>pcre2_get_error_message()</b>. Negative
@@ -2271,7 +2304,7 @@ is attempted.
 </pre>
 The internal recursion limit was reached.
 <a name="extractbynumber"></a></P>
-<br><a name="SEC28" href="#TOC1">EXTRACTING CAPTURED SUBSTRINGS BY NUMBER</a><br>
+<br><a name="SEC30" href="#TOC1">EXTRACTING CAPTURED SUBSTRINGS BY NUMBER</a><br>
 <P>
 <b>int pcre2_substring_length_bynumber(pcre2_match_data *<i>match_data</i>,</b>
 <b>  uint32_t <i>number</i>, PCRE2_SIZE *<i>length</i>);</b>
@@ -2368,7 +2401,7 @@ The substring did not participate in the match. For example, if the pattern is
 (abc)|(def) and the subject is "def", and the ovector contains at least two
 capturing slots, substring number 1 is unset.
 </P>
-<br><a name="SEC29" href="#TOC1">EXTRACTING A LIST OF ALL CAPTURED SUBSTRINGS</a><br>
+<br><a name="SEC31" href="#TOC1">EXTRACTING A LIST OF ALL CAPTURED SUBSTRINGS</a><br>
 <P>
 <b>int pcre2_substring_list_get(pcre2_match_data *<i>match_data</i>,</b>
 <b>"  PCRE2_UCHAR ***<i>listptr</i>, PCRE2_SIZE **<i>lengthsptr</i>);</b>
@@ -2407,7 +2440,7 @@ can be distinguished from a genuine zero-length substring by inspecting the
 appropriate offset in the ovector, which contain PCRE2_UNSET for unset
 substrings, or by calling <b>pcre2_substring_length_bynumber()</b>.
 <a name="extractbyname"></a></P>
-<br><a name="SEC30" href="#TOC1">EXTRACTING CAPTURED SUBSTRINGS BY NAME</a><br>
+<br><a name="SEC32" href="#TOC1">EXTRACTING CAPTURED SUBSTRINGS BY NAME</a><br>
 <P>
 <b>int pcre2_substring_number_from_name(const pcre2_code *<i>code</i>,</b>
 <b>  PCRE2_SPTR <i>name</i>);</b>
@@ -2467,7 +2500,7 @@ names are not included in the compiled code. The matching process uses only
 numbers. For this reason, the use of different names for subpatterns of the
 same number causes an error at compile time.
 </P>
-<br><a name="SEC31" href="#TOC1">CREATING A NEW STRING WITH SUBSTITUTIONS</a><br>
+<br><a name="SEC33" href="#TOC1">CREATING A NEW STRING WITH SUBSTITUTIONS</a><br>
 <P>
 <b>int pcre2_substitute(const pcre2_code *<i>code</i>, PCRE2_SPTR <i>subject</i>,</b>
 <b>  PCRE2_SIZE <i>length</i>, PCRE2_SIZE <i>startoffset</i>,</b>
@@ -2528,7 +2561,7 @@ straight back. PCRE2_ERROR_BADREPLACEMENT is returned for an invalid
 replacement string (unrecognized sequence following a dollar sign), and
 PCRE2_ERROR_NOMEMORY is returned if the output buffer is not big enough.
 </P>
-<br><a name="SEC32" href="#TOC1">DUPLICATE SUBPATTERN NAMES</a><br>
+<br><a name="SEC34" href="#TOC1">DUPLICATE SUBPATTERN NAMES</a><br>
 <P>
 <b>int pcre2_substring_nametable_scan(const pcre2_code *<i>code</i>,</b>
 <b>  PCRE2_SPTR <i>name</i>, PCRE2_SPTR *<i>first</i>, PCRE2_SPTR *<i>last</i>);</b>
@@ -2573,7 +2606,7 @@ The format of the name table is described above in the section entitled
 Given all the relevant entries for the name, you can extract each of their
 numbers, and hence the captured data.
 </P>
-<br><a name="SEC33" href="#TOC1">FINDING ALL POSSIBLE MATCHES AT ONE POSITION</a><br>
+<br><a name="SEC35" href="#TOC1">FINDING ALL POSSIBLE MATCHES AT ONE POSITION</a><br>
 <P>
 The traditional matching function uses a similar algorithm to Perl, which stops
 when it finds the first match at a given point in the subject. If you want to
@@ -2591,7 +2624,7 @@ substring. Then return 1, which forces <b>pcre2_match()</b> to backtrack and try
 other alternatives. Ultimately, when it runs out of matches,
 <b>pcre2_match()</b> will yield PCRE2_ERROR_NOMATCH.
 <a name="dfamatch"></a></P>
-<br><a name="SEC34" href="#TOC1">MATCHING A PATTERN: THE ALTERNATIVE FUNCTION</a><br>
+<br><a name="SEC36" href="#TOC1">MATCHING A PATTERN: THE ALTERNATIVE FUNCTION</a><br>
 <P>
 <b>int pcre2_dfa_match(const pcre2_code *<i>code</i>, PCRE2_SPTR <i>subject</i>,</b>
 <b>  PCRE2_SIZE <i>length</i>, PCRE2_SIZE <i>startoffset</i>,</b>
@@ -2786,13 +2819,13 @@ some plausibility checks are made on the contents of the workspace, which
 should contain data about the previous partial match. If any of these checks
 fail, this error is given.
 </P>
-<br><a name="SEC35" href="#TOC1">SEE ALSO</a><br>
+<br><a name="SEC37" href="#TOC1">SEE ALSO</a><br>
 <P>
 <b>pcre2build</b>(3), <b>pcre2callout</b>(3), <b>pcre2demo(3)</b>,
 <b>pcre2matching</b>(3), <b>pcre2partial</b>(3), <b>pcre2posix</b>(3),
 <b>pcre2sample</b>(3), <b>pcre2stack</b>(3), <b>pcre2unicode</b>(3).
 </P>
-<br><a name="SEC36" href="#TOC1">AUTHOR</a><br>
+<br><a name="SEC38" href="#TOC1">AUTHOR</a><br>
 <P>
 Philip Hazel
 <br>
@@ -2801,9 +2834,9 @@ University Computing Service
 Cambridge, England.
 <br>
 </P>
-<br><a name="SEC37" href="#TOC1">REVISION</a><br>
+<br><a name="SEC39" href="#TOC1">REVISION</a><br>
 <P>
-Last updated: 02 January 2015
+Last updated: 23 January 2015
 <br>
 Copyright &copy; 1997-2015 University of Cambridge.
 <br>

doc/html/pcre2serialize.html

@@ -0,0 +1,184 @@
+<html>
+<head>
+<title>pcre2serialize specification</title>
+</head>
+<body bgcolor="#FFFFFF" text="#00005A" link="#0066FF" alink="#3399FF" vlink="#2222BB">
+<h1>pcre2serialize man page</h1>
+<p>
+Return to the <a href="index.html">PCRE2 index page</a>.
+</p>
+<p>
+This page is part of the PCRE2 HTML documentation. It was generated
+automatically from the original man page. If there is any nonsense in it,
+please consult the man page, in case the conversion went wrong.
+<br>
+<ul>
+<li><a name="TOC1" href="#SEC1">SAVING AND RE-USING PRECOMPILED PCRE2 PATTERNS</a>
+<li><a name="TOC2" href="#SEC2">SAVING COMPILED PATTERNS</a>
+<li><a name="TOC3" href="#SEC3">RE-USING PRECOMPILED PATTERNS</a>
+<li><a name="TOC4" href="#SEC4">AUTHOR</a>
+<li><a name="TOC5" href="#SEC5">REVISION</a>
+</ul>
+<br><a name="SEC1" href="#TOC1">SAVING AND RE-USING PRECOMPILED PCRE2 PATTERNS</a><br>
+<P>
+<b>int32_t pcre2_serialize_decode(pcre2_code **<i>codes</i>,</b>
+<b>  int32_t <i>number_of_codes</i>, const uint32_t *<i>bytes</i>,</b>
+<b>  pcre2_general_context *<i>gcontext</i>);</b>
+<br>
+<br>
+<b>int32_t pcre2_serialize_encode(pcre2_code **<i>codes</i>,</b>
+<b>  int32_t <i>number_of_codes</i>, uint32_t **<i>serialized_bytes</i>,</b>
+<b>  PCRE2_SIZE *<i>serialized_size</i>, pcre2_general_context *<i>gcontext</i>);</b>
+<br>
+<br>
+<b>void pcre2_serialize_free(uint8_t *<i>bytes</i>);</b>
+<br>
+<br>
+<b>int32_t pcre2_serialize_get_number_of_codes(const uint8_t *<i>bytes</i>);</b>
+<br>
+<br>
+If you are running an application that uses a large number of regular
+expression patterns, it may be useful to store them in a precompiled form
+instead of having to compile them every time the application is run. However,
+if you are using the just-in-time optimization feature, it is not possible to
+save and reload the JIT data, because it is position-dependent. In addition,
+the host on which the patterns are reloaded must be running the same version of
+PCRE2, with the same code unit width, and must also have the same endianness,
+pointer width and PCRE2_SIZE type. For example, patterns compiled on a 32-bit
+system using PCRE2's 16-bit library cannot be reloaded on a 64-bit system, nor
+can they be reloaded using the 8-bit library.
+</P>
+<br><a name="SEC2" href="#TOC1">SAVING COMPILED PATTERNS</a><br>
+<P>
+Before compiled patterns can be saved they must be serialized, that is,
+converted to a stream of bytes. A single byte stream may contain any number of
+compiled patterns, but they must all use the same character tables. A single
+copy of the tables is included in the byte stream (its size is 1088 bytes). For
+more details of character tables, see the
+<a href="pcre2api.html#localesupport">section on locale support</a>
+in the
+<a href="pcre2api.html"><b>pcre2api</b></a>
+documentation.
+</P>
+<P>
+The function <b>pcre2_serialize_encode()</b> creates a serialized byte stream
+from a list of compiled patterns. Its first two arguments specify the list,
+being a pointer to a vector of pointers to compiled patterns, and the length of
+the vector. The third and fourth arguments point to variables which are set to
+point to the created byte stream and its length, respectively. The final
+argument is a pointer to a general context, which can be used to specify custom
+memory mangagement functions. If this argument is NULL, <b>malloc()</b> is used
+to obtain memory for the byte stream. The yield of the function is the number
+of serialized patterns, or one of the following negative error codes:
+<pre>
+  PCRE2_ERROR_BADDATA      the number of patterns is zero or less
+  PCRE2_ERROR_BADMAGIC     mismatch of id bytes in one of the patterns
+  PCRE2_ERROR_MEMORY       memory allocation failed
+  PCRE2_ERROR_MIXEDTABLES  the patterns do not all use the same tables
+  PCRE2_ERROR_NULL         the 1st, 3rd, or 4th argument is NULL
+</pre>
+PCRE2_ERROR_BADMAGIC means either that a pattern's code has been corrupted, or
+that a slot in the vector does not point to a compiled pattern.
+</P>
+<P>
+Once a set of patterns has been serialized you can save the data in any
+appropriate manner. Here is sample code that compiles two patterns and writes
+them to a file. It assumes that the variable <i>fd</i> refers to a file that is
+open for output. The error checking that should be present in a real
+application has been omitted for simplicity.
+<pre>
+  int errorcode;
+  uint8_t *bytes;
+  PCRE2_SIZE erroroffset;
+  PCRE2_SIZE bytescount;
+  pcre2_code *list_of_codes[2];
+  list_of_codes[0] = pcre2_compile("first pattern",
+    PCRE2_ZERO_TERMINATED, 0, &errorcode, &erroroffset, NULL);
+  list_of_codes[1] = pcre2_compile("second pattern",
+    PCRE2_ZERO_TERMINATED, 0, &errorcode, &erroroffset, NULL);
+  errorcode = pcre2_serialize_encode(list_of_codes, 2, &bytes,
+    &bytescount, NULL);
+  errorcode = fwrite(bytes, 1, bytescount, fd);
+</pre>
+Note that the serialized data is binary data that may contain any of the 256
+possible byte values. On systems that make a distinction between binary and
+non-binary data, be sure that the file is opened for binary output.
+</P>
+<P>
+Serializing a set of patterns leaves the original data untouched, so they can
+still be used for matching. Their memory must eventually be freed in the usual
+way by calling <b>pcre2_code_free()</b>. When you have finished with the byte
+stream, it too must be freed by calling <b>pcre2_serialize_free()</b>.
+</P>
+<br><a name="SEC3" href="#TOC1">RE-USING PRECOMPILED PATTERNS</a><br>
+<P>
+In order to re-use a set of saved patterns you must first make the serialized
+byte stream available in main memory (for example, by reading from a file). The
+management of this memory block is up to the application. You can use the
+<b>pcre2_serialize_get_number_of_codes()</b> function to find out how many
+compiled patterns are in the serialized data without actually decoding the
+patterns:
+<pre>
+  uint8_t *bytes = &#60;serialized data&#62;;
+  int32_t number_of_codes = pcre2_serialize_get_number_of_codes(bytes);
+</pre>
+The <b>pcre2_serialize_decode()</b> function reads a byte stream and recreates
+the compiled patterns in new memory blocks, setting pointers to them in a
+vector. The first two arguments are a pointer to a suitable vector and its
+length, and the third argument points to a byte stream. The final argument is a
+pointer to a general context, which can be used to specify custom memory
+mangagement functions for the decoded patterns. If this argument is NULL,
+<b>malloc()</b> and <b>free()</b> are used. After deserialization, the byte
+stream is no longer needed and can be discarded.
+<pre>
+  int32_t number_of_codes;
+  pcre2_code *list_of_codes[2];
+  uint8_t *bytes = &#60;serialized data&#62;;
+  int32_t number_of_codes =
+    pcre2_serialize_decode(list_of_codes, 2, bytes, NULL);
+</pre>
+If the vector is not large enough for all the patterns in the byte stream, it
+is filled with those that fit, and the remainder are ignored. The yield of the
+function is the number of decoded patterns, or one of the following negative
+error codes:
+<pre>
+  PCRE2_ERROR_BADDATA   second argument is zero or less
+  PCRE2_ERROR_BADMAGIC  mismatch of id bytes in the data
+  PCRE2_ERROR_BADMODE   mismatch of variable unit size or PCRE2 version
+  PCRE2_ERROR_MEMORY    memory allocation failed
+  PCRE2_ERROR_NULL      first or third argument is NULL
+</pre>
+PCRE2_ERROR_BADMAGIC may mean that the data is corrupt, or that it was compiled
+on a system with different endianness.
+</P>
+<P>
+Decoded patterns can be used for matching in the usual way, and must be freed
+by calling <b>pcre2_code_free()</b> as normal. A single copy of the character
+tables is used by all the decoded patterns. A reference count is used to
+arrange for its memory to be automatically freed when the last pattern is
+freed.
+</P>
+<P>
+If a pattern was processed by <b>pcre2_jit_compile()</b> before being
+serialized, the JIT data is discarded and so is no longer available after a
+save/restore cycle. You can, however, process a restored pattern with
+<b>pcre2_jit_compile()</b> if you wish.
+</P>
+<br><a name="SEC4" href="#TOC1">AUTHOR</a><br>
+<P>
+Philip Hazel
+<br>
+University Computing Service
+<br>
+Cambridge, England.
+<br>
+</P>
+<br><a name="SEC5" href="#TOC1">REVISION</a><br>
+<P>
+Last updated: 20 January 2015
+<br>
+Copyright &copy; 1997-2015 University of Cambridge.
+<br>
+<p>
+Return to the <a href="index.html">PCRE2 index page</a>.
+</p>

doc/html/pcre2test.html

@@ -30,9 +30,10 @@ please consult the man page, in case the conversion went wrong.
 <li><a name="TOC15" href="#SEC15">RESTARTING AFTER A PARTIAL MATCH</a>
 <li><a name="TOC16" href="#SEC16">CALLOUTS</a>
 <li><a name="TOC17" href="#SEC17">NON-PRINTING CHARACTERS</a>
-<li><a name="TOC18" href="#SEC18">SEE ALSO</a>
+<li><a name="TOC18" href="#SEC18">SAVING AND RESTORING COMPILED PATTERNS</a>
-<li><a name="TOC19" href="#SEC19">AUTHOR</a>
+<li><a name="TOC19" href="#SEC19">SEE ALSO</a>
-<li><a name="TOC20" href="#SEC20">REVISION</a>
+<li><a name="TOC20" href="#SEC20">AUTHOR</a>
+<li><a name="TOC21" href="#SEC21">REVISION</a>
 </ul>
 <br><a name="SEC1" href="#TOC1">SYNOPSIS</a><br>
 <P>
@@ -51,10 +52,11 @@ documentation.
 </P>
 <P>
 The input for <b>pcre2test</b> is a sequence of regular expression patterns and
-subject strings to be matched. The output shows the result of each match
+subject strings to be matched. There are also command lines for setting
-attempt. Modifiers on the command line, the patterns, and the subject lines
+defaults and controlling some special actions. The output shows the result of
-specify PCRE2 function options, control how the subject is processed, and what
+each match attempt. Modifiers on external or internal command lines, the
-output is produced.
+patterns, and the subject lines specify PCRE2 function options, control how the
+subject is processed, and what output is produced.
 </P>
 <P>
 As the original fairly simple PCRE library evolved, it acquired many different
@@ -227,9 +229,7 @@ If <b>pcre2test</b> is given two filename arguments, it reads from the first and
 writes to the second. If the first name is "-", input is taken from the
 standard input. If <b>pcre2test</b> is given only one argument, it reads from
 that file and writes to stdout. Otherwise, it reads from stdin and writes to
-stdout. When the input is a terminal, it prompts for each line of input, using
+stdout.
-"re&#62;" to prompt for regular expression patterns, and "data&#62;" to prompt for
-subject lines.
 </P>
 <P>
 When <b>pcre2test</b> is built, a configuration option can specify that it
@@ -242,10 +242,16 @@ the <b>-help</b> option states whether or not <b>readline()</b> will be used.
 The program handles any number of tests, each of which consists of a set of
 input lines. Each set starts with a regular expression pattern, followed by any
 number of subject lines to be matched against that pattern. In between sets of
-test data, command lines that begin with a hash (#) character may appear. This
+test data, command lines that begin with # may appear. This file format, with
-file format, with some restrictions, can also be processed by the
+some restrictions, can also be processed by the <b>perltest.sh</b> script that
-<b>perltest.sh</b> script that is distributed with PCRE2 as a means of checking
+is distributed with PCRE2 as a means of checking that the behaviour of PCRE2
-that the behaviour of PCRE2 and Perl is the same.
+and Perl is the same.
+</P>
+<P>
+When the input is a terminal, <b>pcre2test</b> prompts for each line of input,
+using "re&#62;" to prompt for regular expression patterns, and "data&#62;" to prompt
+for subject lines. Command lines starting with # can be entered only in
+response to the "re&#62;" prompt.
 </P>
 <P>
 Each subject line is matched separately and independently. If you want to do
@@ -263,21 +269,27 @@ still input to be read.
 </P>
 <br><a name="SEC6" href="#TOC1">COMMAND LINES</a><br>
 <P>
-In between sets of test data, a line that begins with a hash (#) character is
+In between sets of test data, a line that begins with # is interpreted as a
-interpreted as a command line. If the first character is followed by white
+command line. If the first character is followed by white space or an
-space or an exclamation mark, the line is treated as a comment, and ignored.
+exclamation mark, the line is treated as a comment, and ignored. Otherwise, the
-Otherwise, the following commands are recognized:
+following commands are recognized:
 <pre>
   #forbid_utf
 </pre>
 Subsequent patterns automatically have the PCRE2_NEVER_UTF and PCRE2_NEVER_UCP
 options set, which locks out the use of UTF and Unicode property features. This
-is a trigger guard that is used in test files to ensure that UTF/Unicode tests
+is a trigger guard that is used in test files to ensure that UTF or Unicode
-are not accidentally added to files that are used when UTF support is not
+property tests are not accidentally added to files that are used when Unicode
-included in the library. This effect can also be obtained by the use of
+support is not included in the library. This effect can also be obtained by the
-<b>#pattern</b>; the difference is that <b>#forbid_utf</b> cannot be unset, and
+use of <b>#pattern</b>; the difference is that <b>#forbid_utf</b> cannot be
-the automatic options are not displayed in pattern information, to avoid
+unset, and the automatic options are not displayed in pattern information, to
-cluttering up test output.
+avoid cluttering up test output.
+<pre>
+  #load &#60;filename&#62;
+</pre>
+This command is used to load a set of precompiled patterns from a file, as
+described in the section entitled "Saving and restoring compiled patterns"
+<a href="#saverestore">below.</a>
 <pre>
   #pattern &#60;modifier-list&#62;
 </pre>
@@ -293,6 +305,18 @@ lines, none of the other command lines are permitted, because they and many
 of the modifiers are specific to <b>pcre2test</b>, and should not be used in
 test files that are also processed by <b>perltest.sh</b>. The <b>#perltest</b>
 command helps detect tests that are accidentally put in the wrong file.
+<pre>
+  #pop [&#60;modifiers&#62;]
+</pre>
+This command is used to manipulate the stack of compiled patterns, as described
+in the section entitled "Saving and restoring compiled patterns"
+<a href="#saverestore">below.</a>
+<pre>
+  #save &#60;filename&#62;
+</pre>
+This command is used to save a set of compiled patterns to a file, as described
+in the section entitled "Saving and restoring compiled patterns"
+<a href="#saverestore">below.</a>
 <pre>
   #subject &#60;modifier-list&#62;
 </pre>
@@ -428,7 +452,7 @@ There are three types of modifier that can appear in pattern lines, two of
 which may also be used in a <b>#pattern</b> command. A pattern's modifier list
 can add to or override default modifiers that were set by a previous
 <b>#pattern</b> command.
-</P>
+<a name="optionmodifiers"></a></P>
 <br><b>
 Setting compilation options
 </b><br>
@@ -465,7 +489,7 @@ As well as turning on the PCRE2_UTF option, the <b>utf</b> modifier causes all
 non-printing characters in output strings to be printed using the \x{hh...}
 notation. Otherwise, those less than 0x100 are output in hex without the curly
 brackets.
-</P>
+<a name="controlmodifiers"></a></P>
 <br><b>
 Setting compilation controls
 </b><br>
@@ -486,8 +510,8 @@ about the pattern:
       memory                    show memory used
       newline=&#60;type&#62;            set newline type
       parens_nest_limit=&#60;n&#62;     set maximum parentheses depth
-      perlcompat                lock out non-Perl modifiers
       posix                     use the POSIX API
+      push                      push compiled pattern onto the stack
       stackguard=&#60;number&#62;       test the stackguard feature
       tables=[0|1|2]            select internal tables
 </pre>
@@ -726,6 +750,22 @@ not affect the compilation process.
 These modifiers may not appear in a <b>#pattern</b> command. If you want them as
 defaults, set them in a <b>#subject</b> command.
 </P>
+<br><b>
+Saving a compiled pattern
+</b><br>
+<P>
+When a pattern with the <b>push</b> modifier is successfully compiled, it is
+pushed onto a stack of compiled patterns, and <b>pcre2test</b> expects the next
+line to contain a new pattern (or a command) instead of a subject line. This
+facility is used when saving compiled patterns to a file, as described in the
+section entitled "Saving and restoring compiled patterns"
+<a href="#saverestore">below.</a>
+The <b>push</b> modifier is incompatible with compilation modifiers such as
+<b>global</b> that act at match time. Any that are specified are ignored, with a
+warning message, except for <b>replace</b>, which causes an error. Note that,
+<b>jitverify</b>, which is allowed, does not carry through to any subsequent
+matching that uses this pattern.
+</P>
 <br><a name="SEC11" href="#TOC1">SUBJECT MODIFIERS</a><br>
 <P>
 The modifiers that can appear in subject lines and the <b>#subject</b>
@@ -1292,14 +1332,75 @@ string, it behaves in the same way, unless a different locale has been set for
 the pattern (using the <b>/locale</b> modifier). In this case, the
 <b>isprint()</b> function is used to distinguish printing and non-printing
 characters.
+<a name="saverestore"></a></P>
+<br><a name="SEC18" href="#TOC1">SAVING AND RESTORING COMPILED PATTERNS</a><br>
+<P>
+It is possible to save compiled patterns on disc or elsewhere, and reload them
+later, subject to a number of restrictions. JIT data cannot be saved. The host
+on which the patterns are reloaded must be running the same version of PCRE2,
+with the same code unit width, and must also have the same endianness, pointer
+width and PCRE2_SIZE type. Before compiled patterns can be saved they must be
+serialized, that is, converted to a stream of bytes. A single byte stream may
+contain any number of compiled patterns, but they must all use the same
+character tables. A single copy of the tables is included in the byte stream
+(its size is 1088 bytes).
 </P>
-<br><a name="SEC18" href="#TOC1">SEE ALSO</a><br>
+<P>
+The functions whose names begin with <b>pcre2_serialize_</b> are used
+for serializing and de-serializing. They are described in the
+<a href="pcre2serialize.html"><b>pcre2serialize</b></a>
+documentation. In this section we describe the features of <b>pcre2test</b> that
+can be used to test these functions.
+</P>
+<P>
+When a pattern with <b>push</b> modifier is successfully compiled, it is pushed
+onto a stack of compiled patterns, and <b>pcre2test</b> expects the next line to
+contain a new pattern (or command) instead of a subject line. By this means, a
+number of patterns can be compiled and retained. The <b>push</b> modifier is
+incompatible with <b>posix</b>, and control modifiers that act at match time are
+ignored (with a message). The <b>jitverify</b> modifier applies only at compile
+time. The command
+<pre>
+  #save &#60;filename&#62;
+</pre>
+causes all the stacked patterns to be serialized and the result written to the
+named file. Afterwards, all the stacked patterns are freed. The command
+<pre>
+  #load &#60;filename&#62;
+</pre>
+reads the data in the file, and then arranges for it to be de-serialized, with
+the resulting compiled patterns added to the pattern stack. The pattern on the
+top of the stack can be retrieved by the #pop command, which must be followed
+by lines of subjects that are to be matched with the pattern, terminated as
+usual by an empty line or end of file. This command may be followed by a
+modifier list containing only
+<a href="#controlmodifiers">control modifiers</a>
+that act after a pattern has been compiled. In particular, <b>hex</b>,
+<b>posix</b>, and <b>push</b> are not allowed, nor are any
+<a href="#optionmodifiers">option-setting modifiers.</a>
+The JIT modifiers are, however permitted. Here is an example that saves and
+reloads two patterns.
+<pre>
+  /abc/push
+  /xyz/push
+  #save tempfile
+  #load tempfile
+  #pop info
+  xyz
+
+  #pop jit,bincode
+  abc
+</pre>
+If <b>jitverify</b> is used with #pop, it does not automatically imply
+<b>jit</b>, which is different behaviour from when it is used on a pattern.
+</P>
+<br><a name="SEC19" href="#TOC1">SEE ALSO</a><br>
 <P>
 <b>pcre2</b>(3), <b>pcre2api</b>(3), <b>pcre2callout</b>(3),
 <b>pcre2jit</b>, <b>pcre2matching</b>(3), <b>pcre2partial</b>(d),
-<b>pcre2pattern</b>(3).
+<b>pcre2pattern</b>(3), <b>pcre2serialize</b>(3).
 </P>
-<br><a name="SEC19" href="#TOC1">AUTHOR</a><br>
+<br><a name="SEC20" href="#TOC1">AUTHOR</a><br>
 <P>
 Philip Hazel
 <br>
@@ -1308,9 +1409,9 @@ University Computing Service
 Cambridge, England.
 <br>
 </P>
-<br><a name="SEC20" href="#TOC1">REVISION</a><br>
+<br><a name="SEC21" href="#TOC1">REVISION</a><br>
 <P>
-Last updated: 02 January 2015
+Last updated: 23 January 2015
 <br>
 Copyright &copy; 1997-2015 University of Cambridge.
 <br>

doc/index.html.src

@@ -65,6 +65,9 @@ first.
 <tr><td><a href="pcre2sample.html">pcre2sample</a></td>
     <td>&nbsp;&nbsp;Discussion of the pcre2demo program</td></tr>
 
+<tr><td><a href="pcre2serialize.html">pcre2serialize</a></td>
+    <td>&nbsp;&nbsp;Serializing functions for saving precompiled patterns</td></tr>
+
 <tr><td><a href="pcre2stack.html">pcre2stack</a></td>
     <td>&nbsp;&nbsp;Discussion of PCRE2's stack usage</td></tr>
 
@@ -177,6 +180,18 @@ in the library.
 <tr><td><a href="pcre2_pattern_info.html">pcre2_pattern_info</a></td>
     <td>&nbsp;&nbsp;Extract information about a pattern</td></tr>
 
+<tr><td><a href="pcre2_serialize_decode.html">pcre2_serialize_decode</a></td>
+    <td>&nbsp;&nbsp;Decode serialized compiled patterns</td></tr>
+
+<tr><td><a href="pcre2_serialize_encode.html">pcre2_serialize_encode</a></td>
+    <td>&nbsp;&nbsp;Serialize compiled patterns for save/restore</td></tr>
+
+<tr><td><a href="pcre2_serialize_free.html">pcre2_serialize_free</a></td>
+    <td>&nbsp;&nbsp;Free serialized compiled patterns</td></tr>
+
+<tr><td><a href="pcre2_serialize_get_number_of_codes.html">pcre2_serialize_get_number_of_codes</a></td>
+    <td>&nbsp;&nbsp;Get number of serialized compiled patterns</td></tr>
+
 <tr><td><a href="pcre2_set_bsr.html">pcre2_set_bsr</a></td>
     <td>&nbsp;&nbsp;Set \R convention</td></tr>
 

doc/pcre2_serialize_decode.3

@@ -0,0 +1,50 @@
+.TH PCRE2_SERIALIZE_DECODE 3 "19 January 2015" "PCRE2 10.10"
+.SH NAME
+PCRE2 - Perl-compatible regular expressions (revised API)
+.SH SYNOPSIS
+.rs
+.sp
+.B #include <pcre2.h>
+.PP
+.nf
+.B int32_t pcre2_serialize_decode(pcre2_code **\fIcodes\fP,
+.B "  int32_t \fInumber_of_codes\fP, const uint32_t *\fIbytes\fP,"
+.B "  pcre2_general_context *\fIgcontext\fP);"
+.fi
+.
+.SH DESCRIPTION
+.rs
+.sp
+This function decodes a serialized set of compiled patterns back into a list of
+individual patterns. Its arguments are:
+.sp
+  \fIcodes\fP            pointer to a vector in which to build the list
+  \fInumber_of_codes\fP  number of slots in the vector
+  \fIbytes\fP            the serialized byte stream
+  \fIgcontext\fP         pointer to a general context or NULL
+.sp
+The \fIbytes\fP argument must point to a block of data that was originally
+created by \fBpcre2_serialize_encode()\fP, though it may have been saved on
+disc or elsewhere in the meantime. If there are more codes in the serialized
+data than slots in the list, only those compiled patterns that will fit are
+decoded. The yield of the function is the number of decoded patterns, or one of
+the following negative error codes:
+.sp
+  PCRE2_ERROR_BADDATA   \fInumber_of_codes\fP is zero or less
+  PCRE2_ERROR_BADMAGIC  mismatch of id bytes in \fIbytes\fP
+  PCRE2_ERROR_BADMODE   mismatch of variable unit size or PCRE version
+  PCRE2_ERROR_MEMORY    memory allocation failed
+  PCRE2_ERROR_NULL      \fIcodes\fP or \fIbytes\fP is NULL
+.sp
+PCRE2_ERROR_BADMAGIC may mean that the data is corrupt, or that it was compiled
+on a system with different endianness.
+.P
+There is a complete description of the PCRE2 native API in the
+.\" HREF
+\fBpcre2api\fP
+.\"
+page and a description of the POSIX API in the
+.\" HREF
+\fBpcre2posix\fP
+.\"
+page.

doc/pcre2_serialize_encode.3

@@ -0,0 +1,49 @@
+.TH PCRE2_SERIALIZE_ENCODE 3 "19 January 2015" "PCRE2 10.10"
+.SH NAME
+PCRE2 - Perl-compatible regular expressions (revised API)
+.SH SYNOPSIS
+.rs
+.sp
+.B #include <pcre2.h>
+.PP
+.nf
+.B int32_t pcre2_serialize_encode(pcre2_code **\fIcodes\fP,
+.B "  int32_t \fInumber_of_codes\fP, uint32_t **\fIserialized_bytes\fP,"
+.B "  PCRE2_SIZE *\fIserialized_size\fP, pcre2_general_context *\fIgcontext\fP);"
+.fi
+.
+.SH DESCRIPTION
+.rs
+.sp
+This function encodes a list of compiled patterns into a byte stream that can
+be saved on disc or elsewhere. Its arguments are:
+.sp
+  \fIcodes\fP             pointer to a vector containing the list
+  \fInumber_of_codes\fP   number of slots in the vector
+  \fIserialized_bytes\fP  set to point to the serialized byte stream
+  \fIserialized_size\fP   set to the number of bytes in the byte stream
+  \fIgcontext\fP          pointer to a general context or NULL
+.sp
+The context argument is used to obtain memory for the byte stream. When the
+serialized data is no longer needed, it must be freed by calling
+\fBpcre2_serialize_free()\fP. The yield of the function is the number of
+serialized patterns, or one of the following negative error codes:
+.sp
+  PCRE2_ERROR_BADDATA      \fInumber_of_codes\fP is zero or less
+  PCRE2_ERROR_BADMAGIC     mismatch of id bytes in one of the patterns
+  PCRE2_ERROR_MEMORY       memory allocation failed
+  PCRE2_ERROR_MIXEDTABLES  the patterns do not all use the same tables
+  PCRE2_ERROR_NULL         an argument other than \fIgcontext\fP is NULL
+.sp
+PCRE2_ERROR_BADMAGIC means either that a pattern's code has been corrupted, or
+that a slot in the vector does not point to a compiled pattern.
+.P
+There is a complete description of the PCRE2 native API in the
+.\" HREF
+\fBpcre2api\fP
+.\"
+page and a description of the POSIX API in the
+.\" HREF
+\fBpcre2posix\fP
+.\"
+page.

doc/pcre2_serialize_free.3

@@ -0,0 +1,28 @@
+.TH PCRE2_SERIALIZE_FREE 3 "19 January 2015" "PCRE2 10.10"
+.SH NAME
+PCRE2 - Perl-compatible regular expressions (revised API)
+.SH SYNOPSIS
+.rs
+.sp
+.B #include <pcre2.h>
+.PP
+.nf
+.B void pcre2_serialize_free(uint8_t *\fIbytes\fP);
+.fi
+.
+.SH DESCRIPTION
+.rs
+.sp
+This function frees the memory that was obtained by
+\fBpcre2_serialize_encode()\fP to hold a serialized byte stream. The argument
+must point to such a byte stream.
+.P
+There is a complete description of the PCRE2 native API in the
+.\" HREF
+\fBpcre2api\fP
+.\"
+page and a description of the POSIX API in the
+.\" HREF
+\fBpcre2posix\fP
+.\"
+page.

doc/pcre2_serialize_get_number_of_codes.3

@@ -0,0 +1,37 @@
+.TH PCRE2_SERIALIZE_GET_NUMBER_OF_CODES 3 "19 January 2015" "PCRE2 10.10"
+.SH NAME
+PCRE2 - Perl-compatible regular expressions (revised API)
+.SH SYNOPSIS
+.rs
+.sp
+.B #include <pcre2.h>
+.PP
+.nf
+.B int32_t pcre2_serialize_get_number_of_codes(const uint8_t *\fIbytes\fP);
+.fi
+.
+.SH DESCRIPTION
+.rs
+.sp
+The \fIbytes\fP argument must point to a serialized byte stream that was
+originally created by \fBpcre2_serialize_encode()\fP (though it may have been
+saved on disc or elsewhere in the meantime). The function returns the number of
+serialized patterns in the byte stream, or one of the following negative error
+codes:
+.sp
+  PCRE2_ERROR_BADMAGIC  mismatch of id bytes in \fIbytes\fP
+  PCRE2_ERROR_BADMODE   mismatch of variable unit size or PCRE version
+  PCRE2_ERROR_NULL      the argument is NULL
+.sp
+PCRE2_ERROR_BADMAGIC may mean that the data is corrupt, or that it was compiled
+on a system with different endianness.
+.P
+There is a complete description of the PCRE2 native API in the
+.\" HREF
+\fBpcre2api\fP
+.\"
+page and a description of the POSIX API in the
+.\" HREF
+\fBpcre2posix\fP
+.\"
+page.

doc/pcre2api.3

@@ -1,4 +1,4 @@
-.TH PCRE2API 3 "13 January 2015" "PCRE2 10.10"
+.TH PCRE2API 3 "23 January 2015" "PCRE2 10.10"
 .SH NAME
 PCRE2 - Perl-compatible regular expressions (revised API)
 .sp
@@ -205,6 +205,24 @@ document for an overview of all the PCRE2 documentation.
 .fi
 .
 .
+.SH "PCRE2 NATIVE API SERIALIZATION FUNCTIONS"
+.rs
+.sp
+.nf
+.B int32_t pcre2_serialize_decode(pcre2_code **\fIcodes\fP,
+.B "  int32_t \fInumber_of_codes\fP, const uint32_t *\fIbytes\fP,"
+.B "  pcre2_general_context *\fIgcontext\fP);"
+.sp
+.B int32_t pcre2_serialize_encode(pcre2_code **\fIcodes\fP,
+.B "  int32_t \fInumber_of_codes\fP, uint32_t **\fIserialized_bytes\fP,"
+.B "  PCRE2_SIZE *\fIserialized_size\fP, pcre2_general_context *\fIgcontext\fP);"
+.sp
+.B void pcre2_serialize_free(uint8_t *\fIbytes\fP);
+.sp
+.B int32_t pcre2_serialize_get_number_of_codes(const uint8_t *\fIbytes\fP);
+.fi
+.
+.
 .SH "PCRE2 NATIVE API AUXILIARY FUNCTIONS"
 .rs
 .sp
@@ -1689,12 +1707,26 @@ set, the call to \fBpcre2_pattern_info()\fP returns the error PCRE2_ERROR_UNSET.
   PCRE2_INFO_SIZE
 .sp
 Return the size of the compiled pattern in bytes (for all three libraries). The
-third argument should point to a \fBsize_t\fP variable. This value does not
+third argument should point to a \fBsize_t\fP variable. This value includes the
-include the size of the \fBpcre2_code\fP structure that is returned by
+size of the general data block that precedes the code units of the compiled
-\fBpcre_compile()\fP. The value that is used when \fBpcre2_compile()\fP is
+pattern itself. The value that is used when \fBpcre2_compile()\fP is getting
-getting memory in which to place the compiled data is the value returned by
+memory in which to place the compiled pattern may be slightly larger than the
-this option plus the size of the \fBpcre2_code\fP structure. Processing a
+value returned by this option, because there are cases where the code that
-pattern with the JIT compiler does not alter the value returned by this option.
+calculates the size has to over-estimate. Processing a pattern with the JIT
+compiler does not alter the value returned by this option.
+.
+.
+.SH "SERIALIZATION AND PRECOMPILING"
+.rs
+.sp
+It is possible to save compiled patterns on disc or elsewhere, and reload them
+later, subject to a number of restrictions. The functions whose names begin
+with \fBpcre2_serialize_\fP are used for this purpose. They are described in
+the
+.\" HREF
+\fBpcre2serialize\fP
+.\"
+documentation.
 .
 .
 .\" HTML <a name="matchdatablock"></a>
@@ -2853,6 +2885,6 @@ Cambridge, England.
 .rs
 .sp
 .nf
-Last updated: 13 January 2015
+Last updated: 23 January 2015
 Copyright (c) 1997-2015 University of Cambridge.
 .fi

doc/pcre2serialize.3

@@ -0,0 +1,170 @@
+.TH PCRE2SERIALIZE 3 "20 January 2015" "PCRE2 10.10"
+.SH NAME
+PCRE2 - Perl-compatible regular expressions (revised API)
+.SH "SAVING AND RE-USING PRECOMPILED PCRE2 PATTERNS"
+.rs
+.sp
+.nf
+.B int32_t pcre2_serialize_decode(pcre2_code **\fIcodes\fP,
+.B "  int32_t \fInumber_of_codes\fP, const uint32_t *\fIbytes\fP,"
+.B "  pcre2_general_context *\fIgcontext\fP);"
+.sp
+.B int32_t pcre2_serialize_encode(pcre2_code **\fIcodes\fP,
+.B "  int32_t \fInumber_of_codes\fP, uint32_t **\fIserialized_bytes\fP,"
+.B "  PCRE2_SIZE *\fIserialized_size\fP, pcre2_general_context *\fIgcontext\fP);"
+.sp
+.B void pcre2_serialize_free(uint8_t *\fIbytes\fP);
+.sp
+.B int32_t pcre2_serialize_get_number_of_codes(const uint8_t *\fIbytes\fP);
+.fi
+.sp
+If you are running an application that uses a large number of regular
+expression patterns, it may be useful to store them in a precompiled form
+instead of having to compile them every time the application is run. However,
+if you are using the just-in-time optimization feature, it is not possible to
+save and reload the JIT data, because it is position-dependent. In addition,
+the host on which the patterns are reloaded must be running the same version of
+PCRE2, with the same code unit width, and must also have the same endianness,
+pointer width and PCRE2_SIZE type. For example, patterns compiled on a 32-bit
+system using PCRE2's 16-bit library cannot be reloaded on a 64-bit system, nor
+can they be reloaded using the 8-bit library.
+.
+.
+.SH "SAVING COMPILED PATTERNS"
+.rs
+.sp
+Before compiled patterns can be saved they must be serialized, that is,
+converted to a stream of bytes. A single byte stream may contain any number of
+compiled patterns, but they must all use the same character tables. A single
+copy of the tables is included in the byte stream (its size is 1088 bytes). For
+more details of character tables, see the
+.\" HTML <a href="pcre2api.html#localesupport">
+.\" </a>
+section on locale support
+.\"
+in the
+.\" HREF
+\fBpcre2api\fP
+.\"
+documentation.
+.P
+The function \fBpcre2_serialize_encode()\fP creates a serialized byte stream
+from a list of compiled patterns. Its first two arguments specify the list,
+being a pointer to a vector of pointers to compiled patterns, and the length of
+the vector. The third and fourth arguments point to variables which are set to
+point to the created byte stream and its length, respectively. The final
+argument is a pointer to a general context, which can be used to specify custom
+memory mangagement functions. If this argument is NULL, \fBmalloc()\fP is used
+to obtain memory for the byte stream. The yield of the function is the number
+of serialized patterns, or one of the following negative error codes:
+.sp
+  PCRE2_ERROR_BADDATA      the number of patterns is zero or less
+  PCRE2_ERROR_BADMAGIC     mismatch of id bytes in one of the patterns
+  PCRE2_ERROR_MEMORY       memory allocation failed
+  PCRE2_ERROR_MIXEDTABLES  the patterns do not all use the same tables
+  PCRE2_ERROR_NULL         the 1st, 3rd, or 4th argument is NULL
+.sp
+PCRE2_ERROR_BADMAGIC means either that a pattern's code has been corrupted, or
+that a slot in the vector does not point to a compiled pattern.
+.P
+Once a set of patterns has been serialized you can save the data in any
+appropriate manner. Here is sample code that compiles two patterns and writes
+them to a file. It assumes that the variable \fIfd\fP refers to a file that is
+open for output. The error checking that should be present in a real
+application has been omitted for simplicity.
+.sp
+  int errorcode;
+  uint8_t *bytes;
+  PCRE2_SIZE erroroffset;
+  PCRE2_SIZE bytescount;
+  pcre2_code *list_of_codes[2];
+  list_of_codes[0] = pcre2_compile("first pattern",
+    PCRE2_ZERO_TERMINATED, 0, &errorcode, &erroroffset, NULL);
+  list_of_codes[1] = pcre2_compile("second pattern",
+    PCRE2_ZERO_TERMINATED, 0, &errorcode, &erroroffset, NULL);
+  errorcode = pcre2_serialize_encode(list_of_codes, 2, &bytes,
+    &bytescount, NULL);
+  errorcode = fwrite(bytes, 1, bytescount, fd);
+.sp
+Note that the serialized data is binary data that may contain any of the 256
+possible byte values. On systems that make a distinction between binary and
+non-binary data, be sure that the file is opened for binary output.
+.P
+Serializing a set of patterns leaves the original data untouched, so they can
+still be used for matching. Their memory must eventually be freed in the usual
+way by calling \fBpcre2_code_free()\fP. When you have finished with the byte
+stream, it too must be freed by calling \fBpcre2_serialize_free()\fP.
+.
+.
+.SH "RE-USING PRECOMPILED PATTERNS"
+.rs
+.sp
+In order to re-use a set of saved patterns you must first make the serialized
+byte stream available in main memory (for example, by reading from a file). The
+management of this memory block is up to the application. You can use the
+\fBpcre2_serialize_get_number_of_codes()\fP function to find out how many
+compiled patterns are in the serialized data without actually decoding the
+patterns:
+.sp
+  uint8_t *bytes = <serialized data>;
+  int32_t number_of_codes = pcre2_serialize_get_number_of_codes(bytes);
+.sp
+The \fBpcre2_serialize_decode()\fP function reads a byte stream and recreates
+the compiled patterns in new memory blocks, setting pointers to them in a
+vector. The first two arguments are a pointer to a suitable vector and its
+length, and the third argument points to a byte stream. The final argument is a
+pointer to a general context, which can be used to specify custom memory
+mangagement functions for the decoded patterns. If this argument is NULL,
+\fBmalloc()\fP and \fBfree()\fP are used. After deserialization, the byte
+stream is no longer needed and can be discarded.
+.sp
+  int32_t number_of_codes;
+  pcre2_code *list_of_codes[2];
+  uint8_t *bytes = <serialized data>;
+  int32_t number_of_codes =
+    pcre2_serialize_decode(list_of_codes, 2, bytes, NULL);
+.sp
+If the vector is not large enough for all the patterns in the byte stream, it
+is filled with those that fit, and the remainder are ignored. The yield of the
+function is the number of decoded patterns, or one of the following negative
+error codes:
+.sp
+  PCRE2_ERROR_BADDATA   second argument is zero or less
+  PCRE2_ERROR_BADMAGIC  mismatch of id bytes in the data
+  PCRE2_ERROR_BADMODE   mismatch of variable unit size or PCRE2 version
+  PCRE2_ERROR_MEMORY    memory allocation failed
+  PCRE2_ERROR_NULL      first or third argument is NULL
+.sp
+PCRE2_ERROR_BADMAGIC may mean that the data is corrupt, or that it was compiled
+on a system with different endianness.
+.P
+Decoded patterns can be used for matching in the usual way, and must be freed
+by calling \fBpcre2_code_free()\fP as normal. A single copy of the character
+tables is used by all the decoded patterns. A reference count is used to
+arrange for its memory to be automatically freed when the last pattern is
+freed.
+.P
+If a pattern was processed by \fBpcre2_jit_compile()\fP before being
+serialized, the JIT data is discarded and so is no longer available after a
+save/restore cycle. You can, however, process a restored pattern with
+\fBpcre2_jit_compile()\fP if you wish.
+.
+.
+.
+.SH AUTHOR
+.rs
+.sp
+.nf
+Philip Hazel
+University Computing Service
+Cambridge, England.
+.fi
+.
+.
+.SH REVISION
+.rs
+.sp
+.nf
+Last updated: 20 January 2015
+Copyright (c) 1997-2015 University of Cambridge.
+.fi

doc/pcre2test.1

@@ -1,4 +1,4 @@
-.TH PCRE2TEST 1 "02 January 2015" "PCRE 10.00"
+.TH PCRE2TEST 1 "23 January 2015" "PCRE 10.10"
 .SH NAME
 pcre2test - a program for testing Perl-compatible regular expressions.
 .SH SYNOPSIS
@@ -21,10 +21,11 @@ options, see the
 documentation.
 .P
 The input for \fBpcre2test\fP is a sequence of regular expression patterns and
-subject strings to be matched. The output shows the result of each match
+subject strings to be matched. There are also command lines for setting
-attempt. Modifiers on the command line, the patterns, and the subject lines
+defaults and controlling some special actions. The output shows the result of
-specify PCRE2 function options, control how the subject is processed, and what
+each match attempt. Modifiers on external or internal command lines, the
-output is produced.
+patterns, and the subject lines specify PCRE2 function options, control how the
+subject is processed, and what output is produced.
 .P
 As the original fairly simple PCRE library evolved, it acquired many different
 features, and as a result, the original \fBpcretest\fP program ended up with a
@@ -185,9 +186,7 @@ If \fBpcre2test\fP is given two filename arguments, it reads from the first and
 writes to the second. If the first name is "-", input is taken from the
 standard input. If \fBpcre2test\fP is given only one argument, it reads from
 that file and writes to stdout. Otherwise, it reads from stdin and writes to
-stdout. When the input is a terminal, it prompts for each line of input, using
+stdout.
-"re>" to prompt for regular expression patterns, and "data>" to prompt for
-subject lines.
 .P
 When \fBpcre2test\fP is built, a configuration option can specify that it
 should be linked with the \fBlibreadline\fP or \fBlibedit\fP library. When this
@@ -198,10 +197,15 @@ the \fB-help\fP option states whether or not \fBreadline()\fP will be used.
 The program handles any number of tests, each of which consists of a set of
 input lines. Each set starts with a regular expression pattern, followed by any
 number of subject lines to be matched against that pattern. In between sets of
-test data, command lines that begin with a hash (#) character may appear. This
+test data, command lines that begin with # may appear. This file format, with
-file format, with some restrictions, can also be processed by the
+some restrictions, can also be processed by the \fBperltest.sh\fP script that
-\fBperltest.sh\fP script that is distributed with PCRE2 as a means of checking
+is distributed with PCRE2 as a means of checking that the behaviour of PCRE2
-that the behaviour of PCRE2 and Perl is the same.
+and Perl is the same.
+.P
+When the input is a terminal, \fBpcre2test\fP prompts for each line of input,
+using "re>" to prompt for regular expression patterns, and "data>" to prompt
+for subject lines. Command lines starting with # can be entered only in
+response to the "re>" prompt.
 .P
 Each subject line is matched separately and independently. If you want to do
 multi-line matches, you have to use the \en escape sequence (or \er or \er\en,
@@ -219,21 +223,30 @@ still input to be read.
 .SH "COMMAND LINES"
 .rs
 .sp
-In between sets of test data, a line that begins with a hash (#) character is
+In between sets of test data, a line that begins with # is interpreted as a
-interpreted as a command line. If the first character is followed by white
+command line. If the first character is followed by white space or an
-space or an exclamation mark, the line is treated as a comment, and ignored.
+exclamation mark, the line is treated as a comment, and ignored. Otherwise, the
-Otherwise, the following commands are recognized:
+following commands are recognized:
 .sp
   #forbid_utf
 .sp
 Subsequent patterns automatically have the PCRE2_NEVER_UTF and PCRE2_NEVER_UCP
 options set, which locks out the use of UTF and Unicode property features. This
-is a trigger guard that is used in test files to ensure that UTF/Unicode tests
+is a trigger guard that is used in test files to ensure that UTF or Unicode
-are not accidentally added to files that are used when UTF support is not
+property tests are not accidentally added to files that are used when Unicode
-included in the library. This effect can also be obtained by the use of
+support is not included in the library. This effect can also be obtained by the
-\fB#pattern\fP; the difference is that \fB#forbid_utf\fP cannot be unset, and
+use of \fB#pattern\fP; the difference is that \fB#forbid_utf\fP cannot be
-the automatic options are not displayed in pattern information, to avoid
+unset, and the automatic options are not displayed in pattern information, to
-cluttering up test output.
+avoid cluttering up test output.
+.sp
+  #load <filename>
+.sp
+This command is used to load a set of precompiled patterns from a file, as
+described in the section entitled "Saving and restoring compiled patterns"
+.\" HTML <a href="#saverestore">
+.\" </a>
+below.
+.\"
 .sp
   #pattern <modifier-list>
 .sp
@@ -249,6 +262,24 @@ lines, none of the other command lines are permitted, because they and many
 of the modifiers are specific to \fBpcre2test\fP, and should not be used in
 test files that are also processed by \fBperltest.sh\fP. The \fB#perltest\fP
 command helps detect tests that are accidentally put in the wrong file.
+.sp
+  #pop [<modifiers>]
+.sp
+This command is used to manipulate the stack of compiled patterns, as described
+in the section entitled "Saving and restoring compiled patterns"
+.\" HTML <a href="#saverestore">
+.\" </a>
+below.
+.\"
+.sp
+  #save <filename>
+.sp
+This command is used to save a set of compiled patterns to a file, as described
+in the section entitled "Saving and restoring compiled patterns"
+.\" HTML <a href="#saverestore">
+.\" </a>
+below.
+.\"
 .sp
   #subject <modifier-list>
 .sp
@@ -387,6 +418,7 @@ can add to or override default modifiers that were set by a previous
 \fB#pattern\fP command.
 .
 .
+.\" HTML <a name="optionmodifiers"></a>
 .SS "Setting compilation options"
 .rs
 .sp
@@ -426,6 +458,7 @@ notation. Otherwise, those less than 0x100 are output in hex without the curly
 brackets.
 .
 .
+.\" HTML <a name="controlmodifiers"></a>
 .SS "Setting compilation controls"
 .rs
 .sp
@@ -445,8 +478,8 @@ about the pattern:
       memory                    show memory used
       newline=<type>            set newline type
       parens_nest_limit=<n>     set maximum parentheses depth
-      perlcompat                lock out non-Perl modifiers
       posix                     use the POSIX API
+      push                      push compiled pattern onto the stack
       stackguard=<number>       test the stackguard feature
       tables=[0|1|2]            select internal tables
 .sp
@@ -683,6 +716,25 @@ These modifiers may not appear in a \fB#pattern\fP command. If you want them as
 defaults, set them in a \fB#subject\fP command.
 .
 .
+.SS "Saving a compiled pattern"
+.rs
+.sp
+When a pattern with the \fBpush\fP modifier is successfully compiled, it is
+pushed onto a stack of compiled patterns, and \fBpcre2test\fP expects the next
+line to contain a new pattern (or a command) instead of a subject line. This
+facility is used when saving compiled patterns to a file, as described in the
+section entitled "Saving and restoring compiled patterns"
+.\" HTML <a href="#saverestore">
+.\" </a>
+below.
+.\"
+The \fBpush\fP modifier is incompatible with compilation modifiers such as
+\fBglobal\fP that act at match time. Any that are specified are ignored, with a
+warning message, except for \fBreplace\fP, which causes an error. Note that,
+\fBjitverify\fP, which is allowed, does not carry through to any subsequent
+matching that uses this pattern.
+.
+.
 .SH "SUBJECT MODIFIERS"
 .rs
 .sp
@@ -1253,12 +1305,83 @@ characters.
 .
 .
 .
+.\" HTML <a name="saverestore"></a>
+.SH "SAVING AND RESTORING COMPILED PATTERNS"
+.rs
+.sp
+It is possible to save compiled patterns on disc or elsewhere, and reload them
+later, subject to a number of restrictions. JIT data cannot be saved. The host
+on which the patterns are reloaded must be running the same version of PCRE2,
+with the same code unit width, and must also have the same endianness, pointer
+width and PCRE2_SIZE type. Before compiled patterns can be saved they must be
+serialized, that is, converted to a stream of bytes. A single byte stream may
+contain any number of compiled patterns, but they must all use the same
+character tables. A single copy of the tables is included in the byte stream
+(its size is 1088 bytes).
+.P
+The functions whose names begin with \fBpcre2_serialize_\fP are used
+for serializing and de-serializing. They are described in the
+.\" HREF
+\fBpcre2serialize\fP
+.\"
+documentation. In this section we describe the features of \fBpcre2test\fP that
+can be used to test these functions.
+.P
+When a pattern with \fBpush\fP modifier is successfully compiled, it is pushed
+onto a stack of compiled patterns, and \fBpcre2test\fP expects the next line to
+contain a new pattern (or command) instead of a subject line. By this means, a
+number of patterns can be compiled and retained. The \fBpush\fP modifier is
+incompatible with \fBposix\fP, and control modifiers that act at match time are
+ignored (with a message). The \fBjitverify\fP modifier applies only at compile
+time. The command
+.sp
+  #save <filename>
+.sp
+causes all the stacked patterns to be serialized and the result written to the
+named file. Afterwards, all the stacked patterns are freed. The command
+.sp
+  #load <filename>
+.sp
+reads the data in the file, and then arranges for it to be de-serialized, with
+the resulting compiled patterns added to the pattern stack. The pattern on the
+top of the stack can be retrieved by the #pop command, which must be followed
+by lines of subjects that are to be matched with the pattern, terminated as
+usual by an empty line or end of file. This command may be followed by a
+modifier list containing only
+.\" HTML <a href="#controlmodifiers">
+.\" </a>
+control modifiers
+.\"
+that act after a pattern has been compiled. In particular, \fBhex\fP,
+\fBposix\fP, and \fBpush\fP are not allowed, nor are any
+.\" HTML <a href="#optionmodifiers">
+.\" </a>
+option-setting modifiers.
+.\"
+The JIT modifiers are, however permitted. Here is an example that saves and
+reloads two patterns.
+.sp
+  /abc/push
+  /xyz/push
+  #save tempfile
+  #load tempfile
+  #pop info
+  xyz
+.sp
+  #pop jit,bincode
+  abc
+.sp
+If \fBjitverify\fP is used with #pop, it does not automatically imply
+\fBjit\fP, which is different behaviour from when it is used on a pattern.
+.
+.
+.
 .SH "SEE ALSO"
 .rs
 .sp
 \fBpcre2\fP(3), \fBpcre2api\fP(3), \fBpcre2callout\fP(3),
 \fBpcre2jit\fP, \fBpcre2matching\fP(3), \fBpcre2partial\fP(d),
-\fBpcre2pattern\fP(3).
+\fBpcre2pattern\fP(3), \fBpcre2serialize\fP(3).
 .
 .
 .SH AUTHOR
@@ -1275,6 +1398,6 @@ Cambridge, England.
 .rs
 .sp
 .nf
-Last updated: 02 January 2015
+Last updated: 23 January 2015
 Copyright (c) 1997-2015 University of Cambridge.
 .fi

doc/pcre2test.txt

@@ -17,10 +17,12 @@ SYNOPSIS
        options, see the pcre2api documentation.
 
        The input for pcre2test is a sequence of  regular  expression  patterns
-       and  subject strings to be matched. The output shows the result of each
+       and  subject  strings  to  be matched. There are also command lines for
-       match attempt. Modifiers on the command line,  the  patterns,  and  the
+       setting defaults and controlling some special actions. The output shows
-       subject  lines  specify PCRE2 function options, control how the subject
+       the  result  of  each  match attempt. Modifiers on external or internal
-       is processed, and what output is produced.
+       command lines, the patterns, and the subject lines specify PCRE2  func-
+       tion  options, control how the subject is processed, and what output is
+       produced.
 
        As the original fairly simple PCRE library evolved,  it  acquired  many
        different  features,  and  as  a  result, the original pcretest program
@@ -173,9 +175,7 @@ DESCRIPTION
        and writes to the second. If the first name is "-", input is taken from
        the  standard  input. If pcre2test is given only one argument, it reads
        from that file and writes to stdout. Otherwise, it reads from stdin and
-       writes  to  stdout.  When  the input is a terminal, it prompts for each
+       writes to stdout.
-       line of input, using "re>" to prompt for regular  expression  patterns,
-       and "data>" to prompt for subject lines.
 
        When  pcre2test  is  built,  a configuration option can specify that it
        should be linked with the libreadline or libedit library. When this  is
@@ -186,11 +186,15 @@ DESCRIPTION
        The  program  handles  any number of tests, each of which consists of a
        set of input lines. Each set starts with a regular expression  pattern,
        followed by any number of subject lines to be matched against that pat-
-       tern. In between sets of test data, command lines  that  begin  with  a
+       tern. In between sets of test data, command lines that begin with # may
-       hash  (#)  character  may  appear. This file format, with some restric-
+       appear. This file format, with some restrictions, can also be processed
-       tions, can also be processed by the perltest.sh script that is distrib-
+       by the perltest.sh script that is distributed with PCRE2 as a means  of
-       uted  with PCRE2 as a means of checking that the behaviour of PCRE2 and
+       checking that the behaviour of PCRE2 and Perl is the same.
-       Perl is the same.
+
+       When the input is a terminal, pcre2test prompts for each line of input,
+       using "re>" to prompt for regular expression patterns, and  "data>"  to
+       prompt  for subject lines. Command lines starting with # can be entered
+       only in response to the "re>" prompt.
 
        Each subject line is matched separately and independently. If you  want
        to do multi-line matches, you have to use the \n escape sequence (or \r
@@ -207,22 +211,28 @@ DESCRIPTION
 
 COMMAND LINES
 
-       In between sets of test data, a line that begins with a hash (#)  char-
+       In between sets of test data, a line that begins with # is  interpreted
-       acter  is interpreted as a command line. If the first character is fol-
+       as a command line. If the first character is followed by white space or
-       lowed by white space or an exclamation mark, the line is treated  as  a
+       an exclamation mark, the line is treated as  a  comment,  and  ignored.
-       comment,  and  ignored.   Otherwise,  the following commands are recog-
+       Otherwise, the following commands are recognized:
-       nized:
 
          #forbid_utf
 
-       Subsequent  patterns  automatically  have   the   PCRE2_NEVER_UTF   and
+       Subsequent   patterns   automatically   have  the  PCRE2_NEVER_UTF  and
        PCRE2_NEVER_UCP options set, which locks out the use of UTF and Unicode
-       property features. This is a trigger guard that is used in  test  files
+       property  features.  This is a trigger guard that is used in test files
-       to  ensure  that  UTF/Unicode tests are not accidentally added to files
+       to ensure that UTF or Unicode property tests are not accidentally added
-       that are used when UTF support is not included  in  the  library.  This
+       to  files  that  are  used  when Unicode support is not included in the
-       effect  can  also be obtained by the use of #pattern; the difference is
+       library. This effect can also be obtained by the use of  #pattern;  the
-       that #forbid_utf cannot be unset, and the  automatic  options  are  not
+       difference  is  that  #forbid_utf  cannot  be  unset, and the automatic
-       displayed in pattern information, to avoid cluttering up test output.
+       options are not displayed in pattern information, to  avoid  cluttering
+       up test output.
+
+         #load <filename>
+
+       This command is used to load a set of precompiled patterns from a file,
+       as described in the section entitled  "Saving  and  restoring  compiled
+       patterns" below.
 
          #pattern <modifier-list>
 
@@ -240,6 +250,18 @@ COMMAND LINES
        #perltest command helps detect tests that are accidentally put  in  the
        wrong file.
 
+         #pop [<modifiers>]
+
+       This  command  is used to manipulate the stack of compiled patterns, as
+       described in the section entitled "Saving and restoring  compiled  pat-
+       terns" below.
+
+         #save <filename>
+
+       This  command  is used to save a set of compiled patterns to a file, as
+       described in the section entitled "Saving and restoring  compiled  pat-
+       terns" below.
+
          #subject <modifier-list>
 
        This  command  sets  a default modifier list that applies to all subse-
@@ -432,8 +454,8 @@ PATTERN MODIFIERS
              memory                    show memory used
              newline=<type>            set newline type
              parens_nest_limit=<n>     set maximum parentheses depth
-             perlcompat                lock out non-Perl modifiers
              posix                     use the POSIX API
+             push                      push compiled pattern onto the stack
              stackguard=<number>       test the stackguard feature
              tables=[0|1|2]            select internal tables
 
@@ -644,6 +666,19 @@ PATTERN MODIFIERS
        These  modifiers may not appear in a #pattern command. If you want them
        as defaults, set them in a #subject command.
 
+   Saving a compiled pattern
+
+       When a pattern with the push modifier is successfully compiled,  it  is
+       pushed  onto  a  stack  of compiled patterns, and pcre2test expects the
+       next line to contain a new pattern (or a command) instead of a  subject
+       line. This facility is used when saving compiled patterns to a file, as
+       described in the section entitled "Saving and restoring  compiled  pat-
+       terns" below.  The push modifier is incompatible with compilation modi-
+       fiers such as global that act at match time. Any that are specified are
+       ignored,  with  a  warning message, except for replace, which causes an
+       error. Note that, jitverify, which is allowed, does not  carry  through
+       to any subsequent matching that uses this pattern.
+
 
 SUBJECT MODIFIERS
 
@@ -652,7 +687,7 @@ SUBJECT MODIFIERS
 
    Setting match options
 
-       The    following   modifiers   set   options   for   pcre2_match()   or
+       The   following   modifiers   set   options   for   pcre2_match()    or
        pcre2_dfa_match(). See pcreapi for a description of their effects.
 
              anchored                  set PCRE2_ANCHORED
@@ -666,20 +701,20 @@ SUBJECT MODIFIERS
              partial_hard (or ph)      set PCRE2_PARTIAL_HARD
              partial_soft (or ps)      set PCRE2_PARTIAL_SOFT
 
-       The partial matching modifiers are provided with abbreviations  because
+       The  partial matching modifiers are provided with abbreviations because
        they appear frequently in tests.
 
-       If  the  /posix  modifier was present on the pattern, causing the POSIX
+       If the /posix modifier was present on the pattern,  causing  the  POSIX
        wrapper API to be used, the only option-setting modifiers that have any
-       effect   are   notbol,   notempty,   and  noteol,  causing  REG_NOTBOL,
+       effect  are  notbol,  notempty,   and   noteol,   causing   REG_NOTBOL,
-       REG_NOTEMPTY, and REG_NOTEOL, respectively, to be passed to  regexec().
+       REG_NOTEMPTY,  and REG_NOTEOL, respectively, to be passed to regexec().
        Any other modifiers cause an error.
 
    Setting match controls
 
-       The  following  modifiers  affect the matching process or request addi-
+       The following modifiers affect the matching process  or  request  addi-
-       tional information. Some of them may also be  specified  on  a  pattern
+       tional  information.  Some  of  them may also be specified on a pattern
-       line  (see  above), in which case they apply to every subject line that
+       line (see above), in which case they apply to every subject  line  that
        is matched against that pattern.
 
              aftertext                 show text after match
@@ -712,23 +747,23 @@ SUBJECT MODIFIERS
 
    Showing more text
 
-       The aftertext modifier requests that as well as outputting the part  of
+       The  aftertext modifier requests that as well as outputting the part of
        the subject string that matched the entire pattern, pcre2test should in
        addition output the remainder of the subject string. This is useful for
        tests where the subject contains multiple copies of the same substring.
-       The allaftertext modifier requests the same action  for  captured  sub-
+       The  allaftertext  modifier  requests the same action for captured sub-
        strings as well as the main matched substring. In each case the remain-
        der is output on the following line with a plus character following the
        capture number.
 
-       The  allusedtext modifier requests that all the text that was consulted
+       The allusedtext modifier requests that all the text that was  consulted
-       during a successful pattern match by the interpreter should  be  shown.
+       during  a  successful pattern match by the interpreter should be shown.
-       This  feature  is not supported for JIT matching, and if requested with
+       This feature is not supported for JIT matching, and if  requested  with
-       JIT it is ignored (with  a  warning  message).  Setting  this  modifier
+       JIT  it  is  ignored  (with  a  warning message). Setting this modifier
        affects the output if there is a lookbehind at the start of a match, or
-       a lookahead at the end, or if \K is used  in  the  pattern.  Characters
+       a  lookahead  at  the  end, or if \K is used in the pattern. Characters
-       that  precede or follow the start and end of the actual match are indi-
+       that precede or follow the start and end of the actual match are  indi-
-       cated in the output by '<' or '>' characters underneath them.  Here  is
+       cated  in  the output by '<' or '>' characters underneath them. Here is
        an example:
 
            re> /(?<=pqr)abc(?=xyz)/
@@ -736,16 +771,16 @@ SUBJECT MODIFIERS
           0: pqrabcxyz
              <<<   >>>
 
-       This  shows  that  the  matched string is "abc", with the preceding and
+       This shows that the matched string is "abc",  with  the  preceding  and
-       following strings "pqr" and "xyz"  having  been  consulted  during  the
+       following  strings  "pqr"  and  "xyz"  having been consulted during the
        match (when processing the assertions).
 
-       The  startchar  modifier  requests  that the starting character for the
+       The startchar modifier requests that the  starting  character  for  the
-       match be indicated, if it is different to  the  start  of  the  matched
+       match  be  indicated,  if  it  is different to the start of the matched
        string. The only time when this occurs is when \K has been processed as
        part of the match. In this situation, the output for the matched string
-       is  displayed  from  the  starting  character instead of from the match
+       is displayed from the starting character  instead  of  from  the  match
-       point, with circumflex characters under  the  earlier  characters.  For
+       point,  with  circumflex  characters  under the earlier characters. For
        example:
 
            re> /abc\Kxyz/
@@ -753,7 +788,7 @@ SUBJECT MODIFIERS
           0: abcxyz
              ^^^
 
-       Unlike  allusedtext, the startchar modifier can be used with JIT.  How-
+       Unlike allusedtext, the startchar modifier can be used with JIT.   How-
        ever, these two modifiers are mutually exclusive.
 
    Showing the value of all capture groups
@@ -761,84 +796,84 @@ SUBJECT MODIFIERS
        The allcaptures modifier requests that the values of all potential cap-
        tured parentheses be output after a match. By default, only those up to
        the highest one actually used in the match are output (corresponding to
-       the  return  code from pcre2_match()). Groups that did not take part in
+       the return code from pcre2_match()). Groups that did not take  part  in
        the match are output as "<unset>".
 
    Testing callouts
 
-       A callout function is supplied when pcre2test calls the library  match-
+       A  callout function is supplied when pcre2test calls the library match-
-       ing  functions, unless callout_none is specified. If callout_capture is
+       ing functions, unless callout_none is specified. If callout_capture  is
        set, the current captured groups are output when a callout occurs.
 
-       The callout_fail modifier can be given one or two numbers. If there  is
+       The  callout_fail modifier can be given one or two numbers. If there is
        only one number, 1 is returned instead of 0 when a callout of that num-
-       ber is reached. If two numbers are given, 1 is  returned  when  callout
+       ber  is  reached.  If two numbers are given, 1 is returned when callout
        <n> is reached for the <m>th time.
 
-       The  callout_data  modifier can be given an unsigned or a negative num-
+       The callout_data modifier can be given an unsigned or a  negative  num-
-       ber.  Any value other than zero is used as a  return  from  pcre2test's
+       ber.   Any  value  other than zero is used as a return from pcre2test's
        callout function.
 
    Finding all matches in a string
 
        Searching for all possible matches within a subject can be requested by
-       the global or /altglobal modifier. After finding a match, the  matching
+       the  global or /altglobal modifier. After finding a match, the matching
-       function  is  called  again to search the remainder of the subject. The
+       function is called again to search the remainder of  the  subject.  The
-       difference between global and altglobal is that  the  former  uses  the
+       difference  between  global  and  altglobal is that the former uses the
-       start_offset  argument  to  pcre2_match() or pcre2_dfa_match() to start
+       start_offset argument to pcre2_match() or  pcre2_dfa_match()  to  start
-       searching at a new point within the entire string (which is  what  Perl
+       searching  at  a new point within the entire string (which is what Perl
        does), whereas the latter passes over a shortened subject. This makes a
        difference to the matching process if the pattern begins with a lookbe-
        hind assertion (including \b or \B).
 
-       If  an  empty  string  is  matched,  the  next  match  is done with the
+       If an empty string  is  matched,  the  next  match  is  done  with  the
        PCRE2_NOTEMPTY_ATSTART and PCRE2_ANCHORED flags set, in order to search
        for another, non-empty, match at the same point in the subject. If this
-       match fails, the start offset is advanced,  and  the  normal  match  is
+       match  fails,  the  start  offset  is advanced, and the normal match is
-       retried.  This  imitates the way Perl handles such cases when using the
+       retried. This imitates the way Perl handles such cases when  using  the
-       /g modifier or the split() function.  Normally,  the  start  offset  is
+       /g  modifier  or  the  split()  function. Normally, the start offset is
-       advanced  by  one  character,  but if the newline convention recognizes
+       advanced by one character, but if  the  newline  convention  recognizes
-       CRLF as a newline, and the current character is CR followed by  LF,  an
+       CRLF  as  a newline, and the current character is CR followed by LF, an
        advance of two characters occurs.
 
    Testing substring extraction functions
 
-       The  copy  and  get  modifiers  can  be  used  to  test  the pcre2_sub-
+       The copy  and  get  modifiers  can  be  used  to  test  the  pcre2_sub-
        string_copy_xxx() and pcre2_substring_get_xxx() functions.  They can be
-       given  more than once, and each can specify a group name or number, for
+       given more than once, and each can specify a group name or number,  for
        example:
 
           abcd\=copy=1,copy=3,get=G1
 
-       If the #subject command is used to set default copy and/or  get  lists,
+       If  the  #subject command is used to set default copy and/or get lists,
-       these  can  be unset by specifying a negative number to cancel all num-
+       these can be unset by specifying a negative number to cancel  all  num-
        bered groups and an empty name to cancel all named groups.
 
-       The getall modifier tests  pcre2_substring_list_get(),  which  extracts
+       The  getall  modifier  tests pcre2_substring_list_get(), which extracts
        all captured substrings.
 
-       If  the  subject line is successfully matched, the substrings extracted
+       If the subject line is successfully matched, the  substrings  extracted
-       by the convenience functions are output with  C,  G,  or  L  after  the
+       by  the  convenience  functions  are  output  with C, G, or L after the
-       string  number  instead  of  a colon. This is in addition to the normal
+       string number instead of a colon. This is in  addition  to  the  normal
-       full list. The string length (that is, the return from  the  extraction
+       full  list.  The string length (that is, the return from the extraction
        function) is given in parentheses after each substring, followed by the
        name when the extraction was by name.
 
    Testing the substitution function
 
-       If the replace modifier is  set,  the  pcre2_substitute()  function  is
+       If  the  replace  modifier  is  set, the pcre2_substitute() function is
-       called  instead  of  one  of  the  matching  functions.  Unlike subject
+       called instead  of  one  of  the  matching  functions.  Unlike  subject
-       strings, pcre2test does not  process  replacement  strings  for  escape
+       strings,  pcre2test  does  not  process  replacement strings for escape
        sequences. In UTF mode, a replacement string is checked to see if it is
        a valid UTF-8 string.  If so, it is correctly converted to a UTF string
-       of  the appropriate code unit width. If it is not a valid UTF-8 string,
+       of the appropriate code unit width. If it is not a valid UTF-8  string,
        the individual code units are copied directly. This provides a means of
        passing an invalid UTF-8 string for testing purposes.
 
-       If  the  global  modifier  is set, PCRE2_SUBSTITUTE_GLOBAL is passed to
+       If the global modifier is set,  PCRE2_SUBSTITUTE_GLOBAL  is  passed  to
        pcre2_substitute().  After  a  successful  substitution,  the  modified
-       string  is  output, preceded by the number of replacements. This may be
+       string is output, preceded by the number of replacements. This  may  be
-       zero if there were no matches. Here is a simple example of a  substitu-
+       zero  if there were no matches. Here is a simple example of a substitu-
        tion test:
 
          /abc/replace=xxx
@@ -847,11 +882,11 @@ SUBJECT MODIFIERS
              =abc=abc=\=global
           2: =xxx=xxx=
 
-       Subject  and  replacement  strings  should be kept relatively short for
+       Subject and replacement strings should be  kept  relatively  short  for
-       substitution tests, as fixed-size buffers are used. To make it easy  to
+       substitution  tests, as fixed-size buffers are used. To make it easy to
-       test  for buffer overflow, if the replacement string starts with a num-
+       test for buffer overflow, if the replacement string starts with a  num-
-       ber in square brackets, that number is passed to pcre2_substitute()  as
+       ber  in square brackets, that number is passed to pcre2_substitute() as
-       the  size of the output buffer, with the replacement string starting at
+       the size of the output buffer, with the replacement string starting  at
        the next character. Here is an example that tests the edge case:
 
          /abc/
@@ -861,123 +896,123 @@ SUBJECT MODIFIERS
          Failed: error -47: no more memory
 
        A replacement string is ignored with POSIX and DFA matching. Specifying
-       partial  matching  provokes  an  error return ("bad option value") from
+       partial matching provokes an error return  ("bad  option  value")  from
        pcre2_substitute().
 
    Setting the JIT stack size
 
-       The jitstack modifier provides a way of setting the maximum stack  size
+       The  jitstack modifier provides a way of setting the maximum stack size
-       that  is  used  by the just-in-time optimization code. It is ignored if
+       that is used by the just-in-time optimization code. It  is  ignored  if
        JIT optimization is not being used. The value is a number of kilobytes.
        Providing a stack that is larger than the default 32K is necessary only
        for very complicated patterns.
 
    Setting match and recursion limits
 
-       The match_limit and recursion_limit modifiers set the appropriate  lim-
+       The  match_limit and recursion_limit modifiers set the appropriate lim-
        its in the match context. These values are ignored when the find_limits
        modifier is specified.
 
    Finding minimum limits
 
-       If the find_limits modifier is present, pcre2test  calls  pcre2_match()
+       If  the  find_limits modifier is present, pcre2test calls pcre2_match()
-       several  times,  setting  different  values  in  the  match context via
+       several times, setting  different  values  in  the  match  context  via
-       pcre2_set_match_limit() and pcre2_set_recursion_limit() until it  finds
+       pcre2_set_match_limit()  and pcre2_set_recursion_limit() until it finds
-       the  minimum values for each parameter that allow pcre2_match() to com-
+       the minimum values for each parameter that allow pcre2_match() to  com-
        plete without error.
 
        If JIT is being used, only the match limit is relevant. If DFA matching
-       is  being used, neither limit is relevant, and this modifier is ignored
+       is being used, neither limit is relevant, and this modifier is  ignored
        (with a warning message).
 
-       The match_limit number is a measure of the amount of backtracking  that
+       The  match_limit number is a measure of the amount of backtracking that
-       takes  place,  and  learning  the minimum value can be instructive. For
+       takes place, and learning the minimum value  can  be  instructive.  For
-       most simple matches, the number is quite small, but for  patterns  with
+       most  simple  matches, the number is quite small, but for patterns with
-       very  large numbers of matching possibilities, it can become large very
+       very large numbers of matching possibilities, it can become large  very
-       quickly   with   increasing   length    of    subject    string.    The
+       quickly    with    increasing    length    of   subject   string.   The
-       match_limit_recursion  number  is  a  measure of how much stack (or, if
+       match_limit_recursion number is a measure of how  much  stack  (or,  if
-       PCRE2 is compiled with NO_RECURSE, how much heap) memory is  needed  to
+       PCRE2  is  compiled with NO_RECURSE, how much heap) memory is needed to
        complete the match attempt.
 
    Showing MARK names
 
 
        The mark modifier causes the names from backtracking control verbs that
-       are returned from calls to pcre2_match() to be displayed. If a mark  is
+       are  returned from calls to pcre2_match() to be displayed. If a mark is
-       returned  for a match, non-match, or partial match, pcre2test shows it.
+       returned for a match, non-match, or partial match, pcre2test shows  it.
-       For a match, it is on a line by itself, tagged with  "MK:".  Otherwise,
+       For  a  match, it is on a line by itself, tagged with "MK:". Otherwise,
        it is added to the non-match message.
 
    Showing memory usage
 
-       The  memory  modifier causes pcre2test to log all memory allocation and
+       The memory modifier causes pcre2test to log all memory  allocation  and
        freeing calls that occur during a match operation.
 
    Setting a starting offset
 
-       The offset modifier sets an offset  in  the  subject  string  at  which
+       The  offset  modifier  sets  an  offset  in the subject string at which
        matching starts. Its value is a number of code units, not characters.
 
    Setting the size of the output vector
 
-       The  ovector  modifier  applies  only  to  the subject line in which it
+       The ovector modifier applies only to  the  subject  line  in  which  it
-       appears, though of course it can also be used to set  a  default  in  a
+       appears,  though  of  course  it can also be used to set a default in a
-       #subject  command. It specifies the number of pairs of offsets that are
+       #subject command. It specifies the number of pairs of offsets that  are
        available for storing matching information. The default is 15.
 
-       A value of zero is useful when testing the POSIX API because it  causes
+       A  value of zero is useful when testing the POSIX API because it causes
        regexec() to be called with a NULL capture vector. When not testing the
-       POSIX API, a value of  zero  is  used  to  cause  pcre2_match_data_cre-
+       POSIX  API,  a  value  of  zero  is used to cause pcre2_match_data_cre-
-       ate_from_pattern()  to  be  called, in order to create a match block of
+       ate_from_pattern() to be called, in order to create a  match  block  of
        exactly the right size for the pattern. (It is not possible to create a
-       match  block  with  a zero-length ovector; there is always at least one
+       match block with a zero-length ovector; there is always  at  least  one
        pair of offsets.)
 
    Passing the subject as zero-terminated
 
        By default, the subject string is passed to a native API matching func-
        tion with its correct length. In order to test the facility for passing
-       a zero-terminated string, the zero_terminate modifier is  provided.  It
+       a  zero-terminated  string, the zero_terminate modifier is provided. It
        causes the length to be passed as PCRE2_ZERO_TERMINATED. (When matching
-       via the POSIX interface, this modifier has no effect, as  there  is  no
+       via  the  POSIX  interface, this modifier has no effect, as there is no
        facility for passing a length.)
 
-       When  testing  pcre2_substitute(), this modifier also has the effect of
+       When testing pcre2_substitute(), this modifier also has the  effect  of
        passing the replacement string as zero-terminated.
 
 
 THE ALTERNATIVE MATCHING FUNCTION
 
-       By default,  pcre2test  uses  the  standard  PCRE2  matching  function,
+       By  default,  pcre2test  uses  the  standard  PCRE2  matching function,
        pcre2_match() to match each subject line. PCRE2 also supports an alter-
-       native matching function, pcre2_dfa_match(), which operates in  a  dif-
+       native  matching  function, pcre2_dfa_match(), which operates in a dif-
-       ferent  way, and has some restrictions. The differences between the two
+       ferent way, and has some restrictions. The differences between the  two
        functions are described in the pcre2matching documentation.
 
-       If the dfa modifier is set, the alternative matching function is  used.
+       If  the dfa modifier is set, the alternative matching function is used.
-       This  function  finds all possible matches at a given point in the sub-
+       This function finds all possible matches at a given point in  the  sub-
-       ject. If, however, the dfa_shortest modifier is set,  processing  stops
+       ject.  If,  however, the dfa_shortest modifier is set, processing stops
-       after  the  first  match is found. This is always the shortest possible
+       after the first match is found. This is always  the  shortest  possible
        match.
 
 
 DEFAULT OUTPUT FROM pcre2test
 
-       This section describes the output when the  normal  matching  function,
+       This  section  describes  the output when the normal matching function,
        pcre2_match(), is being used.
 
-       When  a  match  succeeds,  pcre2test  outputs the list of captured sub-
+       When a match succeeds, pcre2test outputs  the  list  of  captured  sub-
-       strings, starting with number 0 for the string that matched  the  whole
+       strings,  starting  with number 0 for the string that matched the whole
-       pattern.    Otherwise,  it  outputs  "No  match"  when  the  return  is
+       pattern.   Otherwise,  it  outputs  "No  match"  when  the  return   is
-       PCRE2_ERROR_NOMATCH, or "Partial  match:"  followed  by  the  partially
+       PCRE2_ERROR_NOMATCH,  or  "Partial  match:"  followed  by the partially
-       matching  substring  when the return is PCRE2_ERROR_PARTIAL. (Note that
+       matching substring when the return is PCRE2_ERROR_PARTIAL.  (Note  that
-       this is the entire substring that  was  inspected  during  the  partial
+       this  is  the  entire  substring  that was inspected during the partial
-       match;  it  may  include  characters before the actual match start if a
+       match; it may include characters before the actual  match  start  if  a
        lookbehind assertion, \K, \b, or \B was involved.)
 
        For any other return, pcre2test outputs the PCRE2 negative error number
-       and  a  short  descriptive  phrase. If the error is a failed UTF string
+       and a short descriptive phrase. If the error is  a  failed  UTF  string
-       check, the code unit offset of the start of the  failing  character  is
+       check,  the  code  unit offset of the start of the failing character is
        also output. Here is an example of an interactive pcre2test run.
 
          $ pcre2test
@@ -993,8 +1028,8 @@ DEFAULT OUTPUT FROM pcre2test
        Unset capturing substrings that are not followed by one that is set are
        not shown by pcre2test unless the allcaptures modifier is specified. In
        the following example, there are two capturing substrings, but when the
-       first data line is matched, the second, unset substring is  not  shown.
+       first  data  line is matched, the second, unset substring is not shown.
-       An  "internal" unset substring is shown as "<unset>", as for the second
+       An "internal" unset substring is shown as "<unset>", as for the  second
        data line.
 
            re> /(a)|(b)/
@@ -1006,11 +1041,11 @@ DEFAULT OUTPUT FROM pcre2test
           1: <unset>
           2: b
 
-       If the strings contain any non-printing characters, they are output  as
+       If  the strings contain any non-printing characters, they are output as
-       \xhh  escapes  if  the  value is less than 256 and UTF mode is not set.
+       \xhh escapes if the value is less than 256 and UTF  mode  is  not  set.
        Otherwise they are output as \x{hh...} escapes. See below for the defi-
-       nition  of  non-printing characters. If the /aftertext modifier is set,
+       nition of non-printing characters. If the /aftertext modifier  is  set,
-       the output for substring 0 is followed by the the rest of  the  subject
+       the  output  for substring 0 is followed by the the rest of the subject
        string, identified by "0+" like this:
 
            re> /cat/aftertext
@@ -1018,7 +1053,7 @@ DEFAULT OUTPUT FROM pcre2test
           0: cat
           0+ aract
 
-       If  global  matching  is  requested, the results of successive matching
+       If global matching is requested, the  results  of  successive  matching
        attempts are output in sequence, like this:
 
            re> /\Bi(\w\w)/g
@@ -1030,8 +1065,8 @@ DEFAULT OUTPUT FROM pcre2test
           0: ipp
           1: pp
 
-       "No match" is output only if the first match attempt fails. Here is  an
+       "No  match" is output only if the first match attempt fails. Here is an
-       example  of  a  failure  message (the offset 4 that is specified by the
+       example of a failure message (the offset 4 that  is  specified  by  the
        offset modifier is past the end of the subject string):
 
            re> /xyz/
@@ -1039,7 +1074,7 @@ DEFAULT OUTPUT FROM pcre2test
          Error -24 (bad offset value)
 
        Note that whereas patterns can be continued over several lines (a plain
-       ">"  prompt  is used for continuations), subject lines may not. However
+       ">" prompt is used for continuations), subject lines may  not.  However
        newlines can be included in a subject by means of the \n escape (or \r,
        \r\n, etc., depending on the newline sequence setting).
 
@@ -1047,7 +1082,7 @@ DEFAULT OUTPUT FROM pcre2test
 OUTPUT FROM THE ALTERNATIVE MATCHING FUNCTION
 
        When the alternative matching function, pcre2_dfa_match(), is used, the
-       output consists of a list of all the matches that start  at  the  first
+       output  consists  of  a list of all the matches that start at the first
        point in the subject where there is at least one match. For example:
 
            re> /(tang|tangerine|tan)/
@@ -1056,11 +1091,11 @@ OUTPUT FROM THE ALTERNATIVE MATCHING FUNCTION
           1: tang
           2: tan
 
-       Using  the normal matching function on this data finds only "tang". The
+       Using the normal matching function on this data finds only "tang".  The
-       longest matching string is always  given  first  (and  numbered  zero).
+       longest  matching  string  is  always  given first (and numbered zero).
-       After  a  PCRE2_ERROR_PARTIAL  return,  the output is "Partial match:",
+       After a PCRE2_ERROR_PARTIAL return, the  output  is  "Partial  match:",
-       followed by the partially matching substring. Note  that  this  is  the
+       followed  by  the  partially  matching substring. Note that this is the
-       entire  substring  that  was inspected during the partial match; it may
+       entire substring that was inspected during the partial  match;  it  may
        include characters before the actual match start if a lookbehind asser-
        tion, \b, or \B was involved. (\K is not supported for DFA matching.)
 
@@ -1076,16 +1111,16 @@ OUTPUT FROM THE ALTERNATIVE MATCHING FUNCTION
           1: tan
           0: tan
 
-       The alternative matching function does not support  substring  capture,
+       The  alternative  matching function does not support substring capture,
-       so  the  modifiers  that are concerned with captured substrings are not
+       so the modifiers that are concerned with captured  substrings  are  not
        relevant.
 
 
 RESTARTING AFTER A PARTIAL MATCH
 
-       When the alternative matching function has given  the  PCRE2_ERROR_PAR-
+       When  the  alternative matching function has given the PCRE2_ERROR_PAR-
        TIAL return, indicating that the subject partially matched the pattern,
-       you can restart the match with additional subject data by means of  the
+       you  can restart the match with additional subject data by means of the
        dfa_restart modifier. For example:
 
            re> /^\d?\d(jan|feb|mar|apr|may|jun|jul|aug|sep|oct|nov|dec)\d\d$/
@@ -1094,29 +1129,29 @@ RESTARTING AFTER A PARTIAL MATCH
          data> n05\=dfa,dfa_restart
           0: n05
 
-       For  further  information  about partial matching, see the pcre2partial
+       For further information about partial matching,  see  the  pcre2partial
        documentation.
 
 
 CALLOUTS
 
        If the pattern contains any callout requests, pcre2test's callout func-
-       tion  is  called  during  matching. This works with both matching func-
+       tion is called during matching. This works  with  both  matching  func-
        tions. By default, the called function displays the callout number, the
-       start  and  current  positions in the text at the callout time, and the
+       start and current positions in the text at the callout  time,  and  the
        next pattern item to be tested. For example:
 
          --->pqrabcdef
            0    ^  ^     \d
 
-       This output indicates that  callout  number  0  occurred  for  a  match
+       This  output  indicates  that  callout  number  0  occurred for a match
-       attempt  starting  at  the fourth character of the subject string, when
+       attempt starting at the fourth character of the  subject  string,  when
-       the pointer was at the seventh character, and  when  the  next  pattern
+       the  pointer  was  at  the seventh character, and when the next pattern
-       item  was  \d.  Just  one circumflex is output if the start and current
+       item was \d. Just one circumflex is output if  the  start  and  current
        positions are the same.
 
        Callouts numbered 255 are assumed to be automatic callouts, inserted as
-       a  result  of the /auto_callout pattern modifier. In this case, instead
+       a result of the /auto_callout pattern modifier. In this  case,  instead
        of showing the callout number, the offset in the pattern, preceded by a
        plus, is output. For example:
 
@@ -1130,7 +1165,7 @@ CALLOUTS
           0: E*
 
        If a pattern contains (*MARK) items, an additional line is output when-
-       ever a change of latest mark is passed to  the  callout  function.  For
+       ever  a  change  of  latest mark is passed to the callout function. For
        example:
 
            re> /a(*MARK:X)bc/auto_callout
@@ -1144,37 +1179,96 @@ CALLOUTS
          +12 ^  ^
           0: abc
 
-       The  mark  changes between matching "a" and "b", but stays the same for
+       The mark changes between matching "a" and "b", but stays the  same  for
-       the rest of the match, so nothing more is output. If, as  a  result  of
+       the  rest  of  the match, so nothing more is output. If, as a result of
-       backtracking,  the  mark  reverts to being unset, the text "<unset>" is
+       backtracking, the mark reverts to being unset, the  text  "<unset>"  is
        output.
 
-       The callout function in pcre2test returns zero (carry on  matching)  by
+       The  callout  function in pcre2test returns zero (carry on matching) by
-       default,  but you can use a callout_fail modifier in a subject line (as
+       default, but you can use a callout_fail modifier in a subject line  (as
        described above) to change this and other parameters of the callout.
 
        Inserting callouts can be helpful when using pcre2test to check compli-
-       cated  regular expressions. For further information about callouts, see
+       cated regular expressions. For further information about callouts,  see
        the pcre2callout documentation.
 
 
 NON-PRINTING CHARACTERS
 
        When pcre2test is outputting text in the compiled version of a pattern,
-       bytes  other  than 32-126 are always treated as non-printing characters
+       bytes other than 32-126 are always treated as  non-printing  characters
        and are therefore shown as hex escapes.
 
-       When pcre2test is outputting text that is a matched part of  a  subject
+       When  pcre2test  is outputting text that is a matched part of a subject
-       string,  it behaves in the same way, unless a different locale has been
+       string, it behaves in the same way, unless a different locale has  been
-       set for the pattern (using the /locale modifier).  In  this  case,  the
+       set  for  the  pattern  (using the /locale modifier). In this case, the
-       isprint()  function  is  used  to distinguish printing and non-printing
+       isprint() function is used to  distinguish  printing  and  non-printing
        characters.
 
 
+SAVING AND RESTORING COMPILED PATTERNS
+
+       It  is  possible  to  save  compiled patterns on disc or elsewhere, and
+       reload them later, subject to a number of restrictions. JIT data cannot
+       be  saved.  The host on which the patterns are reloaded must be running
+       the same version of PCRE2, with the same code unit width, and must also
+       have  the  same  endianness,  pointer width and PCRE2_SIZE type. Before
+       compiled patterns can be saved they must be serialized, that  is,  con-
+       verted  to a stream of bytes. A single byte stream may contain any num-
+       ber of compiled patterns, but they must  all  use  the  same  character
+       tables. A single copy of the tables is included in the byte stream (its
+       size is 1088 bytes).
+
+       The functions whose names begin  with  pcre2_serialize_  are  used  for
+       serializing  and de-serializing. They are described in the pcre2serial-
+       ize  documentation.  In  this  section  we  describe  the  features  of
+       pcre2test that can be used to test these functions.
+
+       When  a  pattern  with  push  modifier  is successfully compiled, it is
+       pushed onto a stack of compiled patterns,  and  pcre2test  expects  the
+       next  line  to  contain a new pattern (or command) instead of a subject
+       line. By this means, a number of patterns can be compiled and retained.
+       The  push  modifier  is  incompatible with posix, and control modifiers
+       that act at match time are ignored (with a message). The jitverify mod-
+       ifier applies only at compile time. The command
+
+         #save <filename>
+
+       causes all the stacked patterns to be serialized and the result written
+       to the named file. Afterwards, all the stacked patterns are freed.  The
+       command
+
+         #load <filename>
+
+       reads  the  data in the file, and then arranges for it to be de-serial-
+       ized, with the resulting compiled patterns added to the pattern  stack.
+       The  pattern  on the top of the stack can be retrieved by the #pop com-
+       mand, which must be followed by  lines  of  subjects  that  are  to  be
+       matched  with  the pattern, terminated as usual by an empty line or end
+       of file. This command may be followed by  a  modifier  list  containing
+       only  control  modifiers that act after a pattern has been compiled. In
+       particular, hex, posix, and push are not allowed, nor are  any  option-
+       setting  modifiers.   The JIT modifiers are, however permitted. Here is
+       an example that saves and reloads two patterns.
+
+         /abc/push
+         /xyz/push
+         #save tempfile
+         #load tempfile
+         #pop info
+         xyz
+
+         #pop jit,bincode
+         abc
+
+       If jitverify is used with #pop, it does not  automatically  imply  jit,
+       which is different behaviour from when it is used on a pattern.
+
+
 SEE ALSO
 
        pcre2(3),  pcre2api(3),  pcre2callout(3),  pcre2jit,  pcre2matching(3),
-       pcre2partial(d), pcre2pattern(3).
+       pcre2partial(d), pcre2pattern(3), pcre2serialize(3).
 
 
 AUTHOR
@@ -1186,5 +1280,5 @@ AUTHOR
 
 REVISION
 
-       Last updated: 02 January 2015
+       Last updated: 23 January 2015
        Copyright (c) 1997-2015 University of Cambridge.

maint/ManyConfigTests

@@ -293,7 +293,7 @@ if [ $usevalgrind -ne 0 ]; then
 
   for opts in \
     "--disable-stack-for-recursion --disable-shared" \
-    "--with-link-size=3 --disable-shared" \
+    "--with-link-size=3 --enable-pcre2-16 --enable-pcre2-32 --disable-shared" \
     "--disable-unicode --disable-shared"
   do
     opts="--enable-valgrind $opts"

src/config.h.generic

@@ -200,7 +200,7 @@ 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 10.00"
+#define PACKAGE_STRING "PCRE2 10.10-RC1"
 
 /* Define to the one symbol short name of this package. */
 #define PACKAGE_TARNAME "pcre2"
@@ -209,7 +209,7 @@ 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 "10.00"
+#define PACKAGE_VERSION "10.10-RC1"
 
 /* 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
@@ -287,7 +287,7 @@ sure both macros are undefined; an emulation function will then be used. */
 /* #undef SUPPORT_VALGRIND */
 
 /* Version number of package */
-#define VERSION "10.00"
+#define VERSION "10.10-RC1"
 
 /* Define to empty if `const' does not conform to ANSI C. */
 /* #undef const */

src/pcre2.h.generic

@@ -42,9 +42,9 @@ POSSIBILITY OF SUCH DAMAGE.
 /* The current PCRE version information. */
 
 #define PCRE2_MAJOR          10
-#define PCRE2_MINOR          00
+#define PCRE2_MINOR          10
-#define PCRE2_PRERELEASE     
+#define PCRE2_PRERELEASE     -RC1
-#define PCRE2_DATE           2014-01-05
+#define PCRE2_DATE           2014-01-13
 
 /* When an application links to a PCRE DLL in Windows, the symbols that are
 imported have to be identified as such. When building PCRE2, the appropriate
@@ -455,6 +455,18 @@ PCRE2_EXP_DECL void      pcre2_substring_list_free(PCRE2_SPTR *); \
 PCRE2_EXP_DECL int       pcre2_substring_list_get(pcre2_match_data *, \
                            PCRE2_UCHAR ***, PCRE2_SIZE **);
 
+/* Functions for serializing / deserializing compiled patterns. */
+
+#define PCRE2_SERIALIZE_FUNCTIONS \
+PCRE2_EXP_DECL int       pcre2_serialize_encode(const pcre2_code **, \
+                           PCRE2_SIZE, uint8_t **, PCRE2_SIZE *, \
+                           pcre2_general_context *); \
+PCRE2_EXP_DECL int       pcre2_serialize_decode(pcre2_code **, PCRE2_SIZE, \
+                           const uint8_t *, pcre2_general_context *); \
+PCRE2_EXP_DECL int       pcre2_serialize_get_number_of_codes(const uint8_t *, \
+                           PCRE2_SIZE *); \
+PCRE2_EXP_DECL void      pcre2_serialize_free(uint8_t *);
+
 
 /* Convenience function for match + substitute. */
 
@@ -560,6 +572,10 @@ pcre2_compile are called by application code. */
 #define pcre2_match_data_create_from_pattern  PCRE2_SUFFIX(pcre2_match_data_create_from_pattern_)
 #define pcre2_match_data_free                 PCRE2_SUFFIX(pcre2_match_data_free_)
 #define pcre2_pattern_info                    PCRE2_SUFFIX(pcre2_pattern_info_)
+#define pcre2_serialize_decode                PCRE2_SUFFIX(pcre2_serialize_decode_)
+#define pcre2_serialize_encode                PCRE2_SUFFIX(pcre2_serialize_encode_)
+#define pcre2_serialize_free                  PCRE2_SUFFIX(pcre2_serialize_free_)
+#define pcre2_serialize_get_number_of_codes   PCRE2_SUFFIX(pcre2_serialize_get_number_of_codes_)
 #define pcre2_set_bsr                         PCRE2_SUFFIX(pcre2_set_bsr_)
 #define pcre2_set_callout                     PCRE2_SUFFIX(pcre2_set_callout_)
 #define pcre2_set_character_tables            PCRE2_SUFFIX(pcre2_set_character_tables_)
@@ -596,8 +612,9 @@ PCRE2_MATCH_CONTEXT_FUNCTIONS \
 PCRE2_COMPILE_FUNCTIONS \
 PCRE2_PATTERN_INFO_FUNCTIONS \
 PCRE2_MATCH_FUNCTIONS \
-PCRE2_SUBSTITUTE_FUNCTION \
 PCRE2_SUBSTRING_FUNCTIONS \
+PCRE2_SERIALIZE_FUNCTIONS \
+PCRE2_SUBSTITUTE_FUNCTION \
 PCRE2_JIT_FUNCTIONS \
 PCRE2_OTHER_FUNCTIONS
 
@@ -625,6 +642,8 @@ PCRE2_TYPES_STRUCTURES_AND_FUNCTIONS
 #undef PCRE2_PATTERN_INFO_FUNCTIONS
 #undef PCRE2_MATCH_FUNCTIONS
 #undef PCRE2_SUBSTRING_FUNCTIONS
+#undef PCRE2_SERIALIZE_FUNCTIONS
+#undef PCRE2_SUBSTITUTE_FUNCTION
 #undef PCRE2_JIT_FUNCTIONS
 #undef PCRE2_OTHER_FUNCTIONS
 #undef PCRE2_TYPES_STRUCTURES_AND_FUNCTIONS

src/pcre2.h.in

@@ -198,11 +198,13 @@ greater than zero. */
 #define PCRE2_ERROR_UTF32_ERR1      (-27)
 #define PCRE2_ERROR_UTF32_ERR2      (-28)
 
-/* Error codes for pcre2[_dfa]_match(), substring extraction functions, and
+/* Error codes for pcre2[_dfa]_match(), substring extraction functions, context
-context functions. */
+functions, and serializing functions. They are in numerical order. Originally
+they were in alphabetical order too, but now that PCRE2 is released, the
+numbers must not be changed. */
 
 #define PCRE2_ERROR_BADDATA           (-29)
-#define PCRE2_ERROR_BADLENGTH         (-30)
+#define PCRE2_ERROR_MIXEDTABLES       (-30)  /* Name was changed */
 #define PCRE2_ERROR_BADMAGIC          (-31)
 #define PCRE2_ERROR_BADMODE           (-32)
 #define PCRE2_ERROR_BADOFFSET         (-33)
@@ -455,6 +457,17 @@ PCRE2_EXP_DECL void      pcre2_substring_list_free(PCRE2_SPTR *); \
 PCRE2_EXP_DECL int       pcre2_substring_list_get(pcre2_match_data *, \
                            PCRE2_UCHAR ***, PCRE2_SIZE **);
 
+/* Functions for serializing / deserializing compiled patterns. */
+
+#define PCRE2_SERIALIZE_FUNCTIONS \
+PCRE2_EXP_DECL int32_t   pcre2_serialize_encode(const pcre2_code **, \
+                           int32_t, uint8_t **, PCRE2_SIZE *, \
+                           pcre2_general_context *); \
+PCRE2_EXP_DECL int32_t   pcre2_serialize_decode(pcre2_code **, int32_t, \
+                           const uint8_t *, pcre2_general_context *); \
+PCRE2_EXP_DECL int32_t   pcre2_serialize_get_number_of_codes(const uint8_t *); \
+PCRE2_EXP_DECL void      pcre2_serialize_free(uint8_t *);
+
 
 /* Convenience function for match + substitute. */
 
@@ -560,6 +573,10 @@ pcre2_compile are called by application code. */
 #define pcre2_match_data_create_from_pattern  PCRE2_SUFFIX(pcre2_match_data_create_from_pattern_)
 #define pcre2_match_data_free                 PCRE2_SUFFIX(pcre2_match_data_free_)
 #define pcre2_pattern_info                    PCRE2_SUFFIX(pcre2_pattern_info_)
+#define pcre2_serialize_decode                PCRE2_SUFFIX(pcre2_serialize_decode_)
+#define pcre2_serialize_encode                PCRE2_SUFFIX(pcre2_serialize_encode_)
+#define pcre2_serialize_free                  PCRE2_SUFFIX(pcre2_serialize_free_)
+#define pcre2_serialize_get_number_of_codes   PCRE2_SUFFIX(pcre2_serialize_get_number_of_codes_)
 #define pcre2_set_bsr                         PCRE2_SUFFIX(pcre2_set_bsr_)
 #define pcre2_set_callout                     PCRE2_SUFFIX(pcre2_set_callout_)
 #define pcre2_set_character_tables            PCRE2_SUFFIX(pcre2_set_character_tables_)
@@ -596,8 +613,9 @@ PCRE2_MATCH_CONTEXT_FUNCTIONS \
 PCRE2_COMPILE_FUNCTIONS \
 PCRE2_PATTERN_INFO_FUNCTIONS \
 PCRE2_MATCH_FUNCTIONS \
-PCRE2_SUBSTITUTE_FUNCTION \
 PCRE2_SUBSTRING_FUNCTIONS \
+PCRE2_SERIALIZE_FUNCTIONS \
+PCRE2_SUBSTITUTE_FUNCTION \
 PCRE2_JIT_FUNCTIONS \
 PCRE2_OTHER_FUNCTIONS
 
@@ -625,6 +643,8 @@ PCRE2_TYPES_STRUCTURES_AND_FUNCTIONS
 #undef PCRE2_PATTERN_INFO_FUNCTIONS
 #undef PCRE2_MATCH_FUNCTIONS
 #undef PCRE2_SUBSTRING_FUNCTIONS
+#undef PCRE2_SERIALIZE_FUNCTIONS
+#undef PCRE2_SUBSTITUTE_FUNCTION
 #undef PCRE2_JIT_FUNCTIONS
 #undef PCRE2_OTHER_FUNCTIONS
 #undef PCRE2_TYPES_STRUCTURES_AND_FUNCTIONS

src/pcre2_compile.c

@@ -683,10 +683,28 @@ static const uint8_t opcode_possessify[] = {
 PCRE2_EXP_DEFN void PCRE2_CALL_CONVENTION
 pcre2_code_free(pcre2_code *code)
 {
+PCRE2_SIZE* ref_count;
+
 if (code != NULL)
   {
   if (code->executable_jit != NULL)
     PRIV(jit_free)(code->executable_jit, &code->memctl);
+
+  if ((code->flags & PCRE2_DEREF_TABLES) != 0)
+    {
+    /* Decoded tables belong to the codes after deserialization, and they must
+    be freed when there are no more reference to them. The *ref_count should
+    always be > 0. */
+
+    ref_count = (PCRE2_SIZE *)(code->tables + tables_length);
+    if (*ref_count > 0)
+      {
+      (*ref_count)--;
+      if (*ref_count == 0)
+        code->memctl.free((void *)code->tables, code->memctl.memory_data);
+      }
+    }
+
   code->memctl.free(code, code->memctl.memory_data);
   }
 }
@@ -7317,8 +7335,14 @@ for (i = 0; i < cb->names_found; i++)
 
 PUT2(slot, 0, groupno);
 memcpy(slot + IMM2_SIZE, name, CU2BYTES(length));
-slot[IMM2_SIZE + length] = 0;
 cb->names_found++;
+
+/* Add a terminating zero and fill the rest of the slot with zeroes so that
+the memory is all initialized. Otherwise valgrind moans about uninitialized
+memory when saving serialized compiled patterns. */
+
+memset(slot + IMM2_SIZE + length, 0,
+  CU2BYTES(cb->name_entry_size - length - IMM2_SIZE));
 }
 
 
@@ -7356,6 +7380,7 @@ PCRE2_SPTR codestart;                   /* Start of compiled code */
 PCRE2_SPTR ptr;                         /* Current pointer in pattern */
 
 size_t length = 1;                      /* Allow or final END opcode */
+size_t usedlength;                      /* Actual length used */
 size_t re_blocksize;                    /* Size of memory block */
 
 int32_t firstcuflags, reqcuflags;       /* Type of first/req code unit */
@@ -7754,13 +7779,16 @@ overflow. */
 
 if (errorcode == 0 && ptr < cb.end_pattern) errorcode = ERR22;
 *code++ = OP_END;
-if ((size_t)(code - codestart) > length) errorcode = ERR23;
+usedlength = code - codestart;
+if (usedlength > length) errorcode = ERR23;
 
+/* If the estimated length exceeds the really used length, adjust the value of
+re->blocksize, and if valgrind support is configured, mark the extra allocated
+memory as unaddressable, so that any out-of-bound reads can be detected. */
+
+re->blocksize -= CU2BYTES(length - usedlength);
 #ifdef SUPPORT_VALGRIND
-/* If the estimated length exceeds the really used length, mark the extra
+VALGRIND_MAKE_MEM_NOACCESS(code, CU2BYTES(length - usedlength));
-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

src/pcre2_error.c

@@ -7,7 +7,7 @@ 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
+         New API code Copyright (c) 2015 University of Cambridge
 
 -----------------------------------------------------------------------------
 Redistribution and use in source and binary forms, with or without
@@ -200,7 +200,7 @@ static const char match_error_texts[] =
   "UTF-32 error: code points greater than 0x10ffff are not defined\0"
   "bad data value\0"
   /* 30 */
-  "bad length\0"
+  "patterns do not all use the same character tables\0"
   "magic number missing\0"
   "pattern compiled in wrong mode: 8/16/32-bit error\0"
   "bad offset value\0"

src/pcre2_internal.h

@@ -523,6 +523,7 @@ bytes in a code unit in that mode. */
 #define PCRE2_NL_SET        0x00008000  /* newline was set in the pattern */
 #define PCRE2_NOTEMPTY_SET  0x00010000  /* (*NOTEMPTY) used        ) keep */
 #define PCRE2_NE_ATST_SET   0x00020000  /* (*NOTEMPTY_ATSTART) used) together */
+#define PCRE2_DEREF_TABLES  0x00040000  /* Release character tables. */
 
 #define PCRE2_MODE_MASK     (PCRE2_MODE8 | PCRE2_MODE16 | PCRE2_MODE32)
 
@@ -1763,6 +1764,15 @@ typedef struct {
 #define UCD_CASESET(ch)     GET_UCD(ch)->caseset
 #define UCD_OTHERCASE(ch)   ((uint32_t)((int)ch + (int)(GET_UCD(ch)->other_case)))
 
+/* Header for serialized pcre2 codes. */
+
+typedef struct pcre2_serialized_data {
+  uint32_t magic;
+  uint32_t version;
+  uint32_t config;
+  int32_t  number_of_codes;
+} pcre2_serialized_data;
+
 
 
 /* ----------------- Items that need PCRE2_CODE_UNIT_WIDTH ----------------- */

src/pcre2_serialize.c

@@ -0,0 +1,251 @@
+/*************************************************
+*      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) 2015 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 functions for serializing and deserializing
+a sequence of compiled codes. */
+
+
+#ifdef HAVE_CONFIG_H
+#include "config.h"
+#endif
+
+
+#include "pcre2_internal.h"
+
+/* Magic number to provide a small check against being handed junk. */
+
+#define SERIALIZED_DATA_MAGIC 0x50523253u
+
+/* Deserialization is limited to the current PCRE version and
+character width. */
+
+#define SERIALIZED_DATA_VERSION \
+  ((PCRE2_MAJOR) | ((PCRE2_MINOR) << 16))
+
+#define SERIALIZED_DATA_CONFIG \
+  (sizeof(PCRE2_UCHAR) | ((sizeof(void*)) << 8) | ((sizeof(PCRE2_SIZE)) << 16))
+
+
+
+/*************************************************
+*           Serialize compiled patterns          *
+*************************************************/
+
+PCRE2_EXP_DECL int32_t PCRE2_CALL_CONVENTION
+pcre2_serialize_encode(const pcre2_code **codes, int32_t number_of_codes,
+   uint8_t **serialized_bytes, PCRE2_SIZE *serialized_size,
+   pcre2_general_context *gcontext)
+{
+uint8_t *bytes;
+uint8_t *dst_bytes;
+int32_t i;
+PCRE2_SIZE total_size;
+const pcre2_real_code *re;
+const uint8_t *tables;
+pcre2_serialized_data *data;
+
+const pcre2_memctl *memctl = (gcontext != NULL) ?
+  &gcontext->memctl : &PRIV(default_compile_context).memctl;
+
+if (codes == NULL || serialized_bytes == NULL || serialized_size == NULL)
+  return PCRE2_ERROR_NULL;
+
+if (number_of_codes <= 0) return PCRE2_ERROR_BADDATA;
+
+/* Compute total size. */
+total_size = sizeof(pcre2_serialized_data) + tables_length;
+tables = NULL;
+
+for (i = 0; i < number_of_codes; i++)
+  {
+  if (codes[i] == NULL) return PCRE2_ERROR_NULL;
+  re = (const pcre2_real_code *)(codes[i]);
+  if (re->magic_number != MAGIC_NUMBER) return PCRE2_ERROR_BADMAGIC;
+  if (tables == NULL)
+    tables = re->tables;
+  else if (tables != re->tables)
+    return PCRE2_ERROR_MIXEDTABLES;
+  total_size += re->blocksize;
+  }
+  
+/* Initialize the byte stream. */
+bytes = memctl->malloc(total_size + sizeof(pcre2_memctl), memctl->memory_data);
+if (bytes == NULL) return PCRE2_ERROR_NOMEMORY;
+
+/* The controller is stored as a hidden parameter. */
+memcpy(bytes, memctl, sizeof(pcre2_memctl));
+bytes += sizeof(pcre2_memctl);
+
+data = (pcre2_serialized_data *)bytes;
+data->magic = SERIALIZED_DATA_MAGIC;
+data->version = SERIALIZED_DATA_VERSION;
+data->config = SERIALIZED_DATA_CONFIG;
+data->number_of_codes = number_of_codes;
+
+/* Copy all compiled code data. */
+dst_bytes = bytes + sizeof(pcre2_serialized_data);
+memcpy(dst_bytes, tables, tables_length);
+dst_bytes += tables_length;
+
+for (i = 0; i < number_of_codes; i++)
+  {
+  re = (const pcre2_real_code *)(codes[i]);
+  memcpy(dst_bytes, (char *)re, re->blocksize);
+  dst_bytes += re->blocksize;
+  }
+
+*serialized_bytes = bytes;
+*serialized_size = total_size;
+return number_of_codes;
+}
+
+
+/*************************************************
+*          Deserialize compiled patterns         *
+*************************************************/
+
+PCRE2_EXP_DECL int32_t PCRE2_CALL_CONVENTION
+pcre2_serialize_decode(pcre2_code **codes, int32_t number_of_codes,
+   const uint8_t *bytes, pcre2_general_context *gcontext)
+{
+const pcre2_serialized_data *data = (const pcre2_serialized_data *)bytes;
+const pcre2_memctl *memctl = (gcontext != NULL) ?
+  &gcontext->memctl : &PRIV(default_compile_context).memctl;
+
+const uint8_t *src_bytes;
+pcre2_real_code *src_re;
+pcre2_real_code *dst_re;
+uint8_t *tables;
+int32_t i, j;
+
+/* Sanity checks. */
+
+if (data == NULL || codes == NULL) return PCRE2_ERROR_NULL;
+if (number_of_codes <= 0) return PCRE2_ERROR_BADDATA;
+if (data->magic != SERIALIZED_DATA_MAGIC) return PCRE2_ERROR_BADMAGIC;
+if (data->version != SERIALIZED_DATA_VERSION) return PCRE2_ERROR_BADMODE;
+if (data->config != SERIALIZED_DATA_CONFIG) return PCRE2_ERROR_BADMODE;
+
+if (number_of_codes > data->number_of_codes)
+  number_of_codes = data->number_of_codes;
+
+src_bytes = bytes + sizeof(pcre2_serialized_data);
+
+/* Decode tables. The reference count for the tables is stored immediately 
+following them. */
+
+tables = memctl->malloc(tables_length + sizeof(PCRE2_SIZE), memctl->memory_data);
+if (tables == NULL) return PCRE2_ERROR_NOMEMORY;
+
+memcpy(tables, src_bytes, tables_length);
+*(PCRE2_SIZE *)(tables + tables_length) = number_of_codes;
+src_bytes += tables_length;
+
+/* Decode byte stream. */
+
+for (i = 0; i < number_of_codes; i++)
+  {
+  src_re = (pcre2_real_code *)src_bytes;
+
+  /* The allocator provided by gcontext replaces the original one. */
+  dst_re = (pcre2_real_code *)PRIV(memctl_malloc)
+    (src_re->blocksize, (pcre2_memctl *)gcontext);
+  if (dst_re == NULL)
+    {
+    memctl->free(tables, memctl->memory_data);
+    for (j = 0; j < i; j++)
+      {
+      memctl->free(codes[j], memctl->memory_data);
+      codes[j] = NULL;
+      }
+    return PCRE2_ERROR_NOMEMORY;
+    }
+
+  /* The new allocator must be preserved. */
+  memcpy(((uint8_t *)dst_re) + sizeof(pcre2_memctl),
+    src_bytes + sizeof(pcre2_memctl),
+    src_re->blocksize - sizeof(pcre2_memctl));
+
+  /* At the moment only one table is supported. */
+  dst_re->tables = tables;
+  dst_re->executable_jit = NULL;
+  dst_re->flags |= PCRE2_DEREF_TABLES;
+
+  codes[i] = dst_re;
+  src_bytes += src_re->blocksize;
+  }
+
+return number_of_codes;
+}
+
+
+/*************************************************
+*    Get the number of serialized patterns       *
+*************************************************/
+
+PCRE2_EXP_DECL int32_t PCRE2_CALL_CONVENTION
+pcre2_serialize_get_number_of_codes(const uint8_t *bytes)
+{
+const pcre2_serialized_data *data = (const pcre2_serialized_data *)bytes;
+
+if (data == NULL) return PCRE2_ERROR_NULL;
+if (data->magic != SERIALIZED_DATA_MAGIC) return PCRE2_ERROR_BADMAGIC;
+if (data->version != SERIALIZED_DATA_VERSION) return PCRE2_ERROR_BADMODE;
+if (data->config != SERIALIZED_DATA_CONFIG) return PCRE2_ERROR_BADMODE;
+
+return data->number_of_codes;
+}
+
+
+/*************************************************
+*            Free the allocated stream           *
+*************************************************/
+
+PCRE2_EXP_DECL void PCRE2_CALL_CONVENTION
+pcre2_serialize_free(uint8_t *bytes)
+{
+if (bytes != NULL)
+  {
+  pcre2_memctl *memctl = (pcre2_memctl *)(bytes - sizeof(pcre2_memctl));
+  memctl->free(memctl, memctl->memory_data);
+  } 
+}
+
+/* End of pcre2_serialize.c */

src/pcre2test.c

@@ -166,6 +166,7 @@ void vms_setsymbol( char *, char *, int );
 #define JUNK_OFFSET 0xdeadbeef  /* For initializing ovector */
 #define LOCALESIZE 32           /* Size of locale name */
 #define LOOPREPEAT 500000       /* Default loop count for timing */
+#define PATSTACKSIZE 20         /* Pattern stack for save/restore testing */
 #define REPLACE_MODSIZE 96      /* Field for reading 8-bit replacement */
 #define VERSION_SIZE 64         /* Size of buffer for the version strings */
 
@@ -313,6 +314,26 @@ modes, so use the form of the first that is available. */
 #define PCRE2_REAL_MATCH_CONTEXT pcre2_real_match_context_32
 #endif
 
+/* ------------- Structure and table for handling #-commands ------------- */
+
+typedef struct cmdstruct {
+  const char *name;
+  int  value;
+} cmdstruct;
+
+enum { CMD_FORBID_UTF, CMD_LOAD, CMD_PATTERN, CMD_PERLTEST, CMD_POP, CMD_SAVE,
+  CMD_SUBJECT, CMD_UNKNOWN };
+
+static cmdstruct cmdlist[] = {
+  { "forbid_utf",  CMD_FORBID_UTF },
+  { "load",        CMD_LOAD },
+  { "pattern",     CMD_PATTERN },
+  { "perltest",    CMD_PERLTEST },
+  { "pop",         CMD_POP },
+  { "save",        CMD_SAVE },
+  { "subject",     CMD_SUBJECT }};
+
+#define cmdlistcount sizeof(cmdlist)/sizeof(cmdstruct)
 
 /* ------------- Structures and tables for handling modifiers -------------- */
 
@@ -367,8 +388,9 @@ either on a pattern or a data line, so they must all be distinct. */
 #define CTL_MARK               0x00020000u
 #define CTL_MEMORY             0x00040000u
 #define CTL_POSIX              0x00080000u
-#define CTL_STARTCHAR          0x00100000u
+#define CTL_PUSH               0x00100000u
-#define CTL_ZERO_TERMINATE     0x00200000u
+#define CTL_STARTCHAR          0x00200000u
+#define CTL_ZERO_TERMINATE     0x00400000u
 
 #define CTL_BSR_SET          0x80000000u  /* This is informational */
 #define CTL_NL_SET           0x40000000u  /* This is informational */
@@ -426,6 +448,7 @@ typedef struct datctl {    /* Structure for data line modifiers. */
 /* Ids for which context to modify. */
 
 enum { CTX_PAT,            /* Active pattern context */
+       CTX_POPPAT,         /* Ditto, for a popped pattern */
        CTX_DEFPAT,         /* Default pattern context */
        CTX_DAT,            /* Active data (match) context */
        CTX_DEFDAT };       /* Default data (match) context */
@@ -513,6 +536,7 @@ static modstruct modlist[] = {
   { "ph",                  MOD_DAT,  MOD_OPT, PCRE2_PARTIAL_HARD,        DO(options) },
   { "posix",               MOD_PAT,  MOD_CTL, CTL_POSIX,                 PO(control) },
   { "ps",                  MOD_DAT,  MOD_OPT, PCRE2_PARTIAL_SOFT,        DO(options) },
+  { "push",                MOD_PAT,  MOD_CTL, CTL_PUSH,                  PO(control) },
   { "recursion_limit",     MOD_CTM,  MOD_INT, 0,                         MO(recursion_limit) },
   { "replace",             MOD_PND,  MOD_STR, REPLACE_MODSIZE,           PO(replacement) },
   { "stackguard",          MOD_PAT,  MOD_INT, 0,                         PO(stackguard_test) },
@@ -544,6 +568,20 @@ static modstruct modlist[] = {
 
 #define EXCLUSIVE_DAT_CONTROLS (CTL_ALLUSEDTEXT|CTL_STARTCHAR)
 
+/* Control bits that are not ignored with 'push'. */
+
+#define PUSH_SUPPORTED_COMPILE_CONTROLS ( \
+  CTL_BINCODE|CTL_FULLBINCODE|CTL_HEXPAT|CTL_INFO|CTL_JITVERIFY| \
+  CTL_MEMORY|CTL_PUSH|CTL_BSR_SET|CTL_NL_SET)
+
+/* Controls that apply only at compile time with 'push'. */
+
+#define PUSH_COMPILE_ONLY_CONTROLS  CTL_JITVERIFY
+
+/* Controls that are forbidden with #pop. */
+
+#define NOTPOP_CONTROLS (CTL_HEXPAT|CTL_POSIX|CTL_PUSH)
+
 /* Table of single-character abbreviated modifiers. The index field is
 initialized to -1, but the first time the modifier is encountered, it is filled
 in with the index of the full entry in modlist, to save repeated searching when
@@ -671,6 +709,9 @@ static patctl pat_patctl;
 static datctl def_datctl;
 static datctl dat_datctl;
 
+static void *patstack[PATSTACKSIZE];
+static int patstacknext = 0;
+
 #ifdef SUPPORT_PCRE2_8
 static regex_t preg = { NULL, NULL, 0, 0 };
 #endif
@@ -928,6 +969,38 @@ are supported. */
   else \
     pcre2_printint_32(compiled_code32,outfile,a)
 
+#define PCRE2_SERIALIZE_DECODE(r,a,b,c,d) \
+  if (test_mode == PCRE8_MODE) \
+    r = pcre2_serialize_decode_8((pcre2_code_8 **)a,b,c,G(d,8)); \
+  else if (test_mode == PCRE16_MODE) \
+    r = pcre2_serialize_decode_16((pcre2_code_16 **)a,b,c,G(d,16)); \
+  else \
+    r = pcre2_serialize_decode_32((pcre2_code_32 **)a,b,c,G(d,32))
+
+#define PCRE2_SERIALIZE_ENCODE(r,a,b,c,d,e) \
+  if (test_mode == PCRE8_MODE) \
+    r = pcre2_serialize_encode_8((const pcre2_code_8 **)a,b,c,d,G(e,8)); \
+  else if (test_mode == PCRE16_MODE) \
+    r = pcre2_serialize_encode_16((const pcre2_code_16 **)a,b,c,d,G(e,16)); \
+  else \
+    r = pcre2_serialize_encode_32((const pcre2_code_32 **)a,b,c,d,G(e,32))
+
+#define PCRE2_SERIALIZE_FREE(a) \
+  if (test_mode == PCRE8_MODE) \
+    pcre2_serialize_free_8(a); \
+  else if (test_mode == PCRE16_MODE) \
+    pcre2_serialize_free_16(a); \
+  else \
+    pcre2_serialize_free_32(a)
+
+#define PCRE2_SERIALIZE_GET_NUMBER_OF_CODES(r,a) \
+  if (test_mode == PCRE8_MODE) \
+    r = pcre2_serialize_get_number_of_codes_8(a); \
+  else if (test_mode == PCRE16_MODE) \
+    r = pcre2_serialize_get_number_of_codes_16(a); \
+  else \
+    r = pcre2_serialize_get_number_of_codes_32(a); \
+
 #define PCRE2_SET_CALLOUT(a,b,c) \
   if (test_mode == PCRE8_MODE) \
     pcre2_set_callout_8(G(a,8),(int (*)(pcre2_callout_block_8 *, void *))b,c); \
@@ -1297,11 +1370,35 @@ the three different cases. */
     a = G(pcre2_pattern_info_,BITTWO)(G(b,BITTWO),c,d)
 
 #define PCRE2_PRINTINT(a) \
-  if (test_mode == G(G(PCRE,BITONE),_MODE)) \
+ if (test_mode == G(G(PCRE,BITONE),_MODE)) \
     G(pcre2_printint_,BITONE)(G(compiled_code,BITONE),outfile,a); \
   else \
     G(pcre2_printint_,BITTWO)(G(compiled_code,BITTWO),outfile,a)
 
+#define PCRE2_SERIALIZE_DECODE(r,a,b,c,d) \
+ if (test_mode == G(G(PCRE,BITONE),_MODE)) \
+    r = G(pcre2_serialize_decode_,BITONE)((G(pcre2_code_,BITONE) **)a,b,c,G(d,BITONE)); \
+  else \
+    r = G(pcre2_serialize_decode_,BITTWO)((G(pcre2_code_,BITTWO) **)a,b,c,G(d,BITTWO))
+
+#define PCRE2_SERIALIZE_ENCODE(r,a,b,c,d,e) \
+ if (test_mode == G(G(PCRE,BITONE),_MODE)) \
+    r = G(pcre2_serialize_encode_,BITONE)((G(const pcre2_code_,BITONE) **)a,b,c,d,G(e,BITONE)); \
+  else \
+    r = G(pcre2_serialize_encode_,BITTWO)((G(const pcre2_code_,BITTWO) **)a,b,c,d,G(e,BITTWO))
+
+#define PCRE2_SERIALIZE_FREE(a) \
+ if (test_mode == G(G(PCRE,BITONE),_MODE)) \
+    G(pcre2_serialize_free_,BITONE)(a); \
+  else \
+    G(pcre2_serialize_free_,BITTWO)(a)
+
+#define PCRE2_SERIALIZE_GET_NUMBER_OF_CODES(r,a) \
+ if (test_mode == G(G(PCRE,BITONE),_MODE)) \
+    r = G(pcre2_serialize_get_number_of_codes_,BITONE)(a); \
+  else \
+    r = G(pcre2_serialize_get_number_of_codes_,BITTWO)(a)
+
 #define PCRE2_SET_CALLOUT(a,b,c) \
   if (test_mode == G(G(PCRE,BITONE),_MODE)) \
     G(pcre2_set_callout_,BITONE)(G(a,BITONE), \
@@ -1510,6 +1607,13 @@ the three different cases. */
 #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_SERIALIZE_DECODE(r,a,b,c,d) \
+  r = pcre2_serialize_decode_8((pcre2_code_8 **)a,b,c,G(d,8))
+#define PCRE2_SERIALIZE_ENCODE(r,a,b,c,d,e) \
+  r = pcre2_serialize_encode_8((const pcre2_code_8 **)a,b,c,d,G(e,8))
+#define PCRE2_SERIALIZE_FREE(a) pcre2_serialize_free_8(a)
+#define PCRE2_SERIALIZE_GET_NUMBER_OF_CODES(r,a) \
+  r = pcre2_serialize_get_number_of_codes_8(a)
 #define PCRE2_SET_CALLOUT(a,b,c) \
   pcre2_set_callout_8(G(a,8),(int (*)(pcre2_callout_block_8 *, void *))b,c)
 #define PCRE2_SET_CHARACTER_TABLES(a,b) pcre2_set_character_tables_8(G(a,8),b)
@@ -1591,6 +1695,13 @@ the three different cases. */
 #define PCRE2_MATCH_DATA_FREE(a) pcre2_match_data_free_16(G(a,16))
 #define PCRE2_PATTERN_INFO(a,b,c,d) a = pcre2_pattern_info_16(G(b,16),c,d)
 #define PCRE2_PRINTINT(a) pcre2_printint_16(compiled_code16,outfile,a)
+#define PCRE2_SERIALIZE_DECODE(r,a,b,c,d) \
+  r = pcre2_serialize_decode_16((pcre2_code_16 **)a,b,c,G(d,16))
+#define PCRE2_SERIALIZE_ENCODE(r,a,b,c,d,e) \
+  r = pcre2_serialize_encode_16((const pcre2_code_16 **)a,b,c,d,G(e,16))
+#define PCRE2_SERIALIZE_FREE(a) pcre2_serialize_free_16(a)
+#define PCRE2_SERIALIZE_GET_NUMBER_OF_CODES(r,a) \
+  r = pcre2_serialize_get_number_of_codes_16(a)
 #define PCRE2_SET_CALLOUT(a,b,c) \
   pcre2_set_callout_16(G(a,16),(int (*)(pcre2_callout_block_16 *, void *))b,c);
 #define PCRE2_SET_CHARACTER_TABLES(a,b) pcre2_set_character_tables_16(G(a,16),b)
@@ -1672,6 +1783,13 @@ the three different cases. */
 #define PCRE2_MATCH_DATA_FREE(a) pcre2_match_data_free_32(G(a,32))
 #define PCRE2_PATTERN_INFO(a,b,c,d) a = pcre2_pattern_info_32(G(b,32),c,d)
 #define PCRE2_PRINTINT(a) pcre2_printint_32(compiled_code32,outfile,a)
+#define PCRE2_SERIALIZE_DECODE(r,a,b,c,d) \
+  r = pcre2_serialize_decode_32((pcre2_code_32 **)a,b,c,G(d,32))
+#define PCRE2_SERIALIZE_ENCODE(r,a,b,c,d,e) \
+  r = pcre2_serialize_encode_32((const pcre2_code_32 **)a,b,c,d,G(e,32))
+#define PCRE2_SERIALIZE_FREE(a) pcre2_serialize_free_32(a)
+#define PCRE2_SERIALIZE_GET_NUMBER_OF_CODES(r,a) \
+  r = pcre2_serialize_get_number_of_codes_32(a)
 #define PCRE2_SET_CALLOUT(a,b,c) \
   pcre2_set_callout_32(G(a,32),(int (*)(pcre2_callout_block_32 *, void *))b,c);
 #define PCRE2_SET_CHARACTER_TABLES(a,b) pcre2_set_character_tables_32(G(a,32),b)
@@ -2792,6 +2910,7 @@ it is allowed here and find the field that is to be changed.
 Arguments:
   m          the modifier list entry
   ctx        CTX_PAT     => pattern context
+             CTX_POPPAT  => pattern context for popped pattern
              CTX_DEFPAT  => default pattern context
              CTX_DAT     => data context
              CTX_DEFDAT  => default data context
@@ -2837,8 +2956,8 @@ switch (m->which)
   if (dctl != NULL) field = dctl;
   break;
 
-  case MOD_PAT:  /* Pattern modifier */
+  case MOD_PAT:    /* Pattern modifier */
-  case MOD_PATP: /* Allowed for Perl test */
+  case MOD_PATP:   /* Allowed for Perl test */
   if (pctl != NULL) field = pctl;
   break;
 
@@ -2878,6 +2997,7 @@ modifiers that apply to contexts.
 Arguments:
   p          point to modifier string
   ctx        CTX_PAT     => pattern context
+             CTX_POPPAT  => pattern context for popped pattern
              CTX_DEFPAT  => default pattern context
              CTX_DAT     => data context
              CTX_DEFDAT  => default data context
@@ -2902,11 +3022,8 @@ for (;;)
   int index;
   char *endptr;
 
-  /* Skip white space and commas; after a comma we have passed the first
+  /* Skip white space and commas. */
-  item. */
 
-  while (isspace(*p)) p++;
-  if (*p == ',') first = FALSE;
   while (isspace(*p) || *p == ',') p++;
   if (*p == 0) break;
 
@@ -3163,6 +3280,17 @@ for (;;)
     }
 
   p = pp;
+  first = FALSE;
+
+  if (ctx == CTX_POPPAT &&
+     (pctl->options != 0 ||
+      pctl->tables_id != 0 ||
+      pctl->locale[0] != 0 ||
+      (pctl->control & NOTPOP_CONTROLS) != 0))
+    {
+    fprintf(outfile, "** '%s' is not valid here\n", m->name);
+    return FALSE;
+    }
   }
 
 return TRUE;
@@ -3246,7 +3374,7 @@ Returns:      nothing
 static void
 show_controls(uint32_t controls, const char *before)
 {
-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",
+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%s",
   before,
   ((controls & CTL_AFTERTEXT) != 0)? " aftertext" : "",
   ((controls & CTL_ALLAFTERTEXT) != 0)? " allaftertext" : "",
@@ -3268,6 +3396,7 @@ 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",
   ((controls & CTL_MARK) != 0)? " mark" : "",
   ((controls & CTL_MEMORY) != 0)? " memory" : "",
   ((controls & CTL_POSIX) != 0)? " posix" : "",
+  ((controls & CTL_PUSH) != 0)? " push" : "",
   ((controls & CTL_STARTCHAR) != 0)? " startchar" : "",
   ((controls & CTL_ZERO_TERMINATE) != 0)? " zero_terminate" : "");
 }
@@ -3347,6 +3476,40 @@ fprintf(outfile, "%s%s%s%s%s%s%s%s%s%s",
 
 
 
+/*************************************************
+*      Show memory usage info for a pattern      *
+*************************************************/
+
+static void
+show_memory_info(void)
+{
+uint32_t name_count, name_entry_size;
+size_t size, cblock_size;
+
+#ifdef SUPPORT_PCRE2_8
+if (test_mode == 8) cblock_size = sizeof(pcre2_real_code_8);
+#endif
+#ifdef SUPPORT_PCRE2_16
+if (test_mode == 16) cblock_size = sizeof(pcre2_real_code_16);
+#endif
+#ifdef SUPPORT_PCRE2_32
+if (test_mode == 32) cblock_size = sizeof(pcre2_real_code_32);
+#endif
+
+(void)pattern_info(PCRE2_INFO_SIZE, &size, FALSE);
+(void)pattern_info(PCRE2_INFO_NAMECOUNT, &name_count, FALSE);
+(void)pattern_info(PCRE2_INFO_NAMEENTRYSIZE, &name_entry_size, FALSE);
+fprintf(outfile, "Memory allocation (code space): %d\n",
+  (int)(size - name_count*name_entry_size*code_unit_size - cblock_size));
+if (pat_patctl.jit != 0)
+  {
+  (void)pattern_info(PCRE2_INFO_JITSIZE, &size, FALSE);
+  fprintf(outfile, "Memory allocation (JIT code): %d\n", (int)size);
+  }
+}
+
+
+
 /*************************************************
 *        Show information about a pattern        *
 *************************************************/
@@ -3624,12 +3787,79 @@ return PR_OK;
 
 
 
+/*************************************************
+*              Handle serialization error        *
+*************************************************/
+
+/* Print an error message after a serialization failure.
+
+Arguments:
+  rc         the error code
+  msg        an initial message for what failed
+
+Returns:     nothing
+*/
+
+static void
+serial_error(int rc, const char *msg)
+{
+fprintf(outfile, "%s failed: error %d: ", msg, rc);
+PCRE2_GET_ERROR_MESSAGE(rc, rc, pbuffer);
+PCHARSV(CASTVAR(void *, pbuffer), 0, rc, FALSE, outfile);
+fprintf(outfile, "\n");
+}
+
+
+
+/*************************************************
+*        Open file for save/load commands        *
+*************************************************/
+
+/* This function decodes the file name and opens the file.
+
+Arguments:
+  buffptr     point after the #command
+  mode        open mode
+  fptr        points to the FILE variable
+
+Returns:      PR_OK or PR_ABEND
+*/
+
+static int
+open_file(uint8_t *buffptr, const char *mode, FILE **fptr)
+{
+char *endf;
+char *filename = (char *)buffptr;
+while (isspace(*filename)) filename++;
+endf = filename + strlen8(filename);
+while (endf > filename && isspace(endf[-1])) endf--;
+
+if (endf == filename)
+  {
+  fprintf(outfile, "** File name expected after #save\n");
+  return PR_ABEND;
+  }
+
+*endf = 0;
+*fptr = fopen((const char *)filename, mode);
+if (*fptr == NULL)
+  {
+  fprintf(outfile, "** Failed to open '%s'\n", filename);
+  return PR_ABEND;
+  }
+
+return PR_OK;
+}
+
+
+
 /*************************************************
 *               Process command line             *
 *************************************************/
 
 /* This function is called for lines beginning with # and a character that is
-not ! or whitespace, when encountered between tests. The line is in buffer.
+not ! or whitespace, when encountered between tests, which means that there is
+no compiled pattern (compiled_code is NULL). The line is in buffer.
 
 Arguments:  none
 
@@ -3641,33 +3871,176 @@ Returns:    PR_OK     continue processing next line
 static int
 process_command(void)
 {
+FILE *f;
+PCRE2_SIZE serial_size;
+size_t i;
+int rc, cmd, cmdlen;
+const char *cmdname;
+uint8_t *argptr, *serial;
+
 if (restrict_for_perl_test)
   {
   fprintf(outfile, "** #-commands are not allowed after #perltest\n");
   return PR_ABEND;
   }
 
-if (strncmp((char *)buffer, "#forbid_utf", 11) == 0 && isspace(buffer[11]))
+cmd = CMD_UNKNOWN;
+cmdlen = 0;
+
+for (i = 0; i < cmdlistcount; i++)
   {
-  forbid_utf = PCRE2_NEVER_UTF|PCRE2_NEVER_UCP;
+  cmdname = cmdlist[i].name;
+  cmdlen = strlen(cmdname);
+  if (strncmp((char *)(buffer+1), cmdname, cmdlen) == 0 &&
+      isspace(buffer[cmdlen+1]))
+    {
+    cmd = cmdlist[i].value;
+    break;
+    }
   }
-else if (strncmp((char *)buffer, "#pattern", 8) == 0 && isspace(buffer[8]))
+
+argptr = buffer + cmdlen + 1;
+
+switch(cmd)
   {
-  (void)decode_modifiers(buffer + 8, CTX_DEFPAT, &def_patctl, NULL);
+  case CMD_UNKNOWN:
+  fprintf(outfile, "** Unknown command: %s", buffer);
+  break;
+
+  case CMD_FORBID_UTF:
+  forbid_utf = PCRE2_NEVER_UTF|PCRE2_NEVER_UCP;
+  break;
+
+  case CMD_PERLTEST:
+  restrict_for_perl_test = TRUE;
+  break;
+
+  /* Set default pattern modifiers */
+
+  case CMD_PATTERN:
+  (void)decode_modifiers(argptr, CTX_DEFPAT, &def_patctl, NULL);
   if (def_patctl.jit == 0 && (def_patctl.control & CTL_JITVERIFY) != 0)
     def_patctl.jit = 7;
-  }
+  break;
-else if (strncmp((char *)buffer, "#perltest", 9) == 0 && isspace(buffer[9]))
+
-  {
+  /* Set default subject modifiers */
-  restrict_for_perl_test = TRUE;
+
-  }
+  case CMD_SUBJECT:
-else if (strncmp((char *)buffer, "#subject", 8) == 0 && isspace(buffer[8]))
+  (void)decode_modifiers(argptr, CTX_DEFDAT, NULL, &def_datctl);
-  {
+  break;
-  (void)decode_modifiers(buffer + 8, CTX_DEFDAT, NULL, &def_datctl);
+
-  }
+  /* Pop a compiled pattern off the stack. Modifiers that do not affect the
-else
+  compiled pattern (e.g. to give information) are permitted. The default
-  {
+  pattern modifiers are ignored. */
-  fprintf(outfile, "** Unknown command: %s", buffer);
+
+  case CMD_POP:
+  if (patstacknext <= 0)
+    {
+    fprintf(outfile, "** Can't pop off an empty stack\n");
+    return PR_SKIP;
+    }
+  memset(&pat_patctl, 0, sizeof(patctl));   /* Completely unset */
+  if (!decode_modifiers(argptr, CTX_POPPAT, &pat_patctl, NULL))
+    return PR_SKIP;
+  SET(compiled_code, patstack[--patstacknext]);
+  if (pat_patctl.jit != 0)
+    {
+    PCRE2_JIT_COMPILE(compiled_code, pat_patctl.jit);
+    }
+  if ((pat_patctl.control & CTL_MEMORY) != 0) show_memory_info();
+  if ((pat_patctl.control & CTL_ANYINFO) != 0)
+    {
+    rc = show_pattern_info();
+    if (rc != PR_OK) return rc;
+    }
+  break;
+
+  /* Save the stack of compiled patterns to a file, then empty the stack. */
+
+  case CMD_SAVE:
+  if (patstacknext <= 0)
+    {
+    fprintf(outfile, "** No stacked patterns to save\n");
+    return PR_OK;
+    }
+
+  rc = open_file(argptr+1, OUTPUT_MODE, &f);
+  if (rc != PR_OK) return rc;
+
+  PCRE2_SERIALIZE_ENCODE(rc, patstack, patstacknext, &serial, &serial_size,
+    general_context);
+  if (rc < 0)
+    {
+    serial_error(rc, "Serialization");
+    break;
+    }
+
+  /* Write the length at the start of the file to make it straightforward to
+  get the right memory when re-loading. This saves having to read the file size
+  in different operating systems. To allow for different endianness (even
+  though reloading with the opposite endianness does not work), write the
+  length byte-by-byte. */
+
+  for (i = 0; i < 4; i++) fputc((serial_size >> (i*8)) & 255, f);
+  if (fwrite(serial, 1, serial_size, f) != serial_size)
+    {
+    fprintf(outfile, "** Wrong return from fwrite()\n");
+    return PR_ABEND;
+    }
+
+  fclose(f);
+  PCRE2_SERIALIZE_FREE(serial);
+  while(patstacknext > 0)
+    {
+    SET(compiled_code, patstack[--patstacknext]);
+    SUB1(pcre2_code_free, compiled_code);
+    }
+  SET(compiled_code, NULL);
+  break;
+
+  /* Load a set of compiled patterns from a file onto the stack */
+
+  case CMD_LOAD:
+  rc = open_file(argptr+1, INPUT_MODE, &f);
+  if (rc != PR_OK) return rc;
+
+  serial_size = 0;
+  for (i = 0; i < 4; i++) serial_size |= fgetc(f) << (i*8);
+
+  serial = malloc(serial_size);
+  if (serial == NULL)
+    {
+    fprintf(outfile, "** Failed to get memory (size %ld) for #load\n",
+      serial_size);
+    return PR_ABEND;
+    }
+
+  if (fread(serial, 1, serial_size, f) != serial_size)
+    {
+    fprintf(outfile, "** Wrong return from fread()\n");
+    return PR_ABEND;
+    }
+  fclose(f);
+
+  PCRE2_SERIALIZE_GET_NUMBER_OF_CODES(rc, serial);
+  if (rc < 0) serial_error(rc, "Get number of codes"); else
+    {
+    if (rc + patstacknext > PATSTACKSIZE)
+      {
+      fprintf(outfile, "** Not enough space on pattern stack for %d pattern%s\n",
+        rc, (rc == 1)? "" : "s");
+      rc = PATSTACKSIZE - patstacknext;
+      fprintf(outfile, "** Decoding %d pattern%s\n", rc,
+        (rc == 1)? "" : "s");
+      }
+    PCRE2_SERIALIZE_DECODE(rc, patstack + patstacknext, rc, serial,
+      general_context);
+    if (rc < 0) serial_error(rc, "Deserialization");
+      else patstacknext += rc;
+    }
+
+  free(serial);
+  break;
   }
 
 return PR_OK;
@@ -3750,6 +4123,14 @@ if (pat_patctl.jit == 0 &&
     (pat_patctl.control & (CTL_JITVERIFY|CTL_JITFAST)) != 0)
   pat_patctl.jit = 7;
 
+/* POSIX and 'push' do not play together. */
+
+if ((pat_patctl.control & (CTL_POSIX|CTL_PUSH)) == (CTL_POSIX|CTL_PUSH))
+  {
+  fprintf(outfile, "** The POSIX interface is incompatible with 'push'\n");
+  return PR_ABEND;
+  }
+
 /* Now copy the pattern to pbuffer8 for use in 8-bit testing and for reflecting
 in callouts. Convert to binary if required. */
 
@@ -3897,8 +4278,31 @@ if ((pat_patctl.control & CTL_POSIX) != 0)
 #endif  /* SUPPORT_PCRE2_8 */
   }
 
-/* Handle compiling via the native interface, converting the input in non-8-bit
+/* Handle compiling via the native interface. Controls that act later are
-modes. */
+ignored with "push". Replacements are locked out. */
+
+if ((pat_patctl.control & CTL_PUSH) != 0)
+  {
+  if (pat_patctl.replacement[0] != 0)
+    {
+    fprintf(outfile, "** Replacement text is not supported with 'push'.\n");
+    return PR_OK;
+    }
+  if ((pat_patctl.control & ~PUSH_SUPPORTED_COMPILE_CONTROLS) != 0)
+    {
+    show_controls(pat_patctl.control & ~PUSH_SUPPORTED_COMPILE_CONTROLS,
+      "** Ignored when compiled pattern is stacked with 'push':");
+    fprintf(outfile, "\n");
+    }
+  if ((pat_patctl.control & PUSH_COMPILE_ONLY_CONTROLS) != 0)
+    {
+    show_controls(pat_patctl.control & PUSH_COMPILE_ONLY_CONTROLS,
+      "** Applies only to compile when pattern is stacked with 'push':");
+    fprintf(outfile, "\n");
+    }
+  }
+
+/* Convert the input in non-8-bit modes. */
 
 #ifdef SUPPORT_PCRE2_8
 if (test_mode == PCRE8_MODE) errorcode = 0;
@@ -4017,39 +4421,27 @@ if (pat_patctl.jit != 0)
 
 /* Output code size and other information if requested. */
 
-if ((pat_patctl.control & CTL_MEMORY) != 0)
+if ((pat_patctl.control & CTL_MEMORY) != 0) show_memory_info();
-  {
-  uint32_t name_count, name_entry_size;
-  size_t size, cblock_size;
-
-#ifdef SUPPORT_PCRE2_8
-  if (test_mode == 8) cblock_size = sizeof(pcre2_real_code_8);
-#endif
-#ifdef SUPPORT_PCRE2_16
-  if (test_mode == 16) cblock_size = sizeof(pcre2_real_code_16);
-#endif
-#ifdef SUPPORT_PCRE2_32
-  if (test_mode == 32) cblock_size = sizeof(pcre2_real_code_32);
-#endif
-
-  (void)pattern_info(PCRE2_INFO_SIZE, &size, FALSE);
-  (void)pattern_info(PCRE2_INFO_NAMECOUNT, &name_count, FALSE);
-  (void)pattern_info(PCRE2_INFO_NAMEENTRYSIZE, &name_entry_size, FALSE);
-  fprintf(outfile, "Memory allocation (code space): %d\n",
-    (int)(size - name_count*name_entry_size*code_unit_size - cblock_size));
-  if (pat_patctl.jit != 0)
-    {
-    (void)pattern_info(PCRE2_INFO_JITSIZE, &size, FALSE);
-    fprintf(outfile, "Memory allocation (JIT code): %d\n", (int)size);
-    }
-  }
-
 if ((pat_patctl.control & CTL_ANYINFO) != 0)
   {
   int rc = show_pattern_info();
   if (rc != PR_OK) return rc;
   }
 
+/* The "push" control requests that the compiled pattern be remembered on a
+stack. This is mainly for testing the serialization functionality. */
+
+if ((pat_patctl.control & CTL_PUSH) != 0)
+  {
+  if (patstacknext >= PATSTACKSIZE)
+    {
+    fprintf(outfile, "** Too many pushed patterns (max %d)\n", PATSTACKSIZE);
+    return PR_ABEND;
+    }
+  patstack[patstacknext++] = PTR(compiled_code);
+  SET(compiled_code, NULL);
+  }
+
 return PR_OK;
 }
 
@@ -6253,7 +6645,7 @@ if (argc > 1 && strcmp(argv[op], "-") != 0)
   infile = fopen(argv[op], INPUT_MODE);
   if (infile == NULL)
     {
-    printf("** Failed to open %s\n", argv[op]);
+    printf("** Failed to open '%s'\n", argv[op]);
     yield = 1;
     goto EXIT;
     }
@@ -6264,7 +6656,7 @@ if (argc > 2)
   outfile = fopen(argv[op+1], OUTPUT_MODE);
   if (outfile == NULL)
     {
-    printf("** Failed to open %s\n", argv[op+1]);
+    printf("** Failed to open '%s'\n", argv[op+1]);
     yield = 1;
     goto EXIT;
     }
@@ -6399,6 +6791,12 @@ free((void *)locale_tables);
 PCRE2_MATCH_DATA_FREE(match_data);
 SUB1(pcre2_code_free, compiled_code);
 
+while(patstacknext-- > 0)
+  {
+  SET(compiled_code, patstack[patstacknext]);
+  SUB1(pcre2_code_free, compiled_code);
+  }
+
 PCRE2_JIT_FREE_UNUSED_MEMORY(general_context);
 if (jit_stack != NULL)
   {

testdata/testinput15

@@ -6,4 +6,4 @@
 
 /a*/I
 
-# End of testinput14
+# End of testinput15

testdata/testinput16

@@ -161,7 +161,7 @@
 # match to happen via the interpreter, but for fast JIT invalid options are
 # ignored, so an unanchored match happens.
 
-/abcd/jit
+/abcd/
     abcd\=anchored
     fail abcd\=anchored 
     
@@ -169,4 +169,21 @@
     abcd\=anchored 
     succeed abcd\=anchored 
     
+# Push/pop does not lose the JIT information, though jitverify applies only to
+# compilation, but serializing (save/load) discards JIT data completely.
+
+/^abc\Kdef/info,push
+#pop jitverify
+    abcdef
+
+/^abc\Kdef/info,push
+#save testsaved1
+#load testsaved1
+#pop jitverify
+    abcdef
+    
+#load testsaved1
+#pop jit,jitverify
+    abcdef
+    
 # End of testinput16

testdata/testinput19

@@ -0,0 +1,62 @@
+# This set of tests exercises the serialization/deserialization functions in
+# the library. It does not use UTF or JIT.
+
+#forbid_utf
+
+# Compile several patterns, push them onto the stack, and then write them
+# all to a file.
+
+#pattern push
+
+/(?<NAME>(?&NAME_PAT))\s+(?<ADDR>(?&ADDRESS_PAT))
+  (?(DEFINE)
+  (?<NAME_PAT>[a-z]+)
+  (?<ADDRESS_PAT>\d+)
+  )/x
+/^(?:((.)(?1)\2|)|((.)(?3)\4|.))$/i
+
+#save testsaved1
+
+# Do it again for some more patterns.
+
+/(*MARK:A)(*SKIP:B)(C|X)/mark
+/(?:(?<n>foo)|(?<n>bar))\k<n>/dupnames
+
+#save testsaved2
+#pattern -push
+
+# Reload the patterns, then pop them one by one and check them.
+
+#load testsaved1
+#load testsaved2
+
+#pop info
+    foofoo             
+    barbar
+    
+#pop mark
+    C
+    D 
+    
+#pop
+    AmanaplanacanalPanama   
+
+#pop info
+    metcalfe 33
+    
+# Check for an error when different tables are used.
+
+/abc/push,tables=1
+/xyz/push,tables=2
+#save testsaved1
+
+#pop
+    xyz
+
+#pop
+    abc
+
+#pop should give an error
+    pqr
+
+# End of testinput19 

testdata/testoutput15

@@ -14,4 +14,4 @@ Capturing subpattern count = 0
 May match empty string
 Subject length lower bound = 0
 
-# End of testinput14
+# End of testinput15

testdata/testoutput16

@@ -310,7 +310,7 @@ Failed: error -46: JIT stack limit reached
 # match to happen via the interpreter, but for fast JIT invalid options are
 # ignored, so an unanchored match happens.
 
-/abcd/jit
+/abcd/
     abcd\=anchored
  0: abcd
     fail abcd\=anchored 
@@ -322,4 +322,36 @@ No match
     succeed abcd\=anchored 
  0: abcd (JIT)
     
+# Push/pop does not lose the JIT information, though jitverify applies only to
+# compilation, but serializing (save/load) discards JIT data completely.
+
+/^abc\Kdef/info,push
+** Applied only to compile when pattern is stacked with 'push': jitverify
+Capturing subpattern count = 0
+Compile options: <none>
+Overall options: anchored
+Subject length lower bound = 6
+JIT compilation was successful
+#pop jitverify
+    abcdef
+ 0: def (JIT)
+
+/^abc\Kdef/info,push
+** Applied only to compile when pattern is stacked with 'push': jitverify
+Capturing subpattern count = 0
+Compile options: <none>
+Overall options: anchored
+Subject length lower bound = 6
+JIT compilation was successful
+#save testsaved1
+#load testsaved1
+#pop jitverify
+    abcdef
+ 0: def
+    
+#load testsaved1
+#pop jit,jitverify
+    abcdef
+ 0: def (JIT)
+    
 # End of testinput16

testdata/testoutput19

@@ -0,0 +1,100 @@
+# This set of tests exercises the serialization/deserialization functions in
+# the library. It does not use UTF or JIT.
+
+#forbid_utf
+
+# Compile several patterns, push them onto the stack, and then write them
+# all to a file.
+
+#pattern push
+
+/(?<NAME>(?&NAME_PAT))\s+(?<ADDR>(?&ADDRESS_PAT))
+  (?(DEFINE)
+  (?<NAME_PAT>[a-z]+)
+  (?<ADDRESS_PAT>\d+)
+  )/x
+/^(?:((.)(?1)\2|)|((.)(?3)\4|.))$/i
+
+#save testsaved1
+
+# Do it again for some more patterns.
+
+/(*MARK:A)(*SKIP:B)(C|X)/mark
+** Ignored when compiled pattern is stacked with 'push': mark
+/(?:(?<n>foo)|(?<n>bar))\k<n>/dupnames
+
+#save testsaved2
+#pattern -push
+
+# Reload the patterns, then pop them one by one and check them.
+
+#load testsaved1
+#load testsaved2
+
+#pop info
+Capturing subpattern count = 2
+Max back reference = 2
+Named capturing subpatterns:
+  n   1
+  n   2
+Options: dupnames
+Starting code units: b f 
+Subject length lower bound = 6
+    foofoo             
+ 0: foofoo
+ 1: foo
+    barbar
+ 0: barbar
+ 1: <unset>
+ 2: bar
+    
+#pop mark
+    C
+ 0: C
+ 1: C
+MK: A
+    D 
+No match, mark = A
+    
+#pop
+    AmanaplanacanalPanama   
+ 0: AmanaplanacanalPanama
+ 1: <unset>
+ 2: <unset>
+ 3: AmanaplanacanalPanama
+ 4: A
+
+#pop info
+Capturing subpattern count = 4
+Named capturing subpatterns:
+  ADDR          2
+  ADDRESS_PAT   4
+  NAME          1
+  NAME_PAT      3
+Options: extended
+Subject length lower bound = 3
+    metcalfe 33
+ 0: metcalfe 33
+ 1: metcalfe
+ 2: 33
+    
+# Check for an error when different tables are used.
+
+/abc/push,tables=1
+/xyz/push,tables=2
+#save testsaved1
+Serialization failed: error -30: patterns do not all use the same character tables
+
+#pop
+    xyz
+ 0: xyz
+
+#pop
+    abc
+ 0: abc
+
+#pop should give an error
+** Can't pop off an empty stack
+    pqr
+
+# End of testinput19