Implement PCRE2_EXTRA_MATCH_LINE and PCRE2_EXTRA_MATCH_WORD for the benefit

of pcre2grep.

ChangeLog

@@ -189,6 +189,9 @@ pattern lines.
 
 41. Implement PCRE2_LITERAL and use it to support REG_NOSPEC.
 
+42. Implement PCRE2_EXTRA_MATCH_LINE and PCRE2_EXTRA_MATCH_WORD for the benefit 
+of pcre2grep.
+
 
 Version 10.23 14-February-2017
 ------------------------------

doc/pcre2api.3

@@ -1,4 +1,4 @@
-.TH PCRE2API 3 "15 June 2017" "PCRE2 10.30"
+.TH PCRE2API 3 "16 June 2017" "PCRE2 10.30"
 .SH NAME
 PCRE2 - Perl-compatible regular expressions (revised API)
 .sp
@@ -1400,10 +1400,12 @@ If this option is set, all meta-characters in the pattern are disabled, and it
 is treated as a literal string. Matching literal strings with a regular
 expression engine is not the most efficient way of doing it. If you are doing a
 lot of literal matching and are worried about efficiency, you should consider
-using other approaches. The only other options that are allowed with 
+using other approaches. The only other main options that are allowed with
 PCRE2_LITERAL are: PCRE2_ANCHORED, PCRE2_ENDANCHORED, PCRE2_AUTO_CALLOUT,
 PCRE2_CASELESS, PCRE2_FIRSTLINE, PCRE2_NO_START_OPTIMIZE, PCRE2_NO_UTF_CHECK,
-PCRE2_UTF, and PCRE2_USE_OFFSET_LIMIT. Any other options cause an error.
+PCRE2_UTF, and PCRE2_USE_OFFSET_LIMIT. The extra options PCRE2_EXTRA_MATCH_LINE
+and PCRE2_EXTRA_MATCH_WORD are also supported. Any other options cause an
+error.
 .sp
   PCRE2_MATCH_UNSET_BACKREF
 .sp
@@ -1689,6 +1691,24 @@ treated as single-character escapes. For example, \ej is a literal "j" and
 \ex{2z} is treated as the literal string "x{2z}". Setting this option means
 that typos in patterns may go undetected and have unexpected results. This is a
 dangerous option. Use with care.
+.sp
+  PCRE2_EXTRA_MATCH_LINE
+.sp
+This option is provided for use by the \fB-x\fP option of \fBpcre2grep\fP. It
+causes the pattern only to match complete lines. This is achieved by 
+automatically inserting the code for "^(?:" at the start of the compiled
+pattern and ")$" at the end. Thus, when PCRE2_MULTILINE is set, the matched
+line may be in the middle of the subject string. This option can be used with
+PCRE2_LITERAL.
+.sp
+  PCRE2_EXTRA_MATCH_WORD
+.sp
+This option is provided for use by the \fB-w\fP option of \fBpcre2grep\fP. It
+causes the pattern only to match strings that have a word boundary at the start
+and the end. This is achieved by automatically inserting the code for "\eb(?:"
+at the start of the compiled pattern and ")\eb" at the end. The option may be
+used with PCRE2_LITERAL. However, it is ignored if PCRE2_EXTRA_MATCH_LINE is
+also set.
 .
 .
 .SH "COMPILATION ERROR CODES"
@@ -3519,6 +3539,6 @@ Cambridge, England.
 .rs
 .sp
 .nf
-Last updated: 15 June 2017
+Last updated: 16 June 2017
 Copyright (c) 1997-2017 University of Cambridge.
 .fi

doc/pcre2test.1

@@ -1,4 +1,4 @@
-.TH PCRE2TEST 1 "15 June 2017" "PCRE 10.30"
+.TH PCRE2TEST 1 "16 June 2017" "PCRE 10.30"
 .SH NAME
 pcre2test - a program for testing Perl-compatible regular expressions.
 .SH SYNOPSIS
@@ -556,7 +556,9 @@ for a description of the effects of these options.
   /xx extended_more             set PCRE2_EXTENDED_MORE 
       firstline                 set PCRE2_FIRSTLINE
       literal                   set PCRE2_LITERAL 
+      match_line                set PCRE2_EXTRA_MATCH_LINE 
       match_unset_backref       set PCRE2_MATCH_UNSET_BACKREF
+      match_word                set PCRE2_EXTRA_MATCH_WORD 
   /m  multiline                 set PCRE2_MULTILINE
       never_backslash_c         set PCRE2_NEVER_BACKSLASH_C
       never_ucp                 set PCRE2_NEVER_UCP
@@ -1835,6 +1837,6 @@ Cambridge, England.
 .rs
 .sp
 .nf
-Last updated: 15 June 2017
+Last updated: 16 June 2017
 Copyright (c) 1997-2017 University of Cambridge.
 .fi

src/pcre2.h

@@ -142,8 +142,10 @@ D   is inspected during pcre2_dfa_match() execution
 
 /* An additional compile options word is available in the compile context. */
 
-#define PCRE2_EXTRA_ALLOW_SURROGATE_ESCAPES  0x00000001u  /* C       */
+#define PCRE2_EXTRA_ALLOW_SURROGATE_ESCAPES  0x00000001u  /* C */
-#define PCRE2_EXTRA_BAD_ESCAPE_IS_LITERAL    0x00000002u  /* C       */
+#define PCRE2_EXTRA_BAD_ESCAPE_IS_LITERAL    0x00000002u  /* C */
+#define PCRE2_EXTRA_MATCH_WORD               0x00000004u  /* C */
+#define PCRE2_EXTRA_MATCH_LINE               0x00000008u  /* C */
 
 /* These are for pcre2_jit_compile(). */
 

src/pcre2.h.in

@@ -142,8 +142,10 @@ D   is inspected during pcre2_dfa_match() execution
 
 /* An additional compile options word is available in the compile context. */
 
-#define PCRE2_EXTRA_ALLOW_SURROGATE_ESCAPES  0x00000001u  /* C       */
+#define PCRE2_EXTRA_ALLOW_SURROGATE_ESCAPES  0x00000001u  /* C */
-#define PCRE2_EXTRA_BAD_ESCAPE_IS_LITERAL    0x00000002u  /* C       */
+#define PCRE2_EXTRA_BAD_ESCAPE_IS_LITERAL    0x00000002u  /* C */
+#define PCRE2_EXTRA_MATCH_WORD               0x00000004u  /* C */
+#define PCRE2_EXTRA_MATCH_LINE               0x00000008u  /* C */
 
 /* These are for pcre2_jit_compile(). */
 

src/pcre2_compile.c

@@ -690,24 +690,30 @@ static int posix_substitutes[] = {
 #define POSIX_SUBSIZE (sizeof(posix_substitutes) / (2*sizeof(uint32_t)))
 #endif  /* SUPPORT_UNICODE */
 
-/* Masks for checking option settings. */
+/* Masks for checking option settings. When PCRE2_LITERAL is set, only a subset
-
+are allowed. */
-#define PUBLIC_COMPILE_OPTIONS \
-  (PCRE2_ANCHORED|PCRE2_ALLOW_EMPTY_CLASS|PCRE2_ALT_BSUX|PCRE2_ALT_CIRCUMFLEX| \
-   PCRE2_ALT_VERBNAMES|PCRE2_AUTO_CALLOUT|PCRE2_CASELESS|PCRE2_DOLLAR_ENDONLY| \
-   PCRE2_DOTALL|PCRE2_DUPNAMES|PCRE2_ENDANCHORED|PCRE2_EXTENDED| \
-   PCRE2_EXTENDED_MORE|PCRE2_FIRSTLINE|PCRE2_LITERAL| \
-   PCRE2_MATCH_UNSET_BACKREF|PCRE2_MULTILINE|PCRE2_NEVER_BACKSLASH_C| \
-   PCRE2_NEVER_UCP|PCRE2_NEVER_UTF|PCRE2_NO_AUTO_CAPTURE| \
-   PCRE2_NO_AUTO_POSSESS|PCRE2_NO_DOTSTAR_ANCHOR|PCRE2_NO_START_OPTIMIZE| \
-   PCRE2_NO_UTF_CHECK|PCRE2_UCP|PCRE2_UNGREEDY|PCRE2_USE_OFFSET_LIMIT| \
-   PCRE2_UTF)
 
 #define PUBLIC_LITERAL_COMPILE_OPTIONS \
   (PCRE2_ANCHORED|PCRE2_AUTO_CALLOUT|PCRE2_CASELESS|PCRE2_ENDANCHORED| \
    PCRE2_FIRSTLINE|PCRE2_LITERAL|PCRE2_NO_START_OPTIMIZE| \
    PCRE2_NO_UTF_CHECK|PCRE2_USE_OFFSET_LIMIT|PCRE2_UTF)
 
+#define PUBLIC_COMPILE_OPTIONS \
+  (PUBLIC_LITERAL_COMPILE_OPTIONS| \
+   PCRE2_ALLOW_EMPTY_CLASS|PCRE2_ALT_BSUX|PCRE2_ALT_CIRCUMFLEX| \
+   PCRE2_ALT_VERBNAMES|PCRE2_DOLLAR_ENDONLY|PCRE2_DOTALL|PCRE2_DUPNAMES| \
+   PCRE2_EXTENDED|PCRE2_EXTENDED_MORE|PCRE2_MATCH_UNSET_BACKREF| \
+   PCRE2_MULTILINE|PCRE2_NEVER_BACKSLASH_C|PCRE2_NEVER_UCP| \
+   PCRE2_NEVER_UTF|PCRE2_NO_AUTO_CAPTURE|PCRE2_NO_AUTO_POSSESS| \
+   PCRE2_NO_DOTSTAR_ANCHOR|PCRE2_UCP|PCRE2_UNGREEDY)
+
+#define PUBLIC_LITERAL_COMPILE_EXTRA_OPTIONS \
+   (PCRE2_EXTRA_MATCH_LINE|PCRE2_EXTRA_MATCH_WORD)
+
+#define PUBLIC_COMPILE_EXTRA_OPTIONS \
+   (PUBLIC_LITERAL_COMPILE_EXTRA_OPTIONS| \
+    PCRE2_EXTRA_ALLOW_SURROGATE_ESCAPES|PCRE2_EXTRA_BAD_ESCAPE_IS_LITERAL)
+
 /* Compile time error code numbers. They are given names so that they can more
 easily be tracked. When a new number is added, the tables called eint1 and
 eint2 in pcre2posix.c may need to be updated, and a new error text must be
@@ -2304,6 +2310,20 @@ PCRE2_SPTR verbnamestart = NULL;    /* Value avoids compiler warning */
 named_group *ng;
 nest_save *top_nest, *end_nests;
 
+/* Insert leading items for word and line matching (features provided for the
+benefit of pcre2grep). */
+
+if ((cb->cx->extra_options & PCRE2_EXTRA_MATCH_LINE) != 0)
+  {
+  *parsed_pattern++ = META_CIRCUMFLEX;
+  *parsed_pattern++ = META_NOCAPTURE;
+  }
+else if ((cb->cx->extra_options & PCRE2_EXTRA_MATCH_WORD) != 0)
+  {
+  *parsed_pattern++ = META_ESCAPE + ESC_b;
+  *parsed_pattern++ = META_NOCAPTURE;
+  }
+
 /* If the pattern is actually a literal string, process it separately to avoid
 cluttering up the main loop. */
 
@@ -2323,8 +2343,7 @@ if ((options & PCRE2_LITERAL) != 0)
         auto_callout, parsed_pattern, cb);
     PARSED_LITERAL(c, parsed_pattern);
     }
-  *parsed_pattern = META_END;
+  goto PARSED_END;
-  return 0;
   }
 
 /* Process a real regex which may contain meta-characters. */
@@ -4166,9 +4185,24 @@ if (inverbname && ptr >= ptrend)
 
 /* Manage callout for the final item */
 
+PARSED_END:
 parsed_pattern = manage_callouts(ptr, &previous_callout, auto_callout,
   parsed_pattern, cb);
 
+/* Insert trailing items for word and line matching (features provided for the
+benefit of pcre2grep). */
+
+if ((cb->cx->extra_options & PCRE2_EXTRA_MATCH_LINE) != 0)
+  {
+  *parsed_pattern++ = META_KET;
+  *parsed_pattern++ = META_DOLLAR;
+  }
+else if ((cb->cx->extra_options & PCRE2_EXTRA_MATCH_WORD) != 0)
+  {
+  *parsed_pattern++ = META_KET;
+  *parsed_pattern++ = META_ESCAPE + ESC_b;
+  }
+
 /* Terminate the parsed pattern, then return success if all groups are closed.
 Otherwise we have unclosed parentheses. */
 
@@ -4177,6 +4211,7 @@ if (parsed_pattern >= parsed_pattern_end)
   errorcode = ERR63;  /* Internal error (parsed pattern overflow) */
   goto FAILED;
   }
+
 *parsed_pattern = META_END;
 if (nest_depth == 0) return 0;
 
@@ -8984,26 +9019,28 @@ if (pattern == NULL)
   return NULL;
   }
 
+/* A NULL compile context means "use a default context" */
+
+if (ccontext == NULL)
+  ccontext = (pcre2_compile_context *)(&PRIV(default_compile_context));
+
 /* Check that all undefined public option bits are zero. */
 
-if ((options & ~PUBLIC_COMPILE_OPTIONS) != 0)
+if ((options & ~PUBLIC_COMPILE_OPTIONS) != 0 ||
+    (ccontext->extra_options & ~PUBLIC_COMPILE_EXTRA_OPTIONS) != 0)
   {
   *errorptr = ERR17;
   return NULL;
   }
 
 if ((options & PCRE2_LITERAL) != 0 &&
-    (options & ~PUBLIC_LITERAL_COMPILE_OPTIONS) != 0)
+    ((options & ~PUBLIC_LITERAL_COMPILE_OPTIONS) != 0 ||
+     (ccontext->extra_options & ~PUBLIC_LITERAL_COMPILE_EXTRA_OPTIONS) != 0))
   {
   *errorptr = ERR92;
   return NULL;
   }
 
-/* A NULL compile context means "use a default context" */
-
-if (ccontext == NULL)
-  ccontext = (pcre2_compile_context *)(&PRIV(default_compile_context));
-
 /* A zero-terminated pattern is indicated by the special length value
 PCRE2_ZERO_TERMINATED. Check for an overlong pattern. */
 
@@ -9262,10 +9299,10 @@ and comments removed (amongst other things).
 
 In all but one case, when PCRE2_AUTO_CALLOUT is not set, the number of unsigned
 32-bit ints in the parsed pattern is bounded by the length of the pattern plus
-one (for the terminator). The exceptional case is when running in 32-bit,
+one (for the terminator) plus four if PCRE2_EXTRA_WORD or PCRE2_EXTRA_LINE is
-non-UTF mode, when literal characters greater than META_END (0x80000000) have
+set. The exceptional case is when running in 32-bit, non-UTF mode, when literal
-to be coded as two units. In this case, therefore, we scan the pattern to check
+characters greater than META_END (0x80000000) have to be coded as two units. In
-for such values. */
+this case, therefore, we scan the pattern to check for such values. */
 
 #if PCRE2_CODE_UNIT_WIDTH == 32
 if (!utf)
@@ -9282,6 +9319,11 @@ many smaller patterns the vector on the stack (which was set up above) can be
 used. */
 
 parsed_size_needed = patlen - skipatstart + big32count;
+
+if ((ccontext->extra_options &
+     (PCRE2_EXTRA_MATCH_WORD|PCRE2_EXTRA_MATCH_LINE)) != 0)
+  parsed_size_needed += 4;
+
 if ((options & PCRE2_AUTO_CALLOUT) != 0)
   parsed_size_needed = (parsed_size_needed + 1) * 5;
 

src/pcre2test.c

@@ -638,7 +638,9 @@ static modstruct modlist[] = {
   { "locale",                     MOD_PAT,  MOD_STR, LOCALESIZE,                 PO(locale) },
   { "mark",                       MOD_PNDP, MOD_CTL, CTL_MARK,                   PO(control) },
   { "match_limit",                MOD_CTM,  MOD_INT, 0,                          MO(match_limit) },
+  { "match_line",                 MOD_CTC,  MOD_OPT, PCRE2_EXTRA_MATCH_LINE,     CO(extra_options) },
   { "match_unset_backref",        MOD_PAT,  MOD_OPT, PCRE2_MATCH_UNSET_BACKREF,  PO(options) },
+  { "match_word",                 MOD_CTC,  MOD_OPT, PCRE2_EXTRA_MATCH_WORD,     CO(extra_options) },
   { "max_pattern_length",         MOD_CTC,  MOD_SIZ, 0,                          CO(max_pattern_length) },
   { "memory",                     MOD_PD,   MOD_CTL, CTL_MEMORY,                 PD(control) },
   { "multiline",                  MOD_PATP, MOD_OPT, PCRE2_MULTILINE,            PO(options) },

testdata/testinput2

@@ -5305,7 +5305,7 @@ a)"xI
     X\na\\b(c
 
 /a\b?c/literal,use_offset_limit
-    XXXXa\\b?c\=offset_limit=5
+    XXXXa\\b?c\=offset_limit=4
 \= Expect no match
     XXXXa\\b?c\=offset_limit=3
 
@@ -5327,4 +5327,24 @@ a)"xI
 /(*CR)abc/literal
     (*CR)abc
 
+/cat|dog/match_word
+    the cat sat
+\= Expect no match   
+    caterpillar
+    snowcat
+    syndicate
+
+/(cat)|dog/match_line,literal
+    (cat)|dog
+\= Expect no match   
+    the cat sat
+    caterpillar
+    snowcat
+    syndicate
+
+/a whole line/match_line,multiline
+    Rhubarb \na whole line\n custard
+\= Expect no match
+    Not a whole line         
+
 # End of testinput2 

testdata/testoutput2

@@ -16033,7 +16033,7 @@ Failed: error 108 at offset 4: range out of order in character class
 No match
 
 /a\b?c/literal,use_offset_limit
-    XXXXa\\b?c\=offset_limit=5
+    XXXXa\\b?c\=offset_limit=4
  0: a\b?c
 \= Expect no match
     XXXXa\\b?c\=offset_limit=3
@@ -16064,7 +16064,8 @@ Failed: error 192 at offset 0: invalid option bits with PCRE2_LITERAL
  +1     ^^        \
  +2     ^ ^       b
  +3     ^  ^      (
- +4     ^   ^     
+ +4     ^   ^     c
+ +5     ^    ^    
  0: a\b(c
 
 /a\b(c/literal,auto_callout
@@ -16074,13 +16075,45 @@ Failed: error 192 at offset 0: invalid option bits with PCRE2_LITERAL
  +1     ^^        \
  +2     ^ ^       b
  +3     ^  ^      (
- +4     ^   ^     
+ +4     ^   ^     c
+ +5     ^    ^    
  0: a\b(c
 
 /(*CR)abc/literal
     (*CR)abc
  0: (*CR)abc
 
+/cat|dog/match_word
+    the cat sat
+ 0: cat
+\= Expect no match   
+    caterpillar
+No match
+    snowcat
+No match
+    syndicate
+No match
+
+/(cat)|dog/match_line,literal
+    (cat)|dog
+ 0: (cat)|dog
+\= Expect no match   
+    the cat sat
+No match
+    caterpillar
+No match
+    snowcat
+No match
+    syndicate
+No match
+
+/a whole line/match_line,multiline
+    Rhubarb \na whole line\n custard
+ 0: a whole line
+\= Expect no match
+    Not a whole line         
+No match
+
 # End of testinput2 
 Error -65: PCRE2_ERROR_BADDATA (unknown error number)
 Error -62: bad serialized data