From 4c501a0ab62a1978d4583a4a96c557c420aed22f Mon Sep 17 00:00:00 2001 From: Nuno Lopes Date: Sun, 13 Jan 2008 12:44:57 +0000 Subject: [PATCH] upgrade PCRE to version 7.5 --- NEWS | 2 +- ext/pcre/pcrelib/ChangeLog | 137 ++ ext/pcre/pcrelib/NEWS | 8 + ext/pcre/pcrelib/NON-UNIX-USE | 2 +- ext/pcre/pcrelib/README | 20 +- ext/pcre/pcrelib/config.h | 37 +- ext/pcre/pcrelib/pcre.h | 4 +- ext/pcre/pcrelib/pcre_compile.c | 173 +- ext/pcre/pcrelib/pcre_exec.c | 10 +- ext/pcre/pcrelib/pcre_internal.h | 15 +- ext/pcre/pcrelib/pcre_valid_utf8.c | 2 +- ext/pcre/pcrelib/pcregrep.c | 2104 ------------------------- ext/pcre/pcrelib/pcreposix.c | 4 +- ext/pcre/pcrelib/testdata/grepoutput | 15 + ext/pcre/pcrelib/testdata/testinput1 | 16 +- ext/pcre/pcrelib/testdata/testinput2 | 80 +- ext/pcre/pcrelib/testdata/testinput4 | 72 + ext/pcre/pcrelib/testdata/testinput5 | 8 + ext/pcre/pcrelib/testdata/testinput6 | 75 + ext/pcre/pcrelib/testdata/testoutput1 | 24 +- ext/pcre/pcrelib/testdata/testoutput2 | 161 +- ext/pcre/pcrelib/testdata/testoutput4 | 131 ++ ext/pcre/pcrelib/testdata/testoutput5 | 13 + ext/pcre/pcrelib/testdata/testoutput6 | 157 ++ ext/pcre/pcrelib/ucptable.h | 50 +- 25 files changed, 1066 insertions(+), 2254 deletions(-) delete mode 100644 ext/pcre/pcrelib/pcregrep.c diff --git a/NEWS b/NEWS index e6e4e5661c..0039722215 100644 --- a/NEWS +++ b/NEWS @@ -29,7 +29,7 @@ PHP NEWS invoking the date parser. (Scott) - Removed the experimental RPL (master/slave) functions from mysqli. (Andrey) -- Upgraded PCRE to version 7.4 (Nuno) +- Upgraded PCRE to version 7.5 (Nuno) - Improved php.ini handling: (Jani) . Added ".htaccess" style user-defined php.ini files support for CGI/FastCGI diff --git a/ext/pcre/pcrelib/ChangeLog b/ext/pcre/pcrelib/ChangeLog index cff73670ad..a6978962cd 100644 --- a/ext/pcre/pcrelib/ChangeLog +++ b/ext/pcre/pcrelib/ChangeLog @@ -1,6 +1,143 @@ ChangeLog for PCRE ------------------ +Version 7.5 10-Jan-08 +--------------------- + +1. Applied a patch from Craig: "This patch makes it possible to 'ignore' + values in parens when parsing an RE using the C++ wrapper." + +2. Negative specials like \S did not work in character classes in UTF-8 mode. + Characters greater than 255 were excluded from the class instead of being + included. + +3. The same bug as (2) above applied to negated POSIX classes such as + [:^space:]. + +4. PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it + defined or documented. It seems to have been a typo for PCRE_STATIC, so + I have changed it. + +5. The construct (?&) was not diagnosed as a syntax error (it referenced the + first named subpattern) and a construct such as (?&a) would reference the + first named subpattern whose name started with "a" (in other words, the + length check was missing). Both these problems are fixed. "Subpattern name + expected" is now given for (?&) (a zero-length name), and this patch also + makes it give the same error for \k'' (previously it complained that that + was a reference to a non-existent subpattern). + +6. The erroneous patterns (?+-a) and (?-+a) give different error messages; + this is right because (?- can be followed by option settings as well as by + digits. I have, however, made the messages clearer. + +7. Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns + than the number used in the conditional) now cause a compile-time error. + This is actually not compatible with Perl, which accepts such patterns, but + treats the conditional as always being FALSE (as PCRE used to), but it + seems to me that giving a diagnostic is better. + +8. Change "alphameric" to the more common word "alphanumeric" in comments + and messages. + +9. Fix two occurrences of "backslash" in comments that should have been + "backspace". + +10. Remove two redundant lines of code that can never be obeyed (their function + was moved elsewhere). + +11. The program that makes PCRE's Unicode character property table had a bug + which caused it to generate incorrect table entries for sequences of + characters that have the same character type, but are in different scripts. + It amalgamated them into a single range, with the script of the first of + them. In other words, some characters were in the wrong script. There were + thirteen such cases, affecting characters in the following ranges: + + U+002b0 - U+002c1 + U+0060c - U+0060d + U+0061e - U+00612 + U+0064b - U+0065e + U+0074d - U+0076d + U+01800 - U+01805 + U+01d00 - U+01d77 + U+01d9b - U+01dbf + U+0200b - U+0200f + U+030fc - U+030fe + U+03260 - U+0327f + U+0fb46 - U+0fbb1 + U+10450 - U+1049d + +12. The -o option (show only the matching part of a line) for pcregrep was not + compatible with GNU grep in that, if there was more than one match in a + line, it showed only the first of them. It now behaves in the same way as + GNU grep. + +13. If the -o and -v options were combined for pcregrep, it printed a blank + line for every non-matching line. GNU grep prints nothing, and pcregrep now + does the same. The return code can be used to tell if there were any + non-matching lines. + +14. Added --file-offsets and --line-offsets to pcregrep. + +15. The pattern (?=something)(?R) was not being diagnosed as a potentially + infinitely looping recursion. The bug was that positive lookaheads were not + being skipped when checking for a possible empty match (negative lookaheads + and both kinds of lookbehind were skipped). + +16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the + inclusion of to before rather than after the definition of + INVALID_FILE_ATTRIBUTES (patch from David Byron). + +17. Specifying a possessive quantifier with a specific limit for a Unicode + character property caused pcre_compile() to compile bad code, which led at + runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this + are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that + caused the error; without that there was no problem. + +18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2. + +19. Added --enable-pcretest-libreadline. + +20. In pcrecpp.cc, the variable 'count' was incremented twice in + RE::GlobalReplace(). As a result, the number of replacements returned was + double what it should be. I removed one of the increments, but Craig sent a + later patch that removed the other one (the right fix) and added unit tests + that check the return values (which was not done before). + +21. Several CMake things: + + (1) Arranged that, when cmake is used on Unix, the libraries end up with + the names libpcre and libpcreposix, not just pcre and pcreposix. + + (2) The above change means that pcretest and pcregrep are now correctly + linked with the newly-built libraries, not previously installed ones. + + (3) Added PCRE_SUPPORT_LIBREADLINE, PCRE_SUPPORT_LIBZ, PCRE_SUPPORT_LIBBZ2. + +22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.* + crashed when matching a string such as a\x{2029}b (note that \x{2029} is a + UTF-8 newline character). The key issue is that the pattern starts .*; + this means that the match must be either at the beginning, or after a + newline. The bug was in the code for advancing after a failed match and + checking that the new position followed a newline. It was not taking + account of UTF-8 characters correctly. + +23. PCRE was behaving differently from Perl in the way it recognized POSIX + character classes. PCRE was not treating the sequence [:...:] as a + character class unless the ... were all letters. Perl, however, seems to + allow any characters between [: and :], though of course it rejects as + unknown any "names" that contain non-letters, because all the known class + names consist only of letters. Thus, Perl gives an error for [[:1234:]], + for example, whereas PCRE did not - it did not recognize a POSIX character + class. This seemed a bit dangerous, so the code has been changed to be + closer to Perl. The behaviour is not identical to Perl, because PCRE will + diagnose an unknown class for, for example, [[:l\ower:]] where Perl will + treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where + Perl does, and where it didn't before. + +24. Rewrite so as to remove the single use of %n from pcregrep because in some + Windows environments %n is disabled by default. + + Version 7.4 21-Sep-07 --------------------- diff --git a/ext/pcre/pcrelib/NEWS b/ext/pcre/pcrelib/NEWS index 8ef40b03d3..185463db0f 100644 --- a/ext/pcre/pcrelib/NEWS +++ b/ext/pcre/pcrelib/NEWS @@ -1,6 +1,14 @@ News about PCRE releases ------------------------ +Release 7.5 10-Jan-08 +--------------------- + +This is mainly a bug-fix release. However the ability to link pcregrep with +libz or libbz2 and the ability to link pcretest with libreadline have been +added. Also the --line-offsets and --file-offsets options were added to +pcregrep. + Release 7.4 21-Sep-07 --------------------- diff --git a/ext/pcre/pcrelib/NON-UNIX-USE b/ext/pcre/pcrelib/NON-UNIX-USE index fe6cd02302..9239838e70 100644 --- a/ext/pcre/pcrelib/NON-UNIX-USE +++ b/ext/pcre/pcrelib/NON-UNIX-USE @@ -84,7 +84,7 @@ The following are generic comments about building the PCRE C library "by hand". ucptable.h (5) Also ensure that you have the following file, which is #included as source - when building a debugging version of PCRE and is also used by pcretest. + when building a debugging version of PCRE, and is also used by pcretest. pcre_printint.src diff --git a/ext/pcre/pcrelib/README b/ext/pcre/pcrelib/README index dabf243755..ad6ca4dbfd 100644 --- a/ext/pcre/pcrelib/README +++ b/ext/pcre/pcrelib/README @@ -258,6 +258,24 @@ library. You can read more about them in the pcrebuild man page. This automatically implies --enable-rebuild-chartables (see above). +. It is possible to compile pcregrep to use libz and/or libbz2, in order to + read .gz and .bz2 files (respectively), by specifying one or both of + + --enable-pcregrep-libz + --enable-pcregrep-libbz2 + + Of course, the relevant libraries must be installed on your system. + +. It is possible to compile pcretest so that it links with the libreadline + library, by specifying + + --enable-pcretest-libreadline + + If this is done, when pcretest's input is from a terminal, it reads it using + the readline() function. This provides line-editing and history facilities. + Note that libreadline is GPL-licenced, so if you distribute a binary of + pcretest linked in this way, there may be licensing issues. + The "configure" script builds the following files for the basic C library: . Makefile is the makefile that builds the library @@ -725,4 +743,4 @@ The distribution should contain the following files: Philip Hazel Email local part: ph10 Email domain: cam.ac.uk -Last updated: 21 September 2007 +Last updated: 18 December 2007 diff --git a/ext/pcre/pcrelib/config.h b/ext/pcre/pcrelib/config.h index 39f036958e..005e3f8b0c 100644 --- a/ext/pcre/pcrelib/config.h +++ b/ext/pcre/pcrelib/config.h @@ -51,6 +51,11 @@ them both to 0; an emulation function will be used. */ /* Define to 1 if you have the header file. */ /* #undef HAVE_BITS_TYPE_TRAITS_H */ +/* Define to 1 if you have the header file. */ +#ifndef HAVE_BZLIB_H +#define HAVE_BZLIB_H 1 +#endif + /* Define to 1 if you have the header file. */ #ifndef HAVE_DIRENT_H #define HAVE_DIRENT_H 1 @@ -86,6 +91,16 @@ them both to 0; an emulation function will be used. */ #define HAVE_MEMORY_H 1 #endif +/* Define to 1 if you have the header file. */ +#ifndef HAVE_READLINE_HISTORY_H +#define HAVE_READLINE_HISTORY_H 1 +#endif + +/* Define to 1 if you have the header file. */ +#ifndef HAVE_READLINE_READLINE_H +#define HAVE_READLINE_READLINE_H 1 +#endif + /* Define to 1 if you have the header file. */ #ifndef HAVE_STDINT_H #define HAVE_STDINT_H 1 @@ -152,6 +167,11 @@ them both to 0; an emulation function will be used. */ /* Define to 1 if you have the header file. */ /* #undef HAVE_WINDOWS_H */ +/* Define to 1 if you have the header file. */ +#ifndef HAVE_ZLIB_H +#define HAVE_ZLIB_H 1 +#endif + /* Define to 1 if you have the `_strtoi64' function. */ /* #undef HAVE__STRTOI64 */ @@ -231,13 +251,13 @@ them both to 0; an emulation function will be used. */ #define PACKAGE_NAME "PCRE" /* Define to the full name and version of this package. */ -#define PACKAGE_STRING "PCRE 7.4" +#define PACKAGE_STRING "PCRE 7.5" /* Define to the one symbol short name of this package. */ #define PACKAGE_TARNAME "pcre" /* Define to the version of this package. */ -#define PACKAGE_VERSION "7.4" +#define PACKAGE_VERSION "7.5" /* If you are compiling for a system other than a Unix-like system or @@ -271,6 +291,17 @@ them both to 0; an emulation function will be used. */ #define STDC_HEADERS 1 #endif +/* Define to allow pcregrep to be linked with libbz2, so that it is able to + handle .bz2 files. */ +/* #undef SUPPORT_LIBBZ2 */ + +/* Define to allow pcretest to be linked with libreadline. */ +/* #undef SUPPORT_LIBREADLINE */ + +/* Define to allow pcregrep to be linked with libz, so that it is able to + handle .gz files. */ +/* #undef SUPPORT_LIBZ */ + /* Define to enable support for Unicode properties */ /* #undef SUPPORT_UCP */ @@ -279,7 +310,7 @@ them both to 0; an emulation function will be used. */ /* Version number of package */ #ifndef VERSION -#define VERSION "7.4" +#define VERSION "7.5" #endif /* Define to empty if `const' does not conform to ANSI C. */ diff --git a/ext/pcre/pcrelib/pcre.h b/ext/pcre/pcrelib/pcre.h index 58a83c369a..a3cc3c7ef2 100644 --- a/ext/pcre/pcrelib/pcre.h +++ b/ext/pcre/pcrelib/pcre.h @@ -42,9 +42,9 @@ POSSIBILITY OF SUCH DAMAGE. /* The current PCRE version information. */ #define PCRE_MAJOR 7 -#define PCRE_MINOR 4 +#define PCRE_MINOR 5 #define PCRE_PRERELEASE -#define PCRE_DATE 2007-09-21 +#define PCRE_DATE 2008-01-10 /* When an application links to a PCRE DLL in Windows, the symbols that are imported have to be identified as such. When building PCRE, the appropriate diff --git a/ext/pcre/pcrelib/pcre_compile.c b/ext/pcre/pcrelib/pcre_compile.c index d9599d0415..ef33cfc388 100644 --- a/ext/pcre/pcrelib/pcre_compile.c +++ b/ext/pcre/pcrelib/pcre_compile.c @@ -239,7 +239,7 @@ static const char error_texts[] = /* 10 */ "operand of unlimited repeat could match the empty string\0" /** DEAD **/ "internal error: unexpected repeat\0" - "unrecognized character after (?\0" + "unrecognized character after (? or (?-\0" "POSIX named classes are supported only within a class\0" "missing )\0" /* 15 */ @@ -298,7 +298,9 @@ static const char error_texts[] = "(*VERB) with an argument is not supported\0" /* 60 */ "(*VERB) not recognized\0" - "number is too big"; + "number is too big\0" + "subpattern name expected\0" + "digit expected after (?+"; /* Table to identify digits and hex digits. This is used when compiling @@ -494,16 +496,16 @@ ptr--; /* Set pointer back to the last byte */ if (c == 0) *errorcodeptr = ERR1; -/* Non-alphamerics are literals. For digits or letters, do an initial lookup in -a table. A non-zero result is something that can be returned immediately. +/* Non-alphanumerics are literals. For digits or letters, do an initial lookup +in a table. A non-zero result is something that can be returned immediately. Otherwise further processing may be required. */ #ifndef EBCDIC /* ASCII coding */ -else if (c < '0' || c > 'z') {} /* Not alphameric */ +else if (c < '0' || c > 'z') {} /* Not alphanumeric */ else if ((i = escapes[c - '0']) != 0) c = i; #else /* EBCDIC coding */ -else if (c < 'a' || (ebcdic_chartab[c] & 0x0E) == 0) {} /* Not alphameric */ +else if (c < 'a' || (ebcdic_chartab[c] & 0x0E) == 0) {} /* Not alphanumeric */ else if ((i = escapes[c - 0x48]) != 0) c = i; #endif @@ -720,10 +722,10 @@ else break; /* PCRE_EXTRA enables extensions to Perl in the matter of escapes. Any - other alphameric following \ is an error if PCRE_EXTRA was set; otherwise, - for Perl compatibility, it is a literal. This code looks a bit odd, but - there used to be some cases other than the default, and there may be again - in future, so I haven't "optimized" it. */ + other alphanumeric following \ is an error if PCRE_EXTRA was set; + otherwise, for Perl compatibility, it is a literal. This code looks a bit + odd, but there used to be some cases other than the default, and there may + be again in future, so I haven't "optimized" it. */ default: if ((options & PCRE_EXTRA) != 0) switch(c) @@ -1504,8 +1506,9 @@ for (;;) can match the empty string or not. It is called from could_be_empty() below and from compile_branch() when checking for an unlimited repeat of a group that can match nothing. Note that first_significant_code() skips over -assertions. If we hit an unclosed bracket, we return "empty" - this means we've -struck an inner bracket whose current branch will already have been scanned. +backward and negative forward assertions when its final argument is TRUE. If we +hit an unclosed bracket, we return "empty" - this means we've struck an inner +bracket whose current branch will already have been scanned. Arguments: code points to start of search @@ -1527,6 +1530,16 @@ for (code = first_significant_code(code + _pcre_OP_lengths[*code], NULL, 0, TRUE c = *code; + /* Skip over forward assertions; the other assertions are skipped by + first_significant_code() with a TRUE final argument. */ + + if (c == OP_ASSERT) + { + do code += GET(code, 1); while (*code == OP_ALT); + c = *code; + continue; + } + /* Groups with zero repeats can of course be empty; skip them. */ if (c == OP_BRAZERO || c == OP_BRAMINZERO) @@ -1722,29 +1735,48 @@ return TRUE; *************************************************/ /* This function is called when the sequence "[:" or "[." or "[=" is -encountered in a character class. It checks whether this is followed by an -optional ^ and then a sequence of letters, terminated by a matching ":]" or -".]" or "=]". +encountered in a character class. It checks whether this is followed by a +sequence of characters terminated by a matching ":]" or ".]" or "=]". If we +reach an unescaped ']' without the special preceding character, return FALSE. + +Originally, this function only recognized a sequence of letters between the +terminators, but it seems that Perl recognizes any sequence of characters, +though of course unknown POSIX names are subsequently rejected. Perl gives an +"Unknown POSIX class" error for [:f\oo:] for example, where previously PCRE +didn't consider this to be a POSIX class. Likewise for [:1234:]. + +The problem in trying to be exactly like Perl is in the handling of escapes. We +have to be sure that [abc[:x\]pqr] is *not* treated as containing a POSIX +class, but [abc[:x\]pqr:]] is (so that an error can be generated). The code +below handles the special case of \], but does not try to do any other escape +processing. This makes it different from Perl for cases such as [:l\ower:] +where Perl recognizes it as the POSIX class "lower" but PCRE does not recognize +"l\ower". This is a lesser evil that not diagnosing bad classes when Perl does, +I think. -Argument: +Arguments: ptr pointer to the initial [ endptr where to return the end pointer - cd pointer to compile data Returns: TRUE or FALSE */ static BOOL -check_posix_syntax(const uschar *ptr, const uschar **endptr, compile_data *cd) +check_posix_syntax(const uschar *ptr, const uschar **endptr) { int terminator; /* Don't combine these lines; the Solaris cc */ terminator = *(++ptr); /* compiler warns about "non-constant" initializer. */ -if (*(++ptr) == '^') ptr++; -while ((cd->ctypes[*ptr] & ctype_letter) != 0) ptr++; -if (*ptr == terminator && ptr[1] == ']') +for (++ptr; *ptr != 0; ptr++) { - *endptr = ptr; - return TRUE; + if (*ptr == '\\' && ptr[1] == ']') ptr++; else + { + if (*ptr == ']') return FALSE; + if (*ptr == terminator && ptr[1] == ']') + { + *endptr = ptr; + return TRUE; + } + } } return FALSE; } @@ -2381,6 +2413,7 @@ req_caseopt = ((options & PCRE_CASELESS) != 0)? REQ_CASELESS : 0; for (;; ptr++) { BOOL negate_class; + BOOL should_flip_negation; BOOL possessive_quantifier; BOOL is_quantifier; BOOL is_recurse; @@ -2604,7 +2637,7 @@ for (;; ptr++) they are encountered at the top level, so we'll do that too. */ if ((ptr[1] == ':' || ptr[1] == '.' || ptr[1] == '=') && - check_posix_syntax(ptr, &tempptr, cd)) + check_posix_syntax(ptr, &tempptr)) { *errorcodeptr = (ptr[1] == ':')? ERR13 : ERR31; goto FAILED; @@ -2629,6 +2662,12 @@ for (;; ptr++) else break; } + /* If a class contains a negative special such as \S, we need to flip the + negation flag at the end, so that support for characters > 255 works + correctly (they are all included in the class). */ + + should_flip_negation = FALSE; + /* Keep a count of chars with values < 256 so that we can optimize the case of just a single character (as long as it's < 256). However, For higher valued UTF-8 characters, we don't yet do any optimization. */ @@ -2684,7 +2723,7 @@ for (;; ptr++) if (c == '[' && (ptr[1] == ':' || ptr[1] == '.' || ptr[1] == '=') && - check_posix_syntax(ptr, &tempptr, cd)) + check_posix_syntax(ptr, &tempptr)) { BOOL local_negate = FALSE; int posix_class, taboffset, tabopt; @@ -2701,6 +2740,7 @@ for (;; ptr++) if (*ptr == '^') { local_negate = TRUE; + should_flip_negation = TRUE; /* Note negative special */ ptr++; } @@ -2775,7 +2815,7 @@ for (;; ptr++) c = check_escape(&ptr, errorcodeptr, cd->bracount, options, TRUE); if (*errorcodeptr != 0) goto FAILED; - if (-c == ESC_b) c = '\b'; /* \b is backslash in a class */ + if (-c == ESC_b) c = '\b'; /* \b is backspace in a class */ else if (-c == ESC_X) c = 'X'; /* \X is literal X in a class */ else if (-c == ESC_R) c = 'R'; /* \R is literal R in a class */ else if (-c == ESC_Q) /* Handle start of quoted string */ @@ -2803,6 +2843,7 @@ for (;; ptr++) continue; case ESC_D: + should_flip_negation = TRUE; for (c = 0; c < 32; c++) classbits[c] |= ~cbits[c+cbit_digit]; continue; @@ -2811,6 +2852,7 @@ for (;; ptr++) continue; case ESC_W: + should_flip_negation = TRUE; for (c = 0; c < 32; c++) classbits[c] |= ~cbits[c+cbit_word]; continue; @@ -2820,13 +2862,11 @@ for (;; ptr++) continue; case ESC_S: + should_flip_negation = TRUE; for (c = 0; c < 32; c++) classbits[c] |= ~cbits[c+cbit_space]; classbits[1] |= 0x08; /* Perl 5.004 onwards omits VT from \s */ continue; - case ESC_E: /* Perl ignores an orphan \E */ - continue; - default: /* Not recognized; fall through */ break; /* Need "default" setting to stop compiler warning. */ } @@ -3061,7 +3101,7 @@ for (;; ptr++) d = check_escape(&ptr, errorcodeptr, cd->bracount, options, TRUE); if (*errorcodeptr != 0) goto FAILED; - /* \b is backslash; \X is literal X; \R is literal R; any other + /* \b is backspace; \X is literal X; \R is literal R; any other special means the '-' was literal */ if (d < 0) @@ -3325,11 +3365,14 @@ we set the flag only if there is a literal "\r" or "\n" in the class. */ zeroreqbyte = reqbyte; /* If there are characters with values > 255, we have to compile an - extended class, with its own opcode. If there are no characters < 256, - we can omit the bitmap in the actual compiled code. */ + extended class, with its own opcode, unless there was a negated special + such as \S in the class, because in that case all characters > 255 are in + the class, so any that were explicitly given as well can be ignored. If + (when there are explicit characters > 255 that must be listed) there are no + characters < 256, we can omit the bitmap in the actual compiled code. */ #ifdef SUPPORT_UTF8 - if (class_utf8) + if (class_utf8 && !should_flip_negation) { *class_utf8data++ = XCL_END; /* Marks the end of extra data */ *code++ = OP_XCLASS; @@ -3355,20 +3398,19 @@ we set the flag only if there is a literal "\r" or "\n" in the class. */ } #endif - /* If there are no characters > 255, negate the 32-byte map if necessary, - and copy it into the code vector. If this is the first thing in the branch, - there can be no first char setting, whatever the repeat count. Any reqbyte - setting must remain unchanged after any kind of repeat. */ + /* If there are no characters > 255, set the opcode to OP_CLASS or + OP_NCLASS, depending on whether the whole class was negated and whether + there were negative specials such as \S in the class. Then copy the 32-byte + map into the code vector, negating it if necessary. */ + *code++ = (negate_class == should_flip_negation) ? OP_CLASS : OP_NCLASS; if (negate_class) { - *code++ = OP_NCLASS; if (lengthptr == NULL) /* Save time in the pre-compile phase */ for (c = 0; c < 32; c++) code[c] = ~classbits[c]; } else { - *code++ = OP_CLASS; memcpy(code, classbits, 32); } code += 32; @@ -4004,7 +4046,9 @@ we set the flag only if there is a literal "\r" or "\n" in the class. */ int len; if (*tempcode == OP_EXACT || *tempcode == OP_TYPEEXACT || *tempcode == OP_NOTEXACT) - tempcode += _pcre_OP_lengths[*tempcode]; + tempcode += _pcre_OP_lengths[*tempcode] + + ((*tempcode == OP_TYPEEXACT && + (tempcode[3] == OP_PROP || tempcode[3] == OP_NOTPROP))? 2:0); len = code - tempcode; if (len > 0) switch (*tempcode) { @@ -4231,16 +4275,13 @@ we set the flag only if there is a literal "\r" or "\n" in the class. */ *errorcodeptr = ERR58; goto FAILED; } - if (refsign == '-') + recno = (refsign == '-')? + cd->bracount - recno + 1 : recno +cd->bracount; + if (recno <= 0 || recno > cd->final_bracount) { - recno = cd->bracount - recno + 1; - if (recno <= 0) - { - *errorcodeptr = ERR15; - goto FAILED; - } + *errorcodeptr = ERR15; + goto FAILED; } - else recno += cd->bracount; PUT2(code, 2+LINK_SIZE, recno); break; } @@ -4312,9 +4353,10 @@ we set the flag only if there is a literal "\r" or "\n" in the class. */ skipbytes = 1; } - /* Check for the "name" actually being a subpattern number. */ + /* Check for the "name" actually being a subpattern number. We are + in the second pass here, so final_bracount is set. */ - else if (recno > 0) + else if (recno > 0 && recno <= cd->final_bracount) { PUT2(code, 2+LINK_SIZE, recno); } @@ -4508,7 +4550,9 @@ we set the flag only if there is a literal "\r" or "\n" in the class. */ /* We come here from the Python syntax above that handles both references (?P=name) and recursion (?P>name), as well as falling - through from the Perl recursion syntax (?&name). */ + through from the Perl recursion syntax (?&name). We also come here from + the Perl \k or \k'name' back reference syntax and the \k{name} + .NET syntax. */ NAMED_REF_OR_RECURSE: name = ++ptr; @@ -4520,6 +4564,11 @@ we set the flag only if there is a literal "\r" or "\n" in the class. */ if (lengthptr != NULL) { + if (namelen == 0) + { + *errorcodeptr = ERR62; + goto FAILED; + } if (*ptr != terminator) { *errorcodeptr = ERR42; @@ -4533,14 +4582,19 @@ we set the flag only if there is a literal "\r" or "\n" in the class. */ recno = 0; } - /* In the real compile, seek the name in the table */ + /* In the real compile, seek the name in the table. We check the name + first, and then check that we have reached the end of the name in the + table. That way, if the name that is longer than any in the table, + the comparison will fail without reading beyond the table entry. */ else { slot = cd->name_table; for (i = 0; i < cd->names_found; i++) { - if (strncmp((char *)name, (char *)slot+2, namelen) == 0) break; + if (strncmp((char *)name, (char *)slot+2, namelen) == 0 && + slot[2+namelen] == 0) + break; slot += cd->name_entry_size; } @@ -4577,7 +4631,15 @@ we set the flag only if there is a literal "\r" or "\n" in the class. */ { const uschar *called; - if ((refsign = *ptr) == '+') ptr++; + if ((refsign = *ptr) == '+') + { + ptr++; + if ((digitab[*ptr] & ctype_digit) == 0) + { + *errorcodeptr = ERR63; + goto FAILED; + } + } else if (refsign == '-') { if ((digitab[ptr[1]] & ctype_digit) == 0) @@ -5904,7 +5966,7 @@ to compile parts of the pattern into; the compiled code is discarded when it is no longer needed, so hopefully this workspace will never overflow, though there is a test for its doing so. */ -cd->bracount = 0; +cd->bracount = cd->final_bracount = 0; cd->names_found = 0; cd->name_entry_size = 0; cd->name_table = NULL; @@ -5981,6 +6043,7 @@ field. Reset the bracket count and the names_found field. Also reset the hwm field; this time it's used for remembering forward references to subpatterns. */ +cd->final_bracount = cd->bracount; /* Save for checking forward references */ cd->bracount = 0; cd->names_found = 0; cd->name_table = (uschar *)re + re->name_table_offset; diff --git a/ext/pcre/pcrelib/pcre_exec.c b/ext/pcre/pcrelib/pcre_exec.c index 6fc2126c8d..593707fc89 100644 --- a/ext/pcre/pcrelib/pcre_exec.c +++ b/ext/pcre/pcrelib/pcre_exec.c @@ -4668,10 +4668,10 @@ for(;;) if (first_byte_caseless) while (start_match < end_subject && md->lcc[*start_match] != first_byte) - start_match++; + { NEXTCHAR(start_match); } else while (start_match < end_subject && *start_match != first_byte) - start_match++; + { NEXTCHAR(start_match); } } /* Or to just after a linebreak for a multiline match if possible */ @@ -4681,7 +4681,7 @@ for(;;) if (start_match > md->start_subject + start_offset) { while (start_match <= end_subject && !WAS_NEWLINE(start_match)) - start_match++; + { NEXTCHAR(start_match); } /* If we have just passed a CR and the newline option is ANY or ANYCRLF, and we are now at a LF, advance the match position by one more character. @@ -4702,7 +4702,9 @@ for(;;) while (start_match < end_subject) { register unsigned int c = *start_match; - if ((start_bits[c/8] & (1 << (c&7))) == 0) start_match++; else break; + if ((start_bits[c/8] & (1 << (c&7))) == 0) + { NEXTCHAR(start_match); } + else break; } } diff --git a/ext/pcre/pcrelib/pcre_internal.h b/ext/pcre/pcrelib/pcre_internal.h index 417c4bd027..0b2603b2b1 100644 --- a/ext/pcre/pcrelib/pcre_internal.h +++ b/ext/pcre/pcrelib/pcre_internal.h @@ -363,6 +363,7 @@ never be called in byte mode. To make sure it can never even appear when UTF-8 support is omitted, we don't even define it. */ #ifndef SUPPORT_UTF8 +#define NEXTCHAR(p) p++; #define GETCHAR(c, eptr) c = *eptr; #define GETCHARTEST(c, eptr) c = *eptr; #define GETCHARINC(c, eptr) c = *eptr++; @@ -372,6 +373,13 @@ support is omitted, we don't even define it. */ #else /* SUPPORT_UTF8 */ +/* Advance a character pointer one byte in non-UTF-8 mode and by one character +in UTF-8 mode. */ + +#define NEXTCHAR(p) \ + p++; \ + if (utf8) { while((*p & 0xc0) == 0x80) p++; } + /* Get the next UTF-8 character, not advancing the pointer. This is called when we know we are in UTF-8 mode. */ @@ -871,7 +879,7 @@ enum { ERR0, ERR1, ERR2, ERR3, ERR4, ERR5, ERR6, ERR7, ERR8, ERR9, ERR30, ERR31, ERR32, ERR33, ERR34, ERR35, ERR36, ERR37, ERR38, ERR39, ERR40, ERR41, ERR42, ERR43, ERR44, ERR45, ERR46, ERR47, ERR48, ERR49, ERR50, ERR51, ERR52, ERR53, ERR54, ERR55, ERR56, ERR57, ERR58, ERR59, - ERR60, ERR61 }; + ERR60, ERR61, ERR62, ERR63 }; /* The real format of the start of the pcre block; the index of names and the code vector run on as long as necessary after the end. We store an explicit @@ -934,7 +942,8 @@ typedef struct compile_data { uschar *name_table; /* The name/number table */ int names_found; /* Number of entries so far */ int name_entry_size; /* Size of each entry */ - int bracount; /* Count of capturing parens */ + int bracount; /* Count of capturing parens as we compile */ + int final_bracount; /* Saved value after first pass */ int top_backref; /* Maximum back reference */ unsigned int backref_map; /* Bitmap of low back refs */ int external_options; /* External (initial) options */ @@ -1036,7 +1045,7 @@ typedef struct dfa_match_data { #define ctype_letter 0x02 #define ctype_digit 0x04 #define ctype_xdigit 0x08 -#define ctype_word 0x10 /* alphameric or '_' */ +#define ctype_word 0x10 /* alphanumeric or '_' */ #define ctype_meta 0x80 /* regexp meta char or zero (end pattern) */ /* Offsets for the bitmap tables in pcre_cbits. Each table contains a set diff --git a/ext/pcre/pcrelib/pcre_valid_utf8.c b/ext/pcre/pcrelib/pcre_valid_utf8.c index 98ba2f9c7b..4c5ef69c93 100644 --- a/ext/pcre/pcrelib/pcre_valid_utf8.c +++ b/ext/pcre/pcrelib/pcre_valid_utf8.c @@ -60,7 +60,7 @@ an invalid string are then undefined. Originally, this function checked according to RFC 2279, allowing for values in the range 0 to 0x7fffffff, up to 6 bytes long, but ensuring that they were in the canonical format. Once somebody had pointed out RFC 3629 to me (it -obsoletes 2279), additional restrictions were applies. The values are now +obsoletes 2279), additional restrictions were applied. The values are now limited to be between 0 and 0x0010ffff, no more than 4 bytes long, and the subrange 0xd000 to 0xdfff is excluded. diff --git a/ext/pcre/pcrelib/pcregrep.c b/ext/pcre/pcrelib/pcregrep.c deleted file mode 100644 index eb2ce08135..0000000000 --- a/ext/pcre/pcrelib/pcregrep.c +++ /dev/null @@ -1,2104 +0,0 @@ -/************************************************* -* pcregrep program * -*************************************************/ - -/* This is a grep program that uses the PCRE regular expression library to do -its pattern matching. On a Unix or Win32 system it can recurse into -directories. - - Copyright (c) 1997-2007 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. ------------------------------------------------------------------------------ -*/ - -#include "config.h" - -#include -#include -#include -#include -#include -#include - -#include -#include - -#ifdef HAVE_UNISTD_H -#include -#endif - -#include "pcre.h" - -#define FALSE 0 -#define TRUE 1 - -typedef int BOOL; - -#define MAX_PATTERN_COUNT 100 - -#if BUFSIZ > 8192 -#define MBUFTHIRD BUFSIZ -#else -#define MBUFTHIRD 8192 -#endif - -/* Values for the "filenames" variable, which specifies options for file name -output. The order is important; it is assumed that a file name is wanted for -all values greater than FN_DEFAULT. */ - -enum { FN_NONE, FN_DEFAULT, FN_ONLY, FN_NOMATCH_ONLY, FN_FORCE }; - -/* Actions for the -d and -D options */ - -enum { dee_READ, dee_SKIP, dee_RECURSE }; -enum { DEE_READ, DEE_SKIP }; - -/* Actions for special processing options (flag bits) */ - -#define PO_WORD_MATCH 0x0001 -#define PO_LINE_MATCH 0x0002 -#define PO_FIXED_STRINGS 0x0004 - -/* Line ending types */ - -enum { EL_LF, EL_CR, EL_CRLF, EL_ANY, EL_ANYCRLF }; - - - -/************************************************* -* Global variables * -*************************************************/ - -/* Jeffrey Friedl has some debugging requirements that are not part of the -regular code. */ - -#ifdef JFRIEDL_DEBUG -static int S_arg = -1; -static unsigned int jfriedl_XR = 0; /* repeat regex attempt this many times */ -static unsigned int jfriedl_XT = 0; /* replicate text this many times */ -static const char *jfriedl_prefix = ""; -static const char *jfriedl_postfix = ""; -#endif - -static int endlinetype; - -static char *colour_string = (char *)"1;31"; -static char *colour_option = NULL; -static char *dee_option = NULL; -static char *DEE_option = NULL; -static char *newline = NULL; -static char *pattern_filename = NULL; -static char *stdin_name = (char *)"(standard input)"; -static char *locale = NULL; - -static const unsigned char *pcretables = NULL; - -static int pattern_count = 0; -static pcre **pattern_list = NULL; -static pcre_extra **hints_list = NULL; - -static char *include_pattern = NULL; -static char *exclude_pattern = NULL; - -static pcre *include_compiled = NULL; -static pcre *exclude_compiled = NULL; - -static int after_context = 0; -static int before_context = 0; -static int both_context = 0; -static int dee_action = dee_READ; -static int DEE_action = DEE_READ; -static int error_count = 0; -static int filenames = FN_DEFAULT; -static int process_options = 0; - -static BOOL count_only = FALSE; -static BOOL do_colour = FALSE; -static BOOL hyphenpending = FALSE; -static BOOL invert = FALSE; -static BOOL multiline = FALSE; -static BOOL number = FALSE; -static BOOL only_matching = FALSE; -static BOOL quiet = FALSE; -static BOOL silent = FALSE; -static BOOL utf8 = FALSE; - -/* Structure for options and list of them */ - -enum { OP_NODATA, OP_STRING, OP_OP_STRING, OP_NUMBER, OP_OP_NUMBER, - OP_PATLIST }; - -typedef struct option_item { - int type; - int one_char; - void *dataptr; - const char *long_name; - const char *help_text; -} option_item; - -/* Options without a single-letter equivalent get a negative value. This can be -used to identify them. */ - -#define N_COLOUR (-1) -#define N_EXCLUDE (-2) -#define N_HELP (-3) -#define N_INCLUDE (-4) -#define N_LABEL (-5) -#define N_LOCALE (-6) -#define N_NULL (-7) - -static option_item optionlist[] = { - { OP_NODATA, N_NULL, NULL, "", " terminate options" }, - { OP_NODATA, N_HELP, NULL, "help", "display this help and exit" }, - { OP_NUMBER, 'A', &after_context, "after-context=number", "set number of following context lines" }, - { OP_NUMBER, 'B', &before_context, "before-context=number", "set number of prior context lines" }, - { OP_OP_STRING, N_COLOUR, &colour_option, "color=option", "matched text color option" }, - { OP_NUMBER, 'C', &both_context, "context=number", "set number of context lines, before & after" }, - { OP_NODATA, 'c', NULL, "count", "print only a count of matching lines per FILE" }, - { OP_OP_STRING, N_COLOUR, &colour_option, "colour=option", "matched text colour option" }, - { OP_STRING, 'D', &DEE_option, "devices=action","how to handle devices, FIFOs, and sockets" }, - { OP_STRING, 'd', &dee_option, "directories=action", "how to handle directories" }, - { OP_PATLIST, 'e', NULL, "regex(p)", "specify pattern (may be used more than once)" }, - { OP_NODATA, 'F', NULL, "fixed_strings", "patterns are sets of newline-separated strings" }, - { OP_STRING, 'f', &pattern_filename, "file=path", "read patterns from file" }, - { OP_NODATA, 'H', NULL, "with-filename", "force the prefixing filename on output" }, - { OP_NODATA, 'h', NULL, "no-filename", "suppress the prefixing filename on output" }, - { OP_NODATA, 'i', NULL, "ignore-case", "ignore case distinctions" }, - { OP_NODATA, 'l', NULL, "files-with-matches", "print only FILE names containing matches" }, - { OP_NODATA, 'L', NULL, "files-without-match","print only FILE names not containing matches" }, - { OP_STRING, N_LABEL, &stdin_name, "label=name", "set name for standard input" }, - { OP_STRING, N_LOCALE, &locale, "locale=locale", "use the named locale" }, - { OP_NODATA, 'M', NULL, "multiline", "run in multiline mode" }, - { OP_STRING, 'N', &newline, "newline=type", "specify newline type (CR, LF, CRLF, ANYCRLF or ANY)" }, - { OP_NODATA, 'n', NULL, "line-number", "print line number with output lines" }, - { OP_NODATA, 'o', NULL, "only-matching", "show only the part of the line that matched" }, - { OP_NODATA, 'q', NULL, "quiet", "suppress output, just set return code" }, - { OP_NODATA, 'r', NULL, "recursive", "recursively scan sub-directories" }, - { OP_STRING, N_EXCLUDE,&exclude_pattern, "exclude=pattern","exclude matching files when recursing" }, - { OP_STRING, N_INCLUDE,&include_pattern, "include=pattern","include matching files when recursing" }, -#ifdef JFRIEDL_DEBUG - { OP_OP_NUMBER, 'S', &S_arg, "jeffS", "replace matched (sub)string with X" }, -#endif - { OP_NODATA, 's', NULL, "no-messages", "suppress error messages" }, - { OP_NODATA, 'u', NULL, "utf-8", "use UTF-8 mode" }, - { OP_NODATA, 'V', NULL, "version", "print version information and exit" }, - { OP_NODATA, 'v', NULL, "invert-match", "select non-matching lines" }, - { OP_NODATA, 'w', NULL, "word-regex(p)", "force patterns to match only as words" }, - { OP_NODATA, 'x', NULL, "line-regex(p)", "force patterns to match only whole lines" }, - { OP_NODATA, 0, NULL, NULL, NULL } -}; - -/* Tables for prefixing and suffixing patterns, according to the -w, -x, and -F -options. These set the 1, 2, and 4 bits in process_options, respectively. Note -that the combination of -w and -x has the same effect as -x on its own, so we -can treat them as the same. */ - -static const char *prefix[] = { - "", "\\b", "^(?:", "^(?:", "\\Q", "\\b\\Q", "^(?:\\Q", "^(?:\\Q" }; - -static const char *suffix[] = { - "", "\\b", ")$", ")$", "\\E", "\\E\\b", "\\E)$", "\\E)$" }; - -/* UTF-8 tables - used only when the newline setting is "any". */ - -const int utf8_table3[] = { 0xff, 0x1f, 0x0f, 0x07, 0x03, 0x01}; - -const char utf8_table4[] = { - 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, - 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, - 2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2, - 3,3,3,3,3,3,3,3,4,4,4,4,5,5,5,5 }; - - - -/************************************************* -* OS-specific functions * -*************************************************/ - -/* These functions are defined so that they can be made system specific, -although at present the only ones are for Unix, Win32, and for "no support". */ - - -/************* Directory scanning in Unix ***********/ - -#if defined HAVE_SYS_STAT_H && defined HAVE_DIRENT_H && defined HAVE_SYS_TYPES_H -#include -#include -#include - -typedef DIR directory_type; - -static int -isdirectory(char *filename) -{ -struct stat statbuf; -if (stat(filename, &statbuf) < 0) - return 0; /* In the expectation that opening as a file will fail */ -return ((statbuf.st_mode & S_IFMT) == S_IFDIR)? '/' : 0; -} - -static directory_type * -opendirectory(char *filename) -{ -return opendir(filename); -} - -static char * -readdirectory(directory_type *dir) -{ -for (;;) - { - struct dirent *dent = readdir(dir); - if (dent == NULL) return NULL; - if (strcmp(dent->d_name, ".") != 0 && strcmp(dent->d_name, "..") != 0) - return dent->d_name; - } -/* Control never reaches here */ -} - -static void -closedirectory(directory_type *dir) -{ -closedir(dir); -} - - -/************* Test for regular file in Unix **********/ - -static int -isregfile(char *filename) -{ -struct stat statbuf; -if (stat(filename, &statbuf) < 0) - return 1; /* In the expectation that opening as a file will fail */ -return (statbuf.st_mode & S_IFMT) == S_IFREG; -} - - -/************* Test stdout for being a terminal in Unix **********/ - -static BOOL -is_stdout_tty(void) -{ -return isatty(fileno(stdout)); -} - - -/************* Directory scanning in Win32 ***********/ - -/* I (Philip Hazel) have no means of testing this code. It was contributed by -Lionel Fourquaux. David Burgess added a patch to define INVALID_FILE_ATTRIBUTES -when it did not exist. */ - - -#elif HAVE_WINDOWS_H - -#ifndef STRICT -# define STRICT -#endif -#ifndef WIN32_LEAN_AND_MEAN -# define WIN32_LEAN_AND_MEAN -#endif -#ifndef INVALID_FILE_ATTRIBUTES -#define INVALID_FILE_ATTRIBUTES 0xFFFFFFFF -#endif - -#include - -typedef struct directory_type -{ -HANDLE handle; -BOOL first; -WIN32_FIND_DATA data; -} directory_type; - -int -isdirectory(char *filename) -{ -DWORD attr = GetFileAttributes(filename); -if (attr == INVALID_FILE_ATTRIBUTES) - return 0; -return ((attr & FILE_ATTRIBUTE_DIRECTORY) != 0) ? '/' : 0; -} - -directory_type * -opendirectory(char *filename) -{ -size_t len; -char *pattern; -directory_type *dir; -DWORD err; -len = strlen(filename); -pattern = (char *) malloc(len + 3); -dir = (directory_type *) malloc(sizeof(*dir)); -if ((pattern == NULL) || (dir == NULL)) - { - fprintf(stderr, "pcregrep: malloc failed\n"); - exit(2); - } -memcpy(pattern, filename, len); -memcpy(&(pattern[len]), "\\*", 3); -dir->handle = FindFirstFile(pattern, &(dir->data)); -if (dir->handle != INVALID_HANDLE_VALUE) - { - free(pattern); - dir->first = TRUE; - return dir; - } -err = GetLastError(); -free(pattern); -free(dir); -errno = (err == ERROR_ACCESS_DENIED) ? EACCES : ENOENT; -return NULL; -} - -char * -readdirectory(directory_type *dir) -{ -for (;;) - { - if (!dir->first) - { - if (!FindNextFile(dir->handle, &(dir->data))) - return NULL; - } - else - { - dir->first = FALSE; - } - if (strcmp(dir->data.cFileName, ".") != 0 && strcmp(dir->data.cFileName, "..") != 0) - return dir->data.cFileName; - } -#ifndef _MSC_VER -return NULL; /* Keep compiler happy; never executed */ -#endif -} - -void -closedirectory(directory_type *dir) -{ -FindClose(dir->handle); -free(dir); -} - - -/************* Test for regular file in Win32 **********/ - -/* I don't know how to do this, or if it can be done; assume all paths are -regular if they are not directories. */ - -int isregfile(char *filename) -{ -return !isdirectory(filename) -} - - -/************* Test stdout for being a terminal in Win32 **********/ - -/* I don't know how to do this; assume never */ - -static BOOL -is_stdout_tty(void) -{ -FALSE; -} - - -/************* Directory scanning when we can't do it ***********/ - -/* The type is void, and apart from isdirectory(), the functions do nothing. */ - -#else - -typedef void directory_type; - -int isdirectory(char *filename) { return 0; } -directory_type * opendirectory(char *filename) { return (directory_type*)0;} -char *readdirectory(directory_type *dir) { return (char*)0;} -void closedirectory(directory_type *dir) {} - - -/************* Test for regular when we can't do it **********/ - -/* Assume all files are regular. */ - -int isregfile(char *filename) { return 1; } - - -/************* Test stdout for being a terminal when we can't do it **********/ - -static BOOL -is_stdout_tty(void) -{ -return FALSE; -} - - -#endif - - - -#ifndef HAVE_STRERROR -/************************************************* -* Provide strerror() for non-ANSI libraries * -*************************************************/ - -/* Some old-fashioned systems still around (e.g. SunOS4) don't have strerror() -in their libraries, but can provide the same facility by this simple -alternative function. */ - -extern int sys_nerr; -extern char *sys_errlist[]; - -char * -strerror(int n) -{ -if (n < 0 || n >= sys_nerr) return "unknown error number"; -return sys_errlist[n]; -} -#endif /* HAVE_STRERROR */ - - - -/************************************************* -* Find end of line * -*************************************************/ - -/* The length of the endline sequence that is found is set via lenptr. This may -be zero at the very end of the file if there is no line-ending sequence there. - -Arguments: - p current position in line - endptr end of available data - lenptr where to put the length of the eol sequence - -Returns: pointer to the last byte of the line -*/ - -static char * -end_of_line(char *p, char *endptr, int *lenptr) -{ -switch(endlinetype) - { - default: /* Just in case */ - case EL_LF: - while (p < endptr && *p != '\n') p++; - if (p < endptr) - { - *lenptr = 1; - return p + 1; - } - *lenptr = 0; - return endptr; - - case EL_CR: - while (p < endptr && *p != '\r') p++; - if (p < endptr) - { - *lenptr = 1; - return p + 1; - } - *lenptr = 0; - return endptr; - - case EL_CRLF: - for (;;) - { - while (p < endptr && *p != '\r') p++; - if (++p >= endptr) - { - *lenptr = 0; - return endptr; - } - if (*p == '\n') - { - *lenptr = 2; - return p + 1; - } - } - break; - - case EL_ANYCRLF: - while (p < endptr) - { - int extra = 0; - register int c = *((unsigned char *)p); - - if (utf8 && c >= 0xc0) - { - int gcii, gcss; - extra = utf8_table4[c & 0x3f]; /* Number of additional bytes */ - gcss = 6*extra; - c = (c & utf8_table3[extra]) << gcss; - for (gcii = 1; gcii <= extra; gcii++) - { - gcss -= 6; - c |= (p[gcii] & 0x3f) << gcss; - } - } - - p += 1 + extra; - - switch (c) - { - case 0x0a: /* LF */ - *lenptr = 1; - return p; - - case 0x0d: /* CR */ - if (p < endptr && *p == 0x0a) - { - *lenptr = 2; - p++; - } - else *lenptr = 1; - return p; - - default: - break; - } - } /* End of loop for ANYCRLF case */ - - *lenptr = 0; /* Must have hit the end */ - return endptr; - - case EL_ANY: - while (p < endptr) - { - int extra = 0; - register int c = *((unsigned char *)p); - - if (utf8 && c >= 0xc0) - { - int gcii, gcss; - extra = utf8_table4[c & 0x3f]; /* Number of additional bytes */ - gcss = 6*extra; - c = (c & utf8_table3[extra]) << gcss; - for (gcii = 1; gcii <= extra; gcii++) - { - gcss -= 6; - c |= (p[gcii] & 0x3f) << gcss; - } - } - - p += 1 + extra; - - switch (c) - { - case 0x0a: /* LF */ - case 0x0b: /* VT */ - case 0x0c: /* FF */ - *lenptr = 1; - return p; - - case 0x0d: /* CR */ - if (p < endptr && *p == 0x0a) - { - *lenptr = 2; - p++; - } - else *lenptr = 1; - return p; - - case 0x85: /* NEL */ - *lenptr = utf8? 2 : 1; - return p; - - case 0x2028: /* LS */ - case 0x2029: /* PS */ - *lenptr = 3; - return p; - - default: - break; - } - } /* End of loop for ANY case */ - - *lenptr = 0; /* Must have hit the end */ - return endptr; - } /* End of overall switch */ -} - - - -/************************************************* -* Find start of previous line * -*************************************************/ - -/* This is called when looking back for before lines to print. - -Arguments: - p start of the subsequent line - startptr start of available data - -Returns: pointer to the start of the previous line -*/ - -static char * -previous_line(char *p, char *startptr) -{ -switch(endlinetype) - { - default: /* Just in case */ - case EL_LF: - p--; - while (p > startptr && p[-1] != '\n') p--; - return p; - - case EL_CR: - p--; - while (p > startptr && p[-1] != '\n') p--; - return p; - - case EL_CRLF: - for (;;) - { - p -= 2; - while (p > startptr && p[-1] != '\n') p--; - if (p <= startptr + 1 || p[-2] == '\r') return p; - } - return p; /* But control should never get here */ - - case EL_ANY: - case EL_ANYCRLF: - if (*(--p) == '\n' && p > startptr && p[-1] == '\r') p--; - if (utf8) while ((*p & 0xc0) == 0x80) p--; - - while (p > startptr) - { - register int c; - char *pp = p - 1; - - if (utf8) - { - int extra = 0; - while ((*pp & 0xc0) == 0x80) pp--; - c = *((unsigned char *)pp); - if (c >= 0xc0) - { - int gcii, gcss; - extra = utf8_table4[c & 0x3f]; /* Number of additional bytes */ - gcss = 6*extra; - c = (c & utf8_table3[extra]) << gcss; - for (gcii = 1; gcii <= extra; gcii++) - { - gcss -= 6; - c |= (pp[gcii] & 0x3f) << gcss; - } - } - } - else c = *((unsigned char *)pp); - - if (endlinetype == EL_ANYCRLF) switch (c) - { - case 0x0a: /* LF */ - case 0x0d: /* CR */ - return p; - - default: - break; - } - - else switch (c) - { - case 0x0a: /* LF */ - case 0x0b: /* VT */ - case 0x0c: /* FF */ - case 0x0d: /* CR */ - case 0x85: /* NEL */ - case 0x2028: /* LS */ - case 0x2029: /* PS */ - return p; - - default: - break; - } - - p = pp; /* Back one character */ - } /* End of loop for ANY case */ - - return startptr; /* Hit start of data */ - } /* End of overall switch */ -} - - - - - -/************************************************* -* Print the previous "after" lines * -*************************************************/ - -/* This is called if we are about to lose said lines because of buffer filling, -and at the end of the file. The data in the line is written using fwrite() so -that a binary zero does not terminate it. - -Arguments: - lastmatchnumber the number of the last matching line, plus one - lastmatchrestart where we restarted after the last match - endptr end of available data - printname filename for printing - -Returns: nothing -*/ - -static void do_after_lines(int lastmatchnumber, char *lastmatchrestart, - char *endptr, char *printname) -{ -if (after_context > 0 && lastmatchnumber > 0) - { - int count = 0; - while (lastmatchrestart < endptr && count++ < after_context) - { - int ellength; - char *pp = lastmatchrestart; - if (printname != NULL) fprintf(stdout, "%s-", printname); - if (number) fprintf(stdout, "%d-", lastmatchnumber++); - pp = end_of_line(pp, endptr, &ellength); - fwrite(lastmatchrestart, 1, pp - lastmatchrestart, stdout); - lastmatchrestart = pp; - } - hyphenpending = TRUE; - } -} - - - -/************************************************* -* Grep an individual file * -*************************************************/ - -/* This is called from grep_or_recurse() below. It uses a buffer that is three -times the value of MBUFTHIRD. The matching point is never allowed to stray into -the top third of the buffer, thus keeping more of the file available for -context printing or for multiline scanning. For large files, the pointer will -be in the middle third most of the time, so the bottom third is available for -"before" context printing. - -Arguments: - in the fopened FILE stream - printname the file name if it is to be printed for each match - or NULL if the file name is not to be printed - it cannot be NULL if filenames[_nomatch]_only is set - -Returns: 0 if there was at least one match - 1 otherwise (no matches) -*/ - -static int -pcregrep(FILE *in, char *printname) -{ -int rc = 1; -int linenumber = 1; -int lastmatchnumber = 0; -int count = 0; -int offsets[99]; -char *lastmatchrestart = NULL; -char buffer[3*MBUFTHIRD]; -char *ptr = buffer; -char *endptr; -size_t bufflength; -BOOL endhyphenpending = FALSE; - -/* Do the first read into the start of the buffer and set up the pointer to -end of what we have. */ - -bufflength = fread(buffer, 1, 3*MBUFTHIRD, in); -endptr = buffer + bufflength; - -/* Loop while the current pointer is not at the end of the file. For large -files, endptr will be at the end of the buffer when we are in the middle of the -file, but ptr will never get there, because as soon as it gets over 2/3 of the -way, the buffer is shifted left and re-filled. */ - -while (ptr < endptr) - { - int i, endlinelength; - int mrc = 0; - BOOL match = FALSE; - char *t = ptr; - size_t length, linelength; - - /* At this point, ptr is at the start of a line. We need to find the length - of the subject string to pass to pcre_exec(). In multiline mode, it is the - length remainder of the data in the buffer. Otherwise, it is the length of - the next line. After matching, we always advance by the length of the next - line. In multiline mode the PCRE_FIRSTLINE option is used for compiling, so - that any match is constrained to be in the first line. */ - - t = end_of_line(t, endptr, &endlinelength); - linelength = t - ptr - endlinelength; - length = multiline? (size_t)(endptr - ptr) : linelength; - - /* Extra processing for Jeffrey Friedl's debugging. */ - -#ifdef JFRIEDL_DEBUG - if (jfriedl_XT || jfriedl_XR) - { - #include - #include - struct timeval start_time, end_time; - struct timezone dummy; - - if (jfriedl_XT) - { - unsigned long newlen = length * jfriedl_XT + strlen(jfriedl_prefix) + strlen(jfriedl_postfix); - const char *orig = ptr; - ptr = malloc(newlen + 1); - if (!ptr) { - printf("out of memory"); - exit(2); - } - endptr = ptr; - strcpy(endptr, jfriedl_prefix); endptr += strlen(jfriedl_prefix); - for (i = 0; i < jfriedl_XT; i++) { - strncpy(endptr, orig, length); - endptr += length; - } - strcpy(endptr, jfriedl_postfix); endptr += strlen(jfriedl_postfix); - length = newlen; - } - - if (gettimeofday(&start_time, &dummy) != 0) - perror("bad gettimeofday"); - - - for (i = 0; i < jfriedl_XR; i++) - match = (pcre_exec(pattern_list[0], hints_list[0], ptr, length, 0, 0, offsets, 99) >= 0); - - if (gettimeofday(&end_time, &dummy) != 0) - perror("bad gettimeofday"); - - double delta = ((end_time.tv_sec + (end_time.tv_usec / 1000000.0)) - - - (start_time.tv_sec + (start_time.tv_usec / 1000000.0))); - - printf("%s TIMER[%.4f]\n", match ? "MATCH" : "FAIL", delta); - return 0; - } -#endif - - - /* Run through all the patterns until one matches. Note that we don't include - the final newline in the subject string. */ - - for (i = 0; i < pattern_count; i++) - { - mrc = pcre_exec(pattern_list[i], hints_list[i], ptr, length, 0, 0, - offsets, 99); - if (mrc >= 0) { match = TRUE; break; } - if (mrc != PCRE_ERROR_NOMATCH) - { - fprintf(stderr, "pcregrep: pcre_exec() error %d while matching ", mrc); - if (pattern_count > 1) fprintf(stderr, "pattern number %d to ", i+1); - fprintf(stderr, "this line:\n"); - fwrite(ptr, 1, linelength, stderr); /* In case binary zero included */ - fprintf(stderr, "\n"); - if (error_count == 0 && - (mrc == PCRE_ERROR_MATCHLIMIT || mrc == PCRE_ERROR_RECURSIONLIMIT)) - { - fprintf(stderr, "pcregrep: error %d means that a resource limit " - "was exceeded\n", mrc); - fprintf(stderr, "pcregrep: check your regex for nested unlimited loops\n"); - } - if (error_count++ > 20) - { - fprintf(stderr, "pcregrep: too many errors - abandoned\n"); - exit(2); - } - match = invert; /* No more matching; don't show the line again */ - break; - } - } - - /* If it's a match or a not-match (as required), do what's wanted. */ - - if (match != invert) - { - BOOL hyphenprinted = FALSE; - - /* We've failed if we want a file that doesn't have any matches. */ - - if (filenames == FN_NOMATCH_ONLY) return 1; - - /* Just count if just counting is wanted. */ - - if (count_only) count++; - - /* If all we want is a file name, there is no need to scan any more lines - in the file. */ - - else if (filenames == FN_ONLY) - { - fprintf(stdout, "%s\n", printname); - return 0; - } - - /* Likewise, if all we want is a yes/no answer. */ - - else if (quiet) return 0; - - /* The --only-matching option prints just the substring that matched, and - does not pring any context. */ - - else if (only_matching) - { - if (printname != NULL) fprintf(stdout, "%s:", printname); - if (number) fprintf(stdout, "%d:", linenumber); - fwrite(ptr + offsets[0], 1, offsets[1] - offsets[0], stdout); - fprintf(stdout, "\n"); - } - - /* This is the default case when none of the above options is set. We print - the matching lines(s), possibly preceded and/or followed by other lines of - context. */ - - else - { - /* See if there is a requirement to print some "after" lines from a - previous match. We never print any overlaps. */ - - if (after_context > 0 && lastmatchnumber > 0) - { - int ellength; - int linecount = 0; - char *p = lastmatchrestart; - - while (p < ptr && linecount < after_context) - { - p = end_of_line(p, ptr, &ellength); - linecount++; - } - - /* It is important to advance lastmatchrestart during this printing so - that it interacts correctly with any "before" printing below. Print - each line's data using fwrite() in case there are binary zeroes. */ - - while (lastmatchrestart < p) - { - char *pp = lastmatchrestart; - if (printname != NULL) fprintf(stdout, "%s-", printname); - if (number) fprintf(stdout, "%d-", lastmatchnumber++); - pp = end_of_line(pp, endptr, &ellength); - fwrite(lastmatchrestart, 1, pp - lastmatchrestart, stdout); - lastmatchrestart = pp; - } - if (lastmatchrestart != ptr) hyphenpending = TRUE; - } - - /* If there were non-contiguous lines printed above, insert hyphens. */ - - if (hyphenpending) - { - fprintf(stdout, "--\n"); - hyphenpending = FALSE; - hyphenprinted = TRUE; - } - - /* See if there is a requirement to print some "before" lines for this - match. Again, don't print overlaps. */ - - if (before_context > 0) - { - int linecount = 0; - char *p = ptr; - - while (p > buffer && (lastmatchnumber == 0 || p > lastmatchrestart) && - linecount < before_context) - { - linecount++; - p = previous_line(p, buffer); - } - - if (lastmatchnumber > 0 && p > lastmatchrestart && !hyphenprinted) - fprintf(stdout, "--\n"); - - while (p < ptr) - { - int ellength; - char *pp = p; - if (printname != NULL) fprintf(stdout, "%s-", printname); - if (number) fprintf(stdout, "%d-", linenumber - linecount--); - pp = end_of_line(pp, endptr, &ellength); - fwrite(p, 1, pp - p, stdout); - p = pp; - } - } - - /* Now print the matching line(s); ensure we set hyphenpending at the end - of the file if any context lines are being output. */ - - if (after_context > 0 || before_context > 0) - endhyphenpending = TRUE; - - if (printname != NULL) fprintf(stdout, "%s:", printname); - if (number) fprintf(stdout, "%d:", linenumber); - - /* In multiline mode, we want to print to the end of the line in which - the end of the matched string is found, so we adjust linelength and the - line number appropriately, but only when there actually was a match - (invert not set). Because the PCRE_FIRSTLINE option is set, the start of - the match will always be before the first newline sequence. */ - - if (multiline) - { - int ellength; - char *endmatch = ptr; - if (!invert) - { - endmatch += offsets[1]; - t = ptr; - while (t < endmatch) - { - t = end_of_line(t, endptr, &ellength); - if (t <= endmatch) linenumber++; else break; - } - } - endmatch = end_of_line(endmatch, endptr, &ellength); - linelength = endmatch - ptr - ellength; - } - - /*** NOTE: Use only fwrite() to output the data line, so that binary - zeroes are treated as just another data character. */ - - /* This extra option, for Jeffrey Friedl's debugging requirements, - replaces the matched string, or a specific captured string if it exists, - with X. When this happens, colouring is ignored. */ - -#ifdef JFRIEDL_DEBUG - if (S_arg >= 0 && S_arg < mrc) - { - int first = S_arg * 2; - int last = first + 1; - fwrite(ptr, 1, offsets[first], stdout); - fprintf(stdout, "X"); - fwrite(ptr + offsets[last], 1, linelength - offsets[last], stdout); - } - else -#endif - - /* We have to split the line(s) up if colouring. */ - - if (do_colour) - { - fwrite(ptr, 1, offsets[0], stdout); - fprintf(stdout, "%c[%sm", 0x1b, colour_string); - fwrite(ptr + offsets[0], 1, offsets[1] - offsets[0], stdout); - fprintf(stdout, "%c[00m", 0x1b); - fwrite(ptr + offsets[1], 1, (linelength + endlinelength) - offsets[1], - stdout); - } - else fwrite(ptr, 1, linelength + endlinelength, stdout); - } - - /* End of doing what has to be done for a match */ - - rc = 0; /* Had some success */ - - /* Remember where the last match happened for after_context. We remember - where we are about to restart, and that line's number. */ - - lastmatchrestart = ptr + linelength + endlinelength; - lastmatchnumber = linenumber + 1; - } - - /* For a match in multiline inverted mode (which of course did not cause - anything to be printed), we have to move on to the end of the match before - proceeding. */ - - if (multiline && invert && match) - { - int ellength; - char *endmatch = ptr + offsets[1]; - t = ptr; - while (t < endmatch) - { - t = end_of_line(t, endptr, &ellength); - if (t <= endmatch) linenumber++; else break; - } - endmatch = end_of_line(endmatch, endptr, &ellength); - linelength = endmatch - ptr - ellength; - } - - /* Advance to after the newline and increment the line number. */ - - ptr += linelength + endlinelength; - linenumber++; - - /* If we haven't yet reached the end of the file (the buffer is full), and - the current point is in the top 1/3 of the buffer, slide the buffer down by - 1/3 and refill it. Before we do this, if some unprinted "after" lines are - about to be lost, print them. */ - - if (bufflength >= sizeof(buffer) && ptr > buffer + 2*MBUFTHIRD) - { - if (after_context > 0 && - lastmatchnumber > 0 && - lastmatchrestart < buffer + MBUFTHIRD) - { - do_after_lines(lastmatchnumber, lastmatchrestart, endptr, printname); - lastmatchnumber = 0; - } - - /* Now do the shuffle */ - - memmove(buffer, buffer + MBUFTHIRD, 2*MBUFTHIRD); - ptr -= MBUFTHIRD; - bufflength = 2*MBUFTHIRD + fread(buffer + 2*MBUFTHIRD, 1, MBUFTHIRD, in); - endptr = buffer + bufflength; - - /* Adjust any last match point */ - - if (lastmatchnumber > 0) lastmatchrestart -= MBUFTHIRD; - } - } /* Loop through the whole file */ - -/* End of file; print final "after" lines if wanted; do_after_lines sets -hyphenpending if it prints something. */ - -if (!only_matching && !count_only) - { - do_after_lines(lastmatchnumber, lastmatchrestart, endptr, printname); - hyphenpending |= endhyphenpending; - } - -/* Print the file name if we are looking for those without matches and there -were none. If we found a match, we won't have got this far. */ - -if (filenames == FN_NOMATCH_ONLY) - { - fprintf(stdout, "%s\n", printname); - return 0; - } - -/* Print the match count if wanted */ - -if (count_only) - { - if (printname != NULL) fprintf(stdout, "%s:", printname); - fprintf(stdout, "%d\n", count); - } - -return rc; -} - - - -/************************************************* -* Grep a file or recurse into a directory * -*************************************************/ - -/* Given a path name, if it's a directory, scan all the files if we are -recursing; if it's a file, grep it. - -Arguments: - pathname the path to investigate - dir_recurse TRUE if recursing is wanted (-r or -drecurse) - only_one_at_top TRUE if the path is the only one at toplevel - -Returns: 0 if there was at least one match - 1 if there were no matches - 2 there was some kind of error - -However, file opening failures are suppressed if "silent" is set. -*/ - -static int -grep_or_recurse(char *pathname, BOOL dir_recurse, BOOL only_one_at_top) -{ -int rc = 1; -int sep; -FILE *in; - -/* If the file name is "-" we scan stdin */ - -if (strcmp(pathname, "-") == 0) - { - return pcregrep(stdin, - (filenames > FN_DEFAULT || (filenames == FN_DEFAULT && !only_one_at_top))? - stdin_name : NULL); - } - - -/* If the file is a directory, skip if skipping or if we are recursing, scan -each file within it, subject to any include or exclude patterns that were set. -The scanning code is localized so it can be made system-specific. */ - -if ((sep = isdirectory(pathname)) != 0) - { - if (dee_action == dee_SKIP) return 1; - if (dee_action == dee_RECURSE) - { - char buffer[1024]; - char *nextfile; - directory_type *dir = opendirectory(pathname); - - if (dir == NULL) - { - if (!silent) - fprintf(stderr, "pcregrep: Failed to open directory %s: %s\n", pathname, - strerror(errno)); - return 2; - } - - while ((nextfile = readdirectory(dir)) != NULL) - { - int frc, blen; - sprintf(buffer, "%.512s%c%.128s", pathname, sep, nextfile); - blen = strlen(buffer); - - if (exclude_compiled != NULL && - pcre_exec(exclude_compiled, NULL, buffer, blen, 0, 0, NULL, 0) >= 0) - continue; - - if (include_compiled != NULL && - pcre_exec(include_compiled, NULL, buffer, blen, 0, 0, NULL, 0) < 0) - continue; - - frc = grep_or_recurse(buffer, dir_recurse, FALSE); - if (frc > 1) rc = frc; - else if (frc == 0 && rc == 1) rc = 0; - } - - closedirectory(dir); - return rc; - } - } - -/* If the file is not a directory and not a regular file, skip it if that's -been requested. */ - -else if (!isregfile(pathname) && DEE_action == DEE_SKIP) return 1; - -/* Control reaches here if we have a regular file, or if we have a directory -and recursion or skipping was not requested, or if we have anything else and -skipping was not requested. The scan proceeds. If this is the first and only -argument at top level, we don't show the file name, unless we are only showing -the file name, or the filename was forced (-H). */ - -in = fopen(pathname, "r"); -if (in == NULL) - { - if (!silent) - fprintf(stderr, "pcregrep: Failed to open %s: %s\n", pathname, - strerror(errno)); - return 2; - } - -rc = pcregrep(in, (filenames > FN_DEFAULT || - (filenames == FN_DEFAULT && !only_one_at_top))? pathname : NULL); - -fclose(in); -return rc; -} - - - - -/************************************************* -* Usage function * -*************************************************/ - -static int -usage(int rc) -{ -option_item *op; -fprintf(stderr, "Usage: pcregrep [-"); -for (op = optionlist; op->one_char != 0; op++) - { - if (op->one_char > 0) fprintf(stderr, "%c", op->one_char); - } -fprintf(stderr, "] [long options] [pattern] [files]\n"); -fprintf(stderr, "Type `pcregrep --help' for more information.\n"); -return rc; -} - - - - -/************************************************* -* Help function * -*************************************************/ - -static void -help(void) -{ -option_item *op; - -printf("Usage: pcregrep [OPTION]... [PATTERN] [FILE1 FILE2 ...]\n"); -printf("Search for PATTERN in each FILE or standard input.\n"); -printf("PATTERN must be present if neither -e nor -f is used.\n"); -printf("\"-\" can be used as a file name to mean STDIN.\n\n"); -printf("Example: pcregrep -i 'hello.*world' menu.h main.c\n\n"); - -printf("Options:\n"); - -for (op = optionlist; op->one_char != 0; op++) - { - int n; - char s[4]; - if (op->one_char > 0) sprintf(s, "-%c,", op->one_char); else strcpy(s, " "); - printf(" %s --%s%n", s, op->long_name, &n); - n = 30 - n; - if (n < 1) n = 1; - printf("%.*s%s\n", n, " ", op->help_text); - } - -printf("\nWhen reading patterns from a file instead of using a command line option,\n"); -printf("trailing white space is removed and blank lines are ignored.\n"); -printf("There is a maximum of %d patterns.\n", MAX_PATTERN_COUNT); - -printf("\nWith no FILEs, read standard input. If fewer than two FILEs given, assume -h.\n"); -printf("Exit status is 0 if any matches, 1 if no matches, and 2 if trouble.\n"); -} - - - - -/************************************************* -* Handle a single-letter, no data option * -*************************************************/ - -static int -handle_option(int letter, int options) -{ -switch(letter) - { - case N_HELP: help(); exit(0); - case 'c': count_only = TRUE; break; - case 'F': process_options |= PO_FIXED_STRINGS; break; - case 'H': filenames = FN_FORCE; break; - case 'h': filenames = FN_NONE; break; - case 'i': options |= PCRE_CASELESS; break; - case 'l': filenames = FN_ONLY; break; - case 'L': filenames = FN_NOMATCH_ONLY; break; - case 'M': multiline = TRUE; options |= PCRE_MULTILINE|PCRE_FIRSTLINE; break; - case 'n': number = TRUE; break; - case 'o': only_matching = TRUE; break; - case 'q': quiet = TRUE; break; - case 'r': dee_action = dee_RECURSE; break; - case 's': silent = TRUE; break; - case 'u': options |= PCRE_UTF8; utf8 = TRUE; break; - case 'v': invert = TRUE; break; - case 'w': process_options |= PO_WORD_MATCH; break; - case 'x': process_options |= PO_LINE_MATCH; break; - - case 'V': - fprintf(stderr, "pcregrep version %s\n", pcre_version()); - exit(0); - break; - - default: - fprintf(stderr, "pcregrep: Unknown option -%c\n", letter); - exit(usage(2)); - } - -return options; -} - - - - -/************************************************* -* Construct printed ordinal * -*************************************************/ - -/* This turns a number into "1st", "3rd", etc. */ - -static char * -ordin(int n) -{ -static char buffer[8]; -char *p = buffer; -sprintf(p, "%d", n); -while (*p != 0) p++; -switch (n%10) - { - case 1: strcpy(p, "st"); break; - case 2: strcpy(p, "nd"); break; - case 3: strcpy(p, "rd"); break; - default: strcpy(p, "th"); break; - } -return buffer; -} - - - -/************************************************* -* Compile a single pattern * -*************************************************/ - -/* When the -F option has been used, this is called for each substring. -Otherwise it's called for each supplied pattern. - -Arguments: - pattern the pattern string - options the PCRE options - filename the file name, or NULL for a command-line pattern - count 0 if this is the only command line pattern, or - number of the command line pattern, or - linenumber for a pattern from a file - -Returns: TRUE on success, FALSE after an error -*/ - -static BOOL -compile_single_pattern(char *pattern, int options, char *filename, int count) -{ -char buffer[MBUFTHIRD + 16]; -const char *error; -int errptr; - -if (pattern_count >= MAX_PATTERN_COUNT) - { - fprintf(stderr, "pcregrep: Too many %spatterns (max %d)\n", - (filename == NULL)? "command-line " : "", MAX_PATTERN_COUNT); - return FALSE; - } - -sprintf(buffer, "%s%.*s%s", prefix[process_options], MBUFTHIRD, pattern, - suffix[process_options]); -pattern_list[pattern_count] = - pcre_compile(buffer, options, &error, &errptr, pcretables); -if (pattern_list[pattern_count] != NULL) - { - pattern_count++; - return TRUE; - } - -/* Handle compile errors */ - -errptr -= (int)strlen(prefix[process_options]); -if (errptr > (int)strlen(pattern)) errptr = (int)strlen(pattern); - -if (filename == NULL) - { - if (count == 0) - fprintf(stderr, "pcregrep: Error in command-line regex " - "at offset %d: %s\n", errptr, error); - else - fprintf(stderr, "pcregrep: Error in %s command-line regex " - "at offset %d: %s\n", ordin(count), errptr, error); - } -else - { - fprintf(stderr, "pcregrep: Error in regex in line %d of %s " - "at offset %d: %s\n", count, filename, errptr, error); - } - -return FALSE; -} - - - -/************************************************* -* Compile one supplied pattern * -*************************************************/ - -/* When the -F option has been used, each string may be a list of strings, -separated by line breaks. They will be matched literally. - -Arguments: - pattern the pattern string - options the PCRE options - filename the file name, or NULL for a command-line pattern - count 0 if this is the only command line pattern, or - number of the command line pattern, or - linenumber for a pattern from a file - -Returns: TRUE on success, FALSE after an error -*/ - -static BOOL -compile_pattern(char *pattern, int options, char *filename, int count) -{ -if ((process_options & PO_FIXED_STRINGS) != 0) - { - char *eop = pattern + strlen(pattern); - char buffer[MBUFTHIRD]; - for(;;) - { - int ellength; - char *p = end_of_line(pattern, eop, &ellength); - if (ellength == 0) - return compile_single_pattern(pattern, options, filename, count); - sprintf(buffer, "%.*s", (int)(p - pattern - ellength), pattern); - pattern = p; - if (!compile_single_pattern(buffer, options, filename, count)) - return FALSE; - } - } -else return compile_single_pattern(pattern, options, filename, count); -} - - - -/************************************************* -* Main program * -*************************************************/ - -/* Returns 0 if something matched, 1 if nothing matched, 2 after an error. */ - -int -main(int argc, char **argv) -{ -int i, j; -int rc = 1; -int pcre_options = 0; -int cmd_pattern_count = 0; -int hint_count = 0; -int errptr; -BOOL only_one_at_top; -char *patterns[MAX_PATTERN_COUNT]; -const char *locale_from = "--locale"; -const char *error; - -/* Set the default line ending value from the default in the PCRE library; -"lf", "cr", "crlf", and "any" are supported. Anything else is treated as "lf". -*/ - -(void)pcre_config(PCRE_CONFIG_NEWLINE, &i); -switch(i) - { - default: newline = (char *)"lf"; break; - case '\r': newline = (char *)"cr"; break; - case ('\r' << 8) | '\n': newline = (char *)"crlf"; break; - case -1: newline = (char *)"any"; break; - case -2: newline = (char *)"anycrlf"; break; - } - -/* Process the options */ - -for (i = 1; i < argc; i++) - { - option_item *op = NULL; - char *option_data = (char *)""; /* default to keep compiler happy */ - BOOL longop; - BOOL longopwasequals = FALSE; - - if (argv[i][0] != '-') break; - - /* If we hit an argument that is just "-", it may be a reference to STDIN, - but only if we have previously had -e or -f to define the patterns. */ - - if (argv[i][1] == 0) - { - if (pattern_filename != NULL || pattern_count > 0) break; - else exit(usage(2)); - } - - /* Handle a long name option, or -- to terminate the options */ - - if (argv[i][1] == '-') - { - char *arg = argv[i] + 2; - char *argequals = strchr(arg, '='); - - if (*arg == 0) /* -- terminates options */ - { - i++; - break; /* out of the options-handling loop */ - } - - longop = TRUE; - - /* Some long options have data that follows after =, for example file=name. - Some options have variations in the long name spelling: specifically, we - allow "regexp" because GNU grep allows it, though I personally go along - with Jeffrey Friedl and Larry Wall in preferring "regex" without the "p". - These options are entered in the table as "regex(p)". No option is in both - these categories, fortunately. */ - - for (op = optionlist; op->one_char != 0; op++) - { - char *opbra = strchr(op->long_name, '('); - char *equals = strchr(op->long_name, '='); - if (opbra == NULL) /* Not a (p) case */ - { - if (equals == NULL) /* Not thing=data case */ - { - if (strcmp(arg, op->long_name) == 0) break; - } - else /* Special case xxx=data */ - { - int oplen = equals - op->long_name; - int arglen = (argequals == NULL)? (int)strlen(arg) : argequals - arg; - if (oplen == arglen && strncmp(arg, op->long_name, oplen) == 0) - { - option_data = arg + arglen; - if (*option_data == '=') - { - option_data++; - longopwasequals = TRUE; - } - break; - } - } - } - else /* Special case xxxx(p) */ - { - char buff1[24]; - char buff2[24]; - int baselen = opbra - op->long_name; - sprintf(buff1, "%.*s", baselen, op->long_name); - sprintf(buff2, "%s%.*s", buff1, - (int)strlen(op->long_name) - baselen - 2, opbra + 1); - if (strcmp(arg, buff1) == 0 || strcmp(arg, buff2) == 0) - break; - } - } - - if (op->one_char == 0) - { - fprintf(stderr, "pcregrep: Unknown option %s\n", argv[i]); - exit(usage(2)); - } - } - - - /* Jeffrey Friedl's debugging harness uses these additional options which - are not in the right form for putting in the option table because they use - only one hyphen, yet are more than one character long. By putting them - separately here, they will not get displayed as part of the help() output, - but I don't think Jeffrey will care about that. */ - -#ifdef JFRIEDL_DEBUG - else if (strcmp(argv[i], "-pre") == 0) { - jfriedl_prefix = argv[++i]; - continue; - } else if (strcmp(argv[i], "-post") == 0) { - jfriedl_postfix = argv[++i]; - continue; - } else if (strcmp(argv[i], "-XT") == 0) { - sscanf(argv[++i], "%d", &jfriedl_XT); - continue; - } else if (strcmp(argv[i], "-XR") == 0) { - sscanf(argv[++i], "%d", &jfriedl_XR); - continue; - } -#endif - - - /* One-char options; many that have no data may be in a single argument; we - continue till we hit the last one or one that needs data. */ - - else - { - char *s = argv[i] + 1; - longop = FALSE; - while (*s != 0) - { - for (op = optionlist; op->one_char != 0; op++) - { if (*s == op->one_char) break; } - if (op->one_char == 0) - { - fprintf(stderr, "pcregrep: Unknown option letter '%c' in \"%s\"\n", - *s, argv[i]); - exit(usage(2)); - } - if (op->type != OP_NODATA || s[1] == 0) - { - option_data = s+1; - break; - } - pcre_options = handle_option(*s++, pcre_options); - } - } - - /* At this point we should have op pointing to a matched option. If the type - is NO_DATA, it means that there is no data, and the option might set - something in the PCRE options. */ - - if (op->type == OP_NODATA) - { - pcre_options = handle_option(op->one_char, pcre_options); - continue; - } - - /* If the option type is OP_OP_STRING or OP_OP_NUMBER, it's an option that - either has a value or defaults to something. It cannot have data in a - separate item. At the moment, the only such options are "colo(u)r" and - Jeffrey Friedl's special -S debugging option. */ - - if (*option_data == 0 && - (op->type == OP_OP_STRING || op->type == OP_OP_NUMBER)) - { - switch (op->one_char) - { - case N_COLOUR: - colour_option = (char *)"auto"; - break; -#ifdef JFRIEDL_DEBUG - case 'S': - S_arg = 0; - break; -#endif - } - continue; - } - - /* Otherwise, find the data string for the option. */ - - if (*option_data == 0) - { - if (i >= argc - 1 || longopwasequals) - { - fprintf(stderr, "pcregrep: Data missing after %s\n", argv[i]); - exit(usage(2)); - } - option_data = argv[++i]; - } - - /* If the option type is OP_PATLIST, it's the -e option, which can be called - multiple times to create a list of patterns. */ - - if (op->type == OP_PATLIST) - { - if (cmd_pattern_count >= MAX_PATTERN_COUNT) - { - fprintf(stderr, "pcregrep: Too many command-line patterns (max %d)\n", - MAX_PATTERN_COUNT); - return 2; - } - patterns[cmd_pattern_count++] = option_data; - } - - /* Otherwise, deal with single string or numeric data values. */ - - else if (op->type != OP_NUMBER && op->type != OP_OP_NUMBER) - { - *((char **)op->dataptr) = option_data; - } - else - { - char *endptr; - int n = strtoul(option_data, &endptr, 10); - if (*endptr != 0) - { - if (longop) - { - char *equals = strchr(op->long_name, '='); - int nlen = (equals == NULL)? (int)strlen(op->long_name) : - equals - op->long_name; - fprintf(stderr, "pcregrep: Malformed number \"%s\" after --%.*s\n", - option_data, nlen, op->long_name); - } - else - fprintf(stderr, "pcregrep: Malformed number \"%s\" after -%c\n", - option_data, op->one_char); - exit(usage(2)); - } - *((int *)op->dataptr) = n; - } - } - -/* Options have been decoded. If -C was used, its value is used as a default -for -A and -B. */ - -if (both_context > 0) - { - if (after_context == 0) after_context = both_context; - if (before_context == 0) before_context = both_context; - } - -/* If a locale has not been provided as an option, see if the LC_CTYPE or -LC_ALL environment variable is set, and if so, use it. */ - -if (locale == NULL) - { - locale = getenv("LC_ALL"); - locale_from = "LCC_ALL"; - } - -if (locale == NULL) - { - locale = getenv("LC_CTYPE"); - locale_from = "LC_CTYPE"; - } - -/* If a locale has been provided, set it, and generate the tables the PCRE -needs. Otherwise, pcretables==NULL, which causes the use of default tables. */ - -if (locale != NULL) - { - if (setlocale(LC_CTYPE, locale) == NULL) - { - fprintf(stderr, "pcregrep: Failed to set locale %s (obtained from %s)\n", - locale, locale_from); - return 2; - } - pcretables = pcre_maketables(); - } - -/* Sort out colouring */ - -if (colour_option != NULL && strcmp(colour_option, "never") != 0) - { - if (strcmp(colour_option, "always") == 0) do_colour = TRUE; - else if (strcmp(colour_option, "auto") == 0) do_colour = is_stdout_tty(); - else - { - fprintf(stderr, "pcregrep: Unknown colour setting \"%s\"\n", - colour_option); - return 2; - } - if (do_colour) - { - char *cs = getenv("PCREGREP_COLOUR"); - if (cs == NULL) cs = getenv("PCREGREP_COLOR"); - if (cs != NULL) colour_string = cs; - } - } - -/* Interpret the newline type; the default settings are Unix-like. */ - -if (strcmp(newline, "cr") == 0 || strcmp(newline, "CR") == 0) - { - pcre_options |= PCRE_NEWLINE_CR; - endlinetype = EL_CR; - } -else if (strcmp(newline, "lf") == 0 || strcmp(newline, "LF") == 0) - { - pcre_options |= PCRE_NEWLINE_LF; - endlinetype = EL_LF; - } -else if (strcmp(newline, "crlf") == 0 || strcmp(newline, "CRLF") == 0) - { - pcre_options |= PCRE_NEWLINE_CRLF; - endlinetype = EL_CRLF; - } -else if (strcmp(newline, "any") == 0 || strcmp(newline, "ANY") == 0) - { - pcre_options |= PCRE_NEWLINE_ANY; - endlinetype = EL_ANY; - } -else if (strcmp(newline, "anycrlf") == 0 || strcmp(newline, "ANYCRLF") == 0) - { - pcre_options |= PCRE_NEWLINE_ANYCRLF; - endlinetype = EL_ANYCRLF; - } -else - { - fprintf(stderr, "pcregrep: Invalid newline specifier \"%s\"\n", newline); - return 2; - } - -/* Interpret the text values for -d and -D */ - -if (dee_option != NULL) - { - if (strcmp(dee_option, "read") == 0) dee_action = dee_READ; - else if (strcmp(dee_option, "recurse") == 0) dee_action = dee_RECURSE; - else if (strcmp(dee_option, "skip") == 0) dee_action = dee_SKIP; - else - { - fprintf(stderr, "pcregrep: Invalid value \"%s\" for -d\n", dee_option); - return 2; - } - } - -if (DEE_option != NULL) - { - if (strcmp(DEE_option, "read") == 0) DEE_action = DEE_READ; - else if (strcmp(DEE_option, "skip") == 0) DEE_action = DEE_SKIP; - else - { - fprintf(stderr, "pcregrep: Invalid value \"%s\" for -D\n", DEE_option); - return 2; - } - } - -/* Check the values for Jeffrey Friedl's debugging options. */ - -#ifdef JFRIEDL_DEBUG -if (S_arg > 9) - { - fprintf(stderr, "pcregrep: bad value for -S option\n"); - return 2; - } -if (jfriedl_XT != 0 || jfriedl_XR != 0) - { - if (jfriedl_XT == 0) jfriedl_XT = 1; - if (jfriedl_XR == 0) jfriedl_XR = 1; - } -#endif - -/* Get memory to store the pattern and hints lists. */ - -pattern_list = (pcre **)malloc(MAX_PATTERN_COUNT * sizeof(pcre *)); -hints_list = (pcre_extra **)malloc(MAX_PATTERN_COUNT * sizeof(pcre_extra *)); - -if (pattern_list == NULL || hints_list == NULL) - { - fprintf(stderr, "pcregrep: malloc failed\n"); - goto EXIT2; - } - -/* If no patterns were provided by -e, and there is no file provided by -f, -the first argument is the one and only pattern, and it must exist. */ - -if (cmd_pattern_count == 0 && pattern_filename == NULL) - { - if (i >= argc) return usage(2); - patterns[cmd_pattern_count++] = argv[i++]; - } - -/* Compile the patterns that were provided on the command line, either by -multiple uses of -e or as a single unkeyed pattern. */ - -for (j = 0; j < cmd_pattern_count; j++) - { - if (!compile_pattern(patterns[j], pcre_options, NULL, - (j == 0 && cmd_pattern_count == 1)? 0 : j + 1)) - goto EXIT2; - } - -/* Compile the regular expressions that are provided in a file. */ - -if (pattern_filename != NULL) - { - int linenumber = 0; - FILE *f; - char *filename; - char buffer[MBUFTHIRD]; - - if (strcmp(pattern_filename, "-") == 0) - { - f = stdin; - filename = stdin_name; - } - else - { - f = fopen(pattern_filename, "r"); - if (f == NULL) - { - fprintf(stderr, "pcregrep: Failed to open %s: %s\n", pattern_filename, - strerror(errno)); - goto EXIT2; - } - filename = pattern_filename; - } - - while (fgets(buffer, MBUFTHIRD, f) != NULL) - { - char *s = buffer + (int)strlen(buffer); - while (s > buffer && isspace((unsigned char)(s[-1]))) s--; - *s = 0; - linenumber++; - if (buffer[0] == 0) continue; /* Skip blank lines */ - if (!compile_pattern(buffer, pcre_options, filename, linenumber)) - goto EXIT2; - } - - if (f != stdin) fclose(f); - } - -/* Study the regular expressions, as we will be running them many times */ - -for (j = 0; j < pattern_count; j++) - { - hints_list[j] = pcre_study(pattern_list[j], 0, &error); - if (error != NULL) - { - char s[16]; - if (pattern_count == 1) s[0] = 0; else sprintf(s, " number %d", j); - fprintf(stderr, "pcregrep: Error while studying regex%s: %s\n", s, error); - goto EXIT2; - } - hint_count++; - } - -/* If there are include or exclude patterns, compile them. */ - -if (exclude_pattern != NULL) - { - exclude_compiled = pcre_compile(exclude_pattern, 0, &error, &errptr, - pcretables); - if (exclude_compiled == NULL) - { - fprintf(stderr, "pcregrep: Error in 'exclude' regex at offset %d: %s\n", - errptr, error); - goto EXIT2; - } - } - -if (include_pattern != NULL) - { - include_compiled = pcre_compile(include_pattern, 0, &error, &errptr, - pcretables); - if (include_compiled == NULL) - { - fprintf(stderr, "pcregrep: Error in 'include' regex at offset %d: %s\n", - errptr, error); - goto EXIT2; - } - } - -/* If there are no further arguments, do the business on stdin and exit. */ - -if (i >= argc) - { - rc = pcregrep(stdin, (filenames > FN_DEFAULT)? stdin_name : NULL); - goto EXIT; - } - -/* Otherwise, work through the remaining arguments as files or directories. -Pass in the fact that there is only one argument at top level - this suppresses -the file name if the argument is not a directory and filenames are not -otherwise forced. */ - -only_one_at_top = i == argc - 1; /* Catch initial value of i */ - -for (; i < argc; i++) - { - int frc = grep_or_recurse(argv[i], dee_action == dee_RECURSE, - only_one_at_top); - if (frc > 1) rc = frc; - else if (frc == 0 && rc == 1) rc = 0; - } - -EXIT: -if (pattern_list != NULL) - { - for (i = 0; i < pattern_count; i++) free(pattern_list[i]); - free(pattern_list); - } -if (hints_list != NULL) - { - for (i = 0; i < hint_count; i++) free(hints_list[i]); - free(hints_list); - } -return rc; - -EXIT2: -rc = 2; -goto EXIT; -} - -/* End of pcregrep */ diff --git a/ext/pcre/pcrelib/pcreposix.c b/ext/pcre/pcrelib/pcreposix.c index 756ab5f989..6f1f9c58a6 100644 --- a/ext/pcre/pcrelib/pcreposix.c +++ b/ext/pcre/pcrelib/pcreposix.c @@ -122,7 +122,9 @@ static const int eint[] = { REG_INVARG, /* inconsistent NEWLINE options */ REG_BADPAT, /* \g is not followed followed by an (optionally braced) non-zero number */ REG_BADPAT, /* (?+ or (?- must be followed by a non-zero number */ - REG_BADPAT /* number is too big */ + REG_BADPAT, /* number is too big */ + REG_BADPAT, /* subpattern name expected */ + REG_BADPAT /* digit expected after (?+ */ }; /* Table of texts corresponding to POSIX error codes */ diff --git a/ext/pcre/pcrelib/testdata/grepoutput b/ext/pcre/pcrelib/testdata/grepoutput index 7f76d9edc9..3241984c1c 100644 --- a/ext/pcre/pcrelib/testdata/grepoutput +++ b/ext/pcre/pcrelib/testdata/grepoutput @@ -358,10 +358,13 @@ after the binary zero ./testdata/grepinput:597:after the binary zero ---------------------------- Test 42 ------------------------------ 595:before +595:zero 596:zero 597:after +597:zero ---------------------------- Test 43 ------------------------------ 595:before +595:zero 596:zero 597:zero ---------------------------- Test 44 ------------------------------ @@ -387,3 +390,15 @@ PUT NEW DATA ABOVE THIS LINE. over the lazy dog. ---------------------------- Test 51 ------------------------------ fox jumps +---------------------------- Test 52 ------------------------------ +36972,6 +36990,4 +37024,4 +37066,5 +37083,4 +---------------------------- Test 53 ------------------------------ +595:15,6 +595:33,4 +596:28,4 +597:15,5 +597:32,4 diff --git a/ext/pcre/pcrelib/testdata/testinput1 b/ext/pcre/pcrelib/testdata/testinput1 index 79c98fa7bd..91554ffc49 100644 --- a/ext/pcre/pcrelib/testdata/testinput1 +++ b/ext/pcre/pcrelib/testdata/testinput1 @@ -3421,11 +3421,6 @@ /((?m)^b)/ a\nb\nc\n -/(?(1)a|b)/ - -/(?(1)b|a)/ - a - /(x)?(?(1)a|b)/ *** Failers a @@ -4030,4 +4025,15 @@ /( (?(1)0|)* )/x abcd +/[[:abcd:xyz]]/ + a] + :] + +/[abc[:x\]pqr]/ + a + [ + : + ] + p + / End of testinput1 / diff --git a/ext/pcre/pcrelib/testdata/testinput2 b/ext/pcre/pcrelib/testdata/testinput2 index e0018b604f..1a13fa8dd9 100644 --- a/ext/pcre/pcrelib/testdata/testinput2 +++ b/ext/pcre/pcrelib/testdata/testinput2 @@ -398,8 +398,6 @@ /(?(1?)a|b)/ -/(?(1)a|b|c)/ - /[a[:xyz:/ /(?<=x+)y/ @@ -568,15 +566,15 @@ /ab\d+/I -/a(?(1)b)/I +/a(?(1)b)(.)/I -/a(?(1)bag|big)/I +/a(?(1)bag|big)(.)/I -/a(?(1)bag|big)*/I +/a(?(1)bag|big)*(.)/I -/a(?(1)bag|big)+/I +/a(?(1)bag|big)+(.)/I -/a(?(1)b..|b..)/I +/a(?(1)b..|b..)(.)/I /ab\d{0}e/I @@ -977,13 +975,13 @@ /()a/I -/(?(1)ab|ac)/I +/(?(1)ab|ac)(.)/I -/(?(1)abz|acz)/I +/(?(1)abz|acz)(.)/I -/(?(1)abz)/I +/(?(1)abz)(.)/I -/(?(1)abz)123/I +/(?(1)abz)(1)23/I /(a)+/I @@ -2190,8 +2188,8 @@ a random value. /Ix /((?(-2)a))/BZ -/^(?(+1)X|Y)/BZ - Y +/^(?(+1)X|Y)(.)/BZ + Y! /(foo)\Kbar/ foobar @@ -2535,4 +2533,60 @@ a random value. /Ix /(*CRLF)(*BSR_ANYCRLF)(*CR)ab/I +/(?)(?&)/ + +/(?)(?&a)/ + +/(?)(?&aaaaaaaaaaaaaaaaaaaaaaa)/ + +/(?+-a)/ + +/(?-+a)/ + +/(?(-1))/ + +/(?(+10))/ + +/(?(10))/ + +/(?(+2))()()/ + +/(?(2))()()/ + +/\k''/ + +/\k<>/ + +/\k{}/ + +/(?P=)/ + +/(?P>)/ + +/(?!\w)(?R)/ + +/(?=\w)(?R)/ + +/(? a\x0bb\ +/.*a.*=.b.*/8 + QQQ\x{2029}ABCaXYZ=!bPQR + ** Failers + a\x{2029}b + \x61\xe2\x80\xa9\x62 + +/[[:a\x{100}b:]]/8 + / End of testinput5 / diff --git a/ext/pcre/pcrelib/testdata/testinput6 b/ext/pcre/pcrelib/testdata/testinput6 index 53d2b328ff..0204a7acae 100644 --- a/ext/pcre/pcrelib/testdata/testinput6 +++ b/ext/pcre/pcrelib/testdata/testinput6 @@ -832,4 +832,79 @@ was broken in all cases./ /(\p{Yi}{0,3}+\277)*/ +/^[\p{Arabic}]/8 + \x{60e} + \x{656} + \x{657} + \x{658} + \x{659} + \x{65a} + \x{65b} + \x{65c} + \x{65d} + \x{65e} + \x{66a} + \x{6e9} + \x{6ef} + \x{6fa} + ** Failers + \x{600} + \x{650} + \x{651} + \x{652} + \x{653} + \x{654} + \x{655} + \x{65f} + +/^\p{Cyrillic}/8 + \x{1d2b} + +/^\p{Common}/8 + \x{589} + \x{60c} + \x{61f} + \x{964} + \x{965} + \x{970} + +/^\p{Inherited}/8 + \x{64b} + \x{654} + \x{655} + \x{200c} + ** Failers + \x{64a} + \x{656} + +/^\p{Shavian}/8 + \x{10450} + \x{1047f} + +/^\p{Deseret}/8 + \x{10400} + \x{1044f} + +/^\p{Osmanya}/8 + \x{10480} + \x{1049d} + \x{104a0} + \x{104a9} + ** Failers + \x{1049e} + \x{1049f} + \x{104aa} + +/\p{Zl}{2,3}+/8BZ + \xe2\x80\xa8\xe2\x80\xa8 + \x{2028}\x{2028}\x{2028} + +/\p{Zl}/8BZ + +/\p{Lu}{3}+/8BZ + +/\pL{2}+/8BZ + +/\p{Cc}{2}+/8BZ + / End of testinput6 / diff --git a/ext/pcre/pcrelib/testdata/testoutput1 b/ext/pcre/pcrelib/testdata/testoutput1 index 4c0e680d11..9b8b268edb 100644 --- a/ext/pcre/pcrelib/testdata/testoutput1 +++ b/ext/pcre/pcrelib/testdata/testoutput1 @@ -5551,12 +5551,6 @@ No match 0: b 1: b -/(?(1)a|b)/ - -/(?(1)b|a)/ - a - 0: a - /(x)?(?(1)a|b)/ *** Failers No match @@ -6593,4 +6587,22 @@ No match 0: 1: +/[[:abcd:xyz]]/ + a] + 0: a] + :] + 0: :] + +/[abc[:x\]pqr]/ + a + 0: a + [ + 0: [ + : + 0: : + ] + 0: ] + p + 0: p + / End of testinput1 / diff --git a/ext/pcre/pcrelib/testdata/testoutput2 b/ext/pcre/pcrelib/testdata/testoutput2 index 11fcb1eed1..dba227f503 100644 --- a/ext/pcre/pcrelib/testdata/testoutput2 +++ b/ext/pcre/pcrelib/testdata/testoutput2 @@ -109,7 +109,7 @@ Failed: missing ) at offset 4 Failed: missing ) after comment at offset 7 /(?z)abc/ -Failed: unrecognized character after (? at offset 2 +Failed: unrecognized character after (? or (?- at offset 2 /.*b/I Capturing subpattern count = 0 @@ -310,7 +310,7 @@ No match No match /ab(?z)cd/ -Failed: unrecognized character after (? at offset 4 +Failed: unrecognized character after (? or (?- at offset 4 /^abc|def/I Capturing subpattern count = 0 @@ -946,26 +946,23 @@ Failed: missing ) at offset 4 Failed: unrecognized character after (?< at offset 3 /a(?{)b/ -Failed: unrecognized character after (? at offset 3 +Failed: unrecognized character after (? or (?- at offset 3 /a(?{{})b/ -Failed: unrecognized character after (? at offset 3 +Failed: unrecognized character after (? or (?- at offset 3 /a(?{}})b/ -Failed: unrecognized character after (? at offset 3 +Failed: unrecognized character after (? or (?- at offset 3 /a(?{"{"})b/ -Failed: unrecognized character after (? at offset 3 +Failed: unrecognized character after (? or (?- at offset 3 /a(?{"{"}})b/ -Failed: unrecognized character after (? at offset 3 +Failed: unrecognized character after (? or (?- at offset 3 /(?(1?)a|b)/ Failed: malformed number or name after (?( at offset 4 -/(?(1)a|b|c)/ -Failed: conditional group contains more than two branches at offset 10 - /[a[:xyz:/ Failed: missing terminating ] for character class at offset 8 @@ -1599,32 +1596,32 @@ No options First char = 'a' Need char = 'b' -/a(?(1)b)/I -Capturing subpattern count = 0 +/a(?(1)b)(.)/I +Capturing subpattern count = 1 No options First char = 'a' No need char -/a(?(1)bag|big)/I -Capturing subpattern count = 0 +/a(?(1)bag|big)(.)/I +Capturing subpattern count = 1 No options First char = 'a' Need char = 'g' -/a(?(1)bag|big)*/I -Capturing subpattern count = 0 +/a(?(1)bag|big)*(.)/I +Capturing subpattern count = 1 No options First char = 'a' No need char -/a(?(1)bag|big)+/I -Capturing subpattern count = 0 +/a(?(1)bag|big)+(.)/I +Capturing subpattern count = 1 No options First char = 'a' Need char = 'g' -/a(?(1)b..|b..)/I -Capturing subpattern count = 0 +/a(?(1)b..|b..)(.)/I +Capturing subpattern count = 1 No options First char = 'a' Need char = 'b' @@ -1905,7 +1902,7 @@ No need char ------------------------------------------------------------------ Bra ^ - [\x00-/:-@[-`{-\xff] + [\x00-/:-@[-`{-\xff] (neg) Ket End ------------------------------------------------------------------ @@ -1931,7 +1928,7 @@ No need char ------------------------------------------------------------------ Bra ^ - [\x00-@[-`{-\xff] + [\x00-@[-`{-\xff] (neg) Ket End ------------------------------------------------------------------ @@ -1965,7 +1962,7 @@ No need char ------------------------------------------------------------------ Bra ^ - [\x80-\xff] + [\x80-\xff] (neg) Ket End ------------------------------------------------------------------ @@ -1991,7 +1988,7 @@ No need char ------------------------------------------------------------------ Bra ^ - [\x00-\x08\x0a-\x1f!-\xff] + [\x00-\x08\x0a-\x1f!-\xff] (neg) Ket End ------------------------------------------------------------------ @@ -2142,7 +2139,7 @@ No need char ------------------------------------------------------------------ Bra ^ - [ -~\x80-\xff] + [ -~\x80-\xff] (neg) Ket End ------------------------------------------------------------------ @@ -2155,7 +2152,7 @@ No need char ------------------------------------------------------------------ Bra ^ - [\x00-/12:-\xff] + [\x00-/12:-\xff] (neg) Ket End ------------------------------------------------------------------ @@ -2168,7 +2165,7 @@ No need char ------------------------------------------------------------------ Bra ^ - [\x00-\x08\x0a-\x1f!-\xff] + [\x00-\x08\x0a-\x1f!-\xff] (neg) Ket End ------------------------------------------------------------------ @@ -2736,7 +2733,7 @@ No need char /[\S]/DZ ------------------------------------------------------------------ Bra - [\x00-\x08\x0b\x0e-\x1f!-\xff] + [\x00-\x08\x0b\x0e-\x1f!-\xff] (neg) Ket End ------------------------------------------------------------------ @@ -3441,26 +3438,26 @@ No options No first char Need char = 'a' -/(?(1)ab|ac)/I -Capturing subpattern count = 0 +/(?(1)ab|ac)(.)/I +Capturing subpattern count = 1 No options First char = 'a' No need char -/(?(1)abz|acz)/I -Capturing subpattern count = 0 +/(?(1)abz|acz)(.)/I +Capturing subpattern count = 1 No options First char = 'a' Need char = 'z' -/(?(1)abz)/I -Capturing subpattern count = 0 +/(?(1)abz)(.)/I +Capturing subpattern count = 1 No options No first char No need char -/(?(1)abz)123/I -Capturing subpattern count = 0 +/(?(1)abz)(1)23/I +Capturing subpattern count = 1 No options No first char Need char = '3' @@ -8308,7 +8305,7 @@ Failed: reference to non-existent subpattern at offset 6 /((?(-2)a))/BZ Failed: reference to non-existent subpattern at offset 7 -/^(?(+1)X|Y)/BZ +/^(?(+1)X|Y)(.)/BZ ------------------------------------------------------------------ Bra ^ @@ -8318,11 +8315,15 @@ Failed: reference to non-existent subpattern at offset 7 Alt Y Ket + CBra 1 + Any + Ket Ket End ------------------------------------------------------------------ - Y - 0: Y + Y! + 0: Y! + 1: ! /(foo)\Kbar/ foobar @@ -9302,4 +9303,86 @@ Forced newline sequence: CR First char = 'a' Need char = 'b' +/(?)(?&)/ +Failed: subpattern name expected at offset 9 + +/(?)(?&a)/ +Failed: reference to non-existent subpattern at offset 12 + +/(?)(?&aaaaaaaaaaaaaaaaaaaaaaa)/ +Failed: reference to non-existent subpattern at offset 32 + +/(?+-a)/ +Failed: digit expected after (?+ at offset 3 + +/(?-+a)/ +Failed: unrecognized character after (? or (?- at offset 3 + +/(?(-1))/ +Failed: reference to non-existent subpattern at offset 6 + +/(?(+10))/ +Failed: reference to non-existent subpattern at offset 7 + +/(?(10))/ +Failed: reference to non-existent subpattern at offset 6 + +/(?(+2))()()/ + +/(?(2))()()/ + +/\k''/ +Failed: subpattern name expected at offset 3 + +/\k<>/ +Failed: subpattern name expected at offset 3 + +/\k{}/ +Failed: subpattern name expected at offset 3 + +/(?P=)/ +Failed: subpattern name expected at offset 4 + +/(?P>)/ +Failed: subpattern name expected at offset 4 + +/(?!\w)(?R)/ +Failed: recursive call could loop indefinitely at offset 9 + +/(?=\w)(?R)/ +Failed: recursive call could loop indefinitely at offset 9 + +/(? No match +/.*a.*=.b.*/8 + QQQ\x{2029}ABCaXYZ=!bPQR + 0: ABCaXYZ=!bPQR + ** Failers +No match + a\x{2029}b +No match + \x61\xe2\x80\xa9\x62 +No match + +/[[:a\x{100}b:]]/8 +Failed: unknown POSIX class name at offset 3 + / End of testinput5 / diff --git a/ext/pcre/pcrelib/testdata/testoutput6 b/ext/pcre/pcrelib/testdata/testoutput6 index 0a58b844f1..db825b08c1 100644 --- a/ext/pcre/pcrelib/testdata/testoutput6 +++ b/ext/pcre/pcrelib/testdata/testoutput6 @@ -1522,4 +1522,161 @@ No match /(\p{Yi}{0,3}+\277)*/ +/^[\p{Arabic}]/8 + \x{60e} + 0: \x{60e} + \x{656} + 0: \x{656} + \x{657} + 0: \x{657} + \x{658} + 0: \x{658} + \x{659} + 0: \x{659} + \x{65a} + 0: \x{65a} + \x{65b} + 0: \x{65b} + \x{65c} + 0: \x{65c} + \x{65d} + 0: \x{65d} + \x{65e} + 0: \x{65e} + \x{66a} + 0: \x{66a} + \x{6e9} + 0: \x{6e9} + \x{6ef} + 0: \x{6ef} + \x{6fa} + 0: \x{6fa} + ** Failers +No match + \x{600} +No match + \x{650} +No match + \x{651} +No match + \x{652} +No match + \x{653} +No match + \x{654} +No match + \x{655} +No match + \x{65f} +No match + +/^\p{Cyrillic}/8 + \x{1d2b} + 0: \x{1d2b} + +/^\p{Common}/8 + \x{589} + 0: \x{589} + \x{60c} + 0: \x{60c} + \x{61f} + 0: \x{61f} + \x{964} + 0: \x{964} + \x{965} + 0: \x{965} + \x{970} + 0: \x{970} + +/^\p{Inherited}/8 + \x{64b} + 0: \x{64b} + \x{654} + 0: \x{654} + \x{655} + 0: \x{655} + \x{200c} + 0: \x{200c} + ** Failers +No match + \x{64a} +No match + \x{656} +No match + +/^\p{Shavian}/8 + \x{10450} + 0: \x{10450} + \x{1047f} + 0: \x{1047f} + +/^\p{Deseret}/8 + \x{10400} + 0: \x{10400} + \x{1044f} + 0: \x{1044f} + +/^\p{Osmanya}/8 + \x{10480} + 0: \x{10480} + \x{1049d} + 0: \x{1049d} + \x{104a0} + 0: \x{104a0} + \x{104a9} + 0: \x{104a9} + ** Failers +No match + \x{1049e} +No match + \x{1049f} +No match + \x{104aa} +No match + +/\p{Zl}{2,3}+/8BZ +------------------------------------------------------------------ + Bra + prop Zl {2} + prop Zl ?+ + Ket + End +------------------------------------------------------------------ + \xe2\x80\xa8\xe2\x80\xa8 + 0: \x{2028}\x{2028} + \x{2028}\x{2028}\x{2028} + 0: \x{2028}\x{2028}\x{2028} + +/\p{Zl}/8BZ +------------------------------------------------------------------ + Bra + prop Zl + Ket + End +------------------------------------------------------------------ + +/\p{Lu}{3}+/8BZ +------------------------------------------------------------------ + Bra + prop Lu {3} + Ket + End +------------------------------------------------------------------ + +/\pL{2}+/8BZ +------------------------------------------------------------------ + Bra + prop L {2} + Ket + End +------------------------------------------------------------------ + +/\p{Cc}{2}+/8BZ +------------------------------------------------------------------ + Bra + prop Cc {2} + Ket + End +------------------------------------------------------------------ + / End of testinput6 / diff --git a/ext/pcre/pcrelib/ucptable.h b/ext/pcre/pcrelib/ucptable.h index 07eaced8f2..a274d443ee 100644 --- a/ext/pcre/pcrelib/ucptable.h +++ b/ext/pcre/pcrelib/ucptable.h @@ -539,7 +539,8 @@ static const cnode ucp_table[] = { { 0x21000293, 0x14000000 }, { 0x21000294, 0x1c000000 }, { 0x21800295, 0x1400001a }, - { 0x218002b0, 0x18000011 }, + { 0x218002b0, 0x18000008 }, + { 0x098002b9, 0x18000008 }, { 0x098002c2, 0x60000003 }, { 0x098002c6, 0x1800000b }, { 0x098002d2, 0x6000000d }, @@ -1039,15 +1040,18 @@ static const cnode ucp_table[] = { { 0x198005f3, 0x54000001 }, { 0x09800600, 0x04000003 }, { 0x0000060b, 0x5c000000 }, - { 0x0980060c, 0x54000001 }, + { 0x0900060c, 0x54000000 }, + { 0x0000060d, 0x54000000 }, { 0x0080060e, 0x68000001 }, { 0x00800610, 0x30000005 }, { 0x0900061b, 0x54000000 }, - { 0x0080061e, 0x54000001 }, + { 0x0000061e, 0x54000000 }, + { 0x0900061f, 0x54000000 }, { 0x00800621, 0x1c000019 }, { 0x09000640, 0x18000000 }, { 0x00800641, 0x1c000009 }, - { 0x1b80064b, 0x30000013 }, + { 0x1b80064b, 0x3000000a }, + { 0x00800656, 0x30000008 }, { 0x09800660, 0x34000009 }, { 0x0080066a, 0x54000003 }, { 0x0080066e, 0x1c000001 }, @@ -1074,7 +1078,8 @@ static const cnode ucp_table[] = { { 0x31000711, 0x30000000 }, { 0x31800712, 0x1c00001d }, { 0x31800730, 0x3000001a }, - { 0x3180074d, 0x1c000020 }, + { 0x3180074d, 0x1c000002 }, + { 0x00800750, 0x1c00001d }, { 0x37800780, 0x1c000025 }, { 0x378007a6, 0x3000000a }, { 0x370007b1, 0x1c000000 }, @@ -1460,7 +1465,10 @@ static const cnode ucp_table[] = { { 0x1f0017dd, 0x30000000 }, { 0x1f8017e0, 0x34000009 }, { 0x1f8017f0, 0x3c000009 }, - { 0x25801800, 0x54000005 }, + { 0x25801800, 0x54000001 }, + { 0x09801802, 0x54000001 }, + { 0x25001804, 0x54000000 }, + { 0x09001805, 0x54000000 }, { 0x25001806, 0x44000000 }, { 0x25801807, 0x54000003 }, { 0x2580180b, 0x30000002 }, @@ -1513,14 +1521,20 @@ static const cnode ucp_table[] = { { 0x3d801b61, 0x68000009 }, { 0x3d801b6b, 0x30000008 }, { 0x3d801b74, 0x68000008 }, - { 0x21801d00, 0x1400002b }, - { 0x21801d2c, 0x18000035 }, - { 0x21801d62, 0x14000015 }, + { 0x21801d00, 0x14000025 }, + { 0x13801d26, 0x14000004 }, + { 0x0c001d2b, 0x14000000 }, + { 0x21801d2c, 0x18000030 }, + { 0x13801d5d, 0x18000004 }, + { 0x21801d62, 0x14000003 }, + { 0x13801d66, 0x14000004 }, + { 0x21801d6b, 0x1400000c }, { 0x0c001d78, 0x18000000 }, { 0x21801d79, 0x14000003 }, { 0x21001d7d, 0x14000ee6 }, { 0x21801d7e, 0x1400001c }, - { 0x21801d9b, 0x18000024 }, + { 0x21801d9b, 0x18000023 }, + { 0x13001dbf, 0x18000000 }, { 0x1b801dc0, 0x3000000a }, { 0x1b801dfe, 0x30000001 }, { 0x21001e00, 0x24000001 }, @@ -1982,7 +1996,9 @@ static const cnode ucp_table[] = { { 0x13001ffc, 0x2000fff7 }, { 0x13801ffd, 0x60000001 }, { 0x09802000, 0x7400000a }, - { 0x0980200b, 0x04000004 }, + { 0x0900200b, 0x04000000 }, + { 0x1b80200c, 0x04000001 }, + { 0x0980200e, 0x04000001 }, { 0x09802010, 0x44000005 }, { 0x09802016, 0x54000001 }, { 0x09002018, 0x50000000 }, @@ -2615,7 +2631,8 @@ static const cnode ucp_table[] = { { 0x090030a0, 0x44000000 }, { 0x1d8030a1, 0x1c000059 }, { 0x090030fb, 0x54000000 }, - { 0x098030fc, 0x18000002 }, + { 0x090030fc, 0x18000000 }, + { 0x1d8030fd, 0x18000001 }, { 0x1d0030ff, 0x1c000000 }, { 0x03803105, 0x1c000027 }, { 0x17803131, 0x1c00005d }, @@ -2630,7 +2647,8 @@ static const cnode ucp_table[] = { { 0x0980322a, 0x68000019 }, { 0x09003250, 0x68000000 }, { 0x09803251, 0x3c00000e }, - { 0x17803260, 0x6800001f }, + { 0x17803260, 0x6800001d }, + { 0x0980327e, 0x68000001 }, { 0x09803280, 0x3c000009 }, { 0x0980328a, 0x68000026 }, { 0x098032b1, 0x3c00000e }, @@ -2678,7 +2696,8 @@ static const cnode ucp_table[] = { { 0x1900fb3e, 0x1c000000 }, { 0x1980fb40, 0x1c000001 }, { 0x1980fb43, 0x1c000001 }, - { 0x1980fb46, 0x1c00006b }, + { 0x1980fb46, 0x1c000009 }, + { 0x0080fb50, 0x1c000061 }, { 0x0080fbd3, 0x1c00016a }, { 0x0900fd3e, 0x58000000 }, { 0x0900fd3f, 0x48000000 }, @@ -2944,7 +2963,8 @@ static const cnode ucp_table[] = { { 0x0d01044d, 0x1400ffd8 }, { 0x0d01044e, 0x1400ffd8 }, { 0x0d01044f, 0x1400ffd8 }, - { 0x2e810450, 0x1c00004d }, + { 0x2e810450, 0x1c00002f }, + { 0x2c810480, 0x1c00001d }, { 0x2c8104a0, 0x34000009 }, { 0x0b810800, 0x1c000005 }, { 0x0b010808, 0x1c000000 }, -- 2.50.1