fivedev/harbour-core
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

2010-06-26 15:51 UTC+0200 Viktor Szakats (harbour.01 syenar.hu)

Browse Source 
* external/pcre/pcreexec.c
  * external/pcre/pcre.h
  * external/pcre/pcreinal.h
  * external/pcre/pcretabs.c
  * external/pcre/pcredfa.c
  * external/pcre/pcrecomp.c
  * external/pcre/pcre.dif
  * external/pcre/config.h
  * external/pcre/pcrexcls.c
  * external/pcre/chartabs.c
  * external/pcre/pcreprni.h
  * external/pcre/pcrestud.c
  * external/pcre/Makefile
    * PCRE update to 8.10 (from 8.02)
    ; Thanks to Tamas Tevesz for the patch. This update was done 
      using the new patchup tool.
This commit is contained in:
Viktor Szakats
2010-06-26 13:52:28 +00:00
parent 0d9811d574
commit b4a3d7bc8c
14 changed files with 1795 additions and 689 deletions
Download Patch File Download Diff File Expand all files Collapse all files

18
harbour/ChangeLog
View File

@@ -16,6 +16,24 @@
The license applies to all entries newer than 2009-04-28.
*/
2010-06-26 15:51 UTC+0200 Viktor Szakats (harbour.01 syenar.hu)
* external/pcre/pcreexec.c
* external/pcre/pcre.h
* external/pcre/pcreinal.h
* external/pcre/pcretabs.c
* external/pcre/pcredfa.c
* external/pcre/pcrecomp.c
* external/pcre/pcre.dif
* external/pcre/config.h
* external/pcre/pcrexcls.c
* external/pcre/chartabs.c
* external/pcre/pcreprni.h
* external/pcre/pcrestud.c
* external/pcre/Makefile
* PCRE update to 8.10 (from 8.02)
; Thanks to Tamas Tevesz for the patch. This update was done
using the new patchup tool.
2010-06-26 15:25 UTC+0200 Viktor Szakats (harbour.01 syenar.hu)
* utils/hbmk2/hbmk2.prg
+ -hbimplib mode now respects -clean option.

4
harbour/external/pcre/Makefile vendored
View File

@@ -68,8 +68,8 @@ else
endif
# ORIGIN http://www.pcre.org/
# VER 8.02
# URL http://sourceforge.net/projects/pcre/files/pcre/8.02/pcre-8.02.zip/download
# VER 8.10
# URL http://sourceforge.net/projects/pcre/files/pcre/8.10/pcre-8.10.zip/download
# DIFF pcre.dif
#
# MAP LICENCE

2
harbour/external/pcre/chartabs.c vendored
View File

@@ -14,7 +14,7 @@ example ISO-8859-1. When dftables is run, it creates these tables in the
current locale. If PCRE is configured with --enable-rebuild-chartables, this
happens automatically.
The following #includes are present because without the gcc 4.x may remove the
The following #includes are present because without them gcc 4.x may remove the
array definition from the final binary if PCRE is built into a static library
and dead code stripping is activated. This leads to link errors. Pulling in the
header ensures that the array gets flagged as "someone outside this compilation

9
harbour/external/pcre/config.h vendored
View File

@@ -250,13 +250,16 @@ 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 8.02"
#define PACKAGE_STRING "PCRE 8.10"
/* Define to the one symbol short name of this package. */
#define PACKAGE_TARNAME "pcre"
/* Define to the home page for this package. */
#define PACKAGE_URL ""
/* Define to the version of this package. */
#define PACKAGE_VERSION "8.02"
#define PACKAGE_VERSION "8.10"
/* If you are compiling for a system other than a Unix-like system or
@@ -312,7 +315,7 @@ them both to 0; an emulation function will be used. */
/* Version number of package */
#ifndef VERSION
#define VERSION "8.02"
#define VERSION "8.10"
#endif
/* Define to empty if `const' does not conform to ANSI C. */

8
harbour/external/pcre/pcre.dif vendored
View File

@@ -1,6 +1,6 @@
diff -urN pcre.orig/pcrefinf.c pcre/pcrefinf.c
--- pcre.orig/pcrefinf.c 2010-06-13 18:04:52.000000000 +0200
+++ pcre/pcrefinf.c 2010-06-11 04:09:57.000000000 +0200
--- pcre.orig/pcrefinf.c 2010-06-26 14:10:17.887330037 +0200
+++ pcre/pcrefinf.c 2010-06-26 14:10:18.107331394 +0200
@@ -126,7 +126,7 @@
case PCRE_INFO_MINLENGTH:
*((int *)where) =
@@ -11,8 +11,8 @@ diff -urN pcre.orig/pcrefinf.c pcre/pcrefinf.c
case PCRE_INFO_LASTLITERAL:
diff -urN pcre.orig/pcreglob.c pcre/pcreglob.c
--- pcre.orig/pcreglob.c 2010-06-13 18:04:52.000000000 +0200
+++ pcre/pcreglob.c 2010-06-11 04:09:57.000000000 +0200
--- pcre.orig/pcreglob.c 2010-06-26 14:10:17.907330491 +0200
+++ pcre/pcreglob.c 2010-06-26 14:10:18.107331394 +0200
@@ -74,11 +74,17 @@
PCRE_EXP_DATA_DEFN int (*pcre_callout)(pcre_callout_block *) = NULL;

9
harbour/external/pcre/pcre.h vendored
View File

@@ -5,7 +5,7 @@
/* This is the public header file for the PCRE library, to be #included by
applications that call the PCRE functions.
Copyright (c) 1997-2009 University of Cambridge
Copyright (c) 1997-2010 University of Cambridge
-----------------------------------------------------------------------------
Redistribution and use in source and binary forms, with or without
@@ -42,9 +42,9 @@ POSSIBILITY OF SUCH DAMAGE.
/* The current PCRE version information. */
#define PCRE_MAJOR 8
#define PCRE_MINOR 02
#define PCRE_MINOR 10
#define PCRE_PRERELEASE
#define PCRE_DATE 2010-03-19
#define PCRE_DATE 2010-06-25
/* 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
@@ -131,6 +131,7 @@ both, so we keep them all distinct. */
#define PCRE_NO_START_OPTIMISE 0x04000000
#define PCRE_PARTIAL_HARD 0x08000000
#define PCRE_NOTEMPTY_ATSTART 0x10000000
#define PCRE_UCP 0x20000000
/* Exec-time and get/set-time error codes */
@@ -200,6 +201,7 @@ these bits, just add new ones on the end, in order to remain compatible. */
#define PCRE_EXTRA_CALLOUT_DATA 0x0004
#define PCRE_EXTRA_TABLES 0x0008
#define PCRE_EXTRA_MATCH_LIMIT_RECURSION 0x0010
#define PCRE_EXTRA_MARK 0x0020
/* Types */
@@ -225,6 +227,7 @@ typedef struct pcre_extra {
void *callout_data; /* Data passed back in callouts */
const unsigned char *tables; /* Pointer to character tables */
unsigned long int match_limit_recursion; /* Max recursive calls to match() */
unsigned char **mark; /* For passing back a mark pointer */
} pcre_extra;
/* The structure for passing out data via the pcre_callout_function. We use a

750
harbour/external/pcre/pcrecomp.c vendored
View File

File diff suppressed because it is too large Load Diff

217
harbour/external/pcre/pcredfa.c vendored
View File

@@ -106,7 +106,7 @@ never stored, so we push them well clear of the normal opcodes. */
/* This table identifies those opcodes that are followed immediately by a
character that is to be tested in some way. This makes is possible to
character that is to be tested in some way. This makes it possible to
centralize the loading of these characters. In the case of Type * etc, the
"character" is the opcode for \D, \d, \S, \s, \W, or \w, which will always be a
small value. Non-zero values in the table are the offsets from the opcode where
@@ -161,8 +161,9 @@ static const uschar coptable[] = {
0, 0, /* RREF, NRREF */
0, /* DEF */
0, 0, /* BRAZERO, BRAMINZERO */
0, 0, 0, 0, /* PRUNE, SKIP, THEN, COMMIT */
0, 0, 0, 0 /* FAIL, ACCEPT, CLOSE, SKIPZERO */
0, 0, 0, /* MARK, PRUNE, PRUNE_ARG, */
0, 0, 0, 0, /* SKIP, SKIP_ARG, THEN, THEN_ARG, */
0, 0, 0, 0, 0 /* COMMIT, FAIL, ACCEPT, CLOSE, SKIPZERO */
};
/* This table identifies those opcodes that inspect a character. It is used to
@@ -218,8 +219,9 @@ static const uschar poptable[] = {
0, 0, /* RREF, NRREF */
0, /* DEF */
0, 0, /* BRAZERO, BRAMINZERO */
0, 0, 0, 0, /* PRUNE, SKIP, THEN, COMMIT */
0, 0, 0, 0 /* FAIL, ACCEPT, CLOSE, SKIPZERO */
0, 0, 0, /* MARK, PRUNE, PRUNE_ARG, */
0, 0, 0, 0, /* SKIP, SKIP_ARG, THEN, THEN_ARG, */
0, 0, 0, 0, 0 /* COMMIT, FAIL, ACCEPT, CLOSE, SKIPZERO */
};
/* These 2 tables allow for compact code for testing for \D, \d, \S, \s, \W,
@@ -473,7 +475,7 @@ if (*first_op == OP_REVERSE)
{
gone_back = (current_subject - max_back < start_subject)?
current_subject - start_subject : max_back;
(int)(current_subject - start_subject) : max_back;
current_subject -= gone_back;
}
@@ -490,7 +492,7 @@ if (*first_op == OP_REVERSE)
int back = GET(end_code, 2+LINK_SIZE);
if (back <= gone_back)
{
int bstate = end_code - start_code + 2 + 2*LINK_SIZE;
int bstate = (int)(end_code - start_code + 2 + 2*LINK_SIZE);
ADD_NEW_DATA(-bstate, 0, gone_back - back);
}
end_code += GET(end_code, 1);
@@ -526,7 +528,7 @@ else
((*this_start_code == OP_CBRA || *this_start_code == OP_SCBRA)? 2:0);
do
{
ADD_NEW(end_code - start_code + length, 0);
ADD_NEW((int)(end_code - start_code + length), 0);
end_code += GET(end_code, 1);
length = 1 + LINK_SIZE;
}
@@ -753,8 +755,8 @@ for (;;)
if (count > 0) memmove(offsets + 2, offsets, count * sizeof(int));
if (offsetcount >= 2)
{
offsets[0] = current_subject - start_subject;
offsets[1] = ptr - start_subject;
offsets[0] = (int)(current_subject - start_subject);
offsets[1] = (int)(ptr - start_subject);
DPRINTF(("%.*sSet matched string = \"%.*s\"\n", rlevel*2-2, SP,
offsets[1] - offsets[0], current_subject));
}
@@ -776,7 +778,7 @@ for (;;)
/*-----------------------------------------------------------------*/
case OP_ALT:
do { code += GET(code, 1); } while (*code == OP_ALT);
ADD_ACTIVE(code - start_code, 0);
ADD_ACTIVE((int)(code - start_code), 0);
break;
/*-----------------------------------------------------------------*/
@@ -784,7 +786,7 @@ for (;;)
case OP_SBRA:
do
{
ADD_ACTIVE(code - start_code + 1 + LINK_SIZE, 0);
ADD_ACTIVE((int)(code - start_code + 1 + LINK_SIZE), 0);
code += GET(code, 1);
}
while (*code == OP_ALT);
@@ -793,11 +795,11 @@ for (;;)
/*-----------------------------------------------------------------*/
case OP_CBRA:
case OP_SCBRA:
ADD_ACTIVE(code - start_code + 3 + LINK_SIZE, 0);
ADD_ACTIVE((int)(code - start_code + 3 + LINK_SIZE), 0);
code += GET(code, 1);
while (*code == OP_ALT)
{
ADD_ACTIVE(code - start_code + 1 + LINK_SIZE, 0);
ADD_ACTIVE((int)(code - start_code + 1 + LINK_SIZE), 0);
code += GET(code, 1);
}
break;
@@ -808,14 +810,14 @@ for (;;)
ADD_ACTIVE(state_offset + 1, 0);
code += 1 + GET(code, 2);
while (*code == OP_ALT) code += GET(code, 1);
ADD_ACTIVE(code - start_code + 1 + LINK_SIZE, 0);
ADD_ACTIVE((int)(code - start_code + 1 + LINK_SIZE), 0);
break;
/*-----------------------------------------------------------------*/
case OP_SKIPZERO:
code += 1 + GET(code, 2);
while (*code == OP_ALT) code += GET(code, 1);
ADD_ACTIVE(code - start_code + 1 + LINK_SIZE, 0);
ADD_ACTIVE((int)(code - start_code + 1 + LINK_SIZE), 0);
break;
/*-----------------------------------------------------------------*/
@@ -920,13 +922,37 @@ for (;;)
if (utf8) BACKCHAR(temp);
#endif
GETCHARTEST(d, temp);
#ifdef SUPPORT_UCP
if ((md->poptions & PCRE_UCP) != 0)
{
if (d == '_') left_word = TRUE; else
{
int cat = UCD_CATEGORY(d);
left_word = (cat == ucp_L || cat == ucp_N);
}
}
else
#endif
left_word = d < 256 && (ctypes[d] & ctype_word) != 0;
}
else left_word = 0;
else left_word = FALSE;
if (clen > 0)
{
#ifdef SUPPORT_UCP
if ((md->poptions & PCRE_UCP) != 0)
{
if (c == '_') right_word = TRUE; else
{
int cat = UCD_CATEGORY(c);
right_word = (cat == ucp_L || cat == ucp_N);
}
}
else
#endif
right_word = c < 256 && (ctypes[c] & ctype_word) != 0;
else right_word = 0;
}
else right_word = FALSE;
if ((left_word == right_word) == (codevalue == OP_NOT_WORD_BOUNDARY))
{ ADD_ACTIVE(state_offset + 1, 0); }
@@ -953,7 +979,8 @@ for (;;)
break;
case PT_LAMP:
OK = prop->chartype == ucp_Lu || prop->chartype == ucp_Ll || prop->chartype == ucp_Lt;
OK = prop->chartype == ucp_Lu || prop->chartype == ucp_Ll ||
prop->chartype == ucp_Lt;
break;
case PT_GC:
@@ -968,6 +995,30 @@ for (;;)
OK = prop->script == code[2];
break;
/* These are specials for combination cases. */
case PT_ALNUM:
OK = _pcre_ucp_gentype[prop->chartype] == ucp_L ||
_pcre_ucp_gentype[prop->chartype] == ucp_N;
break;
case PT_SPACE: /* Perl space */
OK = _pcre_ucp_gentype[prop->chartype] == ucp_Z ||
c == CHAR_HT || c == CHAR_NL || c == CHAR_FF || c == CHAR_CR;
break;
case PT_PXSPACE: /* POSIX space */
OK = _pcre_ucp_gentype[prop->chartype] == ucp_Z ||
c == CHAR_HT || c == CHAR_NL || c == CHAR_VT ||
c == CHAR_FF || c == CHAR_CR;
break;
case PT_WORD:
OK = _pcre_ucp_gentype[prop->chartype] == ucp_L ||
_pcre_ucp_gentype[prop->chartype] == ucp_N ||
c == CHAR_UNDERSCORE;
break;
/* Should never occur, but keep compilers from grumbling. */
default:
@@ -1122,7 +1173,8 @@ for (;;)
break;
case PT_LAMP:
OK = prop->chartype == ucp_Lu || prop->chartype == ucp_Ll || prop->chartype == ucp_Lt;
OK = prop->chartype == ucp_Lu || prop->chartype == ucp_Ll ||
prop->chartype == ucp_Lt;
break;
case PT_GC:
@@ -1137,6 +1189,30 @@ for (;;)
OK = prop->script == code[3];
break;
/* These are specials for combination cases. */
case PT_ALNUM:
OK = _pcre_ucp_gentype[prop->chartype] == ucp_L ||
_pcre_ucp_gentype[prop->chartype] == ucp_N;
break;
case PT_SPACE: /* Perl space */
OK = _pcre_ucp_gentype[prop->chartype] == ucp_Z ||
c == CHAR_HT || c == CHAR_NL || c == CHAR_FF || c == CHAR_CR;
break;
case PT_PXSPACE: /* POSIX space */
OK = _pcre_ucp_gentype[prop->chartype] == ucp_Z ||
c == CHAR_HT || c == CHAR_NL || c == CHAR_VT ||
c == CHAR_FF || c == CHAR_CR;
break;
case PT_WORD:
OK = _pcre_ucp_gentype[prop->chartype] == ucp_L ||
_pcre_ucp_gentype[prop->chartype] == ucp_N ||
c == CHAR_UNDERSCORE;
break;
/* Should never occur, but keep compilers from grumbling. */
default:
@@ -1344,7 +1420,8 @@ for (;;)
break;
case PT_LAMP:
OK = prop->chartype == ucp_Lu || prop->chartype == ucp_Ll || prop->chartype == ucp_Lt;
OK = prop->chartype == ucp_Lu || prop->chartype == ucp_Ll ||
prop->chartype == ucp_Lt;
break;
case PT_GC:
@@ -1359,6 +1436,30 @@ for (;;)
OK = prop->script == code[3];
break;
/* These are specials for combination cases. */
case PT_ALNUM:
OK = _pcre_ucp_gentype[prop->chartype] == ucp_L ||
_pcre_ucp_gentype[prop->chartype] == ucp_N;
break;
case PT_SPACE: /* Perl space */
OK = _pcre_ucp_gentype[prop->chartype] == ucp_Z ||
c == CHAR_HT || c == CHAR_NL || c == CHAR_FF || c == CHAR_CR;
break;
case PT_PXSPACE: /* POSIX space */
OK = _pcre_ucp_gentype[prop->chartype] == ucp_Z ||
c == CHAR_HT || c == CHAR_NL || c == CHAR_VT ||
c == CHAR_FF || c == CHAR_CR;
break;
case PT_WORD:
OK = _pcre_ucp_gentype[prop->chartype] == ucp_L ||
_pcre_ucp_gentype[prop->chartype] == ucp_N ||
c == CHAR_UNDERSCORE;
break;
/* Should never occur, but keep compilers from grumbling. */
default:
@@ -1591,7 +1692,8 @@ for (;;)
break;
case PT_LAMP:
OK = prop->chartype == ucp_Lu || prop->chartype == ucp_Ll || prop->chartype == ucp_Lt;
OK = prop->chartype == ucp_Lu || prop->chartype == ucp_Ll ||
prop->chartype == ucp_Lt;
break;
case PT_GC:
@@ -1606,6 +1708,30 @@ for (;;)
OK = prop->script == code[5];
break;
/* These are specials for combination cases. */
case PT_ALNUM:
OK = _pcre_ucp_gentype[prop->chartype] == ucp_L ||
_pcre_ucp_gentype[prop->chartype] == ucp_N;
break;
case PT_SPACE: /* Perl space */
OK = _pcre_ucp_gentype[prop->chartype] == ucp_Z ||
c == CHAR_HT || c == CHAR_NL || c == CHAR_FF || c == CHAR_CR;
break;
case PT_PXSPACE: /* POSIX space */
OK = _pcre_ucp_gentype[prop->chartype] == ucp_Z ||
c == CHAR_HT || c == CHAR_NL || c == CHAR_VT ||
c == CHAR_FF || c == CHAR_CR;
break;
case PT_WORD:
OK = _pcre_ucp_gentype[prop->chartype] == ucp_L ||
_pcre_ucp_gentype[prop->chartype] == ucp_N ||
c == CHAR_UNDERSCORE;
break;
/* Should never occur, but keep compilers from grumbling. */
default:
@@ -2233,7 +2359,7 @@ for (;;)
points to the byte after the end of the class. If there is a
quantifier, this is where it will be. */
next_state_offset = ecode - start_code;
next_state_offset = (int)(ecode - start_code);
switch (*ecode)
{
@@ -2304,7 +2430,7 @@ for (;;)
md, /* static match data */
code, /* this subexpression's code */
ptr, /* where we currently are */
ptr - start_subject, /* start offset */
(int)(ptr - start_subject), /* start offset */
local_offsets, /* offset vector */
sizeof(local_offsets)/sizeof(int), /* size of same */
local_workspace, /* workspace vector */
@@ -2315,7 +2441,7 @@ for (;;)
if (rc == PCRE_ERROR_DFA_UITEM) return rc;
if ((rc >= 0) == (codevalue == OP_ASSERT || codevalue == OP_ASSERTBACK))
{ ADD_ACTIVE(endasscode + LINK_SIZE + 1 - start_code, 0); }
{ ADD_ACTIVE((int)(endasscode + LINK_SIZE + 1 - start_code), 0); }
}
break;
@@ -2342,9 +2468,9 @@ for (;;)
cb.callout_number = code[LINK_SIZE+2];
cb.offset_vector = offsets;
cb.subject = (PCRE_SPTR)start_subject;
cb.subject_length = end_subject - start_subject;
cb.start_match = current_subject - start_subject;
cb.current_position = ptr - start_subject;
cb.subject_length = (int)(end_subject - start_subject);
cb.start_match = (int)(current_subject - start_subject);
cb.current_position = (int)(ptr - start_subject);
cb.pattern_position = GET(code, LINK_SIZE + 3);
cb.next_item_length = GET(code, 3 + 2*LINK_SIZE);
cb.capture_top = 1;
@@ -2395,7 +2521,7 @@ for (;;)
md, /* fixed match data */
asscode, /* this subexpression's code */
ptr, /* where we currently are */
ptr - start_subject, /* start offset */
(int)(ptr - start_subject), /* start offset */
local_offsets, /* offset vector */
sizeof(local_offsets)/sizeof(int), /* size of same */
local_workspace, /* workspace vector */
@@ -2407,7 +2533,7 @@ for (;;)
if (rc == PCRE_ERROR_DFA_UITEM) return rc;
if ((rc >= 0) ==
(condcode == OP_ASSERT || condcode == OP_ASSERTBACK))
{ ADD_ACTIVE(endasscode + LINK_SIZE + 1 - start_code, 0); }
{ ADD_ACTIVE((int)(endasscode + LINK_SIZE + 1 - start_code), 0); }
else
{ ADD_ACTIVE(state_offset + codelink + LINK_SIZE + 1, 0); }
}
@@ -2428,7 +2554,7 @@ for (;;)
md, /* fixed match data */
start_code + GET(code, 1), /* this subexpression's code */
ptr, /* where we currently are */
ptr - start_subject, /* start offset */
(int)(ptr - start_subject), /* start offset */
local_offsets, /* offset vector */
sizeof(local_offsets)/sizeof(int), /* size of same */
local_workspace, /* workspace vector */
@@ -2480,7 +2606,7 @@ for (;;)
md, /* fixed match data */
code, /* this subexpression's code */
ptr, /* where we currently are */
ptr - start_subject, /* start offset */
(int)(ptr - start_subject), /* start offset */
local_offsets, /* offset vector */
sizeof(local_offsets)/sizeof(int), /* size of same */
local_workspace, /* workspace vector */
@@ -2497,7 +2623,8 @@ for (;;)
do { end_subpattern += GET(end_subpattern, 1); }
while (*end_subpattern == OP_ALT);
next_state_offset = end_subpattern - start_code + LINK_SIZE + 1;
next_state_offset =
(int)(end_subpattern - start_code + LINK_SIZE + 1);
/* If the end of this subpattern is KETRMAX or KETRMIN, we must
arrange for the repeat state also to be added to the relevant list.
@@ -2505,7 +2632,7 @@ for (;;)
repeat_state_offset = (*end_subpattern == OP_KETRMAX ||
*end_subpattern == OP_KETRMIN)?
end_subpattern - start_code - GET(end_subpattern, 1) : -1;
(int)(end_subpattern - start_code - GET(end_subpattern, 1)) : -1;
/* If we have matched an empty string, add the next state at the
current character pointer. This is important so that the duplicate
@@ -2569,9 +2696,9 @@ for (;;)
cb.callout_number = code[1];
cb.offset_vector = offsets;
cb.subject = (PCRE_SPTR)start_subject;
cb.subject_length = end_subject - start_subject;
cb.start_match = current_subject - start_subject;
cb.current_position = ptr - start_subject;
cb.subject_length = (int)(end_subject - start_subject);
cb.start_match = (int)(current_subject - start_subject);
cb.current_position = (int)(ptr - start_subject);
cb.pattern_position = GET(code, 2);
cb.next_item_length = GET(code, 2 + LINK_SIZE);
cb.capture_top = 1;
@@ -2622,8 +2749,8 @@ for (;;)
{
if (offsetcount >= 2)
{
offsets[0] = md->start_used_ptr - start_subject;
offsets[1] = end_subject - start_subject;
offsets[0] = (int)(md->start_used_ptr - start_subject);
offsets[1] = (int)(end_subject - start_subject);
}
match_count = PCRE_ERROR_PARTIAL;
}
@@ -2982,8 +3109,16 @@ for (;;)
while (current_subject < end_subject)
{
register unsigned int c = *current_subject;
if ((start_bits[c/8] & (1 << (c&7))) == 0) current_subject++;
else break;
if ((start_bits[c/8] & (1 << (c&7))) == 0)
{
current_subject++;
#ifdef SUPPORT_UTF8
if (utf8)
while(current_subject < end_subject &&
(*current_subject & 0xc0) == 0x80) current_subject++;
#endif
}
else break;
}
}
}

1051
harbour/external/pcre/pcreexec.c vendored
View File

File diff suppressed because it is too large Load Diff

76
harbour/external/pcre/pcreinal.h vendored
View File

@@ -475,7 +475,8 @@ know we are in UTF-8 mode. */
} \
}
/* Get the next character, testing for UTF-8 mode, and advancing the pointer */
/* Get the next character, testing for UTF-8 mode, and advancing the pointer.
This is called when we don't know if we are in UTF-8 mode. */
#define GETCHARINCTEST(c, eptr) \
c = *eptr++; \
@@ -512,7 +513,7 @@ if there are extra bytes. This is called when we know we are in UTF-8 mode. */
/* Get the next UTF-8 character, testing for UTF-8 mode, not advancing the
pointer, incrementing length if there are extra bytes. This is called when we
know we are in UTF-8 mode. */
do not know if we are in UTF-8 mode. */
#define GETCHARLENTEST(c, eptr, len) \
c = *eptr; \
@@ -580,7 +581,7 @@ time, run time, or study time, respectively. */
PCRE_DOTALL|PCRE_DOLLAR_ENDONLY|PCRE_EXTRA|PCRE_UNGREEDY|PCRE_UTF8| \
PCRE_NO_AUTO_CAPTURE|PCRE_NO_UTF8_CHECK|PCRE_AUTO_CALLOUT|PCRE_FIRSTLINE| \
PCRE_DUPNAMES|PCRE_NEWLINE_BITS|PCRE_BSR_ANYCRLF|PCRE_BSR_UNICODE| \
PCRE_JAVASCRIPT_COMPAT)
PCRE_JAVASCRIPT_COMPAT|PCRE_UCP)
#define PUBLIC_EXEC_OPTIONS \
(PCRE_ANCHORED|PCRE_NOTBOL|PCRE_NOTEOL|PCRE_NOTEMPTY|PCRE_NOTEMPTY_ATSTART| \
@@ -875,6 +876,7 @@ so that PCRE works on both ASCII and EBCDIC platforms, in non-UTF-mode only. */
#define STRING_COMMIT0 "COMMIT\0"
#define STRING_F0 "F\0"
#define STRING_FAIL0 "FAIL\0"
#define STRING_MARK0 "MARK\0"
#define STRING_PRUNE0 "PRUNE\0"
#define STRING_SKIP0 "SKIP\0"
#define STRING_THEN "THEN"
@@ -904,6 +906,7 @@ so that PCRE works on both ASCII and EBCDIC platforms, in non-UTF-mode only. */
#define STRING_BSR_ANYCRLF_RIGHTPAR "BSR_ANYCRLF)"
#define STRING_BSR_UNICODE_RIGHTPAR "BSR_UNICODE)"
#define STRING_UTF8_RIGHTPAR "UTF8)"
#define STRING_UCP_RIGHTPAR "UCP)"
#else /* SUPPORT_UTF8 */
@@ -1127,6 +1130,7 @@ only. */
#define STRING_COMMIT0 STR_C STR_O STR_M STR_M STR_I STR_T "\0"
#define STRING_F0 STR_F "\0"
#define STRING_FAIL0 STR_F STR_A STR_I STR_L "\0"
#define STRING_MARK0 STR_M STR_A STR_R STR_K "\0"
#define STRING_PRUNE0 STR_P STR_R STR_U STR_N STR_E "\0"
#define STRING_SKIP0 STR_S STR_K STR_I STR_P "\0"
#define STRING_THEN STR_T STR_H STR_E STR_N
@@ -1156,6 +1160,7 @@ only. */
#define STRING_BSR_ANYCRLF_RIGHTPAR STR_B STR_S STR_R STR_UNDERSCORE STR_A STR_N STR_Y STR_C STR_R STR_L STR_F STR_RIGHT_PARENTHESIS
#define STRING_BSR_UNICODE_RIGHTPAR STR_B STR_S STR_R STR_UNDERSCORE STR_U STR_N STR_I STR_C STR_O STR_D STR_E STR_RIGHT_PARENTHESIS
#define STRING_UTF8_RIGHTPAR STR_U STR_T STR_F STR_8 STR_RIGHT_PARENTHESIS
#define STRING_UCP_RIGHTPAR STR_U STR_C STR_P STR_RIGHT_PARENTHESIS
#endif /* SUPPORT_UTF8 */
@@ -1188,9 +1193,13 @@ only. */
#define PT_ANY 0 /* Any property - matches all chars */
#define PT_LAMP 1 /* L& - the union of Lu, Ll, Lt */
#define PT_GC 2 /* General characteristic (e.g. L) */
#define PT_PC 3 /* Particular characteristic (e.g. Lu) */
#define PT_GC 2 /* Specified general characteristic (e.g. L) */
#define PT_PC 3 /* Specified particular characteristic (e.g. Lu) */
#define PT_SC 4 /* Script (e.g. Han) */
#define PT_ALNUM 5 /* Alphanumeric - the union of L and N */
#define PT_SPACE 6 /* Perl space - Z plus 9,10,12,13 */
#define PT_PXSPACE 7 /* POSIX space - Z plus 9,10,11,12,13 */
#define PT_WORD 8 /* Word - L plus N plus underscore */
/* Flag bits and data types for the extended class (OP_XCLASS) for classes that
contain UTF-8 characters with values greater than 255. */
@@ -1207,9 +1216,15 @@ contain UTF-8 characters with values greater than 255. */
/* These are escaped items that aren't just an encoding of a particular data
value such as \n. They must have non-zero values, as check_escape() returns
their negation. Also, they must appear in the same order as in the opcode
definitions below, up to ESC_z. There's a dummy for OP_ANY because it
corresponds to "." rather than an escape sequence, and another for OP_ALLANY
(which is used for [^] in JavaScript compatibility mode).
definitions below, up to ESC_z. There's a dummy for OP_ALLANY because it
corresponds to "." in DOTALL mode rather than an escape sequence. It is also
used for [^] in JavaScript compatibility mode. In non-DOTALL mode, "." behaves
like \N.
The special values ESC_DU, ESC_du, etc. are used instead of ESC_D, ESC_d, etc.
when PCRE_UCP is set, when replacement of \d etc by \p sequences is required.
They must be contiguous, and remain in order so that the replacements can be
looked up from a table.
The final escape must be ESC_REF as subsequent values are used for
backreferences (\1, \2, \3, etc). There are two tests in the code for an escape
@@ -1219,11 +1234,12 @@ put in between that don't consume a character, that code will have to change.
*/
enum { ESC_A = 1, ESC_G, ESC_K, ESC_B, ESC_b, ESC_D, ESC_d, ESC_S, ESC_s,
ESC_W, ESC_w, ESC_dum1, ESC_dum2, ESC_C, ESC_P, ESC_p, ESC_R, ESC_H,
ESC_h, ESC_V, ESC_v, ESC_X, ESC_Z, ESC_z, ESC_E, ESC_Q, ESC_g, ESC_k,
ESC_W, ESC_w, ESC_N, ESC_dum, ESC_C, ESC_P, ESC_p, ESC_R, ESC_H,
ESC_h, ESC_V, ESC_v, ESC_X, ESC_Z, ESC_z,
ESC_E, ESC_Q, ESC_g, ESC_k,
ESC_DU, ESC_du, ESC_SU, ESC_su, ESC_WU, ESC_wu,
ESC_REF };
/* Opcode table: Starting from 1 (i.e. after OP_END), the values up to
OP_EOD must correspond in order to the list of escapes immediately above.
@@ -1247,8 +1263,8 @@ enum {
OP_WHITESPACE, /* 9 \s */
OP_NOT_WORDCHAR, /* 10 \W */
OP_WORDCHAR, /* 11 \w */
OP_ANY, /* 12 Match any character (subject to DOTALL) */
OP_ALLANY, /* 13 Match any character (not subject to DOTALL) */
OP_ANY, /* 12 Match any character except newline */
OP_ALLANY, /* 13 Match any character */
OP_ANYBYTE, /* 14 Match any byte (\C); different to OP_ANY for UTF-8 */
OP_NOTPROP, /* 15 \P (not Unicode property) */
OP_PROP, /* 16 \p (Unicode property) */
@@ -1378,20 +1394,24 @@ enum {
/* These are backtracking control verbs */
OP_PRUNE, /* 107 */
OP_SKIP, /* 108 */
OP_THEN, /* 109 */
OP_COMMIT, /* 110 */
OP_MARK, /* 107 always has an argument */
OP_PRUNE, /* 108 */
OP_PRUNE_ARG, /* 109 same, but with argument */
OP_SKIP, /* 110 */
OP_SKIP_ARG, /* 111 same, but with argument */
OP_THEN, /* 112 */
OP_THEN_ARG, /* 113 same, but with argument */
OP_COMMIT, /* 114 */
/* These are forced failure and success verbs */
OP_FAIL, /* 111 */
OP_ACCEPT, /* 112 */
OP_CLOSE, /* 113 Used before OP_ACCEPT to close open captures */
OP_FAIL, /* 115 */
OP_ACCEPT, /* 116 */
OP_CLOSE, /* 117 Used before OP_ACCEPT to close open captures */
/* This is used to skip a subpattern with a {0} quantifier */
OP_SKIPZERO, /* 114 */
OP_SKIPZERO, /* 118 */
/* This is not an opcode, but is used to check that tables indexed by opcode
are the correct length, in order to catch updating errors - there have been
@@ -1402,7 +1422,7 @@ enum {
/* *** NOTE NOTE NOTE *** Whenever the list above is updated, the two macro
definitions that follow must also be updated to match. There are also tables
called "coptable" cna "poptable" in pcre_dfa_exec.c that must be updated. */
called "coptable" and "poptable" in pcre_dfa_exec.c that must be updated. */
/* This macro defines textual names for all the opcodes. These are used only
@@ -1427,7 +1447,8 @@ for debugging. The macro is referenced only in pcre_printint.c. */
"Once", "Bra", "CBra", "Cond", "SBra", "SCBra", "SCond", \
"Cond ref", "Cond nref", "Cond rec", "Cond nrec", "Cond def", \
"Brazero", "Braminzero", \
"*PRUNE", "*SKIP", "*THEN", "*COMMIT", "*FAIL", "*ACCEPT", \
"*MARK", "*PRUNE", "*PRUNE", "*SKIP", "*SKIP", \
"*THEN", "*THEN", "*COMMIT", "*FAIL", "*ACCEPT", \
"Close", "Skip zero"
@@ -1493,8 +1514,9 @@ in UTF-8 mode. The code that uses this table must know about such things. */
3, 3, /* RREF, NRREF */ \
1, /* DEF */ \
1, 1, /* BRAZERO, BRAMINZERO */ \
1, 1, 1, 1, /* PRUNE, SKIP, THEN, COMMIT, */ \
1, 1, 3, 1 /* FAIL, ACCEPT, CLOSE, SKIPZERO */
3, 1, 3, /* MARK, PRUNE, PRUNE_ARG, */ \
1, 3, 1, 3, /* SKIP, SKIP_ARG, THEN, THEN_ARG, */ \
1, 1, 1, 3, 1 /* COMMIT, FAIL, ACCEPT, CLOSE, SKIPZERO */
/* A magic value for OP_RREF and OP_NRREF to indicate the "any recursion"
@@ -1512,7 +1534,7 @@ 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, ERR62, ERR63, ERR64, ERR65, ERRCOUNT };
ERR60, ERR61, ERR62, ERR63, ERR64, ERR65, ERR66, ERR67, ERRCOUNT };
/* 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
@@ -1655,6 +1677,7 @@ typedef struct match_data {
BOOL noteol; /* NOTEOL flag */
BOOL utf8; /* UTF8 flag */
BOOL jscript_compat; /* JAVASCRIPT_COMPAT flag */
BOOL use_ucp; /* PCRE_UCP flag */
BOOL endonly; /* Dollar not before final \n */
BOOL notempty; /* Empty string match not wanted */
BOOL notempty_atstart; /* Empty string match at start not wanted */
@@ -1674,6 +1697,7 @@ typedef struct match_data {
int eptrn; /* Next free eptrblock */
recursion_info *recursive; /* Linked list of recursion data */
void *callout_data; /* To pass back to callouts */
const uschar *mark; /* Mark pointer to pass back */
} match_data;
/* A similar structure is used for the same purpose by the DFA matching

8
harbour/external/pcre/pcreprni.h vendored
View File

@@ -534,6 +534,14 @@ for(;;)
}
break;
case OP_MARK:
case OP_PRUNE_ARG:
case OP_SKIP_ARG:
case OP_THEN_ARG:
fprintf(f, " %s %s", OP_names[*code], code + 2);
extra += code[1];
break;
/* Anything else is just an item with no data*/
default:

274
harbour/external/pcre/pcrestud.c vendored
View File

@@ -48,6 +48,7 @@ supporting functions. */
#include "pcreinal.h"
#define SET_BIT(c) start_bits[c/8] |= (1 << (c&7))
/* Returns from set_start_bits() */
@@ -413,6 +414,15 @@ for (;;)
#endif
break;
/* Skip these, but we need to add in the name length. */
case OP_MARK:
case OP_PRUNE_ARG:
case OP_SKIP_ARG:
case OP_THEN_ARG:
cc += _pcre_OP_lengths[op] + cc[1];
break;
/* For the record, these are the opcodes that are matched by "default":
OP_ACCEPT, OP_CLOSE, OP_COMMIT, OP_FAIL, OP_PRUNE, OP_SET_SOM, OP_SKIP,
OP_THEN. */
@@ -431,25 +441,121 @@ for (;;)
* Set a bit and maybe its alternate case *
*************************************************/
/* Given a character, set its bit in the table, and also the bit for the other
version of a letter if we are caseless.
/* Given a character, set its first byte's bit in the table, and also the
corresponding bit for the other version of a letter if we are caseless. In
UTF-8 mode, for characters greater than 127, we can only do the caseless thing
when Unicode property support is available.
Arguments:
start_bits points to the bit map
c is the character
p points to the character
caseless the caseless flag
cd the block with char table pointers
utf8 TRUE for UTF-8 mode
Returns: nothing
Returns: pointer after the character
*/
static const uschar *
set_table_bit(uschar *start_bits, const uschar *p, BOOL caseless,
compile_data *cd, BOOL utf8)
{
unsigned int c = *p;
SET_BIT(c);
#ifdef SUPPORT_UTF8
if (utf8 && c > 127)
{
GETCHARINC(c, p);
#ifdef SUPPORT_UCP
if (caseless)
{
uschar buff[8];
c = UCD_OTHERCASE(c);
(void)_pcre_ord2utf8(c, buff);
SET_BIT(buff[0]);
}
#endif
return p;
}
#endif
/* Not UTF-8 mode, or character is less than 127. */
if (caseless && (cd->ctypes[c] & ctype_letter) != 0) SET_BIT(cd->fcc[c]);
return p + 1;
}
/*************************************************
* Set bits for a positive character type *
*************************************************/
/* This function sets starting bits for a character type. In UTF-8 mode, we can
only do a direct setting for bytes less than 128, as otherwise there can be
confusion with bytes in the middle of UTF-8 characters. In a "traditional"
environment, the tables will only recognize ASCII characters anyway, but in at
least one Windows environment, some higher bytes bits were set in the tables.
So we deal with that case by considering the UTF-8 encoding.
Arguments:
start_bits the starting bitmap
cbit type the type of character wanted
table_limit 32 for non-UTF-8; 16 for UTF-8
cd the block with char table pointers
Returns: nothing
*/
static void
set_table_bit(uschar *start_bits, unsigned int c, BOOL caseless,
set_type_bits(uschar *start_bits, int cbit_type, int table_limit,
compile_data *cd)
{
start_bits[c/8] |= (1 << (c&7));
if (caseless && (cd->ctypes[c] & ctype_letter) != 0)
start_bits[cd->fcc[c]/8] |= (1 << (cd->fcc[c]&7));
register int c;
for (c = 0; c < table_limit; c++) start_bits[c] |= cd->cbits[c+cbit_type];
if (table_limit == 32) return;
for (c = 128; c < 256; c++)
{
if ((cd->cbits[c/8] & (1 << (c&7))) != 0)
{
uschar buff[8];
(void)_pcre_ord2utf8(c, buff);
SET_BIT(buff[0]);
}
}
}
/*************************************************
* Set bits for a negative character type *
*************************************************/
/* This function sets starting bits for a negative character type such as \D.
In UTF-8 mode, we can only do a direct setting for bytes less than 128, as
otherwise there can be confusion with bytes in the middle of UTF-8 characters.
Unlike in the positive case, where we can set appropriate starting bits for
specific high-valued UTF-8 characters, in this case we have to set the bits for
all high-valued characters. The lowest is 0xc2, but we overkill by starting at
0xc0 (192) for simplicity.
Arguments:
start_bits the starting bitmap
cbit type the type of character wanted
table_limit 32 for non-UTF-8; 16 for UTF-8
cd the block with char table pointers
Returns: nothing
*/
static void
set_nottype_bits(uschar *start_bits, int cbit_type, int table_limit,
compile_data *cd)
{
register int c;
for (c = 0; c < table_limit; c++) start_bits[c] |= ~cd->cbits[c+cbit_type];
if (table_limit != 32) for (c = 24; c < 32; c++) start_bits[c] = 0xff;
}
@@ -484,6 +590,7 @@ set_start_bits(const uschar *code, uschar *start_bits, BOOL caseless,
{
register int c;
int yield = SSB_DONE;
int table_limit = utf8? 16:32;
#if 0
/* ========================================================================= */
@@ -607,12 +714,7 @@ do
case OP_QUERY:
case OP_MINQUERY:
case OP_POSQUERY:
set_table_bit(start_bits, tcode[1], caseless, cd);
tcode += 2;
#ifdef SUPPORT_UTF8
if (utf8 && tcode[-1] >= 0xc0)
tcode += _pcre_utf8_table4[tcode[-1] & 0x3f];
#endif
tcode = set_table_bit(start_bits, tcode + 1, caseless, cd, utf8);
break;
/* Single-char upto sets the bit and tries the next */
@@ -620,12 +722,7 @@ do
case OP_UPTO:
case OP_MINUPTO:
case OP_POSUPTO:
set_table_bit(start_bits, tcode[3], caseless, cd);
tcode += 4;
#ifdef SUPPORT_UTF8
if (utf8 && tcode[-1] >= 0xc0)
tcode += _pcre_utf8_table4[tcode[-1] & 0x3f];
#endif
tcode = set_table_bit(start_bits, tcode + 3, caseless, cd, utf8);
break;
/* At least one single char sets the bit and stops */
@@ -638,59 +735,86 @@ do
case OP_PLUS:
case OP_MINPLUS:
case OP_POSPLUS:
set_table_bit(start_bits, tcode[1], caseless, cd);
(void)set_table_bit(start_bits, tcode + 1, caseless, cd, utf8);
try_next = FALSE;
break;
/* Single character type sets the bits and stops */
/* Special spacing and line-terminating items. These recognize specific
lists of characters. The difference between VSPACE and ANYNL is that the
latter can match the two-character CRLF sequence, but that is not
relevant for finding the first character, so their code here is
identical. */
case OP_HSPACE:
SET_BIT(0x09);
SET_BIT(0x20);
if (utf8)
{
SET_BIT(0xC2); /* For U+00A0 */
SET_BIT(0xE1); /* For U+1680, U+180E */
SET_BIT(0xE2); /* For U+2000 - U+200A, U+202F, U+205F */
SET_BIT(0xE3); /* For U+3000 */
}
else SET_BIT(0xA0);
try_next = FALSE;
break;
case OP_ANYNL:
case OP_VSPACE:
SET_BIT(0x0A);
SET_BIT(0x0B);
SET_BIT(0x0C);
SET_BIT(0x0D);
if (utf8)
{
SET_BIT(0xC2); /* For U+0085 */
SET_BIT(0xE2); /* For U+2028, U+2029 */
}
else SET_BIT(0x85);
try_next = FALSE;
break;
/* Single character types set the bits and stop. Note that if PCRE_UCP
is set, we do not see these op codes because \d etc are converted to
properties. Therefore, these apply in the case when only characters less
than 256 are recognized to match the types. */
case OP_NOT_DIGIT:
for (c = 0; c < 32; c++)
start_bits[c] |= ~cd->cbits[c+cbit_digit];
set_nottype_bits(start_bits, cbit_digit, table_limit, cd);
try_next = FALSE;
break;
case OP_DIGIT:
for (c = 0; c < 32; c++)
start_bits[c] |= cd->cbits[c+cbit_digit];
set_type_bits(start_bits, cbit_digit, table_limit, cd);
try_next = FALSE;
break;
/* The cbit_space table has vertical tab as whitespace; we have to
discard it. */
ensure it is set as not whitespace. */
case OP_NOT_WHITESPACE:
for (c = 0; c < 32; c++)
{
int d = cd->cbits[c+cbit_space];
if (c == 1) d &= ~0x08;
start_bits[c] |= ~d;
}
set_nottype_bits(start_bits, cbit_space, table_limit, cd);
start_bits[1] |= 0x08;
try_next = FALSE;
break;
/* The cbit_space table has vertical tab as whitespace; we have to
discard it. */
not set it from the table. */
case OP_WHITESPACE:
for (c = 0; c < 32; c++)
{
int d = cd->cbits[c+cbit_space];
if (c == 1) d &= ~0x08;
start_bits[c] |= d;
}
c = start_bits[1]; /* Save in case it was already set */
set_type_bits(start_bits, cbit_space, table_limit, cd);
start_bits[1] = (start_bits[1] & ~0x08) | c;
try_next = FALSE;
break;
case OP_NOT_WORDCHAR:
for (c = 0; c < 32; c++)
start_bits[c] |= ~cd->cbits[c+cbit_word];
set_nottype_bits(start_bits, cbit_word, table_limit, cd);
try_next = FALSE;
break;
case OP_WORDCHAR:
for (c = 0; c < 32; c++)
start_bits[c] |= cd->cbits[c+cbit_word];
set_type_bits(start_bits, cbit_word, table_limit, cd);
try_next = FALSE;
break;
@@ -699,6 +823,7 @@ do
case OP_TYPEPLUS:
case OP_TYPEMINPLUS:
case OP_TYPEPOSPLUS:
tcode++;
break;
@@ -722,52 +847,69 @@ do
case OP_TYPEPOSQUERY:
switch(tcode[1])
{
default:
case OP_ANY:
case OP_ALLANY:
return SSB_FAIL;
case OP_HSPACE:
SET_BIT(0x09);
SET_BIT(0x20);
if (utf8)
{
SET_BIT(0xC2); /* For U+00A0 */
SET_BIT(0xE1); /* For U+1680, U+180E */
SET_BIT(0xE2); /* For U+2000 - U+200A, U+202F, U+205F */
SET_BIT(0xE3); /* For U+3000 */
}
else SET_BIT(0xA0);
break;
case OP_ANYNL:
case OP_VSPACE:
SET_BIT(0x0A);
SET_BIT(0x0B);
SET_BIT(0x0C);
SET_BIT(0x0D);
if (utf8)
{
SET_BIT(0xC2); /* For U+0085 */
SET_BIT(0xE2); /* For U+2028, U+2029 */
}
else SET_BIT(0x85);
break;
case OP_NOT_DIGIT:
for (c = 0; c < 32; c++)
start_bits[c] |= ~cd->cbits[c+cbit_digit];
set_nottype_bits(start_bits, cbit_digit, table_limit, cd);
break;
case OP_DIGIT:
for (c = 0; c < 32; c++)
start_bits[c] |= cd->cbits[c+cbit_digit];
set_type_bits(start_bits, cbit_digit, table_limit, cd);
break;
/* The cbit_space table has vertical tab as whitespace; we have to
discard it. */
ensure it gets set as not whitespace. */
case OP_NOT_WHITESPACE:
for (c = 0; c < 32; c++)
{
int d = cd->cbits[c+cbit_space];
if (c == 1) d &= ~0x08;
start_bits[c] |= ~d;
}
set_nottype_bits(start_bits, cbit_space, table_limit, cd);
start_bits[1] |= 0x08;
break;
/* The cbit_space table has vertical tab as whitespace; we have to
discard it. */
avoid setting it. */
case OP_WHITESPACE:
for (c = 0; c < 32; c++)
{
int d = cd->cbits[c+cbit_space];
if (c == 1) d &= ~0x08;
start_bits[c] |= d;
}
c = start_bits[1]; /* Save in case it was already set */
set_type_bits(start_bits, cbit_space, table_limit, cd);
start_bits[1] = (start_bits[1] & ~0x08) | c;
break;
case OP_NOT_WORDCHAR:
for (c = 0; c < 32; c++)
start_bits[c] |= ~cd->cbits[c+cbit_word];
set_nottype_bits(start_bits, cbit_word, table_limit, cd);
break;
case OP_WORDCHAR:
for (c = 0; c < 32; c++)
start_bits[c] |= cd->cbits[c+cbit_word];
set_type_bits(start_bits, cbit_word, table_limit, cd);
break;
}

22
harbour/external/pcre/pcretabs.c vendored
View File

@@ -243,6 +243,10 @@ strings to make sure that UTF-8 support works on EBCDIC platforms. */
#define STRING_Tifinagh0 STR_T STR_i STR_f STR_i STR_n STR_a STR_g STR_h "\0"
#define STRING_Ugaritic0 STR_U STR_g STR_a STR_r STR_i STR_t STR_i STR_c "\0"
#define STRING_Vai0 STR_V STR_a STR_i "\0"
#define STRING_Xan0 STR_X STR_a STR_n "\0"
#define STRING_Xps0 STR_X STR_p STR_s "\0"
#define STRING_Xsp0 STR_X STR_s STR_p "\0"
#define STRING_Xwd0 STR_X STR_w STR_d "\0"
#define STRING_Yi0 STR_Y STR_i "\0"
#define STRING_Z0 STR_Z "\0"
#define STRING_Zl0 STR_Z STR_l "\0"
@@ -376,6 +380,10 @@ const char _pcre_utt_names[] =
STRING_Tifinagh0
STRING_Ugaritic0
STRING_Vai0
STRING_Xan0
STRING_Xps0
STRING_Xsp0
STRING_Xwd0
STRING_Yi0
STRING_Z0
STRING_Zl0
@@ -509,11 +517,15 @@ const ucp_type_table _pcre_utt[] = {
{ 891, PT_SC, ucp_Tifinagh },
{ 900, PT_SC, ucp_Ugaritic },
{ 909, PT_SC, ucp_Vai },
{ 913, PT_SC, ucp_Yi },
{ 916, PT_GC, ucp_Z },
{ 918, PT_PC, ucp_Zl },
{ 921, PT_PC, ucp_Zp },
{ 924, PT_PC, ucp_Zs }
{ 913, PT_ALNUM, 0 },
{ 917, PT_PXSPACE, 0 },
{ 921, PT_SPACE, 0 },
{ 925, PT_WORD, 0 },
{ 929, PT_SC, ucp_Yi },
{ 932, PT_GC, ucp_Z },
{ 934, PT_PC, ucp_Zl },
{ 937, PT_PC, ucp_Zp },
{ 940, PT_PC, ucp_Zs }
};
const int _pcre_utt_size = sizeof(_pcre_utt)/sizeof(ucp_type_table);

36
harbour/external/pcre/pcrexcls.c vendored
View File

@@ -6,7 +6,7 @@
and semantics are as close as possible to those of the Perl 5 language.
Written by Philip Hazel
Copyright (c) 1997-2009 University of Cambridge
Copyright (c) 1997-2010 University of Cambridge
-----------------------------------------------------------------------------
Redistribution and use in source and binary forms, with or without
@@ -112,12 +112,13 @@ while ((t = *data++) != XCL_END)
break;
case PT_LAMP:
if ((prop->chartype == ucp_Lu || prop->chartype == ucp_Ll || prop->chartype == ucp_Lt) ==
(t == XCL_PROP)) return !negated;
if ((prop->chartype == ucp_Lu || prop->chartype == ucp_Ll ||
prop->chartype == ucp_Lt) == (t == XCL_PROP)) return !negated;
break;
case PT_GC:
if ((data[1] == _pcre_ucp_gentype[prop->chartype]) == (t == XCL_PROP)) return !negated;
if ((data[1] == _pcre_ucp_gentype[prop->chartype]) == (t == XCL_PROP))
return !negated;
break;
case PT_PC:
@@ -128,6 +129,33 @@ while ((t = *data++) != XCL_END)
if ((data[1] == prop->script) == (t == XCL_PROP)) return !negated;
break;
case PT_ALNUM:
if ((_pcre_ucp_gentype[prop->chartype] == ucp_L ||
_pcre_ucp_gentype[prop->chartype] == ucp_N) == (t == XCL_PROP))
return !negated;
break;
case PT_SPACE: /* Perl space */
if ((_pcre_ucp_gentype[prop->chartype] == ucp_Z ||
c == CHAR_HT || c == CHAR_NL || c == CHAR_FF || c == CHAR_CR)
== (t == XCL_PROP))
return !negated;
break;
case PT_PXSPACE: /* POSIX space */
if ((_pcre_ucp_gentype[prop->chartype] == ucp_Z ||
c == CHAR_HT || c == CHAR_NL || c == CHAR_VT ||
c == CHAR_FF || c == CHAR_CR) == (t == XCL_PROP))
return !negated;
break;
case PT_WORD:
if ((_pcre_ucp_gentype[prop->chartype] == ucp_L ||
_pcre_ucp_gentype[prop->chartype] == ucp_N || c == CHAR_UNDERSCORE)
== (t == XCL_PROP))
return !negated;
break;
/* This should never occur, but compilers may mutter if there is no
default. */