TDE
/
tdelibs
mirror of https://mirror.git.trinitydesktop.org/gitea/TDE/tdelibs
0
0
Fork 0
Code Issues Releases Wiki Activity

kjs: use libpcre2 instead of libpcre

Browse Source 
Signed-off-by: Michele Calgaro <michele.calgaro@yahoo.it>
pull/295/head
Michele Calgaro 3 months ago
parent b59d51c679
commit 7740e825a6
Signed by: MicheleC
GPG Key ID: 2A75B7CA8ADED5CF
8 changed files with 163 additions and 178 deletions
Show Stats Download Patch File Download Diff File

16
CMakeLists.txt
Unescape View File

@ -111,7 +111,7 @@ option( WITH_OPENEXR "Enable openexr support" ${WITH_ALL_OPTIONS} )
option( WITH_UTEMPTER "Use utempter for utmp management" ${WITH_ALL_OPTIONS} ) option( WITH_UTEMPTER "Use utempter for utmp management" ${WITH_ALL_OPTIONS} )
option( WITH_AVAHI "Enable AVAHI support" ${WITH_ALL_OPTIONS} ) option( WITH_AVAHI "Enable AVAHI support" ${WITH_ALL_OPTIONS} )
option( WITH_ELFICON "Enable ELF embedded icon support" ${WITH_ALL_OPTIONS} ) option( WITH_ELFICON "Enable ELF embedded icon support" ${WITH_ALL_OPTIONS} )
option( WITH_PCRE "Enable pcre regex support for kjs" ON ) option( WITH_PCRE2 "Enable pcre2 regex support for kjs" ON )
option( WITH_GCC_VISIBILITY "Enable fvisibility and fvisibility-inlines-hidden" ${WITH_ALL_OPTIONS} ) option( WITH_GCC_VISIBILITY "Enable fvisibility and fvisibility-inlines-hidden" ${WITH_ALL_OPTIONS} )
option( WITH_INOTIFY "Enable inotify support for tdeio" ON ) option( WITH_INOTIFY "Enable inotify support for tdeio" ON )
option( WITH_GAMIN "Enable FAM/GAMIN support" ${WITH_ALL_OPTIONS} ) option( WITH_GAMIN "Enable FAM/GAMIN support" ${WITH_ALL_OPTIONS} )
@ -1223,14 +1223,14 @@ if( WITH_ELFICON )
endif( ) endif( )
##### check for pcre ########################### ##### check for pcre2 ###########################
if( WITH_PCRE ) if( WITH_PCRE2 )
pkg_search_module( LIBPCRE libpcre ) pkg_check_modules( LIBPCRE2 libpcre2-8 libpcre2-posix )
if( NOT LIBPCRE_FOUND ) if( NOT LIBPCRE2_FOUND )
message(FATAL_ERROR "\npcre support are requested, but not found on your system" ) tde_message_fatal( "pcre2 support was requested, but not found on your system" )
endif( NOT LIBPCRE_FOUND ) endif( )
set( HAVE_PCREPOSIX 1 ) set( HAVE_PCRE2POSIX 1 )
endif( ) endif( )

4
config.h.cmake
Unescape View File

@ -417,8 +417,8 @@
/* Define to 1 if you have the <paths.h> header file. */ /* Define to 1 if you have the <paths.h> header file. */
#cmakedefine HAVE_PATHS_H 1 #cmakedefine HAVE_PATHS_H 1
/* Define if you have pcreposix libraries and header files. */ /* Define if you have pcre2 libraries and header files. */
#cmakedefine HAVE_PCREPOSIX 1 #cmakedefine HAVE_PCRE2POSIX 1
/* Define to 1 if you have the `poll' function. */ /* Define to 1 if you have the `poll' function. */
#cmakedefine HAVE_POLL 1 #cmakedefine HAVE_POLL 1

8
configure.in.bot
Unescape View File

@ -23,12 +23,12 @@ if test -z "$LIBART_CONFIG"; then
all_tests=bad all_tests=bad
fi fi
if test -z "$LIBPCRE"; then if test -z "$LIBPCRE2"; then
echo "" echo ""
echo "You're missing libpcre." echo "You're missing libpcre2."
echo "Some web pages (using regular expressions in Javascript code) will not " echo "Some web pages (using regular expressions in Javascript code) will not "
echo "work correctly, the regexp support being limited without libpcre. " echo "work correctly, the regexp support being limited without libpcre2. "
echo "If you plan to use Konqueror as a Web Browser, download libpcre from " echo "If you plan to use Konqueror as a Web Browser, download libpcre2 from "
echo "http://www.pcre.org or find a binary package for your platform." echo "http://www.pcre.org or find a binary package for your platform."
echo "" echo ""
all_tests=bad all_tests=bad

6
kjs/CMakeLists.txt
Unescape View File

@ -15,12 +15,12 @@ include_directories(
${CMAKE_CURRENT_BINARY_DIR} ${CMAKE_CURRENT_BINARY_DIR}
${CMAKE_BINARY_DIR} ${CMAKE_BINARY_DIR}
${CMAKE_SOURCE_DIR}/tdecore ${CMAKE_SOURCE_DIR}/tdecore
${LIBPCRE_INCLUDEDIR} ${LIBPCRE2_INCLUDEDIR}
) )
link_directories( link_directories(
${TDECORE_LIBRARY_DIRS} ${TDECORE_LIBRARY_DIRS}
${LIBPCRE_LIBDIR} ${LIBPCRE2_LIBDIR}
) )
@ -62,6 +62,6 @@ tde_add_library( ${target} SHARED
SOURCES ${${target}_SRCS} SOURCES ${${target}_SRCS}
VERSION 1.2.0 VERSION 1.2.0
LINK tdecore-shared LINK tdecore-shared
LINK_PRIVATE ${LIBPCRE_LIBRARIES} LINK_PRIVATE ${LIBPCRE2_LIBRARIES}
DESTINATION ${LIB_INSTALL_DIR} DESTINATION ${LIB_INSTALL_DIR}
) )

6
kjs/Makefile.am
Unescape View File

@ -17,7 +17,7 @@
# Boston, MA 02110-1301, USA. # Boston, MA 02110-1301, USA.
YACC = bison YACC = bison
INCLUDES = $(PCRECFLAGS) $(all_includes) INCLUDES = $(PCRE2CFLAGS) $(all_includes)
lib_LTLIBRARIES = libkjs.la lib_LTLIBRARIES = libkjs.la
@ -50,7 +50,7 @@ endif
libkjs_la_LDFLAGS = -version-info 3:0:2 -no-undefined $(VSCRIPT) \ libkjs_la_LDFLAGS = -version-info 3:0:2 -no-undefined $(VSCRIPT) \
$(USER_LDFLAGS) $(all_libraries) $(USER_LDFLAGS) $(all_libraries)
libkjs_la_LIBADD = -lm $(LIBPCRE) libkjs_la_LIBADD = -lm $(LIBPCRE2)
EXTRA_DIST = grammar.y EXTRA_DIST = grammar.y
@ -93,7 +93,7 @@ CLEANFILES = $(LUT_FILES)
## test program (in one program for easier profiling/memory debugging) ## test program (in one program for easier profiling/memory debugging)
EXTRA_PROGRAMS = testkjs_static EXTRA_PROGRAMS = testkjs_static
testkjs_static_SOURCES = testkjs.cpp testkjs_static_SOURCES = testkjs.cpp
testkjs_static_LDADD = $(LIBPCRE) libkjs.la testkjs_static_LDADD = $(LIBPCRE2) libkjs.la
testkjs_static_LDFLAGS = -static testkjs_static_LDFLAGS = -static
## test program (linked to libkjs) ## test program (linked to libkjs)

51
kjs/configure.in.in
Unescape View File

@ -2,52 +2,55 @@ dnl KDE JavaScript specific configure tests
AC_CHECK_HEADERS(ieeefp.h float.h) AC_CHECK_HEADERS(ieeefp.h float.h)
AC_DEFUN([AC_CHECK_PCREPOSIX], AC_DEFUN(AC_CHECK_PCRE2POSIX],
[ [
dnl define the configure option that disables pcre dnl define the configure option that disables pcre2
AC_ARG_ENABLE(pcre,AC_HELP_STRING([--disable-pcre],[don't require libpcre (poor RegExp support in Javascript)]), AC_ARG_ENABLE(pcre2,AC_HELP_STRING([--disable-pcre],[don't require libpcre (poor RegExp support in Javascript)]),
with_pcre=$enableval, with_pcre=yes) with_pcre2=$enableval, with_pcre2=yes)
if test "$with_pcre" = "yes"; then if test "$with_pcre2" = "yes"; then
KDE_FIND_PATH(pcre-config, PCRE_CONFIG, [${exec_prefix}/bin ${prefix}/bin], [PCRE_CONFIG="" ]) KDE_FIND_PATH(pcre2-config, PCRE2_CONFIG, [${exec_prefix}/bin ${prefix}/bin], [PCRE2_CONFIG="" ])
if test -n "$PCRE_CONFIG" && $PCRE_CONFIG --libs >/dev/null 2>&1; then if test -n "$PCRE2_CONFIG" && $PCRE2_CONFIG --libs8 >/dev/null 2>&1; then
LIBPCRE=`$PCRE_CONFIG --libs-posix | sed -e "s,-L/usr/lib ,," -e "s,[\b-].\+pcreposix[^[:space:]]*\b,,"` LIBPCRE2=`$PCRE2_CONFIG --libs-posix | sed -e "s,-L/usr/lib ,," -e "s,[\b-].\+pcreposix[^[:space:]]*\b,,"`
PCRECFLAGS=`$PCRE_CONFIG --cflags` PCRE2CFLAGS=`$PCRE2_CONFIG --cflags`
else else
LIBPCRE="-lpcre" LIBPCRE2="-lpcre2-8"
PCRECFLAGS= PCRE2CFLAGS=
fi fi
AC_CACHE_VAL(ac_cv_have_pcreposix, [ AC_CACHE_VAL(ac_cv_have_pcre2posix, [
ac_save_libs="$LIBS" ac_save_libs="$LIBS"
LIBS="$LIBPCRE" LIBS="$LIBPCRE2"
ac_CPPFLAGS_save="$CPPFLAGS" ac_CPPFLAGS_save="$CPPFLAGS"
CPPFLAGS="$CPPFLAGS $PCRECFLAGS $all_includes" CPPFLAGS="$CPPFLAGS $PCRE2CFLAGS $all_includes"
ac_LDFLAGS_save="$LDFLAGS" ac_LDFLAGS_save="$LDFLAGS"
LDFLAGS="$LDFLAGS $all_libraries" LDFLAGS="$LDFLAGS $all_libraries"
AC_TRY_LINK( AC_TRY_LINK(
[#include <pcre.h>], [
[regfree(0);], #define PCRE2_CODE_UNIT_WIDTH 8
[ac_cv_have_pcreposix="yes"], #include <pcre2.h>
[ac_cv_have_pcreposix="no"] ],
[pcre2_regfree(0);],
[ac_cv_have_pcre2posix="yes"],
[ac_cv_have_pcre2posix="no"]
) )
LIBS="$ac_save_libs" LIBS="$ac_save_libs"
LDFLAGS="$ac_LDFLAGS_save" LDFLAGS="$ac_LDFLAGS_save"
CPPFLAGS="$ac_CPPFLAGS_save" CPPFLAGS="$ac_CPPFLAGS_save"
]) ])
if test "$ac_cv_have_pcreposix" = "yes"; then if test "$ac_cv_have_pcre2posix" = "yes"; then
AC_DEFINE(HAVE_PCREPOSIX, 1, [Define if you have pcreposix libraries and header files.]) AC_DEFINE(HAVE_PCRE2POSIX, 1, [Define if you have pcre2posix libraries and header files.])
else else
AC_MSG_ERROR([You're missing libpcre. AC_MSG_ERROR([You're missing libpcre.
Download libpcre from http://www.pcre.org or find a binary package for your platform. Download libpcre2 from http://www.pcre.org or find a binary package for your platform.
Alternatively, you can specify --disable-pcre, but some web pages - using regular Alternatively, you can specify --disable-pcre, but some web pages - using regular
expressions in Javascript code - will not work correctly, the regexp support being expressions in Javascript code - will not work correctly, the regexp support being
quite limited if libpcre isn't present.]) quite limited if libpcre isn't present.])
fi fi
fi fi
]) ])
AC_CHECK_PCREPOSIX AC_CHECK_PCRE2POSIX
AC_SUBST(LIBPCRE) AC_SUBST(LIBPCRE2)
AC_SUBST(PCRECFLAGS) AC_SUBST(PCRE2CFLAGS)
AM_CONFIG_HEADER([kjs/global.h]) AM_CONFIG_HEADER([kjs/global.h])

229
kjs/regexp.cpp
Unescape View File

@ -30,21 +30,17 @@
using namespace KJS; using namespace KJS;
#ifdef PCRE_CONFIG_UTF8
RegExp::UTF8SupportState RegExp::utf8Support = RegExp::Unknown; RegExp::UTF8SupportState RegExp::utf8Support = RegExp::Unknown;
#endif
RegExp::RegExp(const UString &p, int f) RegExp::RegExp(const UString &p, int f)
: pat(p), flgs(f), m_notEmpty(false), valid(true), buffer(0), originalPos(0) : pat(p), flgs(f), m_notEmpty(false), valid(true), buffer(0), originalPos(0)
{ {
// Determine whether libpcre has unicode support if need be.. // Determine whether libpcre has unicode support if need be..
#ifdef PCRE_CONFIG_UTF8
if (utf8Support == Unknown) { if (utf8Support == Unknown) {
int supported; uint32_t supported;
pcre_config(PCRE_CONFIG_UTF8, (void*)&supported); pcre2_config(PCRE2_CONFIG_COMPILED_WIDTHS, (void*)&supported);
utf8Support = supported ? Supported : Unsupported; utf8Support = (supported & 0x0001) ? Supported : Unsupported;
} }
#endif
nrSubPatterns = 0; // determined in match() with POSIX regex. nrSubPatterns = 0; // determined in match() with POSIX regex.
@ -63,33 +59,33 @@ RegExp::RegExp(const UString &p, int f)
escape = false; escape = false;
// we only care about \u // we only care about \u
if (c == 'u') { if (c == 'u') {
// standard unicode escape sequence looks like \uxxxx but // standard unicode escape sequence looks like \uxxxx but
// other browsers also accept less then 4 hex digits // other browsers also accept less then 4 hex digits
unsigned short u = 0; unsigned short u = 0;
int j = 0; int j = 0;
for (j = 0; j < 4; ++j) { for (j = 0; j < 4; ++j) {
if (i + 1 < p.size() && Lexer::isHexDigit(p[i + 1].unicode())) { if (i + 1 < p.size() && Lexer::isHexDigit(p[i + 1].unicode())) {
u = (u << 4) + Lexer::convertHex(p[i + 1].unicode()); u = (u << 4) + Lexer::convertHex(p[i + 1].unicode());
++i; ++i;
} else { } else {
// sequence incomplete. restore index. // sequence incomplete. restore index.
// TODO: cleaner way to propagate warning // TODO: cleaner way to propagate warning
fprintf(stderr, "KJS: saw %d digit \\u sequence.\n", j); fprintf(stderr, "KJS: saw %d digit \\u sequence.\n", j);
i -= j; i -= j;
break; break;
} }
} }
if (j < 4) { if (j < 4) {
// sequence was incomplete. treat \u as u which IE always // sequence was incomplete. treat \u as u which IE always
// and FF sometimes does. // and FF sometimes does.
intern.append(UString('u')); intern.append(UString('u'));
} else { } else {
c = UChar(u); c = UChar(u);
switch (u) { switch (u) {
case 0: case 0:
// Make sure to encode 0, to avoid terminating the string // Make sure to encode 0, to avoid terminating the string
intern += UString(nil); intern += UString(nil);
break; break;
case '^': case '^':
case '$': case '$':
case '\\': case '\\':
@ -101,13 +97,13 @@ RegExp::RegExp(const UString &p, int f)
case '{': case '}': case '{': case '}':
case '[': case ']': case '[': case ']':
case '|': case '|':
// escape pattern characters have to remain escaped // escape pattern characters have to remain escaped
intern.append(UString('\\')); intern.append(UString('\\'));
// intentional fallthrough // intentional fallthrough
default: default:
intern += UString(&c, 1); intern += UString(&c, 1);
break; break;
} }
} }
continue; continue;
} }
@ -126,45 +122,46 @@ RegExp::RegExp(const UString &p, int f)
intern = p; intern = p;
} }
#ifdef HAVE_PCREPOSIX #ifdef HAVE_PCRE2POSIX
int pcreflags = 0; uint32_t pcre2flags = 0;
const char *perrormsg; int errorCode;
int errorOffset; PCRE2_SIZE errorOffset;
if (flgs & IgnoreCase) if (flgs & IgnoreCase)
pcreflags |= PCRE_CASELESS; pcre2flags |= PCRE2_CASELESS;
if (flgs & Multiline) if (flgs & Multiline)
pcreflags |= PCRE_MULTILINE; pcre2flags |= PCRE2_MULTILINE;
#ifdef PCRE_CONFIG_UTF8
if (utf8Support == Supported) if (utf8Support == Supported)
pcreflags |= (PCRE_UTF8 | PCRE_NO_UTF8_CHECK); pcre2flags |= (PCRE2_UTF | PCRE2_NO_UTF_CHECK);
#endif
// Fill our buffer with an encoded version, whether utf-8, or, // Fill our buffer with an encoded version, whether utf-8, or,
// if PCRE is incapable, truncated. // if PCRE is incapable, truncated.
prepareMatch(intern); prepareMatch(intern);
pcregex = pcre_compile(buffer, pcreflags, pcregex = pcre2_compile(buffer, PCRE2_ZERO_TERMINATED, pcre2flags,
&perrormsg, &errorOffset, NULL); &errorCode, &errorOffset, NULL);
doneMatch(); // Cleanup buffers doneMatch(); // Cleanup buffers
if (!pcregex) { if (!pcregex) {
#ifndef NDEBUG #ifndef NDEBUG
fprintf(stderr, "KJS: pcre_compile() failed with '%s'\n", perrormsg); PCRE2_UCHAR errorMsg[256];
pcre2_get_error_message(errorCode, errorMsg, sizeof(errorMsg));
fprintf(stderr, "KJS: pcre_compile() failed with '%s'\n", errorMsg);
#endif #endif
valid = false; valid = false;
return; return;
} }
#ifdef PCRE_INFO_CAPTURECOUNT
// Get number of subpatterns that will be returned // Get number of subpatterns that will be returned
int rc = pcre_fullinfo( pcregex, NULL, PCRE_INFO_CAPTURECOUNT, &nrSubPatterns); int rc = pcre2_pattern_info(pcregex, PCRE2_INFO_CAPTURECOUNT, &nrSubPatterns);
if (rc != 0) if (rc != 0)
#endif {
nrSubPatterns = 0; // fallback. We always need the first pair of offsets. nrSubPatterns = 0; // fallback. We always need the first pair of offsets.
}
#else /* HAVE_PCREPOSIX */ match_data = pcre2_match_data_create_from_pattern(pcregex, NULL);
#else
int regflags = 0; int regflags = 0;
#ifdef REG_EXTENDED #ifdef REG_EXTENDED
@ -195,9 +192,15 @@ RegExp::RegExp(const UString &p, int f)
RegExp::~RegExp() RegExp::~RegExp()
{ {
doneMatch(); // Be 100% sure buffers are freed doneMatch(); // Be 100% sure buffers are freed
#ifdef HAVE_PCREPOSIX #ifdef HAVE_PCRE2POSIX
if (match_data)
{
pcre2_match_data_free(match_data);
}
if (pcregex) if (pcregex)
pcre_free(pcregex); {
pcre2_code_free(pcregex);
}
#else #else
/* TODO: is this really okay after an error ? */ /* TODO: is this really okay after an error ? */
regfree(&preg); regfree(&preg);
@ -208,7 +211,7 @@ void RegExp::prepareUtf8(const UString& s)
{ {
// Allocate a buffer big enough to hold all the characters plus \0 // Allocate a buffer big enough to hold all the characters plus \0
const int length = s.size(); const int length = s.size();
buffer = new char[length * 3 + 1]; buffer = new buftype_t[length * 3 + 1];
// Also create buffer for positions. We need one extra character in there, // Also create buffer for positions. We need one extra character in there,
// even past the \0 since the non-empty handling may jump one past the end // even past the \0 since the non-empty handling may jump one past the end
@ -217,7 +220,7 @@ void RegExp::prepareUtf8(const UString& s)
// Convert to runs of 8-bit characters, and generate indeces // Convert to runs of 8-bit characters, and generate indeces
// Note that we do NOT combine surrogate pairs here, as // Note that we do NOT combine surrogate pairs here, as
// regexps operate on them as separate characters // regexps operate on them as separate characters
char *p = buffer; buftype_t *p = buffer;
int *posOut = originalPos; int *posOut = originalPos;
const UChar *d = s.data(); const UChar *d = s.data();
for (int i = 0; i != length; ++i) { for (int i = 0; i != length; ++i) {
@ -225,16 +228,16 @@ void RegExp::prepareUtf8(const UString& s)
int sequenceLen; int sequenceLen;
if (c < 0x80) { if (c < 0x80) {
*p++ = (char)c; *p++ = (buftype_t)c;
sequenceLen = 1; sequenceLen = 1;
} else if (c < 0x800) { } else if (c < 0x800) {
*p++ = (char)((c >> 6) | 0xC0); // C0 is the 2-byte flag for UTF-8 *p++ = (buftype_t)((c >> 6) | 0xC0); // C0 is the 2-byte flag for UTF-8
*p++ = (char)((c | 0x80) & 0xBF); // next 6 bits, with high bit set *p++ = (buftype_t)((c | 0x80) & 0xBF); // next 6 bits, with high bit set
sequenceLen = 2; sequenceLen = 2;
} else { } else {
*p++ = (char)((c >> 12) | 0xE0); // E0 is the 3-byte flag for UTF-8 *p++ = (buftype_t)((c >> 12) | 0xE0); // E0 is the 3-byte flag for UTF-8
*p++ = (char)(((c >> 6) | 0x80) & 0xBF); // next 6 bits, with high bit set *p++ = (buftype_t)(((c >> 6) | 0x80) & 0xBF); // next 6 bits, with high bit set
*p++ = (char)((c | 0x80) & 0xBF); // next 6 bits, with high bit set *p++ = (buftype_t)((c | 0x80) & 0xBF); // next 6 bits, with high bit set
sequenceLen = 3; sequenceLen = 3;
} }
@ -262,7 +265,7 @@ void RegExp::prepareASCII (const UString& s)
// when we don't have utf 8 available -- use // when we don't have utf 8 available -- use
// truncated version, and pray for the best // truncated version, and pray for the best
CString truncated = s.cstring(); CString truncated = s.cstring();
buffer = new char[truncated.size() + 1]; buffer = new buftype_t[truncated.size() + 1];
memcpy(buffer, truncated.c_str(), truncated.size()); memcpy(buffer, truncated.c_str(), truncated.size());
buffer[truncated.size()] = '\0'; // For _compile use buffer[truncated.size()] = '\0'; // For _compile use
bufferSize = truncated.size(); bufferSize = truncated.size();
@ -272,11 +275,9 @@ void RegExp::prepareMatch(const UString &s)
{ {
delete[] originalPos; // Just to be sure.. delete[] originalPos; // Just to be sure..
delete[] buffer; delete[] buffer;
#ifdef PCRE_CONFIG_UTF8
if (utf8Support == Supported) if (utf8Support == Supported)
prepareUtf8(s); prepareUtf8(s);
else else
#endif
prepareASCII(s); prepareASCII(s);
#ifndef NDEBUG #ifndef NDEBUG
@ -308,17 +309,16 @@ UString RegExp::match(const UString &s, int i, int *pos, int **ovector)
if (i > s.size() || s.isNull()) if (i > s.size() || s.isNull())
return UString::null; return UString::null;
#ifdef HAVE_PCREPOSIX #ifdef HAVE_PCRE2POSIX
int ovecsize = (nrSubPatterns+1)*3; // see pcre docu if (!pcregex || !match_data)
if (ovector) *ovector = new int[ovecsize]; return UString::null;
if (!pcregex) if (!ovector)
return UString::null; return UString::null;
int startPos; int startPos;
int nextPos; int nextPos;
if (utf8Support == Supported)
#ifdef PCRE_CONFIG_UTF8 {
if (utf8Support == Supported) {
startPos = i; startPos = i;
while (originalPos[startPos] < i) while (originalPos[startPos] < i)
++startPos; ++startPos;
@ -328,53 +328,59 @@ UString RegExp::match(const UString &s, int i, int *pos, int **ovector)
while (originalPos[nextPos] < (i + 1)) while (originalPos[nextPos] < (i + 1))
++nextPos; ++nextPos;
} }
} else }
#endif else
{ {
startPos = i; startPos = i;
nextPos = i + (i < s.size() ? 1 : 0); nextPos = i + (i < s.size() ? 1 : 0);
} }
int baseFlags = uint32_t baseFlags = (utf8Support == Supported ? PCRE2_NO_UTF_CHECK : 0);
#ifdef PCRE_CONFIG_UTF8 if (m_notEmpty)
utf8Support == Supported ? PCRE_NO_UTF8_CHECK : {
#endif baseFlags |= PCRE2_NOTEMPTY | PCRE2_ANCHORED;
0; }
int numMatches = pcre_exec(pcregex, NULL, buffer, bufferSize, startPos, int numMatches = pcre2_match(pcregex, buffer, PCRE2_ZERO_TERMINATED, startPos, baseFlags, match_data, NULL);
m_notEmpty ? (PCRE_NOTEMPTY | PCRE_ANCHORED | baseFlags) : baseFlags, // see man pcretest if (numMatches <= 0)
ovector ? *ovector : 0L, ovecsize);
if (numMatches < 0)
{ {
// Failed to match. // Failed to match.
if (numMatches == PCRE_ERROR_NOMATCH && (flgs & Global) && m_notEmpty && ovector && startPos < nextPos) if (numMatches == PCRE2_ERROR_NOMATCH && (flgs & Global) && m_notEmpty && startPos < nextPos)
{ {
// We set m_notEmpty ourselves, to look for a non-empty match // We set m_notEmpty ourselves, to look for a non-empty match
// (see man pcretest or pcretest.c for details).
// So we don't stop here, we want to try again at i+1. // So we don't stop here, we want to try again at i+1.
#ifdef KJS_VERBOSE #ifdef KJS_VERBOSE
fprintf(stderr, "No match after m_notEmpty. +1 and keep going.\n"); fprintf(stderr, "No match after m_notEmpty. +1 and keep going.\n");
#endif #endif
m_notEmpty = 0; m_notEmpty = 0;
numMatches = pcre_exec(pcregex, NULL, buffer, bufferSize, nextPos, baseFlags, baseFlags = (utf8Support == Supported ? PCRE2_NO_UTF_CHECK : 0);
ovector ? *ovector : 0L, ovecsize); numMatches = pcre2_match(pcregex, buffer, PCRE2_ZERO_TERMINATED, nextPos, baseFlags, match_data, NULL);
if (numMatches < 0) if (numMatches <= 0)
return UString::null; return UString::null;
} }
else // done else
return UString::null; return UString::null;
} }
// Got a match, proceed with it. PCRE2_SIZE *pcre2_ovector = pcre2_get_ovector_pointer(match_data);
// But fix up the ovector if need be.. if (!pcre2_ovector)
if (ovector && originalPos) { return UString::null;
for (unsigned c = 0; c < 2 * TQMIN((unsigned)numMatches, nrSubPatterns+1); ++c) {
if ((*ovector)[c] != -1) uint32_t pcre2_ovecCount = pcre2_get_ovector_count(match_data);
(*ovector)[c] = originalPos[(*ovector)[c]]; *ovector = new int[pcre2_ovecCount * 2];
if (originalPos)
{
for (size_t c = 0; c < 2 * pcre2_ovecCount; ++c)
{
(*ovector)[c] = (pcre2_ovector[c] != -1) ? originalPos[pcre2_ovector[c]] : -1;
}
}
else
{
for (size_t c = 0; c < 2 * pcre2_ovecCount; ++c)
{
(*ovector)[c] = pcre2_ovector[c];
} }
} }
if (!ovector)
return UString::null; // don't rely on the return value if you pass ovector==0
#else #else
const uint maxMatch = 10; const uint maxMatch = 10;
regmatch_t rmatch[maxMatch]; regmatch_t rmatch[maxMatch];
@ -419,28 +425,3 @@ UString RegExp::match(const UString &s, int i, int *pos, int **ovector)
} }
return s.substr((*ovector)[0], (*ovector)[1] - (*ovector)[0]); return s.substr((*ovector)[0], (*ovector)[1] - (*ovector)[0]);
} }
#if 0 // unused
bool RegExp::test(const UString &s, int)
{
#ifdef HAVE_PCREPOSIX
int ovector[300];
CString buffer(s.cstring());
if (s.isNull() ||
pcre_exec(pcregex, NULL, buffer.c_str(), buffer.size(), 0,
0, ovector, 300) == PCRE_ERROR_NOMATCH)
return false;
else
return true;
#else
char *str = strdup(s.ascii());
int r = regexec(&preg, str, 0, 0, 0);
free(str);
return r == 0;
#endif
}
#endif

21
kjs/regexp.h
Unescape View File

@ -25,13 +25,16 @@
#include "config.h" #include "config.h"
#ifdef HAVE_PCREPOSIX #ifdef HAVE_PCRE2POSIX
#include <pcre.h> #define PCRE2_CODE_UNIT_WIDTH 8
#include <pcre2.h>
typedef PCRE2_UCHAR8 buftype_t;
#else // POSIX regex - not so good... #else // POSIX regex - not so good...
extern "C" { // bug with some libc5 distributions extern "C" { // bug with some libc5 distributions
#include <regex.h> #include <regex.h>
typedef char buftype_t;
} }
#endif //HAVE_PCREPOSIX #endif
#include "ustring.h" #include "ustring.h"
@ -61,7 +64,7 @@ namespace KJS {
bool valid; bool valid;
// Cached encoding info... // Cached encoding info...
char* buffer; buftype_t *buffer;
int* originalPos; int* originalPos;
int bufferSize; int bufferSize;
@ -71,22 +74,20 @@ namespace KJS {
UString originalS; // the original string, used for sanity-checking UString originalS; // the original string, used for sanity-checking
#endif #endif
#ifndef HAVE_PCREPOSIX #ifndef HAVE_PCRE2POSIX
regex_t preg; regex_t preg;
#else #else
pcre *pcregex; pcre2_code *pcregex;
pcre2_match_data *match_data;
enum UTF8SupportState { enum UTF8SupportState {
Unknown, Unknown,
Supported, Supported,
Unsupported Unsupported
}; };
#ifdef PCRE_CONFIG_UTF8
static UTF8SupportState utf8Support; static UTF8SupportState utf8Support;
#endif #endif
#endif uint32_t nrSubPatterns;
unsigned int nrSubPatterns;
RegExp(); RegExp();
}; };

Write Preview
Loading…
Cancel
Save