[libxml2] 480323 add code to plug in ICU converters by default
- From: Daniel Veillard <veillard src gnome org>
- To: commits-list gnome org
- Cc:
- Subject: [libxml2] 480323 add code to plug in ICU converters by default
- Date: Thu, 4 Nov 2010 16:44:24 +0000 (UTC)
commit 48f7dcb7242e236d80465615cb6cfc9cdf4cba46
Author: Giuseppe Iuculano <giuseppe iuculano it>
Date: Thu Nov 4 17:42:42 2010 +0100
480323 add code to plug in ICU converters by default
This is not configured in by default but after some serious massaging
incorporate that patch from Chromium/Chrome.
configure.in | 19 +
encoding.c | 809 ++++++++++++++++++++++++++--------------
include/libxml/encoding.h | 14 +
include/libxml/parser.h | 2 +-
include/libxml/xmlversion.h.in | 9 +
parser.c | 6 +
win32/Makefile.msvc | 3 +
win32/configure.js | 16 +-
8 files changed, 592 insertions(+), 286 deletions(-)
---
diff --git a/configure.in b/configure.in
index fcc8f65..4d43796 100644
--- a/configure.in
+++ b/configure.in
@@ -128,6 +128,8 @@ AC_ARG_WITH(http,
[ --with-http add the HTTP support (on)])
AC_ARG_WITH(iconv,
[ --with-iconv[[=DIR]] add ICONV support (on)])
+AC_ARG_WITH(icu,
+[ --with-icu add ICU support (off)])
AC_ARG_WITH(iso8859x,
[ --with-iso8859x add ISO8859X support if no iconv (on)])
AC_ARG_WITH(legacy,
@@ -1321,6 +1323,23 @@ XML_LIBS="-lxml2 $Z_LIBS $THREAD_LIBS $ICONV_LIBS $M_LIBS $LIBS"
XML_LIBTOOLLIBS="libxml2.la"
AC_SUBST(WITH_ICONV)
+WITH_ICU=0
+if test "$with_icu" != "yes" ; then
+ echo Disabling ICU support
+else
+ ICU_CONFIG=icu-config
+ if ${ICU_CONFIG} --cflags >/dev/null 2>&1
+ then
+ ICU_LIBS=`icu-config --ldflags`
+ LDFLAGS="$LDFLAGS $ICU_LIBS"
+ WITH_ICU=1
+ echo Enabling ICU support
+ else
+ AC_MSG_ERROR([libicu config program icu-config not found])
+ fi
+fi
+AC_SUBST(WITH_ICU)
+
WITH_ISO8859X=1
if test "$WITH_ICONV" != "1" ; then
if test "$with_iso8859x" = "no" ; then
diff --git a/encoding.c b/encoding.c
index a68cf21..d1140bf 100644
--- a/encoding.c
+++ b/encoding.c
@@ -1,7 +1,7 @@
/*
* encoding.c : implements the encoding conversion functions needed for XML
*
- * Related specs:
+ * Related specs:
* rfc2044 (UTF-8 and UTF-16) F. Yergeau Alis Technologies
* rfc2781 UTF-16, an encoding of ISO 10646, P. Hoffman, F. Yergeau
* [ISO-10646] UTF-8 and UTF-16 in Annexes
@@ -58,7 +58,7 @@ static xmlCharEncodingAliasPtr xmlCharEncodingAliases = NULL;
static int xmlCharEncodingAliasesNb = 0;
static int xmlCharEncodingAliasesMax = 0;
-#ifdef LIBXML_ICONV_ENABLED
+#if defined(LIBXML_ICONV_ENABLED) || defined(LIBXML_ICU_ENABLED)
#if 0
#define DEBUG_ENCODING /* Define this to get encoding traces */
#endif
@@ -97,6 +97,54 @@ xmlEncodingErr(xmlParserErrors error, const char *msg, const char *val)
NULL, 0, val, NULL, NULL, 0, 0, msg, val);
}
+#ifdef LIBXML_ICU_ENABLED
+static uconv_t*
+openIcuConverter(const char* name, int toUnicode)
+{
+ UErrorCode status = U_ZERO_ERROR;
+ uconv_t *conv = (uconv_t *) xmlMalloc(sizeof(uconv_t));
+ if (conv == NULL)
+ return NULL;
+
+ conv->uconv = ucnv_open(name, &status);
+ if (U_FAILURE(status))
+ goto error;
+
+ status = U_ZERO_ERROR;
+ if (toUnicode) {
+ ucnv_setToUCallBack(conv->uconv, UCNV_TO_U_CALLBACK_STOP,
+ NULL, NULL, NULL, &status);
+ }
+ else {
+ ucnv_setFromUCallBack(conv->uconv, UCNV_FROM_U_CALLBACK_STOP,
+ NULL, NULL, NULL, &status);
+ }
+ if (U_FAILURE(status))
+ goto error;
+
+ status = U_ZERO_ERROR;
+ conv->utf8 = ucnv_open("UTF-8", &status);
+ if (U_SUCCESS(status))
+ return conv;
+
+error:
+ if (conv->uconv)
+ ucnv_close(conv->uconv);
+ xmlFree(conv);
+ return NULL;
+}
+
+static void
+closeIcuConverter(uconv_t *conv)
+{
+ if (conv != NULL) {
+ ucnv_close(conv->uconv);
+ ucnv_close(conv->utf8);
+ xmlFree(conv);
+ }
+}
+#endif /* LIBXML_ICU_ENABLED */
+
/************************************************************************
* *
* Conversions To/From UTF8 encoding *
@@ -135,12 +183,12 @@ asciiToUTF8(unsigned char* out, int *outlen,
break;
if (c < 0x80) {
*out++ = c;
- } else {
+ } else {
*outlen = out - outstart;
*inlen = processed - base;
return(-1);
}
-
+
processed = (const unsigned char*) in;
}
*outlen = out - outstart;
@@ -206,7 +254,7 @@ UTF8Toascii(unsigned char* out, int *outlen,
if (inend - in < trailing) {
break;
- }
+ }
for ( ; trailing; trailing--) {
if ((in >= inend) || (((d= *in++) & 0xC0) != 0x80))
@@ -379,7 +427,7 @@ UTF8Toisolat1(unsigned char* out, int *outlen,
if (inend - in < trailing) {
break;
- }
+ }
for ( ; trailing; trailing--) {
if (in >= inend)
@@ -487,7 +535,7 @@ UTF16LEToUTF8(unsigned char* out, int *outlen,
else if (c < 0x800) { *out++= ((c >> 6) & 0x1F) | 0xC0; bits= 0; }
else if (c < 0x10000) { *out++= ((c >> 12) & 0x0F) | 0xE0; bits= 6; }
else { *out++= ((c >> 18) & 0x07) | 0xF0; bits= 12; }
-
+
for ( ; bits >= 0; bits-= 6) {
if (out >= outend)
break;
@@ -512,7 +560,7 @@ UTF16LEToUTF8(unsigned char* out, int *outlen,
* block of chars out.
*
* Returns the number of bytes written, or -1 if lack of space, or -2
- * if the transcoding failed.
+ * if the transcoding failed.
*/
static int
UTF8ToUTF16LE(unsigned char* outb, int *outlen,
@@ -558,7 +606,7 @@ UTF8ToUTF16LE(unsigned char* outb, int *outlen,
if (inend - in < trailing) {
break;
- }
+ }
for ( ; trailing; trailing--) {
if ((in >= inend) || (((d= *in++) & 0xC0) != 0x80))
@@ -621,7 +669,7 @@ UTF8ToUTF16LE(unsigned char* outb, int *outlen,
* block of chars out.
*
* Returns the number of bytes written, or -1 if lack of space, or -2
- * if the transcoding failed.
+ * if the transcoding failed.
*/
static int
UTF8ToUTF16(unsigned char* outb, int *outlen,
@@ -693,7 +741,7 @@ UTF16BEToUTF8(unsigned char* out, int *outlen,
in++;
} else {
c= *in++;
- }
+ }
if ((c & 0xFC00) == 0xD800) { /* surrogates */
if (in >= inend) { /* (in > inend) shouldn't happens */
*outlen = out - outstart;
@@ -723,15 +771,15 @@ UTF16BEToUTF8(unsigned char* out, int *outlen,
}
/* assertion: c is a single UTF-4 value */
- if (out >= outend)
+ if (out >= outend)
break;
if (c < 0x80) { *out++= c; bits= -6; }
else if (c < 0x800) { *out++= ((c >> 6) & 0x1F) | 0xC0; bits= 0; }
else if (c < 0x10000) { *out++= ((c >> 12) & 0x0F) | 0xE0; bits= 6; }
else { *out++= ((c >> 18) & 0x07) | 0xF0; bits= 12; }
-
+
for ( ; bits >= 0; bits-= 6) {
- if (out >= outend)
+ if (out >= outend)
break;
*out++= ((c >> bits) & 0x3F) | 0x80;
}
@@ -754,7 +802,7 @@ UTF16BEToUTF8(unsigned char* out, int *outlen,
* block of chars out.
*
* Returns the number of byte written, or -1 by lack of space, or -2
- * if the transcoding failed.
+ * if the transcoding failed.
*/
static int
UTF8ToUTF16BE(unsigned char* outb, int *outlen,
@@ -800,7 +848,7 @@ UTF8ToUTF16BE(unsigned char* outb, int *outlen,
if (inend - in < trailing) {
break;
- }
+ }
for ( ; trailing; trailing--) {
if ((in >= inend) || (((d= *in++) & 0xC0) != 0x80)) break;
@@ -864,13 +912,13 @@ UTF8ToUTF16BE(unsigned char* outb, int *outlen,
*
* Guess the encoding of the entity using the first bytes of the entity content
* according to the non-normative appendix F of the XML-1.0 recommendation.
- *
+ *
* Returns one of the XML_CHAR_ENCODING_... values.
*/
xmlCharEncoding
xmlDetectCharEncoding(const unsigned char* in, int len)
{
- if (in == NULL)
+ if (in == NULL)
return(XML_CHAR_ENCODING_NONE);
if (len >= 4) {
if ((in[0] == 0x00) && (in[1] == 0x00) &&
@@ -951,7 +999,7 @@ xmlCleanupEncodingAliases(void) {
* @alias: the alias name as parsed, in UTF-8 format (ASCII actually)
*
* Lookup an encoding name for the given alias.
- *
+ *
* Returns NULL if not found, otherwise the original name
*/
const char *
@@ -989,7 +1037,7 @@ xmlGetEncodingAlias(const char *alias) {
*
* Registers an alias @alias for an encoding named @name. Existing alias
* will be overwritten.
- *
+ *
* Returns 0 in case of success, -1 in case of error
*/
int
@@ -1009,13 +1057,13 @@ xmlAddEncodingAlias(const char *name, const char *alias) {
if (xmlCharEncodingAliases == NULL) {
xmlCharEncodingAliasesNb = 0;
xmlCharEncodingAliasesMax = 20;
- xmlCharEncodingAliases = (xmlCharEncodingAliasPtr)
+ xmlCharEncodingAliases = (xmlCharEncodingAliasPtr)
xmlMalloc(xmlCharEncodingAliasesMax * sizeof(xmlCharEncodingAlias));
if (xmlCharEncodingAliases == NULL)
return(-1);
} else if (xmlCharEncodingAliasesNb >= xmlCharEncodingAliasesMax) {
xmlCharEncodingAliasesMax *= 2;
- xmlCharEncodingAliases = (xmlCharEncodingAliasPtr)
+ xmlCharEncodingAliases = (xmlCharEncodingAliasPtr)
xmlRealloc(xmlCharEncodingAliases,
xmlCharEncodingAliasesMax * sizeof(xmlCharEncodingAlias));
}
@@ -1046,7 +1094,7 @@ xmlAddEncodingAlias(const char *name, const char *alias) {
* @alias: the alias name as parsed, in UTF-8 format (ASCII actually)
*
* Unregisters an encoding alias @alias
- *
+ *
* Returns 0 in case of success, -1 in case of error
*/
int
@@ -1081,7 +1129,7 @@ xmlDelEncodingAlias(const char *alias) {
* Compare the string to the encoding schemes already known. Note
* that the comparison is case insensitive accordingly to the section
* [XML] 4.3.3 Character Encoding in Entities.
- *
+ *
* Returns one of the XML_CHAR_ENCODING_... values or XML_CHAR_ENCODING_NONE
* if not recognized.
*/
@@ -1118,7 +1166,7 @@ xmlParseCharEncoding(const char* name)
*/
if (!strcmp(upper, "UTF-16")) return(XML_CHAR_ENCODING_UTF16LE);
if (!strcmp(upper, "UTF16")) return(XML_CHAR_ENCODING_UTF16LE);
-
+
if (!strcmp(upper, "ISO-10646-UCS-2")) return(XML_CHAR_ENCODING_UCS2);
if (!strcmp(upper, "UCS-2")) return(XML_CHAR_ENCODING_UCS2);
if (!strcmp(upper, "UCS2")) return(XML_CHAR_ENCODING_UCS2);
@@ -1131,7 +1179,7 @@ xmlParseCharEncoding(const char* name)
if (!strcmp(upper, "UCS-4")) return(XML_CHAR_ENCODING_UCS4LE);
if (!strcmp(upper, "UCS4")) return(XML_CHAR_ENCODING_UCS4LE);
-
+
if (!strcmp(upper, "ISO-8859-1")) return(XML_CHAR_ENCODING_8859_1);
if (!strcmp(upper, "ISO-LATIN-1")) return(XML_CHAR_ENCODING_8859_1);
if (!strcmp(upper, "ISO LATIN 1")) return(XML_CHAR_ENCODING_8859_1);
@@ -1254,7 +1302,7 @@ static xmlCharEncodingHandlerPtr xmlDefaultCharEncodingHandler = NULL;
* Returns the xmlCharEncodingHandlerPtr created (or NULL in case of error).
*/
xmlCharEncodingHandlerPtr
-xmlNewCharEncodingHandler(const char *name,
+xmlNewCharEncodingHandler(const char *name,
xmlCharEncodingInputFunc input,
xmlCharEncodingOutputFunc output) {
xmlCharEncodingHandlerPtr handler;
@@ -1299,6 +1347,7 @@ xmlNewCharEncodingHandler(const char *name,
xmlEncodingErrMemory("xmlNewCharEncodingHandler : out of memory !\n");
return(NULL);
}
+ memset(handler, 0, sizeof(xmlCharEncodingHandler));
handler->input = input;
handler->output = output;
handler->name = up;
@@ -1306,7 +1355,11 @@ xmlNewCharEncodingHandler(const char *name,
#ifdef LIBXML_ICONV_ENABLED
handler->iconv_in = NULL;
handler->iconv_out = NULL;
-#endif /* LIBXML_ICONV_ENABLED */
+#endif
+#ifdef LIBXML_ICU_ENABLED
+ handler->uconv_in = NULL;
+ handler->uconv_out = NULL;
+#endif
/*
* registers and returns the handler.
@@ -1330,7 +1383,7 @@ xmlNewCharEncodingHandler(const char *name,
void
xmlInitCharEncodingHandlers(void) {
unsigned short int tst = 0x1234;
- unsigned char *ptr = (unsigned char *) &tst;
+ unsigned char *ptr = (unsigned char *) &tst;
if (handlers != NULL) return;
@@ -1350,9 +1403,9 @@ xmlInitCharEncodingHandlers(void) {
}
xmlNewCharEncodingHandler("UTF-8", UTF8ToUTF8, UTF8ToUTF8);
#ifdef LIBXML_OUTPUT_ENABLED
- xmlUTF16LEHandler =
+ xmlUTF16LEHandler =
xmlNewCharEncodingHandler("UTF-16LE", UTF16LEToUTF8, UTF8ToUTF16LE);
- xmlUTF16BEHandler =
+ xmlUTF16BEHandler =
xmlNewCharEncodingHandler("UTF-16BE", UTF16BEToUTF8, UTF8ToUTF16BE);
xmlNewCharEncodingHandler("UTF-16", UTF16LEToUTF8, UTF8ToUTF16);
xmlNewCharEncodingHandler("ISO-8859-1", isolat1ToUTF8, UTF8Toisolat1);
@@ -1362,16 +1415,16 @@ xmlInitCharEncodingHandlers(void) {
xmlNewCharEncodingHandler("HTML", NULL, UTF8ToHtml);
#endif
#else
- xmlUTF16LEHandler =
+ xmlUTF16LEHandler =
xmlNewCharEncodingHandler("UTF-16LE", UTF16LEToUTF8, NULL);
- xmlUTF16BEHandler =
+ xmlUTF16BEHandler =
xmlNewCharEncodingHandler("UTF-16BE", UTF16BEToUTF8, NULL);
xmlNewCharEncodingHandler("UTF-16", UTF16LEToUTF8, NULL);
xmlNewCharEncodingHandler("ISO-8859-1", isolat1ToUTF8, NULL);
xmlNewCharEncodingHandler("ASCII", asciiToUTF8, NULL);
xmlNewCharEncodingHandler("US-ASCII", asciiToUTF8, NULL);
#endif /* LIBXML_OUTPUT_ENABLED */
-#ifndef LIBXML_ICONV_ENABLED
+#if !defined(LIBXML_ICONV_ENABLED) && !defined(LIBXML_ICU_ENABLED)
#ifdef LIBXML_ISO8859X_ENABLED
xmlRegisterCharEncodingHandlersISO8859x ();
#endif
@@ -1550,10 +1603,10 @@ xmlGetCharEncodingHandler(xmlCharEncoding enc) {
handler = xmlFindCharEncodingHandler("EUC-JP");
if (handler != NULL) return(handler);
break;
- default:
+ default:
break;
}
-
+
#ifdef DEBUG_ENCODING
xmlGenericError(xmlGenericErrorContext,
"No handler found for encoding %d\n", enc);
@@ -1578,6 +1631,10 @@ xmlFindCharEncodingHandler(const char *name) {
xmlCharEncodingHandlerPtr enc;
iconv_t icv_in, icv_out;
#endif /* LIBXML_ICONV_ENABLED */
+#ifdef LIBXML_ICU_ENABLED
+ xmlCharEncodingHandlerPtr encu;
+ uconv_t *ucv_in, *ucv_out;
+#endif /* LIBXML_ICU_ENABLED */
char upper[100];
int i;
@@ -1632,6 +1689,7 @@ xmlFindCharEncodingHandler(const char *name) {
iconv_close(icv_out);
return(NULL);
}
+ memset(enc, 0, sizeof(xmlCharEncodingHandler));
enc->name = xmlMemStrdup(name);
enc->input = NULL;
enc->output = NULL;
@@ -1647,6 +1705,36 @@ xmlFindCharEncodingHandler(const char *name) {
"iconv : problems with filters for '%s'\n", name);
}
#endif /* LIBXML_ICONV_ENABLED */
+#ifdef LIBXML_ICU_ENABLED
+ /* check whether icu can handle this */
+ ucv_in = openIcuConverter(name, 1);
+ ucv_out = openIcuConverter(name, 0);
+ if (ucv_in != NULL && ucv_out != NULL) {
+ encu = (xmlCharEncodingHandlerPtr)
+ xmlMalloc(sizeof(xmlCharEncodingHandler));
+ if (encu == NULL) {
+ closeIcuConverter(ucv_in);
+ closeIcuConverter(ucv_out);
+ return(NULL);
+ }
+ memset(encu, 0, sizeof(xmlCharEncodingHandler));
+ encu->name = xmlMemStrdup(name);
+ encu->input = NULL;
+ encu->output = NULL;
+ encu->uconv_in = ucv_in;
+ encu->uconv_out = ucv_out;
+#ifdef DEBUG_ENCODING
+ xmlGenericError(xmlGenericErrorContext,
+ "Found ICU converter handler for encoding %s\n", name);
+#endif
+ return encu;
+ } else if (ucv_in != NULL || ucv_out != NULL) {
+ closeIcuConverter(ucv_in);
+ closeIcuConverter(ucv_out);
+ xmlEncodingErr(XML_ERR_INTERNAL_ERROR,
+ "ICU converter : problems with filters for '%s'\n", name);
+ }
+#endif /* LIBXML_ICU_ENABLED */
#ifdef DEBUG_ENCODING
xmlGenericError(xmlGenericErrorContext,
@@ -1684,12 +1772,12 @@ xmlFindCharEncodingHandler(const char *name) {
* @in: a pointer to an array of ISO Latin 1 chars
* @inlen: the length of @in
*
- * Returns 0 if success, or
+ * Returns 0 if success, or
* -1 by lack of space, or
* -2 if the transcoding fails (for *in is not valid utf8 string or
* the result of transformation can't fit into the encoding we want), or
* -3 if there the last byte can't form a single output char.
- *
+ *
* The value of @inlen after return is the number of octets consumed
* as the return value is positive, else unpredictable.
* The value of @outlen after return is the number of ocetes consumed.
@@ -1737,6 +1825,75 @@ xmlIconvWrapper(iconv_t cd, unsigned char *out, int *outlen,
/************************************************************************
* *
+ * ICU based generic conversion functions *
+ * *
+ ************************************************************************/
+
+#ifdef LIBXML_ICU_ENABLED
+/**
+ * xmlUconvWrapper:
+ * @cd: ICU uconverter data structure
+ * @toUnicode : non-zero if toUnicode. 0 otherwise.
+ * @out: a pointer to an array of bytes to store the result
+ * @outlen: the length of @out
+ * @in: a pointer to an array of ISO Latin 1 chars
+ * @inlen: the length of @in
+ *
+ * Returns 0 if success, or
+ * -1 by lack of space, or
+ * -2 if the transcoding fails (for *in is not valid utf8 string or
+ * the result of transformation can't fit into the encoding we want), or
+ * -3 if there the last byte can't form a single output char.
+ *
+ * The value of @inlen after return is the number of octets consumed
+ * as the return value is positive, else unpredictable.
+ * The value of @outlen after return is the number of ocetes consumed.
+ */
+static int
+xmlUconvWrapper(uconv_t *cd, int toUnicode, unsigned char *out, int *outlen,
+ const unsigned char *in, int *inlen) {
+ const char *ucv_in = (const char *) in;
+ char *ucv_out = (char *) out;
+ UErrorCode err = U_ZERO_ERROR;
+
+ if ((out == NULL) || (outlen == NULL) || (inlen == NULL) || (in == NULL)) {
+ if (outlen != NULL) *outlen = 0;
+ return(-1);
+ }
+
+ /*
+ * TODO(jungshik)
+ * 1. is ucnv_convert(To|From)Algorithmic better?
+ * 2. had we better use an explicit pivot buffer?
+ * 3. error returned comes from 'fromUnicode' only even
+ * when toUnicode is true !
+ */
+ if (toUnicode) {
+ /* encoding => UTF-16 => UTF-8 */
+ ucnv_convertEx(cd->utf8, cd->uconv, &ucv_out, ucv_out + *outlen,
+ &ucv_in, ucv_in + *inlen, NULL, NULL, NULL, NULL,
+ 0, TRUE, &err);
+ } else {
+ /* UTF-8 => UTF-16 => encoding */
+ ucnv_convertEx(cd->uconv, cd->utf8, &ucv_out, ucv_out + *outlen,
+ &ucv_in, ucv_in + *inlen, NULL, NULL, NULL, NULL,
+ 0, TRUE, &err);
+ }
+ *inlen = ucv_in - (const char*) in;
+ *outlen = ucv_out - (char *) out;
+ if (U_SUCCESS(err))
+ return 0;
+ if (err == U_BUFFER_OVERFLOW_ERROR)
+ return -1;
+ if (err == U_INVALID_CHAR_FOUND || err == U_ILLEGAL_CHAR_FOUND)
+ return -2;
+ /* if (err == U_TRUNCATED_CHAR_FOUND) */
+ return -3;
+}
+#endif /* LIBXML_ICU_ENABLED */
+
+/************************************************************************
+ * *
* The real API used by libxml for on-the-fly conversion *
* *
************************************************************************/
@@ -1810,6 +1967,16 @@ xmlCharEncFirstLineInt(xmlCharEncodingHandler *handler, xmlBufferPtr out,
if (ret == -1) ret = -3;
}
#endif /* LIBXML_ICONV_ENABLED */
+#ifdef LIBXML_ICU_ENABLED
+ else if (handler->uconv_in != NULL) {
+ ret = xmlUconvWrapper(handler->uconv_in, 1, &out->content[out->use],
+ &written, in->content, &toconv);
+ xmlBufferShrink(in, toconv);
+ out->use += written;
+ out->content[out->use] = 0;
+ if (ret == -1) ret = -3;
+ }
+#endif /* LIBXML_ICU_ENABLED */
#ifdef DEBUG_ENCODING
switch (ret) {
case 0:
@@ -1915,6 +2082,17 @@ xmlCharEncInFunc(xmlCharEncodingHandler * handler, xmlBufferPtr out,
ret = -3;
}
#endif /* LIBXML_ICONV_ENABLED */
+#ifdef LIBXML_ICU_ENABLED
+ else if (handler->uconv_in != NULL) {
+ ret = xmlUconvWrapper(handler->uconv_in, 1, &out->content[out->use],
+ &written, in->content, &toconv);
+ xmlBufferShrink(in, toconv);
+ out->use += written;
+ out->content[out->use] = 0;
+ if (ret == -1)
+ ret = -3;
+ }
+#endif /* LIBXML_ICU_ENABLED */
switch (ret) {
case 0:
#ifdef DEBUG_ENCODING
@@ -1940,7 +2118,7 @@ xmlCharEncInFunc(xmlCharEncodingHandler * handler, xmlBufferPtr out,
case -2: {
char buf[50];
- snprintf(&buf[0], 49, "0x%02X 0x%02X 0x%02X 0x%02X",
+ snprintf(&buf[0], 49, "0x%02X 0x%02X 0x%02X 0x%02X",
in->content[0], in->content[1],
in->content[2], in->content[3]);
buf[49] = 0;
@@ -1962,15 +2140,15 @@ xmlCharEncInFunc(xmlCharEncodingHandler * handler, xmlBufferPtr out,
* @handler: char enconding transformation data structure
* @out: an xmlBuffer for the output.
* @in: an xmlBuffer for the input
- *
+ *
* Generic front-end for the encoding handler output function
- * a first call with @in == NULL has to be made firs to initiate the
+ * a first call with @in == NULL has to be made firs to initiate the
* output in case of non-stateless encoding needing to initiate their
* state or the output (like the BOM in UTF16).
* In case of UTF8 sequence conversion errors for the given encoder,
* the content will be automatically remapped to a CharRef sequence.
- *
- * Returns the number of byte written if success, or
+ *
+ * Returns the number of byte written if success, or
* -1 general error
* -2 if the transcoding fails (for *in is not valid utf8 string or
* the result of transformation can't fit into the encoding we want), or
@@ -1988,7 +2166,7 @@ xmlCharEncOutFunc(xmlCharEncodingHandler *handler, xmlBufferPtr out,
if (out == NULL) return(-1);
retry:
-
+
written = out->size - out->use;
if (written > 0)
@@ -2015,6 +2193,15 @@ retry:
out->content[out->use] = 0;
}
#endif /* LIBXML_ICONV_ENABLED */
+#ifdef LIBXML_ICU_ENABLED
+ else if (handler->uconv_out != NULL) {
+ ret = xmlUconvWrapper(handler->uconv_out, 0,
+ &out->content[out->use],
+ &written, NULL, &toconv);
+ out->use += written;
+ out->content[out->use] = 0;
+ }
+#endif /* LIBXML_ICU_ENABLED */
#ifdef DEBUG_ENCODING
xmlGenericError(xmlGenericErrorContext,
"initialized encoder\n");
@@ -2039,7 +2226,7 @@ retry:
xmlBufferShrink(in, toconv);
out->use += written;
writtentot += written;
- }
+ }
out->content[out->use] = 0;
}
#ifdef LIBXML_ICONV_ENABLED
@@ -2061,6 +2248,26 @@ retry:
}
}
#endif /* LIBXML_ICONV_ENABLED */
+#ifdef LIBXML_ICU_ENABLED
+ else if (handler->uconv_out != NULL) {
+ ret = xmlUconvWrapper(handler->uconv_out, 0,
+ &out->content[out->use],
+ &written, in->content, &toconv);
+ xmlBufferShrink(in, toconv);
+ out->use += written;
+ writtentot += written;
+ out->content[out->use] = 0;
+ if (ret == -1) {
+ if (written > 0) {
+ /*
+ * Can be a limitation of iconv
+ */
+ goto retry;
+ }
+ ret = -3;
+ }
+ }
+#endif /* LIBXML_ICU_ENABLED */
else {
xmlEncodingErr(XML_I18N_NO_OUTPUT,
"xmlCharEncOutFunc: no output function !\n", NULL);
@@ -2122,7 +2329,7 @@ retry:
} else {
char buf[50];
- snprintf(&buf[0], 49, "0x%02X 0x%02X 0x%02X 0x%02X",
+ snprintf(&buf[0], 49, "0x%02X 0x%02X 0x%02X 0x%02X",
in->content[0], in->content[1],
in->content[2], in->content[3]);
buf[49] = 0;
@@ -2141,7 +2348,7 @@ retry:
/**
* xmlCharEncCloseFunc:
* @handler: char enconding transformation data structure
- *
+ *
* Generic front-end for encoding handler close function
*
* Returns 0 if success, or -1 in case of error
@@ -2149,6 +2356,7 @@ retry:
int
xmlCharEncCloseFunc(xmlCharEncodingHandler *handler) {
int ret = 0;
+ int tofree = 0;
if (handler == NULL) return(-1);
if (handler->name == NULL) return(-1);
#ifdef LIBXML_ICONV_ENABLED
@@ -2157,9 +2365,7 @@ xmlCharEncCloseFunc(xmlCharEncodingHandler *handler) {
* and the associated icon resources.
*/
if ((handler->iconv_out != NULL) || (handler->iconv_in != NULL)) {
- if (handler->name != NULL)
- xmlFree(handler->name);
- handler->name = NULL;
+ tofree = 1;
if (handler->iconv_out != NULL) {
if (iconv_close(handler->iconv_out))
ret = -1;
@@ -2170,9 +2376,28 @@ xmlCharEncCloseFunc(xmlCharEncodingHandler *handler) {
ret = -1;
handler->iconv_in = NULL;
}
- xmlFree(handler);
}
#endif /* LIBXML_ICONV_ENABLED */
+#ifdef LIBXML_ICU_ENABLED
+ if ((handler->uconv_out != NULL) || (handler->uconv_in != NULL)) {
+ tofree = 1;
+ if (handler->uconv_out != NULL) {
+ closeIcuConverter(handler->uconv_out);
+ handler->uconv_out = NULL;
+ }
+ if (handler->uconv_in != NULL) {
+ closeIcuConverter(handler->uconv_in);
+ handler->uconv_in = NULL;
+ }
+ }
+#endif
+ if (tofree) {
+ /* free up only dynamic handlers iconv/uconv */
+ if (handler->name != NULL)
+ xmlFree(handler->name);
+ handler->name = NULL;
+ xmlFree(handler);
+ }
#ifdef DEBUG_ENCODING
if (ret)
xmlGenericError(xmlGenericErrorContext,
@@ -2202,7 +2427,7 @@ xmlCharEncCloseFunc(xmlCharEncodingHandler *handler) {
long
xmlByteConsumed(xmlParserCtxtPtr ctxt) {
xmlParserInputPtr in;
-
+
if (ctxt == NULL) return(-1);
in = ctxt->input;
if (in == NULL) return(-1);
@@ -2248,6 +2473,23 @@ xmlByteConsumed(xmlParserCtxtPtr ctxt) {
cur += toconv;
} while (ret == -2);
#endif
+#ifdef LIBXML_ICU_ENABLED
+ } else if (handler->uconv_out != NULL) {
+ do {
+ toconv = in->end - cur;
+ written = 32000;
+ ret = xmlUconvWrapper(handler->uconv_out, 0, &convbuf[0],
+ &written, cur, &toconv);
+ if (ret < 0) {
+ if (written > 0)
+ ret = -2;
+ else
+ return(-1);
+ }
+ unused += written;
+ cur += toconv;
+ } while (ret == -2);
+#endif
} else {
/* could not find a converter */
return(-1);
@@ -2260,7 +2502,7 @@ xmlByteConsumed(xmlParserCtxtPtr ctxt) {
return(in->consumed + (in->cur - in->base));
}
-#ifndef LIBXML_ICONV_ENABLED
+#if !defined(LIBXML_ICONV_ENABLED) && !defined(LIBXML_ICU_ENABLED)
#ifdef LIBXML_ISO8859X_ENABLED
/**
@@ -2324,7 +2566,7 @@ UTF8ToISO8859x(unsigned char* out, int *outlen,
*inlen = processed - instart;
return(-2);
}
- c = c & 0x3F;
+ c = c & 0x3F;
d = d & 0x1F;
d = xlattable [48 + c + xlattable [d] * 64];
if (d == 0) {
@@ -2333,7 +2575,7 @@ UTF8ToISO8859x(unsigned char* out, int *outlen,
*inlen = processed - instart;
return(-2);
}
- *out++ = d;
+ *out++ = d;
} else if (d < 0xF0) {
unsigned char c1;
unsigned char c2;
@@ -2357,10 +2599,10 @@ UTF8ToISO8859x(unsigned char* out, int *outlen,
*inlen = processed - instart;
return(-2);
}
- c1 = c1 & 0x3F;
- c2 = c2 & 0x3F;
+ c1 = c1 & 0x3F;
+ c2 = c2 & 0x3F;
d = d & 0x0F;
- d = xlattable [48 + c2 + xlattable [48 + c1 +
+ d = xlattable [48 + c2 + xlattable [48 + c1 +
xlattable [32 + d] * 64] * 64];
if (d == 0) {
/* not in character set */
@@ -2368,7 +2610,7 @@ UTF8ToISO8859x(unsigned char* out, int *outlen,
*inlen = processed - instart;
return(-2);
}
- *out++ = d;
+ *out++ = d;
} else {
/* cannot transcode >= U+010000 */
*outlen = out - outstart;
@@ -2454,22 +2696,22 @@ ISO8859xToUTF8(unsigned char* out, int *outlen,
************************************************************************/
static unsigned short const xmlunicodetable_ISO8859_2 [128] = {
- 0x0080, 0x0081, 0x0082, 0x0083, 0x0084, 0x0085, 0x0086, 0x0087,
- 0x0088, 0x0089, 0x008a, 0x008b, 0x008c, 0x008d, 0x008e, 0x008f,
- 0x0090, 0x0091, 0x0092, 0x0093, 0x0094, 0x0095, 0x0096, 0x0097,
- 0x0098, 0x0099, 0x009a, 0x009b, 0x009c, 0x009d, 0x009e, 0x009f,
- 0x00a0, 0x0104, 0x02d8, 0x0141, 0x00a4, 0x013d, 0x015a, 0x00a7,
- 0x00a8, 0x0160, 0x015e, 0x0164, 0x0179, 0x00ad, 0x017d, 0x017b,
- 0x00b0, 0x0105, 0x02db, 0x0142, 0x00b4, 0x013e, 0x015b, 0x02c7,
- 0x00b8, 0x0161, 0x015f, 0x0165, 0x017a, 0x02dd, 0x017e, 0x017c,
- 0x0154, 0x00c1, 0x00c2, 0x0102, 0x00c4, 0x0139, 0x0106, 0x00c7,
- 0x010c, 0x00c9, 0x0118, 0x00cb, 0x011a, 0x00cd, 0x00ce, 0x010e,
- 0x0110, 0x0143, 0x0147, 0x00d3, 0x00d4, 0x0150, 0x00d6, 0x00d7,
- 0x0158, 0x016e, 0x00da, 0x0170, 0x00dc, 0x00dd, 0x0162, 0x00df,
- 0x0155, 0x00e1, 0x00e2, 0x0103, 0x00e4, 0x013a, 0x0107, 0x00e7,
- 0x010d, 0x00e9, 0x0119, 0x00eb, 0x011b, 0x00ed, 0x00ee, 0x010f,
- 0x0111, 0x0144, 0x0148, 0x00f3, 0x00f4, 0x0151, 0x00f6, 0x00f7,
- 0x0159, 0x016f, 0x00fa, 0x0171, 0x00fc, 0x00fd, 0x0163, 0x02d9,
+ 0x0080, 0x0081, 0x0082, 0x0083, 0x0084, 0x0085, 0x0086, 0x0087,
+ 0x0088, 0x0089, 0x008a, 0x008b, 0x008c, 0x008d, 0x008e, 0x008f,
+ 0x0090, 0x0091, 0x0092, 0x0093, 0x0094, 0x0095, 0x0096, 0x0097,
+ 0x0098, 0x0099, 0x009a, 0x009b, 0x009c, 0x009d, 0x009e, 0x009f,
+ 0x00a0, 0x0104, 0x02d8, 0x0141, 0x00a4, 0x013d, 0x015a, 0x00a7,
+ 0x00a8, 0x0160, 0x015e, 0x0164, 0x0179, 0x00ad, 0x017d, 0x017b,
+ 0x00b0, 0x0105, 0x02db, 0x0142, 0x00b4, 0x013e, 0x015b, 0x02c7,
+ 0x00b8, 0x0161, 0x015f, 0x0165, 0x017a, 0x02dd, 0x017e, 0x017c,
+ 0x0154, 0x00c1, 0x00c2, 0x0102, 0x00c4, 0x0139, 0x0106, 0x00c7,
+ 0x010c, 0x00c9, 0x0118, 0x00cb, 0x011a, 0x00cd, 0x00ce, 0x010e,
+ 0x0110, 0x0143, 0x0147, 0x00d3, 0x00d4, 0x0150, 0x00d6, 0x00d7,
+ 0x0158, 0x016e, 0x00da, 0x0170, 0x00dc, 0x00dd, 0x0162, 0x00df,
+ 0x0155, 0x00e1, 0x00e2, 0x0103, 0x00e4, 0x013a, 0x0107, 0x00e7,
+ 0x010d, 0x00e9, 0x0119, 0x00eb, 0x011b, 0x00ed, 0x00ee, 0x010f,
+ 0x0111, 0x0144, 0x0148, 0x00f3, 0x00f4, 0x0151, 0x00f6, 0x00f7,
+ 0x0159, 0x016f, 0x00fa, 0x0171, 0x00fc, 0x00fd, 0x0163, 0x02d9,
};
static unsigned char const xmltranscodetable_ISO8859_2 [48 + 6 * 64] = {
@@ -2503,22 +2745,22 @@ static unsigned char const xmltranscodetable_ISO8859_2 [48 + 6 * 64] = {
};
static unsigned short const xmlunicodetable_ISO8859_3 [128] = {
- 0x0080, 0x0081, 0x0082, 0x0083, 0x0084, 0x0085, 0x0086, 0x0087,
- 0x0088, 0x0089, 0x008a, 0x008b, 0x008c, 0x008d, 0x008e, 0x008f,
- 0x0090, 0x0091, 0x0092, 0x0093, 0x0094, 0x0095, 0x0096, 0x0097,
- 0x0098, 0x0099, 0x009a, 0x009b, 0x009c, 0x009d, 0x009e, 0x009f,
- 0x00a0, 0x0126, 0x02d8, 0x00a3, 0x00a4, 0x0000, 0x0124, 0x00a7,
- 0x00a8, 0x0130, 0x015e, 0x011e, 0x0134, 0x00ad, 0x0000, 0x017b,
- 0x00b0, 0x0127, 0x00b2, 0x00b3, 0x00b4, 0x00b5, 0x0125, 0x00b7,
- 0x00b8, 0x0131, 0x015f, 0x011f, 0x0135, 0x00bd, 0x0000, 0x017c,
- 0x00c0, 0x00c1, 0x00c2, 0x0000, 0x00c4, 0x010a, 0x0108, 0x00c7,
- 0x00c8, 0x00c9, 0x00ca, 0x00cb, 0x00cc, 0x00cd, 0x00ce, 0x00cf,
- 0x0000, 0x00d1, 0x00d2, 0x00d3, 0x00d4, 0x0120, 0x00d6, 0x00d7,
- 0x011c, 0x00d9, 0x00da, 0x00db, 0x00dc, 0x016c, 0x015c, 0x00df,
- 0x00e0, 0x00e1, 0x00e2, 0x0000, 0x00e4, 0x010b, 0x0109, 0x00e7,
- 0x00e8, 0x00e9, 0x00ea, 0x00eb, 0x00ec, 0x00ed, 0x00ee, 0x00ef,
- 0x0000, 0x00f1, 0x00f2, 0x00f3, 0x00f4, 0x0121, 0x00f6, 0x00f7,
- 0x011d, 0x00f9, 0x00fa, 0x00fb, 0x00fc, 0x016d, 0x015d, 0x02d9,
+ 0x0080, 0x0081, 0x0082, 0x0083, 0x0084, 0x0085, 0x0086, 0x0087,
+ 0x0088, 0x0089, 0x008a, 0x008b, 0x008c, 0x008d, 0x008e, 0x008f,
+ 0x0090, 0x0091, 0x0092, 0x0093, 0x0094, 0x0095, 0x0096, 0x0097,
+ 0x0098, 0x0099, 0x009a, 0x009b, 0x009c, 0x009d, 0x009e, 0x009f,
+ 0x00a0, 0x0126, 0x02d8, 0x00a3, 0x00a4, 0x0000, 0x0124, 0x00a7,
+ 0x00a8, 0x0130, 0x015e, 0x011e, 0x0134, 0x00ad, 0x0000, 0x017b,
+ 0x00b0, 0x0127, 0x00b2, 0x00b3, 0x00b4, 0x00b5, 0x0125, 0x00b7,
+ 0x00b8, 0x0131, 0x015f, 0x011f, 0x0135, 0x00bd, 0x0000, 0x017c,
+ 0x00c0, 0x00c1, 0x00c2, 0x0000, 0x00c4, 0x010a, 0x0108, 0x00c7,
+ 0x00c8, 0x00c9, 0x00ca, 0x00cb, 0x00cc, 0x00cd, 0x00ce, 0x00cf,
+ 0x0000, 0x00d1, 0x00d2, 0x00d3, 0x00d4, 0x0120, 0x00d6, 0x00d7,
+ 0x011c, 0x00d9, 0x00da, 0x00db, 0x00dc, 0x016c, 0x015c, 0x00df,
+ 0x00e0, 0x00e1, 0x00e2, 0x0000, 0x00e4, 0x010b, 0x0109, 0x00e7,
+ 0x00e8, 0x00e9, 0x00ea, 0x00eb, 0x00ec, 0x00ed, 0x00ee, 0x00ef,
+ 0x0000, 0x00f1, 0x00f2, 0x00f3, 0x00f4, 0x0121, 0x00f6, 0x00f7,
+ 0x011d, 0x00f9, 0x00fa, 0x00fb, 0x00fc, 0x016d, 0x015d, 0x02d9,
};
static unsigned char const xmltranscodetable_ISO8859_3 [48 + 7 * 64] = {
@@ -2556,22 +2798,22 @@ static unsigned char const xmltranscodetable_ISO8859_3 [48 + 7 * 64] = {
};
static unsigned short const xmlunicodetable_ISO8859_4 [128] = {
- 0x0080, 0x0081, 0x0082, 0x0083, 0x0084, 0x0085, 0x0086, 0x0087,
- 0x0088, 0x0089, 0x008a, 0x008b, 0x008c, 0x008d, 0x008e, 0x008f,
- 0x0090, 0x0091, 0x0092, 0x0093, 0x0094, 0x0095, 0x0096, 0x0097,
- 0x0098, 0x0099, 0x009a, 0x009b, 0x009c, 0x009d, 0x009e, 0x009f,
- 0x00a0, 0x0104, 0x0138, 0x0156, 0x00a4, 0x0128, 0x013b, 0x00a7,
- 0x00a8, 0x0160, 0x0112, 0x0122, 0x0166, 0x00ad, 0x017d, 0x00af,
- 0x00b0, 0x0105, 0x02db, 0x0157, 0x00b4, 0x0129, 0x013c, 0x02c7,
- 0x00b8, 0x0161, 0x0113, 0x0123, 0x0167, 0x014a, 0x017e, 0x014b,
- 0x0100, 0x00c1, 0x00c2, 0x00c3, 0x00c4, 0x00c5, 0x00c6, 0x012e,
- 0x010c, 0x00c9, 0x0118, 0x00cb, 0x0116, 0x00cd, 0x00ce, 0x012a,
- 0x0110, 0x0145, 0x014c, 0x0136, 0x00d4, 0x00d5, 0x00d6, 0x00d7,
- 0x00d8, 0x0172, 0x00da, 0x00db, 0x00dc, 0x0168, 0x016a, 0x00df,
- 0x0101, 0x00e1, 0x00e2, 0x00e3, 0x00e4, 0x00e5, 0x00e6, 0x012f,
- 0x010d, 0x00e9, 0x0119, 0x00eb, 0x0117, 0x00ed, 0x00ee, 0x012b,
- 0x0111, 0x0146, 0x014d, 0x0137, 0x00f4, 0x00f5, 0x00f6, 0x00f7,
- 0x00f8, 0x0173, 0x00fa, 0x00fb, 0x00fc, 0x0169, 0x016b, 0x02d9,
+ 0x0080, 0x0081, 0x0082, 0x0083, 0x0084, 0x0085, 0x0086, 0x0087,
+ 0x0088, 0x0089, 0x008a, 0x008b, 0x008c, 0x008d, 0x008e, 0x008f,
+ 0x0090, 0x0091, 0x0092, 0x0093, 0x0094, 0x0095, 0x0096, 0x0097,
+ 0x0098, 0x0099, 0x009a, 0x009b, 0x009c, 0x009d, 0x009e, 0x009f,
+ 0x00a0, 0x0104, 0x0138, 0x0156, 0x00a4, 0x0128, 0x013b, 0x00a7,
+ 0x00a8, 0x0160, 0x0112, 0x0122, 0x0166, 0x00ad, 0x017d, 0x00af,
+ 0x00b0, 0x0105, 0x02db, 0x0157, 0x00b4, 0x0129, 0x013c, 0x02c7,
+ 0x00b8, 0x0161, 0x0113, 0x0123, 0x0167, 0x014a, 0x017e, 0x014b,
+ 0x0100, 0x00c1, 0x00c2, 0x00c3, 0x00c4, 0x00c5, 0x00c6, 0x012e,
+ 0x010c, 0x00c9, 0x0118, 0x00cb, 0x0116, 0x00cd, 0x00ce, 0x012a,
+ 0x0110, 0x0145, 0x014c, 0x0136, 0x00d4, 0x00d5, 0x00d6, 0x00d7,
+ 0x00d8, 0x0172, 0x00da, 0x00db, 0x00dc, 0x0168, 0x016a, 0x00df,
+ 0x0101, 0x00e1, 0x00e2, 0x00e3, 0x00e4, 0x00e5, 0x00e6, 0x012f,
+ 0x010d, 0x00e9, 0x0119, 0x00eb, 0x0117, 0x00ed, 0x00ee, 0x012b,
+ 0x0111, 0x0146, 0x014d, 0x0137, 0x00f4, 0x00f5, 0x00f6, 0x00f7,
+ 0x00f8, 0x0173, 0x00fa, 0x00fb, 0x00fc, 0x0169, 0x016b, 0x02d9,
};
static unsigned char const xmltranscodetable_ISO8859_4 [48 + 6 * 64] = {
@@ -2605,22 +2847,22 @@ static unsigned char const xmltranscodetable_ISO8859_4 [48 + 6 * 64] = {
};
static unsigned short const xmlunicodetable_ISO8859_5 [128] = {
- 0x0080, 0x0081, 0x0082, 0x0083, 0x0084, 0x0085, 0x0086, 0x0087,
- 0x0088, 0x0089, 0x008a, 0x008b, 0x008c, 0x008d, 0x008e, 0x008f,
- 0x0090, 0x0091, 0x0092, 0x0093, 0x0094, 0x0095, 0x0096, 0x0097,
- 0x0098, 0x0099, 0x009a, 0x009b, 0x009c, 0x009d, 0x009e, 0x009f,
- 0x00a0, 0x0401, 0x0402, 0x0403, 0x0404, 0x0405, 0x0406, 0x0407,
- 0x0408, 0x0409, 0x040a, 0x040b, 0x040c, 0x00ad, 0x040e, 0x040f,
- 0x0410, 0x0411, 0x0412, 0x0413, 0x0414, 0x0415, 0x0416, 0x0417,
- 0x0418, 0x0419, 0x041a, 0x041b, 0x041c, 0x041d, 0x041e, 0x041f,
- 0x0420, 0x0421, 0x0422, 0x0423, 0x0424, 0x0425, 0x0426, 0x0427,
- 0x0428, 0x0429, 0x042a, 0x042b, 0x042c, 0x042d, 0x042e, 0x042f,
- 0x0430, 0x0431, 0x0432, 0x0433, 0x0434, 0x0435, 0x0436, 0x0437,
- 0x0438, 0x0439, 0x043a, 0x043b, 0x043c, 0x043d, 0x043e, 0x043f,
- 0x0440, 0x0441, 0x0442, 0x0443, 0x0444, 0x0445, 0x0446, 0x0447,
- 0x0448, 0x0449, 0x044a, 0x044b, 0x044c, 0x044d, 0x044e, 0x044f,
- 0x2116, 0x0451, 0x0452, 0x0453, 0x0454, 0x0455, 0x0456, 0x0457,
- 0x0458, 0x0459, 0x045a, 0x045b, 0x045c, 0x00a7, 0x045e, 0x045f,
+ 0x0080, 0x0081, 0x0082, 0x0083, 0x0084, 0x0085, 0x0086, 0x0087,
+ 0x0088, 0x0089, 0x008a, 0x008b, 0x008c, 0x008d, 0x008e, 0x008f,
+ 0x0090, 0x0091, 0x0092, 0x0093, 0x0094, 0x0095, 0x0096, 0x0097,
+ 0x0098, 0x0099, 0x009a, 0x009b, 0x009c, 0x009d, 0x009e, 0x009f,
+ 0x00a0, 0x0401, 0x0402, 0x0403, 0x0404, 0x0405, 0x0406, 0x0407,
+ 0x0408, 0x0409, 0x040a, 0x040b, 0x040c, 0x00ad, 0x040e, 0x040f,
+ 0x0410, 0x0411, 0x0412, 0x0413, 0x0414, 0x0415, 0x0416, 0x0417,
+ 0x0418, 0x0419, 0x041a, 0x041b, 0x041c, 0x041d, 0x041e, 0x041f,
+ 0x0420, 0x0421, 0x0422, 0x0423, 0x0424, 0x0425, 0x0426, 0x0427,
+ 0x0428, 0x0429, 0x042a, 0x042b, 0x042c, 0x042d, 0x042e, 0x042f,
+ 0x0430, 0x0431, 0x0432, 0x0433, 0x0434, 0x0435, 0x0436, 0x0437,
+ 0x0438, 0x0439, 0x043a, 0x043b, 0x043c, 0x043d, 0x043e, 0x043f,
+ 0x0440, 0x0441, 0x0442, 0x0443, 0x0444, 0x0445, 0x0446, 0x0447,
+ 0x0448, 0x0449, 0x044a, 0x044b, 0x044c, 0x044d, 0x044e, 0x044f,
+ 0x2116, 0x0451, 0x0452, 0x0453, 0x0454, 0x0455, 0x0456, 0x0457,
+ 0x0458, 0x0459, 0x045a, 0x045b, 0x045c, 0x00a7, 0x045e, 0x045f,
};
static unsigned char const xmltranscodetable_ISO8859_5 [48 + 6 * 64] = {
@@ -2654,22 +2896,22 @@ static unsigned char const xmltranscodetable_ISO8859_5 [48 + 6 * 64] = {
};
static unsigned short const xmlunicodetable_ISO8859_6 [128] = {
- 0x0080, 0x0081, 0x0082, 0x0083, 0x0084, 0x0085, 0x0086, 0x0087,
- 0x0088, 0x0089, 0x008a, 0x008b, 0x008c, 0x008d, 0x008e, 0x008f,
- 0x0090, 0x0091, 0x0092, 0x0093, 0x0094, 0x0095, 0x0096, 0x0097,
- 0x0098, 0x0099, 0x009a, 0x009b, 0x009c, 0x009d, 0x009e, 0x009f,
- 0x00a0, 0x0000, 0x0000, 0x0000, 0x00a4, 0x0000, 0x0000, 0x0000,
- 0x0000, 0x0000, 0x0000, 0x0000, 0x060c, 0x00ad, 0x0000, 0x0000,
- 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000,
- 0x0000, 0x0000, 0x0000, 0x061b, 0x0000, 0x0000, 0x0000, 0x061f,
- 0x0000, 0x0621, 0x0622, 0x0623, 0x0624, 0x0625, 0x0626, 0x0627,
- 0x0628, 0x0629, 0x062a, 0x062b, 0x062c, 0x062d, 0x062e, 0x062f,
- 0x0630, 0x0631, 0x0632, 0x0633, 0x0634, 0x0635, 0x0636, 0x0637,
- 0x0638, 0x0639, 0x063a, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000,
- 0x0640, 0x0641, 0x0642, 0x0643, 0x0644, 0x0645, 0x0646, 0x0647,
- 0x0648, 0x0649, 0x064a, 0x064b, 0x064c, 0x064d, 0x064e, 0x064f,
- 0x0650, 0x0651, 0x0652, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000,
- 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000,
+ 0x0080, 0x0081, 0x0082, 0x0083, 0x0084, 0x0085, 0x0086, 0x0087,
+ 0x0088, 0x0089, 0x008a, 0x008b, 0x008c, 0x008d, 0x008e, 0x008f,
+ 0x0090, 0x0091, 0x0092, 0x0093, 0x0094, 0x0095, 0x0096, 0x0097,
+ 0x0098, 0x0099, 0x009a, 0x009b, 0x009c, 0x009d, 0x009e, 0x009f,
+ 0x00a0, 0x0000, 0x0000, 0x0000, 0x00a4, 0x0000, 0x0000, 0x0000,
+ 0x0000, 0x0000, 0x0000, 0x0000, 0x060c, 0x00ad, 0x0000, 0x0000,
+ 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000,
+ 0x0000, 0x0000, 0x0000, 0x061b, 0x0000, 0x0000, 0x0000, 0x061f,
+ 0x0000, 0x0621, 0x0622, 0x0623, 0x0624, 0x0625, 0x0626, 0x0627,
+ 0x0628, 0x0629, 0x062a, 0x062b, 0x062c, 0x062d, 0x062e, 0x062f,
+ 0x0630, 0x0631, 0x0632, 0x0633, 0x0634, 0x0635, 0x0636, 0x0637,
+ 0x0638, 0x0639, 0x063a, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000,
+ 0x0640, 0x0641, 0x0642, 0x0643, 0x0644, 0x0645, 0x0646, 0x0647,
+ 0x0648, 0x0649, 0x064a, 0x064b, 0x064c, 0x064d, 0x064e, 0x064f,
+ 0x0650, 0x0651, 0x0652, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000,
+ 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000,
};
static unsigned char const xmltranscodetable_ISO8859_6 [48 + 5 * 64] = {
@@ -2699,22 +2941,22 @@ static unsigned char const xmltranscodetable_ISO8859_6 [48 + 5 * 64] = {
};
static unsigned short const xmlunicodetable_ISO8859_7 [128] = {
- 0x0080, 0x0081, 0x0082, 0x0083, 0x0084, 0x0085, 0x0086, 0x0087,
- 0x0088, 0x0089, 0x008a, 0x008b, 0x008c, 0x008d, 0x008e, 0x008f,
- 0x0090, 0x0091, 0x0092, 0x0093, 0x0094, 0x0095, 0x0096, 0x0097,
- 0x0098, 0x0099, 0x009a, 0x009b, 0x009c, 0x009d, 0x009e, 0x009f,
- 0x00a0, 0x2018, 0x2019, 0x00a3, 0x0000, 0x0000, 0x00a6, 0x00a7,
- 0x00a8, 0x00a9, 0x0000, 0x00ab, 0x00ac, 0x00ad, 0x0000, 0x2015,
- 0x00b0, 0x00b1, 0x00b2, 0x00b3, 0x0384, 0x0385, 0x0386, 0x00b7,
- 0x0388, 0x0389, 0x038a, 0x00bb, 0x038c, 0x00bd, 0x038e, 0x038f,
- 0x0390, 0x0391, 0x0392, 0x0393, 0x0394, 0x0395, 0x0396, 0x0397,
- 0x0398, 0x0399, 0x039a, 0x039b, 0x039c, 0x039d, 0x039e, 0x039f,
- 0x03a0, 0x03a1, 0x0000, 0x03a3, 0x03a4, 0x03a5, 0x03a6, 0x03a7,
- 0x03a8, 0x03a9, 0x03aa, 0x03ab, 0x03ac, 0x03ad, 0x03ae, 0x03af,
- 0x03b0, 0x03b1, 0x03b2, 0x03b3, 0x03b4, 0x03b5, 0x03b6, 0x03b7,
- 0x03b8, 0x03b9, 0x03ba, 0x03bb, 0x03bc, 0x03bd, 0x03be, 0x03bf,
- 0x03c0, 0x03c1, 0x03c2, 0x03c3, 0x03c4, 0x03c5, 0x03c6, 0x03c7,
- 0x03c8, 0x03c9, 0x03ca, 0x03cb, 0x03cc, 0x03cd, 0x03ce, 0x0000,
+ 0x0080, 0x0081, 0x0082, 0x0083, 0x0084, 0x0085, 0x0086, 0x0087,
+ 0x0088, 0x0089, 0x008a, 0x008b, 0x008c, 0x008d, 0x008e, 0x008f,
+ 0x0090, 0x0091, 0x0092, 0x0093, 0x0094, 0x0095, 0x0096, 0x0097,
+ 0x0098, 0x0099, 0x009a, 0x009b, 0x009c, 0x009d, 0x009e, 0x009f,
+ 0x00a0, 0x2018, 0x2019, 0x00a3, 0x0000, 0x0000, 0x00a6, 0x00a7,
+ 0x00a8, 0x00a9, 0x0000, 0x00ab, 0x00ac, 0x00ad, 0x0000, 0x2015,
+ 0x00b0, 0x00b1, 0x00b2, 0x00b3, 0x0384, 0x0385, 0x0386, 0x00b7,
+ 0x0388, 0x0389, 0x038a, 0x00bb, 0x038c, 0x00bd, 0x038e, 0x038f,
+ 0x0390, 0x0391, 0x0392, 0x0393, 0x0394, 0x0395, 0x0396, 0x0397,
+ 0x0398, 0x0399, 0x039a, 0x039b, 0x039c, 0x039d, 0x039e, 0x039f,
+ 0x03a0, 0x03a1, 0x0000, 0x03a3, 0x03a4, 0x03a5, 0x03a6, 0x03a7,
+ 0x03a8, 0x03a9, 0x03aa, 0x03ab, 0x03ac, 0x03ad, 0x03ae, 0x03af,
+ 0x03b0, 0x03b1, 0x03b2, 0x03b3, 0x03b4, 0x03b5, 0x03b6, 0x03b7,
+ 0x03b8, 0x03b9, 0x03ba, 0x03bb, 0x03bc, 0x03bd, 0x03be, 0x03bf,
+ 0x03c0, 0x03c1, 0x03c2, 0x03c3, 0x03c4, 0x03c5, 0x03c6, 0x03c7,
+ 0x03c8, 0x03c9, 0x03ca, 0x03cb, 0x03cc, 0x03cd, 0x03ce, 0x0000,
};
static unsigned char const xmltranscodetable_ISO8859_7 [48 + 7 * 64] = {
@@ -2752,22 +2994,22 @@ static unsigned char const xmltranscodetable_ISO8859_7 [48 + 7 * 64] = {
};
static unsigned short const xmlunicodetable_ISO8859_8 [128] = {
- 0x0080, 0x0081, 0x0082, 0x0083, 0x0084, 0x0085, 0x0086, 0x0087,
- 0x0088, 0x0089, 0x008a, 0x008b, 0x008c, 0x008d, 0x008e, 0x008f,
- 0x0090, 0x0091, 0x0092, 0x0093, 0x0094, 0x0095, 0x0096, 0x0097,
- 0x0098, 0x0099, 0x009a, 0x009b, 0x009c, 0x009d, 0x009e, 0x009f,
- 0x00a0, 0x0000, 0x00a2, 0x00a3, 0x00a4, 0x00a5, 0x00a6, 0x00a7,
- 0x00a8, 0x00a9, 0x00d7, 0x00ab, 0x00ac, 0x00ad, 0x00ae, 0x00af,
- 0x00b0, 0x00b1, 0x00b2, 0x00b3, 0x00b4, 0x00b5, 0x00b6, 0x00b7,
- 0x00b8, 0x00b9, 0x00f7, 0x00bb, 0x00bc, 0x00bd, 0x00be, 0x0000,
- 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000,
- 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000,
- 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000,
- 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x2017,
- 0x05d0, 0x05d1, 0x05d2, 0x05d3, 0x05d4, 0x05d5, 0x05d6, 0x05d7,
- 0x05d8, 0x05d9, 0x05da, 0x05db, 0x05dc, 0x05dd, 0x05de, 0x05df,
- 0x05e0, 0x05e1, 0x05e2, 0x05e3, 0x05e4, 0x05e5, 0x05e6, 0x05e7,
- 0x05e8, 0x05e9, 0x05ea, 0x0000, 0x0000, 0x200e, 0x200f, 0x0000,
+ 0x0080, 0x0081, 0x0082, 0x0083, 0x0084, 0x0085, 0x0086, 0x0087,
+ 0x0088, 0x0089, 0x008a, 0x008b, 0x008c, 0x008d, 0x008e, 0x008f,
+ 0x0090, 0x0091, 0x0092, 0x0093, 0x0094, 0x0095, 0x0096, 0x0097,
+ 0x0098, 0x0099, 0x009a, 0x009b, 0x009c, 0x009d, 0x009e, 0x009f,
+ 0x00a0, 0x0000, 0x00a2, 0x00a3, 0x00a4, 0x00a5, 0x00a6, 0x00a7,
+ 0x00a8, 0x00a9, 0x00d7, 0x00ab, 0x00ac, 0x00ad, 0x00ae, 0x00af,
+ 0x00b0, 0x00b1, 0x00b2, 0x00b3, 0x00b4, 0x00b5, 0x00b6, 0x00b7,
+ 0x00b8, 0x00b9, 0x00f7, 0x00bb, 0x00bc, 0x00bd, 0x00be, 0x0000,
+ 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000,
+ 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000,
+ 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000,
+ 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x2017,
+ 0x05d0, 0x05d1, 0x05d2, 0x05d3, 0x05d4, 0x05d5, 0x05d6, 0x05d7,
+ 0x05d8, 0x05d9, 0x05da, 0x05db, 0x05dc, 0x05dd, 0x05de, 0x05df,
+ 0x05e0, 0x05e1, 0x05e2, 0x05e3, 0x05e4, 0x05e5, 0x05e6, 0x05e7,
+ 0x05e8, 0x05e9, 0x05ea, 0x0000, 0x0000, 0x200e, 0x200f, 0x0000,
};
static unsigned char const xmltranscodetable_ISO8859_8 [48 + 7 * 64] = {
@@ -2805,22 +3047,22 @@ static unsigned char const xmltranscodetable_ISO8859_8 [48 + 7 * 64] = {
};
static unsigned short const xmlunicodetable_ISO8859_9 [128] = {
- 0x0080, 0x0081, 0x0082, 0x0083, 0x0084, 0x0085, 0x0086, 0x0087,
- 0x0088, 0x0089, 0x008a, 0x008b, 0x008c, 0x008d, 0x008e, 0x008f,
- 0x0090, 0x0091, 0x0092, 0x0093, 0x0094, 0x0095, 0x0096, 0x0097,
- 0x0098, 0x0099, 0x009a, 0x009b, 0x009c, 0x009d, 0x009e, 0x009f,
- 0x00a0, 0x00a1, 0x00a2, 0x00a3, 0x00a4, 0x00a5, 0x00a6, 0x00a7,
- 0x00a8, 0x00a9, 0x00aa, 0x00ab, 0x00ac, 0x00ad, 0x00ae, 0x00af,
- 0x00b0, 0x00b1, 0x00b2, 0x00b3, 0x00b4, 0x00b5, 0x00b6, 0x00b7,
- 0x00b8, 0x00b9, 0x00ba, 0x00bb, 0x00bc, 0x00bd, 0x00be, 0x00bf,
- 0x00c0, 0x00c1, 0x00c2, 0x00c3, 0x00c4, 0x00c5, 0x00c6, 0x00c7,
- 0x00c8, 0x00c9, 0x00ca, 0x00cb, 0x00cc, 0x00cd, 0x00ce, 0x00cf,
- 0x011e, 0x00d1, 0x00d2, 0x00d3, 0x00d4, 0x00d5, 0x00d6, 0x00d7,
- 0x00d8, 0x00d9, 0x00da, 0x00db, 0x00dc, 0x0130, 0x015e, 0x00df,
- 0x00e0, 0x00e1, 0x00e2, 0x00e3, 0x00e4, 0x00e5, 0x00e6, 0x00e7,
- 0x00e8, 0x00e9, 0x00ea, 0x00eb, 0x00ec, 0x00ed, 0x00ee, 0x00ef,
- 0x011f, 0x00f1, 0x00f2, 0x00f3, 0x00f4, 0x00f5, 0x00f6, 0x00f7,
- 0x00f8, 0x00f9, 0x00fa, 0x00fb, 0x00fc, 0x0131, 0x015f, 0x00ff,
+ 0x0080, 0x0081, 0x0082, 0x0083, 0x0084, 0x0085, 0x0086, 0x0087,
+ 0x0088, 0x0089, 0x008a, 0x008b, 0x008c, 0x008d, 0x008e, 0x008f,
+ 0x0090, 0x0091, 0x0092, 0x0093, 0x0094, 0x0095, 0x0096, 0x0097,
+ 0x0098, 0x0099, 0x009a, 0x009b, 0x009c, 0x009d, 0x009e, 0x009f,
+ 0x00a0, 0x00a1, 0x00a2, 0x00a3, 0x00a4, 0x00a5, 0x00a6, 0x00a7,
+ 0x00a8, 0x00a9, 0x00aa, 0x00ab, 0x00ac, 0x00ad, 0x00ae, 0x00af,
+ 0x00b0, 0x00b1, 0x00b2, 0x00b3, 0x00b4, 0x00b5, 0x00b6, 0x00b7,
+ 0x00b8, 0x00b9, 0x00ba, 0x00bb, 0x00bc, 0x00bd, 0x00be, 0x00bf,
+ 0x00c0, 0x00c1, 0x00c2, 0x00c3, 0x00c4, 0x00c5, 0x00c6, 0x00c7,
+ 0x00c8, 0x00c9, 0x00ca, 0x00cb, 0x00cc, 0x00cd, 0x00ce, 0x00cf,
+ 0x011e, 0x00d1, 0x00d2, 0x00d3, 0x00d4, 0x00d5, 0x00d6, 0x00d7,
+ 0x00d8, 0x00d9, 0x00da, 0x00db, 0x00dc, 0x0130, 0x015e, 0x00df,
+ 0x00e0, 0x00e1, 0x00e2, 0x00e3, 0x00e4, 0x00e5, 0x00e6, 0x00e7,
+ 0x00e8, 0x00e9, 0x00ea, 0x00eb, 0x00ec, 0x00ed, 0x00ee, 0x00ef,
+ 0x011f, 0x00f1, 0x00f2, 0x00f3, 0x00f4, 0x00f5, 0x00f6, 0x00f7,
+ 0x00f8, 0x00f9, 0x00fa, 0x00fb, 0x00fc, 0x0131, 0x015f, 0x00ff,
};
static unsigned char const xmltranscodetable_ISO8859_9 [48 + 5 * 64] = {
@@ -2850,22 +3092,22 @@ static unsigned char const xmltranscodetable_ISO8859_9 [48 + 5 * 64] = {
};
static unsigned short const xmlunicodetable_ISO8859_10 [128] = {
- 0x0080, 0x0081, 0x0082, 0x0083, 0x0084, 0x0085, 0x0086, 0x0087,
- 0x0088, 0x0089, 0x008a, 0x008b, 0x008c, 0x008d, 0x008e, 0x008f,
- 0x0090, 0x0091, 0x0092, 0x0093, 0x0094, 0x0095, 0x0096, 0x0097,
- 0x0098, 0x0099, 0x009a, 0x009b, 0x009c, 0x009d, 0x009e, 0x009f,
- 0x00a0, 0x0104, 0x0112, 0x0122, 0x012a, 0x0128, 0x0136, 0x00a7,
- 0x013b, 0x0110, 0x0160, 0x0166, 0x017d, 0x00ad, 0x016a, 0x014a,
- 0x00b0, 0x0105, 0x0113, 0x0123, 0x012b, 0x0129, 0x0137, 0x00b7,
- 0x013c, 0x0111, 0x0161, 0x0167, 0x017e, 0x2015, 0x016b, 0x014b,
- 0x0100, 0x00c1, 0x00c2, 0x00c3, 0x00c4, 0x00c5, 0x00c6, 0x012e,
- 0x010c, 0x00c9, 0x0118, 0x00cb, 0x0116, 0x00cd, 0x00ce, 0x00cf,
- 0x00d0, 0x0145, 0x014c, 0x00d3, 0x00d4, 0x00d5, 0x00d6, 0x0168,
- 0x00d8, 0x0172, 0x00da, 0x00db, 0x00dc, 0x00dd, 0x00de, 0x00df,
- 0x0101, 0x00e1, 0x00e2, 0x00e3, 0x00e4, 0x00e5, 0x00e6, 0x012f,
- 0x010d, 0x00e9, 0x0119, 0x00eb, 0x0117, 0x00ed, 0x00ee, 0x00ef,
- 0x00f0, 0x0146, 0x014d, 0x00f3, 0x00f4, 0x00f5, 0x00f6, 0x0169,
- 0x00f8, 0x0173, 0x00fa, 0x00fb, 0x00fc, 0x00fd, 0x00fe, 0x0138,
+ 0x0080, 0x0081, 0x0082, 0x0083, 0x0084, 0x0085, 0x0086, 0x0087,
+ 0x0088, 0x0089, 0x008a, 0x008b, 0x008c, 0x008d, 0x008e, 0x008f,
+ 0x0090, 0x0091, 0x0092, 0x0093, 0x0094, 0x0095, 0x0096, 0x0097,
+ 0x0098, 0x0099, 0x009a, 0x009b, 0x009c, 0x009d, 0x009e, 0x009f,
+ 0x00a0, 0x0104, 0x0112, 0x0122, 0x012a, 0x0128, 0x0136, 0x00a7,
+ 0x013b, 0x0110, 0x0160, 0x0166, 0x017d, 0x00ad, 0x016a, 0x014a,
+ 0x00b0, 0x0105, 0x0113, 0x0123, 0x012b, 0x0129, 0x0137, 0x00b7,
+ 0x013c, 0x0111, 0x0161, 0x0167, 0x017e, 0x2015, 0x016b, 0x014b,
+ 0x0100, 0x00c1, 0x00c2, 0x00c3, 0x00c4, 0x00c5, 0x00c6, 0x012e,
+ 0x010c, 0x00c9, 0x0118, 0x00cb, 0x0116, 0x00cd, 0x00ce, 0x00cf,
+ 0x00d0, 0x0145, 0x014c, 0x00d3, 0x00d4, 0x00d5, 0x00d6, 0x0168,
+ 0x00d8, 0x0172, 0x00da, 0x00db, 0x00dc, 0x00dd, 0x00de, 0x00df,
+ 0x0101, 0x00e1, 0x00e2, 0x00e3, 0x00e4, 0x00e5, 0x00e6, 0x012f,
+ 0x010d, 0x00e9, 0x0119, 0x00eb, 0x0117, 0x00ed, 0x00ee, 0x00ef,
+ 0x00f0, 0x0146, 0x014d, 0x00f3, 0x00f4, 0x00f5, 0x00f6, 0x0169,
+ 0x00f8, 0x0173, 0x00fa, 0x00fb, 0x00fc, 0x00fd, 0x00fe, 0x0138,
};
static unsigned char const xmltranscodetable_ISO8859_10 [48 + 7 * 64] = {
@@ -2903,22 +3145,22 @@ static unsigned char const xmltranscodetable_ISO8859_10 [48 + 7 * 64] = {
};
static unsigned short const xmlunicodetable_ISO8859_11 [128] = {
- 0x0080, 0x0081, 0x0082, 0x0083, 0x0084, 0x0085, 0x0086, 0x0087,
- 0x0088, 0x0089, 0x008a, 0x008b, 0x008c, 0x008d, 0x008e, 0x008f,
- 0x0090, 0x0091, 0x0092, 0x0093, 0x0094, 0x0095, 0x0096, 0x0097,
- 0x0098, 0x0099, 0x009a, 0x009b, 0x009c, 0x009d, 0x009e, 0x009f,
- 0x00a0, 0x0e01, 0x0e02, 0x0e03, 0x0e04, 0x0e05, 0x0e06, 0x0e07,
- 0x0e08, 0x0e09, 0x0e0a, 0x0e0b, 0x0e0c, 0x0e0d, 0x0e0e, 0x0e0f,
- 0x0e10, 0x0e11, 0x0e12, 0x0e13, 0x0e14, 0x0e15, 0x0e16, 0x0e17,
- 0x0e18, 0x0e19, 0x0e1a, 0x0e1b, 0x0e1c, 0x0e1d, 0x0e1e, 0x0e1f,
- 0x0e20, 0x0e21, 0x0e22, 0x0e23, 0x0e24, 0x0e25, 0x0e26, 0x0e27,
- 0x0e28, 0x0e29, 0x0e2a, 0x0e2b, 0x0e2c, 0x0e2d, 0x0e2e, 0x0e2f,
- 0x0e30, 0x0e31, 0x0e32, 0x0e33, 0x0e34, 0x0e35, 0x0e36, 0x0e37,
- 0x0e38, 0x0e39, 0x0e3a, 0x0000, 0x0000, 0x0000, 0x0000, 0x0e3f,
- 0x0e40, 0x0e41, 0x0e42, 0x0e43, 0x0e44, 0x0e45, 0x0e46, 0x0e47,
- 0x0e48, 0x0e49, 0x0e4a, 0x0e4b, 0x0e4c, 0x0e4d, 0x0e4e, 0x0e4f,
- 0x0e50, 0x0e51, 0x0e52, 0x0e53, 0x0e54, 0x0e55, 0x0e56, 0x0e57,
- 0x0e58, 0x0e59, 0x0e5a, 0x0e5b, 0x0000, 0x0000, 0x0000, 0x0000,
+ 0x0080, 0x0081, 0x0082, 0x0083, 0x0084, 0x0085, 0x0086, 0x0087,
+ 0x0088, 0x0089, 0x008a, 0x008b, 0x008c, 0x008d, 0x008e, 0x008f,
+ 0x0090, 0x0091, 0x0092, 0x0093, 0x0094, 0x0095, 0x0096, 0x0097,
+ 0x0098, 0x0099, 0x009a, 0x009b, 0x009c, 0x009d, 0x009e, 0x009f,
+ 0x00a0, 0x0e01, 0x0e02, 0x0e03, 0x0e04, 0x0e05, 0x0e06, 0x0e07,
+ 0x0e08, 0x0e09, 0x0e0a, 0x0e0b, 0x0e0c, 0x0e0d, 0x0e0e, 0x0e0f,
+ 0x0e10, 0x0e11, 0x0e12, 0x0e13, 0x0e14, 0x0e15, 0x0e16, 0x0e17,
+ 0x0e18, 0x0e19, 0x0e1a, 0x0e1b, 0x0e1c, 0x0e1d, 0x0e1e, 0x0e1f,
+ 0x0e20, 0x0e21, 0x0e22, 0x0e23, 0x0e24, 0x0e25, 0x0e26, 0x0e27,
+ 0x0e28, 0x0e29, 0x0e2a, 0x0e2b, 0x0e2c, 0x0e2d, 0x0e2e, 0x0e2f,
+ 0x0e30, 0x0e31, 0x0e32, 0x0e33, 0x0e34, 0x0e35, 0x0e36, 0x0e37,
+ 0x0e38, 0x0e39, 0x0e3a, 0x0000, 0x0000, 0x0000, 0x0000, 0x0e3f,
+ 0x0e40, 0x0e41, 0x0e42, 0x0e43, 0x0e44, 0x0e45, 0x0e46, 0x0e47,
+ 0x0e48, 0x0e49, 0x0e4a, 0x0e4b, 0x0e4c, 0x0e4d, 0x0e4e, 0x0e4f,
+ 0x0e50, 0x0e51, 0x0e52, 0x0e53, 0x0e54, 0x0e55, 0x0e56, 0x0e57,
+ 0x0e58, 0x0e59, 0x0e5a, 0x0e5b, 0x0000, 0x0000, 0x0000, 0x0000,
};
static unsigned char const xmltranscodetable_ISO8859_11 [48 + 6 * 64] = {
@@ -2952,22 +3194,22 @@ static unsigned char const xmltranscodetable_ISO8859_11 [48 + 6 * 64] = {
};
static unsigned short const xmlunicodetable_ISO8859_13 [128] = {
- 0x0080, 0x0081, 0x0082, 0x0083, 0x0084, 0x0085, 0x0086, 0x0087,
- 0x0088, 0x0089, 0x008a, 0x008b, 0x008c, 0x008d, 0x008e, 0x008f,
- 0x0090, 0x0091, 0x0092, 0x0093, 0x0094, 0x0095, 0x0096, 0x0097,
- 0x0098, 0x0099, 0x009a, 0x009b, 0x009c, 0x009d, 0x009e, 0x009f,
- 0x00a0, 0x201d, 0x00a2, 0x00a3, 0x00a4, 0x201e, 0x00a6, 0x00a7,
- 0x00d8, 0x00a9, 0x0156, 0x00ab, 0x00ac, 0x00ad, 0x00ae, 0x00c6,
- 0x00b0, 0x00b1, 0x00b2, 0x00b3, 0x201c, 0x00b5, 0x00b6, 0x00b7,
- 0x00f8, 0x00b9, 0x0157, 0x00bb, 0x00bc, 0x00bd, 0x00be, 0x00e6,
- 0x0104, 0x012e, 0x0100, 0x0106, 0x00c4, 0x00c5, 0x0118, 0x0112,
- 0x010c, 0x00c9, 0x0179, 0x0116, 0x0122, 0x0136, 0x012a, 0x013b,
- 0x0160, 0x0143, 0x0145, 0x00d3, 0x014c, 0x00d5, 0x00d6, 0x00d7,
- 0x0172, 0x0141, 0x015a, 0x016a, 0x00dc, 0x017b, 0x017d, 0x00df,
- 0x0105, 0x012f, 0x0101, 0x0107, 0x00e4, 0x00e5, 0x0119, 0x0113,
- 0x010d, 0x00e9, 0x017a, 0x0117, 0x0123, 0x0137, 0x012b, 0x013c,
- 0x0161, 0x0144, 0x0146, 0x00f3, 0x014d, 0x00f5, 0x00f6, 0x00f7,
- 0x0173, 0x0142, 0x015b, 0x016b, 0x00fc, 0x017c, 0x017e, 0x2019,
+ 0x0080, 0x0081, 0x0082, 0x0083, 0x0084, 0x0085, 0x0086, 0x0087,
+ 0x0088, 0x0089, 0x008a, 0x008b, 0x008c, 0x008d, 0x008e, 0x008f,
+ 0x0090, 0x0091, 0x0092, 0x0093, 0x0094, 0x0095, 0x0096, 0x0097,
+ 0x0098, 0x0099, 0x009a, 0x009b, 0x009c, 0x009d, 0x009e, 0x009f,
+ 0x00a0, 0x201d, 0x00a2, 0x00a3, 0x00a4, 0x201e, 0x00a6, 0x00a7,
+ 0x00d8, 0x00a9, 0x0156, 0x00ab, 0x00ac, 0x00ad, 0x00ae, 0x00c6,
+ 0x00b0, 0x00b1, 0x00b2, 0x00b3, 0x201c, 0x00b5, 0x00b6, 0x00b7,
+ 0x00f8, 0x00b9, 0x0157, 0x00bb, 0x00bc, 0x00bd, 0x00be, 0x00e6,
+ 0x0104, 0x012e, 0x0100, 0x0106, 0x00c4, 0x00c5, 0x0118, 0x0112,
+ 0x010c, 0x00c9, 0x0179, 0x0116, 0x0122, 0x0136, 0x012a, 0x013b,
+ 0x0160, 0x0143, 0x0145, 0x00d3, 0x014c, 0x00d5, 0x00d6, 0x00d7,
+ 0x0172, 0x0141, 0x015a, 0x016a, 0x00dc, 0x017b, 0x017d, 0x00df,
+ 0x0105, 0x012f, 0x0101, 0x0107, 0x00e4, 0x00e5, 0x0119, 0x0113,
+ 0x010d, 0x00e9, 0x017a, 0x0117, 0x0123, 0x0137, 0x012b, 0x013c,
+ 0x0161, 0x0144, 0x0146, 0x00f3, 0x014d, 0x00f5, 0x00f6, 0x00f7,
+ 0x0173, 0x0142, 0x015b, 0x016b, 0x00fc, 0x017c, 0x017e, 0x2019,
};
static unsigned char const xmltranscodetable_ISO8859_13 [48 + 7 * 64] = {
@@ -3005,22 +3247,22 @@ static unsigned char const xmltranscodetable_ISO8859_13 [48 + 7 * 64] = {
};
static unsigned short const xmlunicodetable_ISO8859_14 [128] = {
- 0x0080, 0x0081, 0x0082, 0x0083, 0x0084, 0x0085, 0x0086, 0x0087,
- 0x0088, 0x0089, 0x008a, 0x008b, 0x008c, 0x008d, 0x008e, 0x008f,
- 0x0090, 0x0091, 0x0092, 0x0093, 0x0094, 0x0095, 0x0096, 0x0097,
- 0x0098, 0x0099, 0x009a, 0x009b, 0x009c, 0x009d, 0x009e, 0x009f,
- 0x00a0, 0x1e02, 0x1e03, 0x00a3, 0x010a, 0x010b, 0x1e0a, 0x00a7,
- 0x1e80, 0x00a9, 0x1e82, 0x1e0b, 0x1ef2, 0x00ad, 0x00ae, 0x0178,
- 0x1e1e, 0x1e1f, 0x0120, 0x0121, 0x1e40, 0x1e41, 0x00b6, 0x1e56,
- 0x1e81, 0x1e57, 0x1e83, 0x1e60, 0x1ef3, 0x1e84, 0x1e85, 0x1e61,
- 0x00c0, 0x00c1, 0x00c2, 0x00c3, 0x00c4, 0x00c5, 0x00c6, 0x00c7,
- 0x00c8, 0x00c9, 0x00ca, 0x00cb, 0x00cc, 0x00cd, 0x00ce, 0x00cf,
- 0x0174, 0x00d1, 0x00d2, 0x00d3, 0x00d4, 0x00d5, 0x00d6, 0x1e6a,
- 0x00d8, 0x00d9, 0x00da, 0x00db, 0x00dc, 0x00dd, 0x0176, 0x00df,
- 0x00e0, 0x00e1, 0x00e2, 0x00e3, 0x00e4, 0x00e5, 0x00e6, 0x00e7,
- 0x00e8, 0x00e9, 0x00ea, 0x00eb, 0x00ec, 0x00ed, 0x00ee, 0x00ef,
- 0x0175, 0x00f1, 0x00f2, 0x00f3, 0x00f4, 0x00f5, 0x00f6, 0x1e6b,
- 0x00f8, 0x00f9, 0x00fa, 0x00fb, 0x00fc, 0x00fd, 0x0177, 0x00ff,
+ 0x0080, 0x0081, 0x0082, 0x0083, 0x0084, 0x0085, 0x0086, 0x0087,
+ 0x0088, 0x0089, 0x008a, 0x008b, 0x008c, 0x008d, 0x008e, 0x008f,
+ 0x0090, 0x0091, 0x0092, 0x0093, 0x0094, 0x0095, 0x0096, 0x0097,
+ 0x0098, 0x0099, 0x009a, 0x009b, 0x009c, 0x009d, 0x009e, 0x009f,
+ 0x00a0, 0x1e02, 0x1e03, 0x00a3, 0x010a, 0x010b, 0x1e0a, 0x00a7,
+ 0x1e80, 0x00a9, 0x1e82, 0x1e0b, 0x1ef2, 0x00ad, 0x00ae, 0x0178,
+ 0x1e1e, 0x1e1f, 0x0120, 0x0121, 0x1e40, 0x1e41, 0x00b6, 0x1e56,
+ 0x1e81, 0x1e57, 0x1e83, 0x1e60, 0x1ef3, 0x1e84, 0x1e85, 0x1e61,
+ 0x00c0, 0x00c1, 0x00c2, 0x00c3, 0x00c4, 0x00c5, 0x00c6, 0x00c7,
+ 0x00c8, 0x00c9, 0x00ca, 0x00cb, 0x00cc, 0x00cd, 0x00ce, 0x00cf,
+ 0x0174, 0x00d1, 0x00d2, 0x00d3, 0x00d4, 0x00d5, 0x00d6, 0x1e6a,
+ 0x00d8, 0x00d9, 0x00da, 0x00db, 0x00dc, 0x00dd, 0x0176, 0x00df,
+ 0x00e0, 0x00e1, 0x00e2, 0x00e3, 0x00e4, 0x00e5, 0x00e6, 0x00e7,
+ 0x00e8, 0x00e9, 0x00ea, 0x00eb, 0x00ec, 0x00ed, 0x00ee, 0x00ef,
+ 0x0175, 0x00f1, 0x00f2, 0x00f3, 0x00f4, 0x00f5, 0x00f6, 0x1e6b,
+ 0x00f8, 0x00f9, 0x00fa, 0x00fb, 0x00fc, 0x00fd, 0x0177, 0x00ff,
};
static unsigned char const xmltranscodetable_ISO8859_14 [48 + 10 * 64] = {
@@ -3070,22 +3312,22 @@ static unsigned char const xmltranscodetable_ISO8859_14 [48 + 10 * 64] = {
};
static unsigned short const xmlunicodetable_ISO8859_15 [128] = {
- 0x0080, 0x0081, 0x0082, 0x0083, 0x0084, 0x0085, 0x0086, 0x0087,
- 0x0088, 0x0089, 0x008a, 0x008b, 0x008c, 0x008d, 0x008e, 0x008f,
- 0x0090, 0x0091, 0x0092, 0x0093, 0x0094, 0x0095, 0x0096, 0x0097,
- 0x0098, 0x0099, 0x009a, 0x009b, 0x009c, 0x009d, 0x009e, 0x009f,
- 0x00a0, 0x00a1, 0x00a2, 0x00a3, 0x20ac, 0x00a5, 0x0160, 0x00a7,
- 0x0161, 0x00a9, 0x00aa, 0x00ab, 0x00ac, 0x00ad, 0x00ae, 0x00af,
- 0x00b0, 0x00b1, 0x00b2, 0x00b3, 0x017d, 0x00b5, 0x00b6, 0x00b7,
- 0x017e, 0x00b9, 0x00ba, 0x00bb, 0x0152, 0x0153, 0x0178, 0x00bf,
- 0x00c0, 0x00c1, 0x00c2, 0x00c3, 0x00c4, 0x00c5, 0x00c6, 0x00c7,
- 0x00c8, 0x00c9, 0x00ca, 0x00cb, 0x00cc, 0x00cd, 0x00ce, 0x00cf,
- 0x00d0, 0x00d1, 0x00d2, 0x00d3, 0x00d4, 0x00d5, 0x00d6, 0x00d7,
- 0x00d8, 0x00d9, 0x00da, 0x00db, 0x00dc, 0x00dd, 0x00de, 0x00df,
- 0x00e0, 0x00e1, 0x00e2, 0x00e3, 0x00e4, 0x00e5, 0x00e6, 0x00e7,
- 0x00e8, 0x00e9, 0x00ea, 0x00eb, 0x00ec, 0x00ed, 0x00ee, 0x00ef,
- 0x00f0, 0x00f1, 0x00f2, 0x00f3, 0x00f4, 0x00f5, 0x00f6, 0x00f7,
- 0x00f8, 0x00f9, 0x00fa, 0x00fb, 0x00fc, 0x00fd, 0x00fe, 0x00ff,
+ 0x0080, 0x0081, 0x0082, 0x0083, 0x0084, 0x0085, 0x0086, 0x0087,
+ 0x0088, 0x0089, 0x008a, 0x008b, 0x008c, 0x008d, 0x008e, 0x008f,
+ 0x0090, 0x0091, 0x0092, 0x0093, 0x0094, 0x0095, 0x0096, 0x0097,
+ 0x0098, 0x0099, 0x009a, 0x009b, 0x009c, 0x009d, 0x009e, 0x009f,
+ 0x00a0, 0x00a1, 0x00a2, 0x00a3, 0x20ac, 0x00a5, 0x0160, 0x00a7,
+ 0x0161, 0x00a9, 0x00aa, 0x00ab, 0x00ac, 0x00ad, 0x00ae, 0x00af,
+ 0x00b0, 0x00b1, 0x00b2, 0x00b3, 0x017d, 0x00b5, 0x00b6, 0x00b7,
+ 0x017e, 0x00b9, 0x00ba, 0x00bb, 0x0152, 0x0153, 0x0178, 0x00bf,
+ 0x00c0, 0x00c1, 0x00c2, 0x00c3, 0x00c4, 0x00c5, 0x00c6, 0x00c7,
+ 0x00c8, 0x00c9, 0x00ca, 0x00cb, 0x00cc, 0x00cd, 0x00ce, 0x00cf,
+ 0x00d0, 0x00d1, 0x00d2, 0x00d3, 0x00d4, 0x00d5, 0x00d6, 0x00d7,
+ 0x00d8, 0x00d9, 0x00da, 0x00db, 0x00dc, 0x00dd, 0x00de, 0x00df,
+ 0x00e0, 0x00e1, 0x00e2, 0x00e3, 0x00e4, 0x00e5, 0x00e6, 0x00e7,
+ 0x00e8, 0x00e9, 0x00ea, 0x00eb, 0x00ec, 0x00ed, 0x00ee, 0x00ef,
+ 0x00f0, 0x00f1, 0x00f2, 0x00f3, 0x00f4, 0x00f5, 0x00f6, 0x00f7,
+ 0x00f8, 0x00f9, 0x00fa, 0x00fb, 0x00fc, 0x00fd, 0x00fe, 0x00ff,
};
static unsigned char const xmltranscodetable_ISO8859_15 [48 + 6 * 64] = {
@@ -3119,22 +3361,22 @@ static unsigned char const xmltranscodetable_ISO8859_15 [48 + 6 * 64] = {
};
static unsigned short const xmlunicodetable_ISO8859_16 [128] = {
- 0x0080, 0x0081, 0x0082, 0x0083, 0x0084, 0x0085, 0x0086, 0x0087,
- 0x0088, 0x0089, 0x008a, 0x008b, 0x008c, 0x008d, 0x008e, 0x008f,
- 0x0090, 0x0091, 0x0092, 0x0093, 0x0094, 0x0095, 0x0096, 0x0097,
- 0x0098, 0x0099, 0x009a, 0x009b, 0x009c, 0x009d, 0x009e, 0x009f,
- 0x00a0, 0x0104, 0x0105, 0x0141, 0x20ac, 0x201e, 0x0160, 0x00a7,
- 0x0161, 0x00a9, 0x0218, 0x00ab, 0x0179, 0x00ad, 0x017a, 0x017b,
- 0x00b0, 0x00b1, 0x010c, 0x0142, 0x017d, 0x201d, 0x00b6, 0x00b7,
- 0x017e, 0x010d, 0x0219, 0x00bb, 0x0152, 0x0153, 0x0178, 0x017c,
- 0x00c0, 0x00c1, 0x00c2, 0x0102, 0x00c4, 0x0106, 0x00c6, 0x00c7,
- 0x00c8, 0x00c9, 0x00ca, 0x00cb, 0x00cc, 0x00cd, 0x00ce, 0x00cf,
- 0x0110, 0x0143, 0x00d2, 0x00d3, 0x00d4, 0x0150, 0x00d6, 0x015a,
- 0x0170, 0x00d9, 0x00da, 0x00db, 0x00dc, 0x0118, 0x021a, 0x00df,
- 0x00e0, 0x00e1, 0x00e2, 0x0103, 0x00e4, 0x0107, 0x00e6, 0x00e7,
- 0x00e8, 0x00e9, 0x00ea, 0x00eb, 0x00ec, 0x00ed, 0x00ee, 0x00ef,
- 0x0111, 0x0144, 0x00f2, 0x00f3, 0x00f4, 0x0151, 0x00f6, 0x015b,
- 0x0171, 0x00f9, 0x00fa, 0x00fb, 0x00fc, 0x0119, 0x021b, 0x00ff,
+ 0x0080, 0x0081, 0x0082, 0x0083, 0x0084, 0x0085, 0x0086, 0x0087,
+ 0x0088, 0x0089, 0x008a, 0x008b, 0x008c, 0x008d, 0x008e, 0x008f,
+ 0x0090, 0x0091, 0x0092, 0x0093, 0x0094, 0x0095, 0x0096, 0x0097,
+ 0x0098, 0x0099, 0x009a, 0x009b, 0x009c, 0x009d, 0x009e, 0x009f,
+ 0x00a0, 0x0104, 0x0105, 0x0141, 0x20ac, 0x201e, 0x0160, 0x00a7,
+ 0x0161, 0x00a9, 0x0218, 0x00ab, 0x0179, 0x00ad, 0x017a, 0x017b,
+ 0x00b0, 0x00b1, 0x010c, 0x0142, 0x017d, 0x201d, 0x00b6, 0x00b7,
+ 0x017e, 0x010d, 0x0219, 0x00bb, 0x0152, 0x0153, 0x0178, 0x017c,
+ 0x00c0, 0x00c1, 0x00c2, 0x0102, 0x00c4, 0x0106, 0x00c6, 0x00c7,
+ 0x00c8, 0x00c9, 0x00ca, 0x00cb, 0x00cc, 0x00cd, 0x00ce, 0x00cf,
+ 0x0110, 0x0143, 0x00d2, 0x00d3, 0x00d4, 0x0150, 0x00d6, 0x015a,
+ 0x0170, 0x00d9, 0x00da, 0x00db, 0x00dc, 0x0118, 0x021a, 0x00df,
+ 0x00e0, 0x00e1, 0x00e2, 0x0103, 0x00e4, 0x0107, 0x00e6, 0x00e7,
+ 0x00e8, 0x00e9, 0x00ea, 0x00eb, 0x00ec, 0x00ed, 0x00ee, 0x00ef,
+ 0x0111, 0x0144, 0x00f2, 0x00f3, 0x00f4, 0x0151, 0x00f6, 0x015b,
+ 0x0171, 0x00f9, 0x00fa, 0x00fb, 0x00fc, 0x0119, 0x021b, 0x00ff,
};
static unsigned char const xmltranscodetable_ISO8859_16 [48 + 9 * 64] = {
@@ -3333,4 +3575,3 @@ xmlRegisterCharEncodingHandlersISO8859x (void) {
#define bottom_encoding
#include "elfgcchack.h"
-
diff --git a/include/libxml/encoding.h b/include/libxml/encoding.h
index c74b25f..3a3b9b2 100644
--- a/include/libxml/encoding.h
+++ b/include/libxml/encoding.h
@@ -27,6 +27,9 @@
#ifdef LIBXML_ICONV_ENABLED
#include <iconv.h>
#endif
+#ifdef LIBXML_ICU_ENABLED
+#include <unicode/ucnv.h>
+#endif
#ifdef __cplusplus
extern "C" {
#endif
@@ -125,6 +128,13 @@ typedef int (* xmlCharEncodingOutputFunc)(unsigned char *out, int *outlen,
* Block defining the handlers for non UTF-8 encodings.
* If iconv is supported, there are two extra fields.
*/
+#ifdef LIBXML_ICU_ENABLED
+struct _uconv_t {
+ UConverter *uconv; /* for conversion between an encoding and UTF-16 */
+ UConverter *utf8; /* for conversion between UTF-8 and UTF-16 */
+};
+typedef struct _uconv_t uconv_t;
+#endif
typedef struct _xmlCharEncodingHandler xmlCharEncodingHandler;
typedef xmlCharEncodingHandler *xmlCharEncodingHandlerPtr;
@@ -136,6 +146,10 @@ struct _xmlCharEncodingHandler {
iconv_t iconv_in;
iconv_t iconv_out;
#endif /* LIBXML_ICONV_ENABLED */
+#ifdef LIBXML_ICU_ENABLED
+ uconv_t *uconv_in;
+ uconv_t *uconv_out;
+#endif /* LIBXML_ICU_ENABLED */
};
#ifdef __cplusplus
diff --git a/include/libxml/parser.h b/include/libxml/parser.h
index dd79c42..47b3df1 100644
--- a/include/libxml/parser.h
+++ b/include/libxml/parser.h
@@ -1222,6 +1222,7 @@ typedef enum {
XML_WITH_DEBUG_MEM = 29,
XML_WITH_DEBUG_RUN = 30,
XML_WITH_ZLIB = 31,
+ XML_WITH_ICU = 32,
XML_WITH_NONE = 99999 /* just to be sure of allocation size */
} xmlFeature;
@@ -1232,4 +1233,3 @@ XMLPUBFUN int XMLCALL
}
#endif
#endif /* __XML_PARSER_H__ */
-
diff --git a/include/libxml/xmlversion.h.in b/include/libxml/xmlversion.h.in
index 4739f3a..9585654 100644
--- a/include/libxml/xmlversion.h.in
+++ b/include/libxml/xmlversion.h.in
@@ -269,6 +269,15 @@ XMLPUBFUN void XMLCALL xmlCheckVersion(int version);
#endif
/**
+ * LIBXML_ICU_ENABLED:
+ *
+ * Whether icu support is available
+ */
+#if @WITH_ICU@
+#define LIBXML_ICU_ENABLED
+#endif
+
+/**
* LIBXML_ISO8859X_ENABLED:
*
* Whether ISO-8859-* support is made available in case iconv is not
diff --git a/parser.c b/parser.c
index 5f8411e..d1c7888 100644
--- a/parser.c
+++ b/parser.c
@@ -954,6 +954,12 @@ xmlHasFeature(xmlFeature feature)
#else
return(0);
#endif
+ case XML_WITH_ICU:
+#ifdef LIBXML_ICU_ENABLED
+ return(1);
+#else
+ return(0);
+#endif
default:
break;
}
diff --git a/win32/Makefile.msvc b/win32/Makefile.msvc
index 908c088..bd0eb9d 100644
--- a/win32/Makefile.msvc
+++ b/win32/Makefile.msvc
@@ -71,6 +71,9 @@ LIBS = $(LIBS) wsock32.lib ws2_32.lib
!if "$(WITH_ICONV)" == "1"
LIBS = $(LIBS) iconv.lib
!endif
++!if "$(WITH_ICU)" == "1"
++LIBS = $(LIBS) icu.lib
++!endif
!if "$(WITH_ZLIB)" == "1"
LIBS = $(LIBS) zdll.lib
!endif
diff --git a/win32/configure.js b/win32/configure.js
index 3c960a6..342fd9d 100644
--- a/win32/configure.js
+++ b/win32/configure.js
@@ -40,6 +40,7 @@ var withXpath = true;
var withXptr = true;
var withXinclude = true;
var withIconv = true;
+var withIcu = false;
var withIso8859x = false;
var withZlib = false;
var withDebug = true;
@@ -124,6 +125,7 @@ function usage()
txt += " xptr: Enable XPointer support (" + (withXptr? "yes" : "no") + ")\n";
txt += " xinclude: Enable XInclude support (" + (withXinclude? "yes" : "no") + ")\n";
txt += " iconv: Enable iconv support (" + (withIconv? "yes" : "no") + ")\n";
+ txt += " icu: Enable icu support (" + (withIcu? "yes" : "no") + ")\n";
txt += " iso8859x: Enable ISO8859X support (" + (withIso8859x? "yes" : "no") + ")\n";
txt += " zlib: Enable zlib support (" + (withZlib? "yes" : "no") + ")\n";
txt += " xml_debug: Enable XML debbugging module (" + (withDebug? "yes" : "no") + ")\n";
@@ -233,6 +235,7 @@ function discoverVersion()
vf.WriteLine("WITH_XPTR=" + (withXptr? "1" : "0"));
vf.WriteLine("WITH_XINCLUDE=" + (withXinclude? "1" : "0"));
vf.WriteLine("WITH_ICONV=" + (withIconv? "1" : "0"));
+ vf.WriteLine("WITH_ICU=" + (withIcu? "1" : "0"));
vf.WriteLine("WITH_ISO8859X=" + (withIso8859x? "1" : "0"));
vf.WriteLine("WITH_ZLIB=" + (withZlib? "1" : "0"));
vf.WriteLine("WITH_DEBUG=" + (withDebug? "1" : "0"));
@@ -319,6 +322,8 @@ function configureLibxml()
of.WriteLine(s.replace(/\ WITH_XINCLUDE\@/, withXinclude? "1" : "0"));
} else if (s.search(/\ WITH_ICONV\@/) != -1) {
of.WriteLine(s.replace(/\ WITH_ICONV\@/, withIconv? "1" : "0"));
+ } else if (s.search(/\ WITH_ICU\@/) != -1) {
+ of.WriteLine(s.replace(/\ WITH_ICU\@/, withIcu? "1" : "0"));
} else if (s.search(/\ WITH_ISO8859X\@/) != -1) {
of.WriteLine(s.replace(/\ WITH_ISO8859X\@/, withIso8859x? "1" : "0"));
} else if (s.search(/\ WITH_ZLIB\@/) != -1) {
@@ -462,6 +467,8 @@ for (i = 0; (i < WScript.Arguments.length) && (error == 0); i++) {
withXinclude = strToBool(arg.substring(opt.length + 1, arg.length));
else if (opt == "iconv")
withIconv = strToBool(arg.substring(opt.length + 1, arg.length));
+ else if (opt == "icu")
+ withIcu = strToBool(arg.substring(opt.length + 1, arg.length));
else if (opt == "iso8859x")
withIso8859x = strToBool(arg.substring(opt.length + 1, arg.length));
else if (opt == "zlib")
@@ -611,7 +618,13 @@ if (compiler == "mingw")
makefile = ".\\Makefile.mingw";
else if (compiler == "bcb")
makefile = ".\\Makefile.bcb";
-fso.CopyFile(makefile, ".\\Makefile", true);
+var new_makefile = ".\\Makefile";
+var f = fso.FileExists(new_makefile);
+if (f) {
+ var t = fso.GetFile(new_makefile);
+ t.Attributes =0;
+}
+fso.CopyFile(makefile, new_makefile, true);
WScript.Echo("Created Makefile.");
// Create the config.h.
var confighsrc = "..\\include\\win32config.h";
@@ -640,6 +653,7 @@ txtOut += " XPath support: " + boolToStr(withXpath) + "\n";
txtOut += " XPointer support: " + boolToStr(withXptr) + "\n";
txtOut += " XInclude support: " + boolToStr(withXinclude) + "\n";
txtOut += " iconv support: " + boolToStr(withIconv) + "\n";
+txtOut += " icu support: " + boolToStr(withIcu) + "\n";
txtOut += " iso8859x support: " + boolToStr(withIso8859x) + "\n";
txtOut += " zlib support: " + boolToStr(withZlib) + "\n";
txtOut += " Debugging module: " + boolToStr(withDebug) + "\n";
[
Date Prev][
Date Next] [
Thread Prev][
Thread Next]
[
Thread Index]
[
Date Index]
[
Author Index]
