Merge commit '4b3f860e5420ade19a825dc7f4d75789489535ba'

Sibras · Sibras · commit 920899a23c5f · 2024-10-19T19:55:08.000+11:00
diff --git a/HTMLparser.c b/HTMLparser.c
@@ -6185,6 +6185,12 @@ htmlCtxtReset(htmlParserCtxtPtr ctxt)
     ctxt->extSubURI = NULL;
     DICT_FREE(ctxt->extSubSystem);
     ctxt->extSubSystem = NULL;
+
+    if (ctxt->directory != NULL) {
+        xmlFree(ctxt->directory);
+        ctxt->directory = NULL;
+    }
+
     if (ctxt->myDoc != NULL)
         xmlFreeDoc(ctxt->myDoc);
     ctxt->myDoc = NULL;
diff --git a/HTMLtree.c b/HTMLtree.c
@@ -385,22 +385,17 @@ htmlFindOutputEncoder(const char *encoding) {
     xmlCharEncodingHandler *handler = NULL;
 
     if (encoding != NULL) {
-	xmlCharEncoding enc;
+        int res;
 
-	enc = xmlParseCharEncoding(encoding);
-	if (enc != XML_CHAR_ENCODING_UTF8) {
-	    xmlOpenCharEncodingHandler(encoding, /* output */ 1, &handler);
-	    if (handler == NULL)
-		htmlSaveErr(XML_SAVE_UNKNOWN_ENCODING, NULL, encoding);
-	}
+        res = xmlOpenCharEncodingHandler(encoding, /* output */ 1,
+                                         &handler);
+        if (res != XML_ERR_OK)
+            htmlSaveErr(XML_SAVE_UNKNOWN_ENCODING, NULL, encoding);
     } else {
         /*
-         * Fallback to HTML or ASCII when the encoding is unspecified
+         * Fallback to HTML when the encoding is unspecified
          */
-        if (handler == NULL)
-            xmlOpenCharEncodingHandler("HTML", /* output */ 1, &handler);
-        if (handler == NULL)
-            xmlOpenCharEncodingHandler("ascii", /* output */ 1, &handler);
+        xmlOpenCharEncodingHandler("HTML", /* output */ 1, &handler);
     }
 
     return(handler);
diff --git a/NEWS b/NEWS
@@ -1,5 +1,24 @@
 NEWS file for libxml2
 
+v2.13.2: Jul 4 2024
+
+### Regressions
+
+- tree: Fix handling of empty strings in xmlNodeParseContent
+- valid: Restore ID lookup
+- parser: Reenable ctxt->directory
+- uri: Handle filesystem paths in xmlBuildRelativeURISafe
+- encoding: Make xmlFindCharEncodingHandler return UTF-8 handler
+- encoding: Fix encoding lookup with xmlOpenCharEncodingHandler
+- include: Define ATTRIBUTE_UNUSED for clang
+- uri: Fix xmlBuildURI with NULL base
+
+### Improvements
+
+- uri: Enable Windows paths on Cygwin
+- tests: Clarify licence of test/intsubset2.xml
+
+
 v2.13.1: Jun 19 2024
 
 ### Regressions
@@ -57,8 +76,12 @@ to be enabled by passing --with-zlib, --with-lzma or --with-http to
 configure. In legacy mode (--with-legacy) these options are enabled
 by default as before.
 
-Support for FTP and xpointer() XPath extensions will be removed in
-the next release.
+Support for FTP will be removed in the next release.
+
+Support for the range and point extensions of the xpointer() scheme
+will be removed in the next release. The rest of the XPointer
+implementation won't be affected. The xpointer() scheme will behave
+like the xpath1() scheme.
 
 Several more legacy symbols were deprecated. Users of the old "SAX1"
 API functions are encouraged to upgrade to the new "SAX2" API,
diff --git a/SAX2.c b/SAX2.c
@@ -412,6 +412,13 @@ xmlSAX2ResolveEntity(void *ctx, const xmlChar *publicId, const xmlChar *systemId
     if (ctxt->input != NULL)
 	base = BAD_CAST ctxt->input->filename;
 
+    /*
+     * We don't really need the 'directory' struct member, but some
+     * users set it manually to a base URI for memory streams.
+     */
+    if (base == NULL)
+        base = BAD_CAST ctxt->directory;
+
     if ((xmlStrlen(systemId) > XML_MAX_URI_LENGTH) ||
         (xmlStrlen(base) > XML_MAX_URI_LENGTH)) {
         xmlFatalErr(ctxt, XML_ERR_RESOURCE_LIMIT, "URI too long");
@@ -574,6 +581,13 @@ xmlSAX2EntityDecl(void *ctx, const xmlChar *name, int type,
             }
         }
 
+        /*
+         * We don't really need the 'directory' struct member, but some
+         * users set it manually to a base URI for memory streams.
+         */
+        if (base == NULL)
+            base = ctxt->directory;
+
         res = xmlBuildURISafe(systemId, (const xmlChar *) base, &URI);
 
         if (URI == NULL) {
@@ -1255,6 +1269,7 @@ xmlSAX2AttributeInternal(void *ctx, const xmlChar *fullname,
     } else
 #endif /* LIBXML_VALID_ENABLED */
            if (((ctxt->loadsubset & XML_SKIP_IDS) == 0) &&
+               (ctxt->input->entity == NULL) &&
                /* Don't create IDs containing entity references */
                (ret->children != NULL) &&
                (ret->children->type == XML_TEXT_NODE) &&
@@ -1987,6 +2002,7 @@ xmlSAX2AttributeNs(xmlParserCtxtPtr ctxt,
     } else
 #endif /* LIBXML_VALID_ENABLED */
            if (((ctxt->loadsubset & XML_SKIP_IDS) == 0) &&
+               (ctxt->input->entity == NULL) &&
                /* Don't create IDs containing entity references */
                (ret->children != NULL) &&
                (ret->children->type == XML_TEXT_NODE) &&
diff --git a/configure.ac b/configure.ac
@@ -3,7 +3,7 @@ AC_PREREQ([2.63])
 
 m4_define([MAJOR_VERSION], 2)
 m4_define([MINOR_VERSION], 13)
-m4_define([MICRO_VERSION], 1)
+m4_define([MICRO_VERSION], 2)
 
 AC_INIT([libxml2],[MAJOR_VERSION.MINOR_VERSION.MICRO_VERSION])
 AC_CONFIG_SRCDIR([entities.c])
diff --git a/encoding.c b/encoding.c
@@ -1279,9 +1279,8 @@ DECLARE_ISO_FUNCS(16)
     { (char *) name, in, out EMPTY_ICONV EMPTY_UCONV }
 
 static const xmlCharEncodingHandler defaultHandlers[] = {
-    MAKE_HANDLER("UTF-8", UTF8ToUTF8, UTF8ToUTF8)
 #ifdef LIBXML_OUTPUT_ENABLED
-    ,MAKE_HANDLER("UTF-16LE", UTF16LEToUTF8, UTF8ToUTF16LE)
+    MAKE_HANDLER("UTF-16LE", UTF16LEToUTF8, UTF8ToUTF16LE)
     ,MAKE_HANDLER("UTF-16BE", UTF16BEToUTF8, UTF8ToUTF16BE)
     ,MAKE_HANDLER("UTF-16", UTF16LEToUTF8, UTF8ToUTF16)
     ,MAKE_HANDLER("ISO-8859-1", isolat1ToUTF8, UTF8Toisolat1)
@@ -1291,7 +1290,7 @@ static const xmlCharEncodingHandler defaultHandlers[] = {
     ,MAKE_HANDLER("HTML", NULL, UTF8ToHtml)
 #endif
 #else
-    ,MAKE_HANDLER("UTF-16LE", UTF16LEToUTF8, NULL)
+    MAKE_HANDLER("UTF-16LE", UTF16LEToUTF8, NULL)
     ,MAKE_HANDLER("UTF-16BE", UTF16BEToUTF8, NULL)
     ,MAKE_HANDLER("UTF-16", UTF16LEToUTF8, NULL)
     ,MAKE_HANDLER("ISO-8859-1", isolat1ToUTF8, NULL)
@@ -1321,10 +1320,13 @@ static const xmlCharEncodingHandler defaultHandlers[] = {
 #define NUM_DEFAULT_HANDLERS \
     (sizeof(defaultHandlers) / sizeof(defaultHandlers[0]))
 
-static const xmlCharEncodingHandler *xmlUTF16LEHandler = &defaultHandlers[1];
-static const xmlCharEncodingHandler *xmlUTF16BEHandler = &defaultHandlers[2];
-static const xmlCharEncodingHandler *xmlLatin1Handler = &defaultHandlers[4];
-static const xmlCharEncodingHandler *xmlAsciiHandler = &defaultHandlers[5];
+static const xmlCharEncodingHandler xmlUTF8Handler =
+    MAKE_HANDLER("UTF-8", UTF8ToUTF8, UTF8ToUTF8);
+
+static const xmlCharEncodingHandler *xmlUTF16LEHandler = &defaultHandlers[0];
+static const xmlCharEncodingHandler *xmlUTF16BEHandler = &defaultHandlers[1];
+static const xmlCharEncodingHandler *xmlLatin1Handler = &defaultHandlers[3];
+static const xmlCharEncodingHandler *xmlAsciiHandler = &defaultHandlers[4];
 
 /* the size should be growable, but it's not a big deal ... */
 #define MAX_ENCODING_HANDLERS 50
@@ -1922,6 +1924,9 @@ xmlGetCharEncodingHandler(xmlCharEncoding enc) {
  *
  * The handler must be closed with xmlCharEncCloseFunc.
  *
+ * If the encoding is UTF-8, a NULL handler and no error code will
+ * be returned.
+ *
  * Available since 2.13.0.
  *
  * Returns an xmlParserErrors error code.
@@ -1941,6 +1946,10 @@ xmlOpenCharEncodingHandler(const char *name, int output,
     if (name == NULL)
         return(XML_ERR_ARGUMENT);
 
+    if ((xmlStrcasecmp(BAD_CAST name, BAD_CAST "UTF-8") == 0) ||
+        (xmlStrcasecmp(BAD_CAST name, BAD_CAST "UTF8") == 0))
+        return(XML_ERR_OK);
+
     /*
      * Do the alias resolution
      */
@@ -1957,6 +1966,9 @@ xmlOpenCharEncodingHandler(const char *name, int output,
 
     /*
      * Fallback using the canonical names
+     *
+     * TODO: We should make sure that the name of the returned
+     * handler equals norig.
      */
     enc = xmlParseCharEncoding(norig);
     return(xmlLookupCharEncodingHandler(enc, out));
@@ -1976,6 +1988,14 @@ xmlCharEncodingHandlerPtr
 xmlFindCharEncodingHandler(const char *name) {
     xmlCharEncodingHandler *ret;
 
+    /*
+     * This handler shouldn't be used, but we must return a non-NULL
+     * handler.
+     */
+    if ((xmlStrcasecmp(BAD_CAST name, BAD_CAST "UTF-8") == 0) ||
+        (xmlStrcasecmp(BAD_CAST name, BAD_CAST "UTF8") == 0))
+        return((xmlCharEncodingHandlerPtr) &xmlUTF8Handler);
+
     xmlOpenCharEncodingHandler(name, 0, &ret);
     return(ret);
 }
diff --git a/include/libxml/xmlexports.h b/include/libxml/xmlexports.h
@@ -42,7 +42,7 @@
  */
 
 #ifndef ATTRIBUTE_UNUSED
-  #if __GNUC__ * 100 + __GNUC_MINOR__ >= 207
+  #if __GNUC__ * 100 + __GNUC_MINOR__ >= 207 || defined(__clang__)
     #define ATTRIBUTE_UNUSED __attribute__((unused))
   #else
     #define ATTRIBUTE_UNUSED
diff --git a/meson.build b/meson.build
@@ -1,7 +1,7 @@
 project(
     'libxml2',
     'c',
-    version: '2.13.0',
+    version: '2.13.2',
     license: 'MIT',
     default_options: ['buildtype=debug', 'warning_level=3'],
     meson_version: '>= 0.61',
diff --git a/parser.c b/parser.c
@@ -1940,8 +1940,11 @@ xmlCtxtGrowAttrs(xmlParserCtxtPtr ctxt, int nr) {
 int
 inputPush(xmlParserCtxtPtr ctxt, xmlParserInputPtr value)
 {
+    char *directory = NULL;
+
     if ((ctxt == NULL) || (value == NULL))
         return(-1);
+
     if (ctxt->inputNr >= ctxt->inputMax) {
         size_t newSize = ctxt->inputMax * 2;
         xmlParserInputPtr *tmp;
@@ -1955,9 +1958,24 @@ inputPush(xmlParserCtxtPtr ctxt, xmlParserInputPtr value)
         ctxt->inputTab = tmp;
         ctxt->inputMax = newSize;
     }
+
+    if ((ctxt->inputNr == 0) && (value->filename != NULL)) {
+        directory = xmlParserGetDirectory(value->filename);
+        if (directory == NULL) {
+            xmlErrMemory(ctxt);
+            return(-1);
+        }
+    }
+
     ctxt->inputTab[ctxt->inputNr] = value;
     ctxt->input = value;
-    return (ctxt->inputNr++);
+
+    if (ctxt->inputNr == 0) {
+        xmlFree(ctxt->directory);
+        ctxt->directory = directory;
+    }
+
+    return(ctxt->inputNr++);
 }
 /**
  * inputPop:
@@ -13269,6 +13287,12 @@ xmlCtxtReset(xmlParserCtxtPtr ctxt)
     ctxt->extSubURI = NULL;
     DICT_FREE(ctxt->extSubSystem);
     ctxt->extSubSystem = NULL;
+
+    if (ctxt->directory != NULL) {
+        xmlFree(ctxt->directory);
+        ctxt->directory = NULL;
+    }
+
     if (ctxt->myDoc != NULL)
         xmlFreeDoc(ctxt->myDoc);
     ctxt->myDoc = NULL;
diff --git a/parserInternals.c b/parserInternals.c
@@ -2469,6 +2469,7 @@ xmlFreeParserCtxt(xmlParserCtxtPtr ctxt)
     if (ctxt->sax != NULL)
 #endif /* LIBXML_SAX1_ENABLED */
         xmlFree(ctxt->sax);
+    if (ctxt->directory != NULL) xmlFree(ctxt->directory);
     if (ctxt->vctxt.nodeTab != NULL) xmlFree(ctxt->vctxt.nodeTab);
     if (ctxt->atts != NULL) xmlFree((xmlChar * *)ctxt->atts);
     if (ctxt->dict != NULL) xmlDictFree(ctxt->dict);
diff --git a/result/intsubset2.xml b/result/intsubset2.xml
@@ -1,4 +1,13 @@
 <?xml version="1.0"?>
+<!--
+Copyright (C) Electronic Dictionary Research and Development Group
+Released under Creative Commons Attribution-ShareAlike Licence (V4.0)
+
+This file only contains the kanjidic2 DTD without the actual database.
+
+http://nihongo.monash.edu/kanjidic2/index.html
+http://www.edrdg.org/edrdg/licence.html
+-->
 <!DOCTYPE kanjidic2 [
 <!-- Version 1.3
 	This is the DTD of the XML-format kanji file combining information from
diff --git a/result/intsubset2.xml.rde b/result/intsubset2.xml.rde
@@ -1,3 +1,12 @@
+0 8 #comment 0 1 
+Copyright (C) Electronic Dictionary Research and Development Group
+Released under Creative Commons Attribution-ShareAlike Licence (V4.0)
+
+This file only contains the kanjidic2 DTD without the actual database.
+
+http://nihongo.monash.edu/kanjidic2/index.html
+http://www.edrdg.org/edrdg/licence.html
+
 0 10 kanjidic2 0 0
 0 1 kanjidic2 0 0
 1 14 #text 0 1 
diff --git a/result/intsubset2.xml.rdr b/result/intsubset2.xml.rdr
@@ -1,3 +1,12 @@
+0 8 #comment 0 1 
+Copyright (C) Electronic Dictionary Research and Development Group
+Released under Creative Commons Attribution-ShareAlike Licence (V4.0)
+
+This file only contains the kanjidic2 DTD without the actual database.
+
+http://nihongo.monash.edu/kanjidic2/index.html
+http://www.edrdg.org/edrdg/licence.html
+
 0 10 kanjidic2 0 0
 0 1 kanjidic2 0 0
 1 14 #text 0 1 
diff --git a/result/intsubset2.xml.sax b/result/intsubset2.xml.sax
@@ -1,5 +1,14 @@
 SAX.setDocumentLocator()
 SAX.startDocument()
+SAX.comment(
+Copyright (C) Electronic Dictionary Research and Development Group
+Released under Creative Commons Attribution-ShareAlike Licence (V4.0)
+
+This file only contains the kanjidic2 DTD without the actual database.
+
+http://nihongo.monash.edu/kanjidic2/index.html
+http://www.edrdg.org/edrdg/licence.html
+)
 SAX.internalSubset(kanjidic2, , )
 SAX.comment( Version 1.3
 	This is the DTD of the XML-format kanji file combining information from
diff --git a/result/intsubset2.xml.sax2 b/result/intsubset2.xml.sax2
@@ -1,5 +1,14 @@
 SAX.setDocumentLocator()
 SAX.startDocument()
+SAX.comment(
+Copyright (C) Electronic Dictionary Research and Development Group
+Released under Creative Commons Attribution-ShareAlike Licence (V4.0)
+
+This file only contains the kanjidic2 DTD without the actual database.
+
+http://nihongo.monash.edu/kanjidic2/index.html
+http://www.edrdg.org/edrdg/licence.html
+)
 SAX.internalSubset(kanjidic2, , )
 SAX.comment( Version 1.3
 	This is the DTD of the XML-format kanji file combining information from
diff --git a/result/noent/intsubset2.xml b/result/noent/intsubset2.xml
@@ -1,4 +1,13 @@
 <?xml version="1.0"?>
+<!--
+Copyright (C) Electronic Dictionary Research and Development Group
+Released under Creative Commons Attribution-ShareAlike Licence (V4.0)
+
+This file only contains the kanjidic2 DTD without the actual database.
+
+http://nihongo.monash.edu/kanjidic2/index.html
+http://www.edrdg.org/edrdg/licence.html
+-->
 <!DOCTYPE kanjidic2 [
 <!-- Version 1.3
 	This is the DTD of the XML-format kanji file combining information from
diff --git a/result/noent/intsubset2.xml.sax2 b/result/noent/intsubset2.xml.sax2
@@ -1,5 +1,14 @@
 SAX.setDocumentLocator()
 SAX.startDocument()
+SAX.comment(
+Copyright (C) Electronic Dictionary Research and Development Group
+Released under Creative Commons Attribution-ShareAlike Licence (V4.0)
+
+This file only contains the kanjidic2 DTD without the actual database.
+
+http://nihongo.monash.edu/kanjidic2/index.html
+http://www.edrdg.org/edrdg/licence.html
+)
 SAX.internalSubset(kanjidic2, , )
 SAX.comment( Version 1.3
 	This is the DTD of the XML-format kanji file combining information from
diff --git a/result/xmlid/id_tst4.xml b/result/xmlid/id_tst4.xml
diff --git a/test/intsubset2.xml b/test/intsubset2.xml
diff --git a/testparser.c b/testparser.c
diff --git a/tree.c b/tree.c
diff --git a/uri.c b/uri.c
diff --git a/valid.c b/valid.c
diff --git a/xmlsave.c b/xmlsave.c
