better handling of conditional features more testing on parser contexts

* gentest.py testapi.c: better handling of conditional features * HTMLparser.c SAX2.c parserInternals.c xmlwriter.c: more testing on parser contexts closed leaks, error messages Daniel
2025-10-21 14:53:44 +03:00 · 2004-11-09 14:59:59 +00:00
parent 645a924a9d
commit a521d28751
7 changed files with 631 additions and 158 deletions
--- a/6
+++ b/6
@@ -1,3 +1,9 @@
+Tue Nov  9 15:59:50 CET 2004 Daniel Veillard <daniel@veillard.com>
+
+	* gentest.py testapi.c: better handling of conditional features
+	* HTMLparser.c SAX2.c parserInternals.c xmlwriter.c: more testing
+	  on parser contexts closed leaks, error messages
+
 Tue Nov  9 10:21:37 GMT 2004 William Brack <wbrack@mmm.com.hk>

 	* xpath.c: fixed problem concerning XPath context corruption
--- a/HTMLparser.c
+++ b/HTMLparser.c
@@ -5675,8 +5675,12 @@ htmlCtxtReset(htmlParserCtxtPtr ctxt)
    ctxt->input = NULL;

    ctxt->spaceNr = 0;
-    ctxt->spaceTab[0] = -1;
-    ctxt->space = &ctxt->spaceTab[0];
+    if (ctxt->spaceTab != NULL) {
+	ctxt->spaceTab[0] = -1;
+	ctxt->space = &ctxt->spaceTab[0];
+    } else {
+	ctxt->space = NULL;
+    }


    ctxt->nodeNr = 0;
--- a/SAX2.c
+++ b/SAX2.c
@@ -1421,7 +1421,7 @@ xmlSAX2StartElement(void *ctx, const xmlChar *fullname, const xmlChar **atts)
    const xmlChar *value;
    int i;

-    if ((ctx == NULL) || (fullname == NULL)) return;
+    if ((ctx == NULL) || (fullname == NULL) || (ctxt->myDoc == NULL)) return;
    parent = ctxt->node;
 #ifdef DEBUG_SAX
    xmlGenericError(xmlGenericErrorContext,
--- a/gentest.py
+++ b/gentest.py
@@ -20,6 +20,80 @@ skipped_modules = [ "SAX", "xlink", "threads", "globals",
  "DOCBparser",
 ]

+#
+# defines for each module
+#
+modules_defines = {
+    "HTMLparser": "LIBXML_HTML_ENABLED",
+    "catalog": "LIBXML_CATALOG_ENABLED",
+    "xmlreader": "LIBXML_READER_ENABLED",
+    "relaxng": "LIBXML_SCHEMAS_ENABLED",
+    "schemasInternals": "LIBXML_SCHEMAS_ENABLED",
+    "xmlschemas": "LIBXML_SCHEMAS_ENABLED",
+    "xmlschemastypes": "LIBXML_SCHEMAS_ENABLED",
+    "xpath": "LIBXML_XPATH_ENABLED",
+    "xpathInternals": "LIBXML_XPATH_ENABLED",
+    "xinclude": "LIBXML_XINCLUDE_ENABLED",
+    "xpointer": "LIBXML_XPTR_ENABLED",
+    "xmlregexp" : "LIBXML_REGEXP_ENABLED",
+    "xmlautomata" : "LIBXML_AUTOMATA_ENABLED",
+    "xmlsave" : "LIBXML_OUTPUT_ENABLED",
+    "DOCBparser" : "LIBXML_DOCB_ENABLED",
+}
+
+#
+# defines for specific functions
+#
+function_defines = {
+    "htmlDefaultSAXHandlerInit": "LIBXML_HTML_ENABLED",
+    "xmlSAX2EndElement" : "LIBXML_SAX1_ENABLED",
+    "xmlSAX2StartElement" : "LIBXML_SAX1_ENABLED",
+    "xmlSAXDefaultVersion" : "LIBXML_SAX1_ENABLED",
+    "UTF8Toisolat1" : "LIBXML_OUTPUT_ENABLED",
+    "xmlCleanupPredefinedEntities": "LIBXML_LEGACY_ENABLED",
+    "xmlInitializePredefinedEntities": "LIBXML_LEGACY_ENABLED",
+    "xmlSetFeature": "LIBXML_LEGACY_ENABLED",
+    "xmlGetFeature": "LIBXML_LEGACY_ENABLED",
+    "xmlGetFeaturesList": "LIBXML_LEGACY_ENABLED",
+    "xmlIOParseDTD": "LIBXML_VALID_ENABLED",
+    "xmlParseDTD": "LIBXML_VALID_ENABLED",
+    "xmlParseDoc": "LIBXML_SAX1_ENABLED",
+    "xmlParseMemory": "LIBXML_SAX1_ENABLED",
+    "xmlRecoverDoc": "LIBXML_SAX1_ENABLED",
+    "xmlParseFile": "LIBXML_SAX1_ENABLED",
+    "xmlRecoverFile": "LIBXML_SAX1_ENABLED",
+    "xmlRecoverMemory": "LIBXML_SAX1_ENABLED",
+    "xmlSAXParseFileWithData": "LIBXML_SAX1_ENABLED",
+    "xmlSAXParseMemory": "LIBXML_SAX1_ENABLED",
+    "xmlSAXUserParseMemory": "LIBXML_SAX1_ENABLED",
+    "xmlSAXParseDoc": "LIBXML_SAX1_ENABLED",
+    "xmlSAXParseDTD": "LIBXML_SAX1_ENABLED",
+    "xmlSAXUserParseFile": "LIBXML_SAX1_ENABLED",
+    "xmlParseEntity": "LIBXML_SAX1_ENABLED",
+    "xmlParseExternalEntity": "LIBXML_SAX1_ENABLED",
+    "xmlSAXParseMemoryWithData": "LIBXML_SAX1_ENABLED",
+    "xmlParseBalancedChunkMemory": "LIBXML_SAX1_ENABLED",
+    "xmlParseBalancedChunkMemoryRecover": "LIBXML_SAX1_ENABLED",
+    "xmlSetupParserForBuffer": "LIBXML_SAX1_ENABLED",
+    "xmlStopParser": "LIBXML_PUSH_ENABLED",
+    "xmlAttrSerializeTxtContent": "LIBXML_OUTPUT_ENABLED",
+    "xmlSAXParseFile": "LIBXML_SAX1_ENABLED",
+    "xmlSAXParseEntity": "LIBXML_SAX1_ENABLED",
+    "xmlNewTextChild": "LIBXML_TREE_ENABLED",
+    "xmlNewDocRawNode": "LIBXML_TREE_ENABLED",
+    "xmlNewProp": "LIBXML_TREE_ENABLED",
+    "xmlReconciliateNs": "LIBXML_TREE_ENABLED",
+    "xmlValidateNCName": "LIBXML_TREE_ENABLED",
+    "xmlValidateNMToken": "LIBXML_TREE_ENABLED",
+    "xmlValidateName": "LIBXML_TREE_ENABLED",
+    "xmlNewChild": "LIBXML_TREE_ENABLED",
+    "xmlValidateQName": "LIBXML_TREE_ENABLED",
+    "xmlSprintfElementContent": "LIBXML_OUTPUT_ENABLED",
+    "xmlValidGetPotentialChildren" : "LIBXML_VALID_ENABLED",
+    "xmlValidGetValidElements" : "LIBXML_VALID_ENABLED",
+    "docbDefaultSAXHandlerInit" : "LIBXML_DOCB_ENABLED",
+}
+
 #
 # Some function really need to be skipped for the tests.
 #
@@ -97,14 +171,26 @@ skipped_memcheck = [ "xmlLoadCatalog", "xmlAddEncodingAlias",
 # Extra code needed for some test cases
 #
 extra_pre_call = {
-   "xmlSAXUserParseFile":
-       "if (sax == (xmlSAXHandlerPtr)&xmlDefaultSAXHandler) user_data = NULL;",
-   "xmlSAXUserParseMemory":
-       "if (sax == (xmlSAXHandlerPtr)&xmlDefaultSAXHandler) user_data = NULL;",
-   "xmlParseBalancedChunkMemory":
-       "if (sax == (xmlSAXHandlerPtr)&xmlDefaultSAXHandler) user_data = NULL;",
-   "xmlParseBalancedChunkMemoryRecover":
-       "if (sax == (xmlSAXHandlerPtr)&xmlDefaultSAXHandler) user_data = NULL;",
+   "xmlSAXUserParseFile": """
+#ifdef LIBXML_SAX1_ENABLED
+        if (sax == (xmlSAXHandlerPtr)&xmlDefaultSAXHandler) user_data = NULL;
+#endif
+""",
+   "xmlSAXUserParseMemory": """
+#ifdef LIBXML_SAX1_ENABLED
+        if (sax == (xmlSAXHandlerPtr)&xmlDefaultSAXHandler) user_data = NULL;
+#endif
+""",
+   "xmlParseBalancedChunkMemory": """
+#ifdef LIBXML_SAX1_ENABLED
+        if (sax == (xmlSAXHandlerPtr)&xmlDefaultSAXHandler) user_data = NULL;
+#endif
+""",
+   "xmlParseBalancedChunkMemoryRecover": """
+#ifdef LIBXML_SAX1_ENABLED
+        if (sax == (xmlSAXHandlerPtr)&xmlDefaultSAXHandler) user_data = NULL;
+#endif
+""",
   "xmlParserInputBufferCreateFd":
       "if (fd >= 0) fd = -1;",
 }
@@ -151,8 +237,13 @@ extra_post_call = {
   "xmlCopyNamespace": "if (ret_val != NULL) xmlFreeNs(ret_val);",
   "xmlCopyNamespaceList": "if (ret_val != NULL) xmlFreeNsList(ret_val);",
   "xmlNewTextWriter": "if (ret_val != NULL) out = NULL;",
-   "xmlNewTextWriterPushParser": "if (ret_val != NULL) ctxt = NULL;",
+   "xmlNewTextWriterPushParser": "if (ctxt != NULL) {xmlFreeDoc(ctxt->myDoc); ctxt->myDoc = NULL;} if (ret_val != NULL) ctxt = NULL;",
   "xmlNewIOInputStream": "if (ret_val != NULL) input = NULL;",
+   "htmlParseChunk": "if (ctxt != NULL) {xmlFreeDoc(ctxt->myDoc); ctxt->myDoc = NULL;}",
+   "htmlParseDocument": "if (ctxt != NULL) {xmlFreeDoc(ctxt->myDoc); ctxt->myDoc = NULL;}",
+   "xmlParseDocument": "if (ctxt != NULL) {xmlFreeDoc(ctxt->myDoc); ctxt->myDoc = NULL;}",
+   "xmlParseChunk": "if (ctxt != NULL) {xmlFreeDoc(ctxt->myDoc); ctxt->myDoc = NULL;}",
+   "xmlParseExtParsedEnt": "if (ctxt != NULL) {xmlFreeDoc(ctxt->myDoc); ctxt->myDoc = NULL;}",
 }

 modules = []
@@ -217,9 +308,6 @@ def type_convert(str, name, info, module, function, pos):
    res = string.replace(str, " *", "_ptr")
 #    res = string.replace(str, "*", "_ptr")
    res = string.replace(res, " ", "_")
-    res = string.replace(res, "htmlNode", "xmlNode")
-    res = string.replace(res, "htmlDoc", "xmlDoc")
-    res = string.replace(res, "htmlParser", "xmlParser")
    if res == 'const_char_ptr':
        if string.find(name, "file") != -1 or \
           string.find(name, "uri") != -1 or \
@@ -301,6 +389,10 @@ def is_known_param_type(name, rtype):
 	else:
 	    crtype = rtype

+        define = 0
+	if modules_defines.has_key(module):
+	    test.write("#ifdef %s\n" % (modules_defines[module]))
+	    define = 1
        test.write("""
 #define gen_nb_%s 1
 static %s gen_%s(int no ATTRIBUTE_UNUSED, int nr ATTRIBUTE_UNUSED) {
@@ -309,6 +401,8 @@ static %s gen_%s(int no ATTRIBUTE_UNUSED, int nr ATTRIBUTE_UNUSED) {
 static void des_%s(int no ATTRIBUTE_UNUSED, %s val ATTRIBUTE_UNUSED, int nr ATTRIBUTE_UNUSED) {
 }
 """ % (name, crtype, name, name, rtype))
+        if define == 1:
+	    test.write("#endif\n\n")
        add_generated_param_type(name)
        return 1

@@ -386,6 +480,8 @@ for enum in enums:
    name = enum.xpathEval('string(@name)')
    if name == None:
        continue;
+    module = enum.xpathEval('string(@file)')
+    define = 0

    if is_known_param_type(name, name) == 0:
 	values = ctxt.xpathEval("/api/symbols/enum[@type='%s']" % name)
@@ -402,6 +498,9 @@ for enum in enums:
 	if vals == []:
 	    print "Didn't found any value for enum %s" % (name)
 	    continue
+	if modules_defines.has_key(module):
+	    test.write("#ifdef %s\n" % (modules_defines[module]))
+	    define = 1
 	test.write("#define gen_nb_%s %d\n" % (name, len(vals)))
 	test.write("""static %s gen_%s(int no, int nr ATTRIBUTE_UNUSED) {\n""" %
 	           (name, name))
@@ -415,6 +514,9 @@ for enum in enums:
 	known_param_types.append(name)

    if is_known_return_type(name) == 0:
+	if define == 0 and modules_defines.has_key(module):
+	    test.write("#ifdef %s\n" % (modules_defines[module]))
+	    define = 1
        test.write("""static void des_%s(int no ATTRIBUTE_UNUSED, %s val ATTRIBUTE_UNUSED, int nr ATTRIBUTE_UNUSED) {
 }
 static void desret_%s(%s val ATTRIBUTE_UNUSED) {
@@ -422,6 +524,8 @@ static void desret_%s(%s val ATTRIBUTE_UNUSED) {

 """ % (name, name, name, name))
 	known_return_types.append(name)
+    if define == 1:
+        test.write("#endif\n\n")

 #
 # Load the interfaces
@@ -567,6 +671,11 @@ test_%s(void) {
    except:
        pass

+    define = 0
+    if function_defines.has_key(name):
+        test.write("#ifdef %s\n" % (function_defines[name]))
+	define = 1
+    
    # Declare the memory usage counter
    no_mem = is_skipped_memcheck(name)
    if no_mem == 0:
@@ -672,6 +781,8 @@ test_%s(void) {
    while nb_cond > 0:
        test.write("#endif\n")
 	nb_cond = nb_cond -1
+    if define == 1:
+        test.write("#endif\n")

    nb_tests = nb_tests + 1;

--- a/parserInternals.c
+++ b/parserInternals.c
@@ -574,15 +574,20 @@ encoding_error:
     * to ISO-Latin-1 (if you don't like this policy, just declare the
     * encoding !)
     */
-    __xmlErrEncoding(ctxt, XML_ERR_INVALID_CHAR,
-		   "Input is not proper UTF-8, indicate encoding !\n",
-		   NULL, NULL);
-    if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL) &&
-        (ctxt->input != NULL)) {
-        ctxt->sax->error(ctxt->userData,
-                         "Bytes: 0x%02X 0x%02X 0x%02X 0x%02X\n",
-                         ctxt->input->cur[0], ctxt->input->cur[1],
-                         ctxt->input->cur[2], ctxt->input->cur[3]);
+    if ((ctxt == NULL) || (ctxt->input == NULL) ||
+        (ctxt->input->end - ctxt->input->cur < 4)) {
+	__xmlErrEncoding(ctxt, XML_ERR_INVALID_CHAR,
+		     "Input is not proper UTF-8, indicate encoding !\n",
+		     NULL, NULL);
+    } else {
+        char buffer[150];
+
+	snprintf(buffer, 149, "Bytes: 0x%02X 0x%02X 0x%02X 0x%02X\n",
+			ctxt->input->cur[0], ctxt->input->cur[1],
+			ctxt->input->cur[2], ctxt->input->cur[3]);
+	__xmlErrEncoding(ctxt, XML_ERR_INVALID_CHAR,
+		     "Input is not proper UTF-8, indicate encoding !\n%s",
+		     BAD_CAST buffer, NULL);
    }
    ctxt->charset = XML_CHAR_ENCODING_8859_1;
    ctxt->input->cur++;
@@ -722,14 +727,15 @@ encoding_error:
     * to ISO-Latin-1 (if you don't like this policy, just declare the
     * encoding !)
     */
-    __xmlErrEncoding(ctxt, XML_ERR_INVALID_CHAR,
-		   "Input is not proper UTF-8, indicate encoding !\n",
-		   NULL, NULL);
-    if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL) &&
-        (ctxt->input != NULL)) {
-	ctxt->sax->error(ctxt->userData, "Bytes: 0x%02X 0x%02X 0x%02X 0x%02X\n",
+    {
+        char buffer[150];
+
+	snprintf(buffer, 149, "Bytes: 0x%02X 0x%02X 0x%02X 0x%02X\n",
 			ctxt->input->cur[0], ctxt->input->cur[1],
 			ctxt->input->cur[2], ctxt->input->cur[3]);
+	__xmlErrEncoding(ctxt, XML_ERR_INVALID_CHAR,
+		     "Input is not proper UTF-8, indicate encoding !\n%s",
+		     BAD_CAST buffer, NULL);
    }
    ctxt->charset = XML_CHAR_ENCODING_8859_1; 
    *len = 1;
@@ -817,6 +823,16 @@ xmlStringCurrentChar(xmlParserCtxtPtr ctxt, const xmlChar * cur, int *len)
    return ((int) *cur);
 encoding_error:

+    /*
+     * An encoding problem may arise from a truncated input buffer
+     * splitting a character in the middle. In that case do not raise
+     * an error but return 0 to endicate an end of stream problem
+     */
+    if ((ctxt == NULL) || (ctxt->input == NULL) ||
+        (ctxt->input->end - ctxt->input->cur < 4)) {
+	*len = 0;
+	return(0);
+    }
    /*
     * If we detect an UTF8 error that probably mean that the
     * input encoding didn't get properly advertised in the
@@ -824,15 +840,15 @@ encoding_error:
     * to ISO-Latin-1 (if you don't like this policy, just declare the
     * encoding !)
     */
-    __xmlErrEncoding(ctxt, XML_ERR_INVALID_CHAR,
-		   "Input is not proper UTF-8, indicate encoding !\n",
-		   NULL, NULL);
-    if ((ctxt != NULL) && (ctxt->sax != NULL) && (ctxt->sax->error != NULL) &&
-        (ctxt->input != NULL)) {
-	ctxt->sax->error(ctxt->userData,
-			 "Bytes: 0x%02X 0x%02X 0x%02X 0x%02X\n",
-			 ctxt->input->cur[0], ctxt->input->cur[1],
-			 ctxt->input->cur[2], ctxt->input->cur[3]);
+    {
+        char buffer[150];
+
+	snprintf(buffer, 149, "Bytes: 0x%02X 0x%02X 0x%02X 0x%02X\n",
+			ctxt->input->cur[0], ctxt->input->cur[1],
+			ctxt->input->cur[2], ctxt->input->cur[3]);
+	__xmlErrEncoding(ctxt, XML_ERR_INVALID_CHAR,
+		     "Input is not proper UTF-8, indicate encoding !\n%s",
+		     BAD_CAST buffer, NULL);
    }
    *len = 1;
    return ((int) *cur);
@@ -1526,6 +1542,8 @@ xmlNewInputFromFile(xmlParserCtxtPtr ctxt, const char *filename) {
 int
 xmlInitParserCtxt(xmlParserCtxtPtr ctxt)
 {
+    xmlParserInputPtr input;
+
    if(ctxt==NULL) {
        xmlErrInternal(NULL, "Got NULL parser context\n", NULL);
        return(-1);
@@ -1563,6 +1581,9 @@ xmlInitParserCtxt(xmlParserCtxtPtr ctxt)
 	ctxt->input = NULL;
 	return(-1);
    }
+    while ((input = inputPop(ctxt)) != NULL) { /* Non consuming */
+        xmlFreeInputStream(input);
+    }
    ctxt->inputNr = 0;
    ctxt->input = NULL;

--- a/testapi.c
+++ b/testapi.c
--- a/xmlwriter.c
+++ b/xmlwriter.c
@@ -71,6 +71,7 @@ struct _xmlTextWriter {
    xmlChar *ichar;             /* indent character */
    char qchar;                 /* character used for quoting attribute values */
    xmlParserCtxtPtr ctxt;
+    int no_doc_free;
 };

 static void xmlFreeTextWriterStackEntry(xmlLinkPtr lk);
@@ -198,6 +199,7 @@ xmlNewTextWriter(xmlOutputBufferPtr out)
                        "xmlNewTextWriter : out of memory!\n");
        return NULL;
    }
+    ret->no_doc_free = 0;

    return ret;
 }
@@ -376,8 +378,10 @@ xmlNewTextWriterDoc(xmlDocPtr * doc, int compression)

    xmlSetDocCompressMode(ctxt->myDoc, compression);

-    if (doc != NULL)
+    if (doc != NULL) {
        *doc = ctxt->myDoc;
+	ret->no_doc_free = 1;
+    }

    return ret;
 }
@@ -434,6 +438,7 @@ xmlNewTextWriterTree(xmlDocPtr doc, xmlNodePtr node, int compression)

    ctxt->myDoc = doc;
    ctxt->node = node;
+    ret->no_doc_free = 1;

    xmlSetDocCompressMode(doc, compression);

@@ -461,8 +466,13 @@ xmlFreeTextWriter(xmlTextWriterPtr writer)
    if (writer->nsstack != NULL)
        xmlListDelete(writer->nsstack);

-    if (writer->ctxt != NULL)
+    if (writer->ctxt != NULL) {
+        if ((writer->ctxt->myDoc != NULL) && (writer->no_doc_free == 0)) {
+	    xmlFreeDoc(writer->ctxt->myDoc);
+	    writer->ctxt->myDoc = NULL;
+	}
        xmlFreeParserCtxt(writer->ctxt);
+    }

    if (writer->ichar != NULL)
        xmlFree(writer->ichar);