added the same htmlRead APIs than their XML counterparts new parser
* HTMLparser.c testHTML.c xmllint.c include/libxml/HTMLparser.h:
added the same htmlRead APIs than their XML counterparts
* include/libxml/parser.h: new parser options, not yet implemented,
added an options field to the context.
* tree.c: patch from Shaun McCance to fix bug #123238 when ]]>
is found within a cdata section.
* result/noent/cdata2 result/cdata2 result/cdata2.rdr
result/cdata2.sax test/cdata2: add one more cdata test
Daniel
diff --git a/include/libxml/HTMLparser.h b/include/libxml/HTMLparser.h
index 7e008bd..66f2d80 100644
--- a/include/libxml/HTMLparser.h
+++ b/include/libxml/HTMLparser.h
@@ -154,6 +154,88 @@
int size,
int terminate);
+/*
+ * New set of simpler/more flexible APIs
+ */
+/**
+ * xmlParserOption:
+ *
+ * This is the set of XML parser options that can be passed down
+ * to the xmlReadDoc() and similar calls.
+ */
+typedef enum {
+ HTML_PARSE_NOERROR = 1<<5, /* suppress error reports */
+ HTML_PARSE_NOWARNING= 1<<6, /* suppress warning reports */
+ HTML_PARSE_PEDANTIC = 1<<7, /* pedantic error reporting */
+ HTML_PARSE_NOBLANKS = 1<<8, /* remove blank nodes */
+ HTML_PARSE_NONET = 1<<11 /* Forbid network access */
+} htmlParserOption;
+
+XMLPUBFUN void XMLCALL
+ htmlCtxtReset (htmlParserCtxtPtr ctxt);
+XMLPUBFUN int XMLCALL
+ htmlCtxtUseOptions (htmlParserCtxtPtr ctxt,
+ int options);
+XMLPUBFUN htmlDocPtr XMLCALL
+ htmlReadDoc (const xmlChar *cur,
+ const char *URL,
+ const char *encoding,
+ int options);
+XMLPUBFUN htmlDocPtr XMLCALL
+ htmlReadFile (const char *URL,
+ const char *encoding,
+ int options);
+XMLPUBFUN htmlDocPtr XMLCALL
+ htmlReadMemory (const char *buffer,
+ int size,
+ const char *URL,
+ const char *encoding,
+ int options);
+XMLPUBFUN htmlDocPtr XMLCALL
+ htmlReadFd (int fd,
+ const char *URL,
+ const char *encoding,
+ int options);
+XMLPUBFUN htmlDocPtr XMLCALL
+ htmlReadIO (xmlInputReadCallback ioread,
+ xmlInputCloseCallback ioclose,
+ void *ioctx,
+ const char *URL,
+ const char *encoding,
+ int options);
+XMLPUBFUN htmlDocPtr XMLCALL
+ htmlCtxtReadDoc (xmlParserCtxtPtr ctxt,
+ const xmlChar *cur,
+ const char *URL,
+ const char *encoding,
+ int options);
+XMLPUBFUN htmlDocPtr XMLCALL
+ htmlCtxtReadFile (xmlParserCtxtPtr ctxt,
+ const char *filename,
+ const char *encoding,
+ int options);
+XMLPUBFUN htmlDocPtr XMLCALL
+ htmlCtxtReadMemory (xmlParserCtxtPtr ctxt,
+ const char *buffer,
+ int size,
+ const char *URL,
+ const char *encoding,
+ int options);
+XMLPUBFUN htmlDocPtr XMLCALL
+ htmlCtxtReadFd (xmlParserCtxtPtr ctxt,
+ int fd,
+ const char *URL,
+ const char *encoding,
+ int options);
+XMLPUBFUN htmlDocPtr XMLCALL
+ htmlCtxtReadIO (xmlParserCtxtPtr ctxt,
+ xmlInputReadCallback ioread,
+ xmlInputCloseCallback ioclose,
+ void *ioctx,
+ const char *URL,
+ const char *encoding,
+ int options);
+
/* NRK/Jan2003: further knowledge of HTML structure
*/
typedef enum {
diff --git a/include/libxml/parser.h b/include/libxml/parser.h
index e095bab..73dd71f 100644
--- a/include/libxml/parser.h
+++ b/include/libxml/parser.h
@@ -262,15 +262,16 @@
xmlHashTablePtr attsDefault; /* defaulted attributes if any */
xmlHashTablePtr attsSpecial; /* non-CDATA attributes if any */
int nsWellFormed; /* is the document XML Nanespace okay */
+ int options; /* Extra options */
/*
* Those fields are needed only for treaming parsing so far
*/
- int dictNames; /* Use dictionary names for the tree */
- int freeElemsNr; /* number of freed element nodes */
- xmlNodePtr freeElems; /* List of freed element nodes */
- int freeAttrsNr; /* number of freed attributes nodes */
- xmlAttrPtr freeAttrs; /* List of freed attributes nodes */
+ int dictNames; /* Use dictionary names for the tree */
+ int freeElemsNr; /* number of freed element nodes */
+ xmlNodePtr freeElems; /* List of freed element nodes */
+ int freeAttrsNr; /* number of freed attributes nodes */
+ xmlAttrPtr freeAttrs; /* List of freed attributes nodes */
};
/**
@@ -1045,7 +1046,9 @@
XML_PARSE_SAX1 = 1<<9, /* use the SAX1 interface internally */
XML_PARSE_XINCLUDE = 1<<10,/* Implement XInclude substitition */
XML_PARSE_NONET = 1<<11,/* Forbid network access */
- XML_PARSE_NODICT = 1<<12 /* Do not reuse the context dictionnary */
+ XML_PARSE_NODICT = 1<<12,/* Do not reuse the context dictionnary */
+ XML_PARSE_NSCLEAN = 1<<13,/* remove redundant namespaces declarations */
+ XML_PARSE_NOCDATA = 1<<14 /* merge CDATA as text nodes */
} xmlParserOption;
XMLPUBFUN void XMLCALL