mirror of
https://gitlab.gnome.org/GNOME/libxml2.git
synced 2025-10-30 10:45:36 +03:00
f41fbbf6a9
- xpath.c result/XPath/tests/chaptersprefol: bugfixes on order and on predicate - HTMLparser.[ch] HTMLtree.c result/HTML/doc3.htm.err result/HTML/doc3.htm.sax result/HTML/wired.html: sometimes one really want to have tags closed on output even if we accept unclosed ones on input Daniel
116 lines
3.3 KiB
C
116 lines
3.3 KiB
C
/*
|
|
* HTMLparser.h : inf=terface for an HTML 4.0 non-verifying parser
|
|
*
|
|
* See Copyright for the status of this software.
|
|
*
|
|
* Daniel.Veillard@w3.org
|
|
*/
|
|
|
|
#ifndef __HTML_PARSER_H__
|
|
#define __HTML_PARSER_H__
|
|
#include <libxml/parser.h>
|
|
|
|
#ifdef __cplusplus
|
|
extern "C" {
|
|
#endif
|
|
|
|
/*
|
|
* Most of the back-end structures from XML and HTML are shared
|
|
*/
|
|
typedef xmlParserCtxt htmlParserCtxt;
|
|
typedef xmlParserCtxtPtr htmlParserCtxtPtr;
|
|
typedef xmlParserNodeInfo htmlParserNodeInfo;
|
|
typedef xmlSAXHandler htmlSAXHandler;
|
|
typedef xmlSAXHandlerPtr htmlSAXHandlerPtr;
|
|
typedef xmlParserInput htmlParserInput;
|
|
typedef xmlParserInputPtr htmlParserInputPtr;
|
|
typedef xmlDocPtr htmlDocPtr;
|
|
typedef xmlNodePtr htmlNodePtr;
|
|
|
|
/*
|
|
* Internal description of an HTML element
|
|
*/
|
|
typedef struct _htmlElemDesc htmlElemDesc;
|
|
typedef htmlElemDesc *htmlElemDescPtr;
|
|
struct _htmlElemDesc {
|
|
const char *name; /* The tag name */
|
|
char startTag; /* Whether the start tag can be implied */
|
|
char endTag; /* Whether the end tag can be implied */
|
|
char saveEndTag; /* Whether the end tag should be saved */
|
|
char empty; /* Is this an empty element ? */
|
|
char depr; /* Is this a deprecated element ? */
|
|
char dtd; /* 1: only in Loose DTD, 2: only Frameset one */
|
|
const char *desc; /* the description */
|
|
};
|
|
|
|
/*
|
|
* Internal description of an HTML entity
|
|
*/
|
|
typedef struct _htmlEntityDesc htmlEntityDesc;
|
|
typedef htmlEntityDesc *htmlEntityDescPtr;
|
|
struct _htmlEntityDesc {
|
|
int value; /* the UNICODE value for the character */
|
|
const char *name; /* The entity name */
|
|
const char *desc; /* the description */
|
|
};
|
|
|
|
/*
|
|
* There is only few public functions.
|
|
*/
|
|
htmlElemDescPtr htmlTagLookup (const xmlChar *tag);
|
|
htmlEntityDescPtr htmlEntityLookup(const xmlChar *name);
|
|
htmlEntityDescPtr htmlEntityValueLookup(int value);
|
|
|
|
int htmlIsAutoClosed(htmlDocPtr doc,
|
|
htmlNodePtr elem);
|
|
int htmlAutoCloseTag(htmlDocPtr doc,
|
|
const xmlChar *name,
|
|
htmlNodePtr elem);
|
|
htmlEntityDescPtr htmlParseEntityRef(htmlParserCtxtPtr ctxt,
|
|
xmlChar **str);
|
|
int htmlParseCharRef(htmlParserCtxtPtr ctxt);
|
|
void htmlParseElement(htmlParserCtxtPtr ctxt);
|
|
|
|
htmlDocPtr htmlSAXParseDoc (xmlChar *cur,
|
|
const char *encoding,
|
|
htmlSAXHandlerPtr sax,
|
|
void *userData);
|
|
htmlDocPtr htmlParseDoc (xmlChar *cur,
|
|
const char *encoding);
|
|
htmlDocPtr htmlSAXParseFile(const char *filename,
|
|
const char *encoding,
|
|
htmlSAXHandlerPtr sax,
|
|
void *userData);
|
|
htmlDocPtr htmlParseFile (const char *filename,
|
|
const char *encoding);
|
|
int UTF8ToHtml (unsigned char* out,
|
|
int *outlen,
|
|
const unsigned char* in,
|
|
int *inlen);
|
|
int htmlEncodeEntities(unsigned char* out,
|
|
int *outlen,
|
|
const unsigned char* in,
|
|
int *inlen, int quoteChar);
|
|
int htmlIsScriptAttribute(const xmlChar *name);
|
|
int htmlHandleOmittedElem(int val);
|
|
|
|
/**
|
|
* Interfaces for the Push mode
|
|
*/
|
|
void htmlFreeParserCtxt (htmlParserCtxtPtr ctxt);
|
|
htmlParserCtxtPtr htmlCreatePushParserCtxt(htmlSAXHandlerPtr sax,
|
|
void *user_data,
|
|
const char *chunk,
|
|
int size,
|
|
const char *filename,
|
|
xmlCharEncoding enc);
|
|
int htmlParseChunk (htmlParserCtxtPtr ctxt,
|
|
const char *chunk,
|
|
int size,
|
|
int terminate);
|
|
#ifdef __cplusplus
|
|
}
|
|
#endif
|
|
|
|
#endif /* __HTML_PARSER_H__ */
|