marcocosta10000
723ca1506b
Adds `_fetch_page_content()`, which downloads a URL with a browser-like User-Agent and uses BeautifulSoup with the lxml parser to strip scripts, navigation, and other non-content markup, returning up to 3000 characters of readable text. The response validator now also compares generated replies against this scraped text, flagging named entities and claims that do not appear in the retrieved page content. Dependencies: adds beautifulsoup4==4.12.3 and pins lxml==5.3.0 for parser compatibility.
147 lines
3.7 KiB
C
147 lines
3.7 KiB
C
/*
|
|
* Summary: API to build regexp automata
|
|
* Description: the API to build regexp automata
|
|
*
|
|
* Copy: See Copyright for the status of this software.
|
|
*
|
|
* Author: Daniel Veillard
|
|
*/
|
|
|
|
#ifndef __XML_AUTOMATA_H__
|
|
#define __XML_AUTOMATA_H__
|
|
|
|
#include <libxml/xmlversion.h>
|
|
|
|
#ifdef LIBXML_REGEXP_ENABLED
|
|
#ifdef LIBXML_AUTOMATA_ENABLED
|
|
|
|
#include <libxml/xmlstring.h>
|
|
|
|
#ifdef __cplusplus
|
|
extern "C" {
|
|
#endif
|
|
|
|
/**
|
|
* xmlAutomataPtr:
|
|
*
|
|
* A libxml automata description, It can be compiled into a regexp
|
|
*/
|
|
typedef struct _xmlAutomata xmlAutomata;
|
|
typedef xmlAutomata *xmlAutomataPtr;
|
|
|
|
/**
|
|
* xmlAutomataStatePtr:
|
|
*
|
|
* A state int the automata description,
|
|
*/
|
|
typedef struct _xmlAutomataState xmlAutomataState;
|
|
typedef xmlAutomataState *xmlAutomataStatePtr;
|
|
|
|
/*
|
|
* Building API
|
|
*/
|
|
XMLPUBFUN xmlAutomataPtr
|
|
xmlNewAutomata (void);
|
|
XMLPUBFUN void
|
|
xmlFreeAutomata (xmlAutomataPtr am);
|
|
|
|
XMLPUBFUN xmlAutomataStatePtr
|
|
xmlAutomataGetInitState (xmlAutomataPtr am);
|
|
XMLPUBFUN int
|
|
xmlAutomataSetFinalState (xmlAutomataPtr am,
|
|
xmlAutomataStatePtr state);
|
|
XMLPUBFUN xmlAutomataStatePtr
|
|
xmlAutomataNewState (xmlAutomataPtr am);
|
|
XMLPUBFUN xmlAutomataStatePtr
|
|
xmlAutomataNewTransition (xmlAutomataPtr am,
|
|
xmlAutomataStatePtr from,
|
|
xmlAutomataStatePtr to,
|
|
const xmlChar *token,
|
|
void *data);
|
|
XMLPUBFUN xmlAutomataStatePtr
|
|
xmlAutomataNewTransition2 (xmlAutomataPtr am,
|
|
xmlAutomataStatePtr from,
|
|
xmlAutomataStatePtr to,
|
|
const xmlChar *token,
|
|
const xmlChar *token2,
|
|
void *data);
|
|
XMLPUBFUN xmlAutomataStatePtr
|
|
xmlAutomataNewNegTrans (xmlAutomataPtr am,
|
|
xmlAutomataStatePtr from,
|
|
xmlAutomataStatePtr to,
|
|
const xmlChar *token,
|
|
const xmlChar *token2,
|
|
void *data);
|
|
|
|
XMLPUBFUN xmlAutomataStatePtr
|
|
xmlAutomataNewCountTrans (xmlAutomataPtr am,
|
|
xmlAutomataStatePtr from,
|
|
xmlAutomataStatePtr to,
|
|
const xmlChar *token,
|
|
int min,
|
|
int max,
|
|
void *data);
|
|
XMLPUBFUN xmlAutomataStatePtr
|
|
xmlAutomataNewCountTrans2 (xmlAutomataPtr am,
|
|
xmlAutomataStatePtr from,
|
|
xmlAutomataStatePtr to,
|
|
const xmlChar *token,
|
|
const xmlChar *token2,
|
|
int min,
|
|
int max,
|
|
void *data);
|
|
XMLPUBFUN xmlAutomataStatePtr
|
|
xmlAutomataNewOnceTrans (xmlAutomataPtr am,
|
|
xmlAutomataStatePtr from,
|
|
xmlAutomataStatePtr to,
|
|
const xmlChar *token,
|
|
int min,
|
|
int max,
|
|
void *data);
|
|
XMLPUBFUN xmlAutomataStatePtr
|
|
xmlAutomataNewOnceTrans2 (xmlAutomataPtr am,
|
|
xmlAutomataStatePtr from,
|
|
xmlAutomataStatePtr to,
|
|
const xmlChar *token,
|
|
const xmlChar *token2,
|
|
int min,
|
|
int max,
|
|
void *data);
|
|
XMLPUBFUN xmlAutomataStatePtr
|
|
xmlAutomataNewAllTrans (xmlAutomataPtr am,
|
|
xmlAutomataStatePtr from,
|
|
xmlAutomataStatePtr to,
|
|
int lax);
|
|
XMLPUBFUN xmlAutomataStatePtr
|
|
xmlAutomataNewEpsilon (xmlAutomataPtr am,
|
|
xmlAutomataStatePtr from,
|
|
xmlAutomataStatePtr to);
|
|
XMLPUBFUN xmlAutomataStatePtr
|
|
xmlAutomataNewCountedTrans (xmlAutomataPtr am,
|
|
xmlAutomataStatePtr from,
|
|
xmlAutomataStatePtr to,
|
|
int counter);
|
|
XMLPUBFUN xmlAutomataStatePtr
|
|
xmlAutomataNewCounterTrans (xmlAutomataPtr am,
|
|
xmlAutomataStatePtr from,
|
|
xmlAutomataStatePtr to,
|
|
int counter);
|
|
XMLPUBFUN int
|
|
xmlAutomataNewCounter (xmlAutomataPtr am,
|
|
int min,
|
|
int max);
|
|
|
|
XMLPUBFUN struct _xmlRegexp *
|
|
xmlAutomataCompile (xmlAutomataPtr am);
|
|
XMLPUBFUN int
|
|
xmlAutomataIsDeterminist (xmlAutomataPtr am);
|
|
|
|
#ifdef __cplusplus
|
|
}
|
|
#endif
|
|
|
|
#endif /* LIBXML_AUTOMATA_ENABLED */
|
|
#endif /* LIBXML_REGEXP_ENABLED */
|
|
|
|
#endif /* __XML_AUTOMATA_H__ */
|