XML/parser.c - diff

Return to parser.c CVS log

Up to [Public] / XML

Diff for /XML/parser.c between versions 1.15 and 1.16

version 1.15, 1998/05/25 23:58:37	version 1.16, 1998/06/19 04:48:27
Line 127	Line 127
/*	/*
* Forward definition for recusive behaviour.	* Forward definition for recusive behaviour.
*/	*/
xmlNodePtr xmlParseElement(CHAR **p, xmlDocPtr doc);	xmlNodePtr xmlParseElement(xmlParserCtxtPtr ctxt);

/*	/*
* xmlHandleData : this routine represent's the specific application	* xmlHandleData : this routine represent's the specific application
Line 227 CHAR xmlStrchr(const CHAR str, CHAR va	Line 227 CHAR xmlStrchr(const CHAR str, CHAR va
* xmlParseName : parse an XML name.	* xmlParseName : parse an XML name.
*/	*/

CHAR xmlParseName(CHAR *p) {	CHAR *xmlParseName(xmlParserCtxtPtr ctxt) {
CHAR cur = p, q, ret = NULL;	CHAR q, ret = NULL;

/*	/*
* Name ::= (Letter \| '_') (NameChar)*	* Name ::= (Letter \| '_') (NameChar)*
*/	*/
if (!IS_LETTER(cur) && (cur != '_')) return(NULL);	if (!IS_LETTER(ctxt->cur[0]) && (ctxt->cur[0] != '_')) return(NULL);
q = cur++;	q = ctxt->cur++;
while ((IS_LETTER(cur)) \|\| (IS_DIGIT(cur)) \|\|	while ((IS_LETTER(ctxt->cur[0])) \|\| (IS_DIGIT(ctxt->cur[0])) \|\|
(cur == '.') \|\| (cur == '-') \|\| (*cur == '_') \|\|	(ctxt->cur[0] == '.') \|\| (ctxt->cur[0] == '-') \|\| (ctxt->cur[0] == '_') \|\|
(*cur == ':') \|\|	(ctxt->cur[0] == ':') \|\|
(IS_COMBINING(cur)) \|\| (IS_IGNORABLE(cur)) \|\|	(IS_COMBINING(ctxt->cur[0])) \|\| (IS_IGNORABLE(ctxt->cur[0])) \|\|
(IS_EXTENDER(*cur)))	(IS_EXTENDER(ctxt->cur[0])))
cur++;	ctxt->cur++;

ret = xmlStrndup(q, cur - q);	ret = xmlStrndup(q, ctxt->cur - q);

*p = cur;
return(ret);	return(ret);
}	}

/*	/*
* Parse and return a string between quotes or doublequotes	* Parse and return a string between quotes or doublequotes
*/	*/
CHAR xmlParseQuotedString(CHAR *p) {	CHAR *xmlParseQuotedString(xmlParserCtxtPtr ctxt) {
CHAR *ret = NULL;	CHAR *ret = NULL;
CHAR cur = p, *q;	CHAR *q;

if (*cur == '"') {	if (ctxt->cur[0] == '"') {
cur++;	ctxt->cur++;
q = cur;	q = ctxt->cur;
while (IS_CHAR(cur) && (cur != '"')) cur++;	while (IS_CHAR(ctxt->cur[0]) && (ctxt->cur[0] != '"')) ctxt->cur++;
if (*cur != '"')	if (ctxt->cur[0] != '"')
fprintf(stderr, "String not closed \"%.50s\n", q);	fprintf(stderr, "String not closed \"%.50s\n", q);
else {	else {
ret = xmlStrndup(q, cur - q);	ret = xmlStrndup(q, ctxt->cur - q);
cur++;	ctxt->cur++;
}	}
} else if (*cur == '\''){	} else if (ctxt->cur[0] == '\''){
cur++;	ctxt->cur++;
q = cur;	q = ctxt->cur;
while (IS_CHAR(cur) && (cur != '\'')) cur++;	while (IS_CHAR(ctxt->cur[0]) && (ctxt->cur[0] != '\'')) ctxt->cur++;
if (*cur != '\'')	if (ctxt->cur[0] != '\'')
fprintf(stderr, "String not closed '%.50s\n", q);	fprintf(stderr, "String not closed '%.50s\n", q);
else {	else {
ret = xmlStrndup(q, cur - q);	ret = xmlStrndup(q, ctxt->cur - q);
cur++;	ctxt->cur++;
}	}
}	}
*p = cur;
return(ret);	return(ret);
}	}

/*	/*
* Skip an XML (SGML) comment <!-- .... -->	* Skip an XML (SGML) comment <!-- .... -->
	*
	* TODO !!!! Save the comment in the tree !!!
*/	*/
void xmlParserSkipComment(CHAR **p) {	void xmlParserSkipComment(xmlParserCtxtPtr ctxt) {
CHAR cur = p, q, r, *start;	CHAR q, r, *start;

/*	/*
* An extra check may avoid errors and isn't that costly !	* An extra check may avoid errors and isn't that costly !
*/	*/
if ((cur[0] != '<') \|\| (cur[1] != '!') \|\|	if ((ctxt->cur[0] != '<') \|\| (ctxt->cur[1] != '!') \|\|
(cur[2] != '-') \|\| (cur[3] != '-')) return;	(ctxt->cur[2] != '-') \|\| (ctxt->cur[3] != '-')) return;

cur += 4;	ctxt->cur += 4;
start = q = cur;	start = q = ctxt->cur;
cur++;	ctxt->cur++;
r = cur;	r = ctxt->cur;
cur++;	ctxt->cur++;
while (IS_CHAR(*cur) &&	while (IS_CHAR(ctxt->cur[0]) &&
((cur == ':') \|\| (cur != '>') \|\| (r != '-') \|\| (q != '-'))) {	((ctxt->cur[0] == ':') \|\| (ctxt->cur[0] != '>') \|\|
cur++;r++;q++;	(r != '-') \|\| (q != '-'))) {
	ctxt->cur++;r++;q++;
}	}
if (!IS_CHAR(*cur)) {	if (!IS_CHAR(ctxt->cur[0])) {
fprintf(stderr, "Comment not terminated <!--%.50s\n", start);	fprintf(stderr, "Comment not terminated <!--%.50s\n", start);
*p = start;	ctxt->cur = start; /* !!! We shouldn't really try to recover !!! */
} else {	} else {
cur++;	ctxt->cur++;
*p = cur;
}	}
}	}

Line 314 void xmlParserSkipComment(CHAR **p) {	Line 314 void xmlParserSkipComment(CHAR **p) {
* xmlParseNamespace: parse specific '<?namespace ...' constructs.	* xmlParseNamespace: parse specific '<?namespace ...' constructs.
*/	*/

void xmlParseNamespace(CHAR **p, xmlDocPtr doc) {	void xmlParseNamespace(xmlParserCtxtPtr ctxt) {
CHAR cur = p;
CHAR *href = NULL;	CHAR *href = NULL;
CHAR *AS = NULL;	CHAR *AS = NULL;
int garbage = 0;	int garbage = 0;
Line 323 void xmlParseNamespace(CHAR **p, xmlDocP	Line 322 void xmlParseNamespace(CHAR **p, xmlDocP
/*	/*
* We know that 'namespace' is here.	* We know that 'namespace' is here.
*/	*/
cur += 9;	ctxt->cur += 9;
SKIP_BLANKS(cur);	SKIP_BLANKS(ctxt->cur);

while (IS_CHAR(cur) && (cur != '>')) {	while (IS_CHAR(ctxt->cur[0]) && (ctxt->cur[0] != '>')) {
/*	/*
* We can have 'href' or 'AS' attributes.	* We can have 'href' or 'AS' attributes.
*/	*/
if ((cur[0] == 'h') && (cur[1] == 'r') && (cur[2] == 'e') &&	if ((ctxt->cur[0] == 'h') && (ctxt->cur[1] == 'r') &&
(cur[3] == 'f')) {	(ctxt->cur[2] == 'e') && (ctxt->cur[3] == 'f')) {
garbage = 0;	garbage = 0;
cur += 4;	ctxt->cur += 4;
SKIP_BLANKS(cur);	SKIP_BLANKS(ctxt->cur);

if (*cur != '=') continue;	if (ctxt->cur[0] != '=') continue;
cur++;	ctxt->cur++;
SKIP_BLANKS(cur);	SKIP_BLANKS(ctxt->cur);

href = xmlParseQuotedString(&cur);	href = xmlParseQuotedString(ctxt);
SKIP_BLANKS(cur);	SKIP_BLANKS(ctxt->cur);
} else if ((cur[0] == 'A') && (cur[1] == 'S')) {	} else if ((ctxt->cur[0] == 'A') && (ctxt->cur[1] == 'S')) {
garbage = 0;	garbage = 0;
cur += 2;	ctxt->cur += 2;
SKIP_BLANKS(cur);	SKIP_BLANKS(ctxt->cur);

if (*cur != '=') continue;	if (ctxt->cur[0] != '=') continue;
cur++;	ctxt->cur++;
SKIP_BLANKS(cur);	SKIP_BLANKS(ctxt->cur);

AS = xmlParseQuotedString(&cur);	AS = xmlParseQuotedString(ctxt);
SKIP_BLANKS(cur);	SKIP_BLANKS(ctxt->cur);
} else if ((cur[0] == '?') && (cur[1] == '>')) {	} else if ((ctxt->cur[0] == '?') && (ctxt->cur[1] == '>')) {
garbage = 0;	garbage = 0;
cur ++;	ctxt->cur ++;
} else {	} else {
/*	/*
* Found garbage when parsing the namespace	* Found garbage when parsing the namespace
*/	*/
if (!garbage) fprintf(stderr,	if (!garbage) fprintf(stderr,
"\nxmlParseNamespace found garbage: ");	"\nxmlParseNamespace found garbage: ");
fprintf(stderr, "%c", *cur);	fprintf(stderr, "%c", ctxt->cur[0]);
cur++;	ctxt->cur++;
}	}
}	}

MOVETO_ENDTAG(cur);	MOVETO_ENDTAG(ctxt->cur);
cur++;	ctxt->cur++;

/*	/*
* Register the DTD.	* Register the DTD.
*/	*/
if (href != NULL)	if (href != NULL)
xmlNewDtd(doc, href, AS);	xmlNewDtd(ctxt->doc, href, AS);

if (AS != NULL) free(AS);	if (AS != NULL) free(AS);
if (href != NULL) free(href);	if (href != NULL) free(href);

*p = cur;
}	}

/*	/*
* xmlParsePI: parse an XML Processing Instruction.	* xmlParsePI: parse an XML Processing Instruction.
*/	*/

void xmlParsePI(CHAR **p, xmlDocPtr doc) {	void xmlParsePI(xmlParserCtxtPtr ctxt) {
CHAR cur = p;	if ((ctxt->cur[0] == '<') && (ctxt->cur[1] == '?')) {

if ((cur[0] == '<') && (cur[1] == '?')) {
/*	/*
* this is a Processing Instruction.	* this is a Processing Instruction.
*/	*/
cur += 2;	ctxt->cur += 2;

/*	/*
* Special for WebDav, support for the Processing Instruction	* Special for WebDav, support for the Processing Instruction
* '<?namespace ...' contruct in the header of the XML document.	* '<?namespace ...' contruct in the header of the XML document.
*/	*/
if ((cur[0] == 'n') && (cur[1] == 'a') &&	if ((ctxt->cur[0] == 'n') && (ctxt->cur[1] == 'a') &&
(cur[2] == 'm') && (cur[3] == 'e') &&	(ctxt->cur[2] == 'm') && (ctxt->cur[3] == 'e') &&
(cur[4] == 's') && (cur[5] == 'p') &&	(ctxt->cur[4] == 's') && (ctxt->cur[5] == 'p') &&
(cur[6] == 'a') && (cur[7] == 'c') &&	(ctxt->cur[6] == 'a') && (ctxt->cur[7] == 'c') &&
(cur[8] == 'e')) {	(ctxt->cur[8] == 'e')) {
xmlParseNamespace(&cur, doc);	xmlParseNamespace(ctxt);
} else {	} else {
/* Unknown PI, ignore it ! */	/* Unknown PI, ignore it ! */
fprintf(stderr, "xmlParsePI : skipping unknown PI %30s\n", cur);	fprintf(stderr, "xmlParsePI : skipping unknown PI %30s\n",
MOVETO_ENDTAG(cur);	ctxt->cur);
cur++;	MOVETO_ENDTAG(ctxt->cur);
	ctxt->cur++;
}	}
}	}
*p = cur;
}	}

/*	/*
Line 421 void xmlParsePI(CHAR **p, xmlDocPtr doc)	Line 416 void xmlParsePI(CHAR **p, xmlDocPtr doc)
* Attribute ::= Name Eq AttValue	* Attribute ::= Name Eq AttValue
*/	*/

void xmlParseAttribute(CHAR **p, xmlNodePtr node) {	void xmlParseAttribute(xmlParserCtxtPtr ctxt, xmlNodePtr node) {
CHAR cur = p, q, name, *value = NULL;	CHAR q, name, *value = NULL;

if (!IS_LETTER(cur) && (cur != '_')) {	if (!IS_LETTER(ctxt->cur[0]) && (ctxt->cur[0] != '_')) {
return;	return;
}	}
q = cur++;	q = ctxt->cur++;
while ((IS_LETTER(cur)) \|\| (IS_DIGIT(cur)) \|\|	while ((IS_LETTER(ctxt->cur[0])) \|\| (IS_DIGIT(ctxt->cur[0])) \|\|
(cur == '.') \|\| (cur == '-') \|\| (*cur == '_') \|\|	(ctxt->cur[0] == '.') \|\| (ctxt->cur[0] == '-') \|\|
(*cur == ':') \|\|	(ctxt->cur[0] == '_') \|\| (ctxt->cur[0] == ':') \|\|
(IS_COMBINING(cur)) \|\| (IS_IGNORABLE(cur)) \|\|	(IS_COMBINING(ctxt->cur[0])) \|\| (IS_IGNORABLE(ctxt->cur[0])) \|\|
(IS_EXTENDER(*cur)))	(IS_EXTENDER(ctxt->cur[0])))
cur++;	ctxt->cur++;
name = xmlStrndup(q, cur - q);	name = xmlStrndup(q, ctxt->cur - q);

/*	/*
* We should have the equal, we are laxist here and allow attributes	* We should have the equal, we are laxist here and allow attributes
* without values and extra spaces.	* without values and extra spaces.
*/	*/
SKIP_BLANKS(cur);	SKIP_BLANKS(ctxt->cur);
if (*cur == '=') {	if (ctxt->cur[0] == '=') {
cur++;	ctxt->cur++;
SKIP_BLANKS(cur);	SKIP_BLANKS(ctxt->cur);
if ((cur != '\'') && (cur != '"')) {	if ((ctxt->cur[0] != '\'') && (ctxt->cur[0] != '"')) {
fprintf(stderr, "Quotes were expected for attribute value %.20s\n",	fprintf(stderr, "Quotes were expected for attribute value %.20s\n",
q);	q);
} else	} else
value = xmlParseQuotedString(&cur);	value = xmlParseQuotedString(ctxt);
}	}

/*	/*
Line 456 void xmlParseAttribute(CHAR **p, xmlNode	Line 451 void xmlParseAttribute(CHAR **p, xmlNode
*/	*/
if (name != NULL)	if (name != NULL)
xmlNewProp(node, name, value);	xmlNewProp(node, name, value);

*p = cur;
}	}

/*	/*
* xmlParseStartTag: parse a start of tag.	* xmlParseStartTag: parse a start of tag.
*/	*/

xmlNodePtr xmlParseStartTag(CHAR **p, xmlDocPtr doc) {	xmlNodePtr xmlParseStartTag(xmlParserCtxtPtr ctxt) {
CHAR cur = p, q, ns, *name;	CHAR q, ns, *name;
xmlDtdPtr dtd = NULL;	xmlDtdPtr dtd = NULL;
xmlNodePtr ret = NULL;	xmlNodePtr ret = NULL;

Line 483 xmlNodePtr xmlParseStartTag(CHAR **p, xm	Line 476 xmlNodePtr xmlParseStartTag(CHAR **p, xm
*	*
* STag ::= '<' QName (S Attribute)* S? '>'	* STag ::= '<' QName (S Attribute)* S? '>'
*/	*/
if (*cur != '<') return(NULL);	if (ctxt->cur[0] != '<') return(NULL);
cur++;	ctxt->cur++;

if (!IS_LETTER(cur) && (cur != '_')) return(NULL);	if (!IS_LETTER(ctxt->cur[0]) && (ctxt->cur[0] != '_')) return(NULL);
q = cur++;	q = ctxt->cur++;
while ((IS_LETTER(cur)) \|\| (IS_DIGIT(cur)) \|\|	while ((IS_LETTER(ctxt->cur[0])) \|\| (IS_DIGIT(ctxt->cur[0])) \|\|
(cur == '.') \|\| (cur == '-') \|\| (*cur == '_') \|\|	(ctxt->cur[0] == '.') \|\| (ctxt->cur[0] == '-') \|\|
(IS_COMBINING(cur)) \|\| (IS_IGNORABLE(cur)) \|\|	(ctxt->cur[0] == '_') \|\|
(IS_EXTENDER(*cur)))	(IS_COMBINING(ctxt->cur[0])) \|\| (IS_IGNORABLE(ctxt->cur[0])) \|\|
cur++;	(IS_EXTENDER(ctxt->cur[0])))
	ctxt->cur++;

if (*cur == ':') {	if (ctxt->cur[0] == ':') {
ns = xmlStrndup(q, cur - q);	ns = xmlStrndup(q, ctxt->cur - q);

cur++; /* skip the column */	ctxt->cur++; /* skip the column */
if (!IS_LETTER(cur) && (cur != '_')) {	if (!IS_LETTER(ctxt->cur[0]) && (ctxt->cur[0] != '_')) {
fprintf(stderr,	fprintf(stderr,
"Start tag : no element name after namespace identifier %.20s\n",	"Start tag : no element name after namespace identifier %.20s\n",
q);	q);
free(ns);	free(ns);
*p = cur;
return(NULL);	return(NULL);
}	}
q = cur++;	q = ctxt->cur++;
while ((IS_LETTER(cur)) \|\| (IS_DIGIT(cur)) \|\|	while ((IS_LETTER(ctxt->cur[0])) \|\| (IS_DIGIT(ctxt->cur[0])) \|\|
(cur == '.') \|\| (cur == '-') \|\| (*cur == '_') \|\|	(ctxt->cur[0] == '.') \|\| (ctxt->cur[0] == '-') \|\|
(*cur == ':') \|\|	(ctxt->cur[0] == '_') \|\| (ctxt->cur[0] == ':') \|\|
(IS_COMBINING(cur)) \|\| (IS_IGNORABLE(cur)) \|\|	(IS_COMBINING(ctxt->cur[0])) \|\| (IS_IGNORABLE(ctxt->cur[0])) \|\|
(IS_EXTENDER(*cur)))	(IS_EXTENDER(ctxt->cur[0])))
cur++;	ctxt->cur++;
name = xmlStrndup(q, cur - q);	name = xmlStrndup(q, ctxt->cur - q);

/*	/*
* Search the DTD associated to ns.	* Search the DTD associated to ns.
*/	*/
dtd = xmlSearchDtd(doc, ns);	dtd = xmlSearchDtd(ctxt->doc, ns);
if (dtd == NULL)	if (dtd == NULL)
fprintf(stderr, "Start tag : Couldn't find namespace %s\n", ns);	fprintf(stderr, "Start tag : Couldn't find namespace %s\n", ns);
free(ns);	free(ns);
} else	} else
name = xmlStrndup(q, cur - q);	name = xmlStrndup(q, ctxt->cur - q);

ret = xmlNewNode(dtd, name, NULL);	ret = xmlNewNode(dtd, name, NULL);

Line 532 xmlNodePtr xmlParseStartTag(CHAR **p, xm	Line 525 xmlNodePtr xmlParseStartTag(CHAR **p, xm
*	*
* (S Attribute)* S?	* (S Attribute)* S?
*/	*/
SKIP_BLANKS(cur);	SKIP_BLANKS(ctxt->cur);
while ((IS_CHAR(*cur)) &&	while ((IS_CHAR(ctxt->cur[0])) &&
(*cur != '>') &&	(ctxt->cur[0] != '>') &&
((cur[0] != '/') \|\| (cur[1] != '>'))) {	((ctxt->cur[0] != '/') \|\| (ctxt->cur[1] != '>'))) {
if (IS_LETTER(cur) \|\| (cur == '_'))	if (IS_LETTER(ctxt->cur[0]) \|\| (ctxt->cur[0] == '_'))
xmlParseAttribute(&cur, ret);	xmlParseAttribute(ctxt, ret);
else {	else {
/* We should warn TODO !!! */	/* We should warn TODO !!! */
cur++;	ctxt->cur++;
}	}
SKIP_BLANKS(cur);	SKIP_BLANKS(ctxt->cur);
}	}

*p = cur;
return(ret);	return(ret);
}	}

Line 554 xmlNodePtr xmlParseStartTag(CHAR **p, xm	Line 546 xmlNodePtr xmlParseStartTag(CHAR **p, xm
* already been read.	* already been read.
*/	*/

void xmlParseEndTag(CHAR *p, xmlDocPtr doc, xmlDtdPtr dtdPtr, CHAR **tagPtr) {	void xmlParseEndTag(xmlParserCtxtPtr ctxt, xmlDtdPtr dtdPtr, CHAR *tagPtr) {
CHAR cur = p, q, ns, *name;	CHAR q, ns, *name;
xmlDtdPtr dtd = NULL;	xmlDtdPtr dtd = NULL;

*dtdPtr = NULL;	*dtdPtr = NULL;
Line 575 void xmlParseEndTag(CHAR **p, xmlDocPtr	Line 567 void xmlParseEndTag(CHAR **p, xmlDocPtr
*	*
* ETag ::= '</' Name S? '>'	* ETag ::= '</' Name S? '>'
*/	*/
if (!IS_LETTER(cur) && (cur != '_')) return;	if (!IS_LETTER(ctxt->cur[0]) && (ctxt->cur[0] != '_')) return;
q = cur++;	q = ctxt->cur++;
while ((IS_LETTER(cur)) \|\| (IS_DIGIT(cur)) \|\|	while ((IS_LETTER(ctxt->cur[0])) \|\| (IS_DIGIT(ctxt->cur[0])) \|\|
(cur == '.') \|\| (cur == '-') \|\| (*cur == '_') \|\|	(ctxt->cur[0] == '.') \|\| (ctxt->cur[0] == '-') \|\|
(IS_COMBINING(cur)) \|\| (IS_IGNORABLE(cur)) \|\|	(ctxt->cur[0] == '_') \|\|
(IS_EXTENDER(*cur)))	(IS_COMBINING(ctxt->cur[0])) \|\| (IS_IGNORABLE(ctxt->cur[0])) \|\|
cur++;	(IS_EXTENDER(ctxt->cur[0])))
	ctxt->cur++;

if (*cur == ':') {	if (ctxt->cur[0] == ':') {
ns = xmlStrndup(q, cur - q);	ns = xmlStrndup(q, ctxt->cur - q);

cur++; /* skip the column */	ctxt->cur++; /* skip the column */
if (!IS_LETTER(cur) && (cur != '_')) {	if (!IS_LETTER(ctxt->cur[0]) && (ctxt->cur[0] != '_')) {
fprintf(stderr,	fprintf(stderr,
"End tag : no element name after namespace identifier %.20s\n",	"End tag : no element name after namespace identifier %.20s\n",
q);	q);
free(ns);	free(ns);
*p = cur;
return;	return;
}	}
q = cur++;	q = ctxt->cur++;
while ((IS_LETTER(cur)) \|\| (IS_DIGIT(cur)) \|\|	while ((IS_LETTER(ctxt->cur[0])) \|\| (IS_DIGIT(ctxt->cur[0])) \|\|
(cur == '.') \|\| (cur == '-') \|\| (*cur == '_') \|\|	(ctxt->cur[0] == '.') \|\| (ctxt->cur[0] == '-') \|\|
(*cur == ':') \|\|	(ctxt->cur[0] == '_') \|\| (ctxt->cur[0] == ':') \|\|
(IS_COMBINING(cur)) \|\| (IS_IGNORABLE(cur)) \|\|	(IS_COMBINING(ctxt->cur[0])) \|\| (IS_IGNORABLE(ctxt->cur[0])) \|\|
(IS_EXTENDER(*cur)))	(IS_EXTENDER(ctxt->cur[0])))
cur++;	ctxt->cur++;
name = xmlStrndup(q, cur - q);	name = xmlStrndup(q, ctxt->cur - q);

/*	/*
* Search the DTD associated to ns.	* Search the DTD associated to ns.
*/	*/
dtd = xmlSearchDtd(doc, ns);	dtd = xmlSearchDtd(ctxt->doc, ns);
if (dtd == NULL)	if (dtd == NULL)
fprintf(stderr, "End tag : Couldn't find namespace %s\n", ns);	fprintf(stderr, "End tag : Couldn't find namespace %s\n", ns);
free(ns);	free(ns);
} else	} else
name = xmlStrndup(q, cur - q);	name = xmlStrndup(q, ctxt->cur - q);

*dtdPtr = dtd;	*dtdPtr = dtd;
*tagPtr = name;	*tagPtr = name;
Line 620 void xmlParseEndTag(CHAR **p, xmlDocPtr	Line 612 void xmlParseEndTag(CHAR **p, xmlDocPtr
/*	/*
* We should definitely be at the ending "S? '>'" part	* We should definitely be at the ending "S? '>'" part
*/	*/
SKIP_BLANKS(cur);	SKIP_BLANKS(ctxt->cur);
if ((!IS_CHAR(cur)) \|\| (cur != '>')) {	if ((!IS_CHAR(ctxt->cur[0])) \|\| (ctxt->cur[0] != '>')) {
fprintf(stderr, "End tag : expected '>', got %.20s\n", cur);	fprintf(stderr, "End tag : expected '>', got %.20s\n", ctxt->cur);
/*	/*
* Note : skipping to the next '>' is probably otherkill,	* Note : skipping to the next '>' is probably otherkill,
* especially in case the '>' is hust missing.	* especially in case the '>' is hust missing.
*	*
* Otherwise add:	* Otherwise add:
* MOVETO_ENDTAG(cur);	* MOVETO_ENDTAG(ctxt->cur);
*/	*/
} else	} else
cur++;	ctxt->cur++;

*p = cur;
return;	return;
}	}

/*	/*
* xmlParseCDSect: escaped pure raw content.	* xmlParseCDSect: escaped pure raw content.
*/	*/
CHAR xmlParseCDSect(CHAR *p) {	CHAR *xmlParseCDSect(xmlParserCtxtPtr ctxt) {
CHAR cur = p, r, s, base, ret;	CHAR r, s, base, ret;

base = cur;	base = ctxt->cur;
if (!IS_CHAR(*cur)) {	if (!IS_CHAR(ctxt->cur[0])) {
fprintf(stderr, "CData section not finished : %.20s\n", base);	fprintf(stderr, "CData section not finished : %.20s\n", base);
return(NULL);	return(NULL);
}	}
r = cur++;	r = ctxt->cur++;
if (!IS_CHAR(*cur)) {	if (!IS_CHAR(ctxt->cur[0])) {
fprintf(stderr, "CData section not finished : %.20s\n", base);	fprintf(stderr, "CData section not finished : %.20s\n", base);
return(NULL);	return(NULL);
}	}
s = cur++;	s = ctxt->cur++;
while (IS_CHAR(*cur) &&	while (IS_CHAR(ctxt->cur[0]) &&
((r != ']') \|\| (s != ']') \|\| (*cur != '>'))) {	((r != ']') \|\| (s != ']') \|\| (ctxt->cur[0] != '>'))) {
r++;s++;cur++;	r++;s++;ctxt->cur++;
}	}
if (!IS_CHAR(*cur)) {	if (!IS_CHAR(ctxt->cur[0])) {
fprintf(stderr, "CData section not finished : %.20s\n", base);	fprintf(stderr, "CData section not finished : %.20s\n", base);
return(NULL);	return(NULL);
}	}
ret = xmlStrndup(base, cur-base);	ret = xmlStrndup(base, ctxt->cur-base);
*p = cur;
return(ret);	return(ret);
}	}

Line 678 CHAR xmlParseCDSect(CHAR *p) {	Line 669 CHAR xmlParseCDSect(CHAR *p) {
* PI : starts by '<?'	* PI : starts by '<?'
*/	*/

xmlNodePtr xmlParseContent(CHAR **p, xmlDocPtr doc, xmlNodePtr node) {	xmlNodePtr xmlParseContent(xmlParserCtxtPtr ctxt, xmlNodePtr node) {
CHAR cur = p, q, data = NULL;	CHAR q, data = NULL;
xmlNodePtr ret = NULL;	xmlNodePtr ret = NULL;

/*	/*
* First case : a Processing Instruction.	* First case : a Processing Instruction.
*/	*/
if ((cur[0] == '<') && (cur[1] == '?')) {	if ((ctxt->cur[0] == '<') && (ctxt->cur[1] == '?')) {
xmlParsePI(&cur, doc);	xmlParsePI(ctxt);
}	}
/*	/*
* Second case : a CDSection	* Second case : a CDSection
*/	*/
if ((cur[0] == '<') && (cur[1] == '!') && (cur[2] == '[') &&	if ((ctxt->cur[0] == '<') && (ctxt->cur[1] == '!') &&
(cur[3] == 'C') && (cur[4] == 'D') && (cur[5] == 'A') &&	(ctxt->cur[2] == '[') && (ctxt->cur[3] == 'C') &&
(cur[6] == 'T') && (cur[7] == 'A') && (cur[8] == '[')) {	(ctxt->cur[4] == 'D') && (ctxt->cur[5] == 'A') &&
cur += 9;	(ctxt->cur[6] == 'T') && (ctxt->cur[7] == 'A') &&
data = xmlParseCDSect(&cur);	(ctxt->cur[8] == '[')) {
	ctxt->cur += 9;
	data = xmlParseCDSect(ctxt);
}	}
/*	/*
* Third case : a sub-element.	* Third case : a sub-element.
*/	*/
else if (cur[0] == '<') {	else if (ctxt->cur[0] == '<') {
ret = xmlParseElement(&cur, doc);	ret = xmlParseElement(ctxt);
}	}
/*	/*
* Last case, text. Note that References are handled directly.	* Last case, text. Note that References are handled directly.
*/	*/
else {	else {
q = cur;	q = ctxt->cur;
while (IS_CHAR(cur) && (cur != '<')) cur++;	while (IS_CHAR(ctxt->cur[0]) && (ctxt->cur[0] != '<')) ctxt->cur++;

if (!IS_CHAR(*cur)) {	if (!IS_CHAR(ctxt->cur[0])) {
fprintf(stderr, "Truncated content : %.50s\n", q);	fprintf(stderr, "Truncated content : %.50s\n", q);
*p = cur;
return(NULL);	return(NULL);
}	}

/*	/*
* Do the Entities decoding...	* Do the Entities decoding...
*/	*/
data = xmlStrdup(xmlDecodeEntities(doc, q, cur - q));	data = xmlStrdup(xmlDecodeEntities(ctxt->doc, q, ctxt->cur - q));
}	}

/*	/*
Line 736 xmlNodePtr xmlParseContent(CHAR **p, xml	Line 728 xmlNodePtr xmlParseContent(CHAR **p, xml
}	}
}	}

*p = cur;
return(ret);	return(ret);
}	}

Line 744 xmlNodePtr xmlParseContent(CHAR **p, xml	Line 735 xmlNodePtr xmlParseContent(CHAR **p, xml
* xmlParseElement: parse an XML element	* xmlParseElement: parse an XML element
*/	*/

xmlNodePtr xmlParseElement(CHAR **p, xmlDocPtr doc) {	xmlNodePtr xmlParseElement(xmlParserCtxtPtr ctxt) {
CHAR cur = p;
xmlNodePtr ret, child;	xmlNodePtr ret, child;
CHAR openTag = p;	CHAR *openTag = ctxt->cur;
CHAR closeTag = p;	CHAR *closeTag = ctxt->cur;

ret = xmlParseStartTag(&cur, doc);	ret = xmlParseStartTag(ctxt);
if (ret == NULL) {	if (ret == NULL) {
*p = cur;
return(NULL);	return(NULL);
}	}

/*	/*
* Check for an Empty Element.	* Check for an Empty Element.
*/	*/
if ((cur[0] == '/') && (cur[1] == '>')) {	if ((ctxt->cur[0] == '/') && (ctxt->cur[1] == '>')) {
cur += 2;	ctxt->cur += 2;
*p = cur;
return(ret);	return(ret);
}	}
if (cur[0] == '>') cur++;	if (ctxt->cur[0] == '>') ctxt->cur++;
else {	else {
fprintf(stderr, "Couldn't find end of Start Tag %.30s\n", *p);	fprintf(stderr, "Couldn't find end of Start Tag %.30s\n", openTag);
*p = cur;	return(NULL);
return(ret);
}	}

/*	/*
Line 783 xmlNodePtr xmlParseElement(CHAR **p, xml	Line 770 xmlNodePtr xmlParseElement(CHAR **p, xml
*	*
* The loop stops upon detection of an end of tag '</'	* The loop stops upon detection of an end of tag '</'
*/	*/
while ((IS_CHAR(cur[0])) && ((cur[0] != '<') \|\| (cur[1] != '/'))) {	while ((IS_CHAR(ctxt->cur[0])) &&
child = xmlParseContent(&cur, doc, ret);	((ctxt->cur[0] != '<') \|\| (ctxt->cur[1] != '/'))) {
	child = xmlParseContent(ctxt, ret);
if (child != NULL)	if (child != NULL)
xmlAddChild(ret, child);	xmlAddChild(ret, child);
}	}
if (!IS_CHAR(cur[0])) {	if (!IS_CHAR(ctxt->cur[0])) {
fprintf(stderr, "Premature end of data in tag %.30s\n", *p);	fprintf(stderr, "Premature end of data in tag %.30s\n", openTag);
*p = cur;	return(NULL);
return(ret);
}	}

/*	/*
* parse the end of tag : '</' has been detected.	* parse the end of tag : '</' has been detected.
*/	*/
cur += 2;	ctxt->cur += 2;
if (cur == '>') cur++; / simplified closing </> */	if (ctxt->cur[0] == '>') ctxt->cur++; /* simplified closing </> */
else {	else {
CHAR *endTag;	CHAR *endTag;
xmlDtdPtr endDtd;	xmlDtdPtr endDtd;

xmlParseEndTag(&cur, doc, &endDtd, &endTag);	xmlParseEndTag(ctxt, &endDtd, &endTag);

/*	/*
* Check that the Name in the ETag is the same as in the STag.	* Check that the Name in the ETag is the same as in the STag.
Line 818 xmlNodePtr xmlParseElement(CHAR **p, xml	Line 805 xmlNodePtr xmlParseElement(CHAR **p, xml
}	}
}	}

*p = cur;
return(ret);	return(ret);
}	}

Line 826 xmlNodePtr xmlParseElement(CHAR **p, xml	Line 812 xmlNodePtr xmlParseElement(CHAR **p, xml
* xmlParseXMLDecl: parse an XML declaration header	* xmlParseXMLDecl: parse an XML declaration header
*/	*/

xmlDocPtr xmlParseXMLDecl(CHAR **p) {	void xmlParseXMLDecl(xmlParserCtxtPtr ctxt) {
CHAR cur = p;
CHAR *version;	CHAR *version;
xmlDocPtr ret;

/*	/*
* We know that '<?XML' is here.	* We know that '<?XML' is here.
*/	*/
cur += 5;	ctxt->cur += 5;

/*	/*
* Parse the version info	* Parse the version info
*/	*/
SKIP_BLANKS(cur);	SKIP_BLANKS(ctxt->cur);

/*	/*
* We should have 'version=' here !	* We should have 'version=' here !
*/	*/
if ((cur[0] == 'v') && (cur[1] == 'e') && (cur[2] == 'r') &&	if ((ctxt->cur[0] == 'v') && (ctxt->cur[1] == 'e') &&
(cur[3] == 's') && (cur[4] == 'i') && (cur[5] == 'o') &&	(ctxt->cur[2] == 'r') && (ctxt->cur[3] == 's') &&
(cur[6] == 'n') && (cur[7] == '=')) {	(ctxt->cur[4] == 'i') && (ctxt->cur[5] == 'o') &&
cur += 8;	(ctxt->cur[6] == 'n') && (ctxt->cur[7] == '=')) {
version = xmlParseQuotedString(&cur);	ctxt->cur += 8;
	version = xmlParseQuotedString(ctxt);
if (version == NULL)	if (version == NULL)
ret = xmlNewDoc(XML_DEFAULT_VERSION);	ctxt->doc = xmlNewDoc(XML_DEFAULT_VERSION);
else {	else {
ret = xmlNewDoc(version);	ctxt->doc = xmlNewDoc(version);
free(version);	free(version);
}	}
} else {	} else {
ret = xmlNewDoc(XML_DEFAULT_VERSION);	ctxt->doc = xmlNewDoc(XML_DEFAULT_VERSION);
}	}

/*	/*
* We should check for Required Markup Declaration TODO !!!!	* We should check for Required Markup Declaration TODO !!!!
*/	*/
MOVETO_ENDTAG(cur);	MOVETO_ENDTAG(ctxt->cur);
cur++;	ctxt->cur++;

*p = cur;
return(ret);
}	}

/*	/*
Line 874 xmlDocPtr xmlParseXMLDecl(CHAR **p) {	Line 857 xmlDocPtr xmlParseXMLDecl(CHAR **p) {
* (Comment \| PI \| S)*	* (Comment \| PI \| S)*
*/	*/

void xmlParseMisc(CHAR **p, xmlDocPtr doc) {	void xmlParseMisc(xmlParserCtxtPtr ctxt) {
CHAR cur = p;	while (((ctxt->cur[0] == '<') && (ctxt->cur[1] == '?')) \|\|
	((ctxt->cur[0] == '<') && (ctxt->cur[1] == '!') &&
while (((cur[0] == '<') && (cur[1] == '?')) \|\|	(ctxt->cur[2] == '-') && (ctxt->cur[2] == '-')) \|\|
((cur[0] == '<') && (cur[1] == '!') &&	IS_BLANK(ctxt->cur[0])) {
(cur[2] == '-') && (cur[2] == '-')) \|\|	if ((ctxt->cur[0] == '<') && (ctxt->cur[1] == '?')) {
IS_BLANK(*cur)) {	xmlParsePI(ctxt);
if ((cur[0] == '<') && (cur[1] == '?')) {	} else if (IS_BLANK(ctxt->cur[0])) {
xmlParsePI(&cur, doc);	ctxt->cur++;
} else if (IS_BLANK(*cur)) {
cur++;
} else	} else
xmlParserSkipComment(&cur);	xmlParserSkipComment(ctxt);
}	}

*p = cur;
}	}

/*	/*
* xmlParseDoc : parse an XML document and build a tree.	* xmlParseDocument : parse an XML document and build a tree.
*/	*/

xmlDocPtr xmlParseDoc(CHAR *cur) {	int xmlParseDocument(xmlParserCtxtPtr ctxt) {
xmlDocPtr ret;

/*	/*
* We should check for encoding here and plug-in some	* We should check for encoding here and plug-in some
* conversion code TODO !!!!	* conversion code TODO !!!!
Line 907 xmlDocPtr xmlParseDoc(CHAR *cur) {	Line 884 xmlDocPtr xmlParseDoc(CHAR *cur) {
/*	/*
* Wipe out everything which is before the first '<'	* Wipe out everything which is before the first '<'
*/	*/
SKIP_BLANKS(cur);	SKIP_BLANKS(ctxt->cur);

/*	/*
* Check for the XMLDecl in the Prolog.	* Check for the XMLDecl in the Prolog.
*/	*/
if ((cur[0] == '<') && (cur[1] == '?') &&	if ((ctxt->cur[0] == '<') && (ctxt->cur[1] == '?') &&
(cur[2] == 'X') && (cur[3] == 'M') &&	(ctxt->cur[2] == 'X') && (ctxt->cur[3] == 'M') &&
(cur[4] == 'L')) {	(ctxt->cur[4] == 'L')) {
ret = xmlParseXMLDecl(&cur);	xmlParseXMLDecl(ctxt);
/* SKIP_EOL(cur); */	/* SKIP_EOL(cur); */
SKIP_BLANKS(cur);	SKIP_BLANKS(ctxt->cur);
} else {	} else {
ret = xmlNewDoc(XML_DEFAULT_VERSION);	ctxt->doc = xmlNewDoc(XML_DEFAULT_VERSION);
}	}

/*	/*
* The Misc part of the Prolog	* The Misc part of the Prolog
* (Comment \| PI \| S) *	* (Comment \| PI \| S) *
*/	*/
xmlParseMisc(&cur, ret);	xmlParseMisc(ctxt);

/*	/*
* Time to start parsing	* Time to start parsing
*/	*/
ret->root = xmlParseElement(&cur, ret);	ctxt->doc->root = xmlParseElement(ctxt);

	return(0);
	}

	/*
	* xmlParseDoc : parse an XML in-memory document and build a tree.
	*/

	xmlDocPtr xmlParseDoc(CHAR *cur) {
	xmlDocPtr ret;
	xmlParserCtxtPtr ctxt;

	if (cur == NULL) return(NULL);

	ctxt = (xmlParserCtxtPtr) malloc(sizeof(xmlParserCtxt));
	if (ctxt == NULL) {
	perror("malloc");
	return(NULL);
	}

	ctxt->filename = NULL;
	ctxt->base = cur;
	ctxt->cur = cur;
	ctxt->line = 1;
	ctxt->col = 1;
	ctxt->doc = NULL;

	xmlParseDocument(ctxt);
	ret = ctxt->doc;
	if (ctxt->filename != NULL)
	free(ctxt->filename);
	free(ctxt);

return(ret);	return(ret);
}	}

Line 946 xmlDocPtr xmlParseFile(const char *filen	Line 955 xmlDocPtr xmlParseFile(const char *filen
int res;	int res;
struct stat buf;	struct stat buf;
char *buffer;	char *buffer;
	xmlParserCtxtPtr ctxt;

res = stat(filename, &buf);	res = stat(filename, &buf);
if (res < 0) return(NULL);	if (res < 0) return(NULL);
Line 971 xmlDocPtr xmlParseFile(const char *filen	Line 981 xmlDocPtr xmlParseFile(const char *filen
}	}
close(input);	close(input);

	ctxt = (xmlParserCtxtPtr) malloc(sizeof(xmlParserCtxt));
	if (ctxt == NULL) {
	perror("malloc");
	return(NULL);
	}
buffer[buf.st_size] = '\0';	buffer[buf.st_size] = '\0';
ret = xmlParseDoc(buffer);
	ctxt->filename = strdup(filename);
	ctxt->base = buffer;
	ctxt->cur = buffer;
	ctxt->line = 1;
	ctxt->col = 1;
	ctxt->doc = NULL;

	xmlParseDocument(ctxt);
	ret = ctxt->doc;
free(buffer);	free(buffer);
	if (ctxt->filename != NULL)
	free(ctxt->filename);
	free(ctxt);

return(ret);	return(ret);
}	}

Webmaster

Removed from v.1.15
changed lines
	Added in v.1.16