xmlCreateEntityParserCtxtInternal(const xmlChar *URL, const xmlChar *ID,
const xmlChar *base, xmlParserCtxtPtr pctx);
+static void xmlHaltParser(xmlParserCtxtPtr ctxt);
+
/************************************************************************
* *
* Arbitrary limits set in the parser. See XML_PARSE_HUGE *
return (0);
if (ctxt->lastError.code == XML_ERR_ENTITY_LOOP)
return (1);
+
+ /*
+ * This may look absurd but is needed to detect
+ * entities problems
+ */
+ if ((ent != NULL) && (ent->etype != XML_INTERNAL_PREDEFINED_ENTITY) &&
+ (ent->content != NULL) && (ent->checked == 0) &&
+ (ctxt->errNo != XML_ERR_ENTITY_LOOP)) {
+ unsigned long oldnbent = ctxt->nbentities;
+ xmlChar *rep;
+
+ ent->checked = 1;
+
+ ++ctxt->depth;
+ rep = xmlStringDecodeEntities(ctxt, ent->content,
+ XML_SUBSTITUTE_REF, 0, 0, 0);
+ --ctxt->depth;
+ if (ctxt->errNo == XML_ERR_ENTITY_LOOP) {
+ ent->content[0] = 0;
+ }
+
+ ent->checked = (ctxt->nbentities - oldnbent + 1) * 2;
+ if (rep != NULL) {
+ if (xmlStrchr(rep, '<'))
+ ent->checked |= 1;
+ xmlFree(rep);
+ rep = NULL;
+ }
+ }
if (replacement != 0) {
if (replacement < XML_MAX_TEXT_LENGTH)
return(0);
return (0);
} else {
/*
- * strange we got no data for checking just return
+ * strange we got no data for checking
*/
- return (0);
+ if (((ctxt->lastError.code != XML_ERR_UNDECLARED_ENTITY) &&
+ (ctxt->lastError.code != XML_WAR_UNDECLARED_ENTITY)) ||
+ (ctxt->nbentities <= 10000))
+ return (0);
}
xmlFatalErr(ctxt, XML_ERR_ENTITY_LOOP, NULL);
return (1);
xmlFatalErr(xmlParserCtxtPtr ctxt, xmlParserErrors error, const char *info)
{
const char *errmsg;
- char errstr[129] = "";
if ((ctxt != NULL) && (ctxt->disableSAX != 0) &&
(ctxt->instate == XML_PARSER_EOF))
default:
errmsg = "Unregistered error message";
}
- if (info == NULL)
- snprintf(errstr, 128, "%s\n", errmsg);
- else
- snprintf(errstr, 128, "%s: %%s\n", errmsg);
if (ctxt != NULL)
ctxt->errNo = error;
- __xmlRaiseError(NULL, NULL, NULL, ctxt, NULL, XML_FROM_PARSER, error,
- XML_ERR_FATAL, NULL, 0, info, NULL, NULL, 0, 0, &errstr[0],
- info);
+ if (info == NULL) {
+ __xmlRaiseError(NULL, NULL, NULL, ctxt, NULL, XML_FROM_PARSER, error,
+ XML_ERR_FATAL, NULL, 0, info, NULL, NULL, 0, 0, "%s\n",
+ errmsg);
+ } else {
+ __xmlRaiseError(NULL, NULL, NULL, ctxt, NULL, XML_FROM_PARSER, error,
+ XML_ERR_FATAL, NULL, 0, info, NULL, NULL, 0, 0, "%s: %s\n",
+ errmsg, info);
+ }
if (ctxt != NULL) {
ctxt->wellFormed = 0;
if (ctxt->recovery == 0)
*
* Handle a fatal parser error, i.e. violating Well-Formedness constraints
*/
-static void
+static void LIBXML_ATTR_FORMAT(3,0)
xmlFatalErrMsg(xmlParserCtxtPtr ctxt, xmlParserErrors error,
const char *msg)
{
*
* Handle a warning.
*/
-static void
+static void LIBXML_ATTR_FORMAT(3,0)
xmlWarningMsg(xmlParserCtxtPtr ctxt, xmlParserErrors error,
const char *msg, const xmlChar *str1, const xmlChar *str2)
{
*
* Handle a validity error.
*/
-static void
+static void LIBXML_ATTR_FORMAT(3,0)
xmlValidityError(xmlParserCtxtPtr ctxt, xmlParserErrors error,
const char *msg, const xmlChar *str1, const xmlChar *str2)
{
*
* Handle a fatal parser error, i.e. violating Well-Formedness constraints
*/
-static void
+static void LIBXML_ATTR_FORMAT(3,0)
xmlFatalErrMsgInt(xmlParserCtxtPtr ctxt, xmlParserErrors error,
const char *msg, int val)
{
*
* Handle a fatal parser error, i.e. violating Well-Formedness constraints
*/
-static void
+static void LIBXML_ATTR_FORMAT(3,0)
xmlFatalErrMsgStrIntStr(xmlParserCtxtPtr ctxt, xmlParserErrors error,
const char *msg, const xmlChar *str1, int val,
const xmlChar *str2)
*
* Handle a fatal parser error, i.e. violating Well-Formedness constraints
*/
-static void
+static void LIBXML_ATTR_FORMAT(3,0)
xmlFatalErrMsgStr(xmlParserCtxtPtr ctxt, xmlParserErrors error,
const char *msg, const xmlChar * val)
{
*
* Handle a non fatal parser error
*/
-static void
+static void LIBXML_ATTR_FORMAT(3,0)
xmlErrMsgStr(xmlParserCtxtPtr ctxt, xmlParserErrors error,
const char *msg, const xmlChar * val)
{
*
* Handle a fatal parser error, i.e. violating Well-Formedness constraints
*/
-static void
+static void LIBXML_ATTR_FORMAT(3,0)
xmlNsErr(xmlParserCtxtPtr ctxt, xmlParserErrors error,
const char *msg,
const xmlChar * info1, const xmlChar * info2,
*
* Handle a namespace warning error
*/
-static void
+static void LIBXML_ATTR_FORMAT(3,0)
xmlNsWarn(xmlParserCtxtPtr ctxt, xmlParserErrors error,
const char *msg,
const xmlChar * info1, const xmlChar * info2,
xmlFatalErrMsgInt(ctxt, XML_ERR_INTERNAL_ERROR,
"Excessive depth in document: %d use XML_PARSE_HUGE option\n",
xmlParserMaxDepth);
- ctxt->instate = XML_PARSER_EOF;
+ xmlHaltParser(ctxt);
return(-1);
}
ctxt->nodeTab[ctxt->nodeNr] = value;
#define CUR (*ctxt->input->cur)
#define NXT(val) ctxt->input->cur[(val)]
#define CUR_PTR ctxt->input->cur
+#define BASE_PTR ctxt->input->base
#define CMP4( s, c1, c2, c3, c4 ) \
( ((unsigned char *) s)[ 0 ] == c1 && ((unsigned char *) s)[ 1 ] == c2 && \
xmlGROW (ctxt);
static void xmlGROW (xmlParserCtxtPtr ctxt) {
- if ((((ctxt->input->end - ctxt->input->cur) > XML_MAX_LOOKUP_LIMIT) ||
- ((ctxt->input->cur - ctxt->input->base) > XML_MAX_LOOKUP_LIMIT)) &&
+ unsigned long curEnd = ctxt->input->end - ctxt->input->cur;
+ unsigned long curBase = ctxt->input->cur - ctxt->input->base;
+
+ if (((curEnd > (unsigned long) XML_MAX_LOOKUP_LIMIT) ||
+ (curBase > (unsigned long) XML_MAX_LOOKUP_LIMIT)) &&
((ctxt->input->buf) && (ctxt->input->buf->readcallback != (xmlInputReadCallback) xmlNop)) &&
((ctxt->options & XML_PARSE_HUGE) == 0)) {
xmlFatalErr(ctxt, XML_ERR_INTERNAL_ERROR, "Huge input lookup");
- ctxt->instate = XML_PARSER_EOF;
+ xmlHaltParser(ctxt);
+ return;
}
xmlParserInputGrow(ctxt->input, INPUT_CHUNK);
+ if ((ctxt->input->cur > ctxt->input->end) ||
+ (ctxt->input->cur < ctxt->input->base)) {
+ xmlHaltParser(ctxt);
+ xmlFatalErr(ctxt, XML_ERR_INTERNAL_ERROR, "cur index out of bound");
+ return;
+ }
if ((ctxt->input->cur != NULL) && (*ctxt->input->cur == 0) &&
(xmlParserInputGrow(ctxt->input, INPUT_CHUNK) <= 0))
xmlPopInput(ctxt);
while (IS_BLANK_CH(*cur)) {
if (*cur == '\n') {
ctxt->input->line++; ctxt->input->col = 1;
+ } else {
+ ctxt->input->col++;
}
cur++;
res++;
int cur;
do {
cur = CUR;
- while (IS_BLANK_CH(cur)) { /* CHECKED tstblanks.xml */
+ while ((IS_BLANK_CH(cur) && /* CHECKED tstblanks.xml */
+ (ctxt->instate != XML_PARSER_EOF))) {
NEXT;
cur = CUR;
res++;
* Need to handle support of entities branching here
*/
if (*ctxt->input->cur == '%') xmlParserHandlePEReference(ctxt);
- } while (IS_BLANK(cur)); /* CHECKED tstblanks.xml */
+ } while ((IS_BLANK(cur)) && /* CHECKED tstblanks.xml */
+ (ctxt->instate != XML_PARSER_EOF));
}
return(res);
}
name, NULL);
ctxt->valid = 0;
}
+ xmlParserEntityCheck(ctxt, 0, NULL, 0);
} else if (ctxt->input->free != deallocblankswrapper) {
input = xmlNewBlanksWrapperInputStream(ctxt, entity);
if (xmlPushInput(ctxt, input) < 0)
xmlCharEncoding enc;
/*
+ * Note: external parameter entities will not be loaded, it
+ * is not required for a non-validating parser, unless the
+ * option of validating, or substituting entities were
+ * given. Doing so is far more secure as the parser will
+ * only process data coming from the document entity by
+ * default.
+ */
+ if ((entity->etype == XML_EXTERNAL_PARAMETER_ENTITY) &&
+ ((ctxt->options & XML_PARSE_NOENT) == 0) &&
+ ((ctxt->options & XML_PARSE_DTDVALID) == 0) &&
+ ((ctxt->options & XML_PARSE_DTDLOAD) == 0) &&
+ ((ctxt->options & XML_PARSE_DTDATTR) == 0) &&
+ (ctxt->replaceEntities == 0) &&
+ (ctxt->validate == 0))
+ return;
+
+ /*
* handle the extra spaces added before and after
* c.f. http://www.w3.org/TR/REC-xml#as-PE
* this is done independently.
if ((ctxt->lastError.code == XML_ERR_ENTITY_LOOP) ||
(ctxt->lastError.code == XML_ERR_INTERNAL_ERROR))
goto int_error;
+ xmlParserEntityCheck(ctxt, 0, ent, 0);
if (ent != NULL)
ctxt->nbentities += ent->checked / 2;
if ((ent != NULL) &&
0, 0, 0);
ctxt->depth--;
+ if ((ctxt->lastError.code == XML_ERR_ENTITY_LOOP) ||
+ (ctxt->lastError.code == XML_ERR_INTERNAL_ERROR))
+ goto int_error;
+
if (rep != NULL) {
current = rep;
while (*current != 0) { /* non input consuming loop */
ent = xmlParseStringPEReference(ctxt, &str);
if (ctxt->lastError.code == XML_ERR_ENTITY_LOOP)
goto int_error;
+ xmlParserEntityCheck(ctxt, 0, ent, 0);
if (ent != NULL)
ctxt->nbentities += ent->checked / 2;
if (ent != NULL) {
if (ent->content == NULL) {
- xmlLoadEntityContent(ctxt, ent);
+ /*
+ * Note: external parsed entities will not be loaded,
+ * it is not required for a non-validating parser to
+ * complete external PEreferences coming from the
+ * internal subset
+ */
+ if (((ctxt->options & XML_PARSE_NOENT) != 0) ||
+ ((ctxt->options & XML_PARSE_DTDVALID) != 0) ||
+ (ctxt->validate != 0)) {
+ xmlLoadEntityContent(ctxt, ent);
+ } else {
+ xmlWarningMsg(ctxt, XML_ERR_ENTITY_PROCESSING,
+ "not validating will not read content for PE entity %s\n",
+ ent->name, NULL);
+ }
}
ctxt->depth++;
rep = xmlStringDecodeEntities(ctxt, ent->content, what,
int len = 0, l;
int c;
int count = 0;
+ size_t startPosition = 0;
#ifdef DEBUG
nbParseNCNameComplex++;
* Handler for more complex cases
*/
GROW;
+ startPosition = CUR_PTR - BASE_PTR;
c = CUR_CHAR(l);
if ((c == ' ') || (c == '>') || (c == '/') || /* accelerators */
(!xmlIsNameStartChar(ctxt, c) || (c == ':'))) {
c = CUR_CHAR(l);
if (c == 0) {
count = 0;
+ /*
+ * when shrinking to extend the buffer we really need to preserve
+ * the part of the name we already parsed. Hence rolling back
+ * by current lenght.
+ */
+ ctxt->input->cur -= l;
GROW;
+ ctxt->input->cur += l;
if (ctxt->instate == XML_PARSER_EOF)
return(NULL);
c = CUR_CHAR(l);
xmlFatalErr(ctxt, XML_ERR_NAME_TOO_LONG, "NCName");
return(NULL);
}
- return(xmlDictLookup(ctxt->dict, ctxt->input->cur - len, len));
+ return(xmlDictLookup(ctxt->dict, (BASE_PTR + startPosition), len));
}
/**
static const xmlChar *
xmlParseNCName(xmlParserCtxtPtr ctxt) {
- const xmlChar *in;
+ const xmlChar *in, *e;
const xmlChar *ret;
int count = 0;
* Accelerator for simple ASCII names
*/
in = ctxt->input->cur;
- if (((*in >= 0x61) && (*in <= 0x7A)) ||
- ((*in >= 0x41) && (*in <= 0x5A)) ||
- (*in == '_')) {
+ e = ctxt->input->end;
+ if ((((*in >= 0x61) && (*in <= 0x7A)) ||
+ ((*in >= 0x41) && (*in <= 0x5A)) ||
+ (*in == '_')) && (in < e)) {
in++;
- while (((*in >= 0x61) && (*in <= 0x7A)) ||
- ((*in >= 0x41) && (*in <= 0x5A)) ||
- ((*in >= 0x30) && (*in <= 0x39)) ||
- (*in == '_') || (*in == '-') ||
- (*in == '.'))
+ while ((((*in >= 0x61) && (*in <= 0x7A)) ||
+ ((*in >= 0x41) && (*in <= 0x5A)) ||
+ ((*in >= 0x30) && (*in <= 0x39)) ||
+ (*in == '_') || (*in == '-') ||
+ (*in == '.')) && (in < e))
in++;
+ if (in >= e)
+ goto complex;
if ((*in > 0) && (*in < 0x80)) {
count = in - ctxt->input->cur;
if ((count > XML_MAX_NAME_LENGTH) &&
return(ret);
}
}
+complex:
return(xmlParseNCNameComplex(ctxt));
}
}
/* failure (or end of input buffer), check with full function */
ret = xmlParseName (ctxt);
- /* strings coming from the dictionnary direct compare possible */
+ /* strings coming from the dictionary direct compare possible */
if (ret == other) {
return (const xmlChar*) 1;
}
* an entity declaration, it is bypassed and left as is.
* so XML_SUBSTITUTE_REF is not set here.
*/
+ ++ctxt->depth;
ret = xmlStringDecodeEntities(ctxt, buf, XML_SUBSTITUTE_PEREF,
0, 0, 0);
+ --ctxt->depth;
if (orig != NULL)
*orig = buf;
else
} else if ((ent != NULL) &&
(ctxt->replaceEntities != 0)) {
if (ent->etype != XML_INTERNAL_PREDEFINED_ENTITY) {
+ ++ctxt->depth;
rep = xmlStringDecodeEntities(ctxt, ent->content,
XML_SUBSTITUTE_REF,
0, 0, 0);
+ --ctxt->depth;
if (rep != NULL) {
current = rep;
while (*current != 0) { /* non input consuming */
(ent->content != NULL) && (ent->checked == 0)) {
unsigned long oldnbent = ctxt->nbentities;
+ ++ctxt->depth;
rep = xmlStringDecodeEntities(ctxt, ent->content,
XML_SUBSTITUTE_REF, 0, 0, 0);
+ --ctxt->depth;
ent->checked = (ctxt->nbentities - oldnbent + 1) * 2;
if (rep != NULL) {
}
if ((name != NULL) && (xmlStrchr(name, ':') != NULL)) {
xmlNsErr(ctxt, XML_NS_ERR_COLON,
- "colon are forbidden from PI names '%s'\n", name, NULL, NULL);
+ "colons are forbidden from PI names '%s'\n", name, NULL, NULL);
}
return(name);
}
}
if (xmlStrchr(name, ':') != NULL) {
xmlNsErr(ctxt, XML_NS_ERR_COLON,
- "colon are forbidden from notation names '%s'\n",
+ "colons are forbidden from notation names '%s'\n",
name, NULL, NULL);
}
SKIP_BLANKS;
skipped = SKIP_BLANKS;
if (skipped == 0) {
xmlFatalErrMsg(ctxt, XML_ERR_SPACE_REQUIRED,
- "Space required after '%'\n");
+ "Space required after '%%'\n");
}
isParameter = 1;
}
}
if (xmlStrchr(name, ':') != NULL) {
xmlNsErr(ctxt, XML_NS_ERR_COLON,
- "colon are forbidden from entities names '%s'\n",
+ "colons are forbidden from entities names '%s'\n",
name, NULL, NULL);
}
skipped = SKIP_BLANKS;
if (RAW != '>') {
xmlFatalErrMsgStr(ctxt, XML_ERR_ENTITY_NOT_FINISHED,
"xmlParseEntityDecl: entity %s not terminated\n", name);
+ xmlHaltParser(ctxt);
} else {
if (input != ctxt->input) {
xmlFatalErrMsg(ctxt, XML_ERR_ENTITY_BOUNDARY,
if (!IS_BLANK_CH(CUR)) {
xmlFatalErrMsg(ctxt, XML_ERR_SPACE_REQUIRED,
"Space required after 'ELEMENT'\n");
+ return(-1);
}
SKIP_BLANKS;
name = xmlParseName(ctxt);
SKIP_BLANKS;
if (RAW != '[') {
xmlFatalErr(ctxt, XML_ERR_CONDSEC_INVALID, NULL);
+ xmlHaltParser(ctxt);
+ return;
} else {
if (ctxt->input->id != id) {
xmlValidityError(ctxt, XML_ERR_ENTITY_BOUNDARY,
if ((CUR_PTR == check) && (cons == ctxt->input->consumed)) {
xmlFatalErr(ctxt, XML_ERR_EXT_SUBSET_NOT_FINISHED, NULL);
+ xmlHaltParser(ctxt);
break;
}
}
SKIP_BLANKS;
if (RAW != '[') {
xmlFatalErr(ctxt, XML_ERR_CONDSEC_INVALID, NULL);
+ xmlHaltParser(ctxt);
+ return;
} else {
if (ctxt->input->id != id) {
xmlValidityError(ctxt, XML_ERR_ENTITY_BOUNDARY,
} else {
xmlFatalErr(ctxt, XML_ERR_CONDSEC_INVALID_KEYWORD, NULL);
+ xmlHaltParser(ctxt);
+ return;
}
if (RAW == 0)
"All markup of the conditional section is not in the same entity\n",
NULL, NULL);
}
- SKIP(3);
+ if ((ctxt-> instate != XML_PARSER_EOF) &&
+ ((ctxt->input->cur + 3) <= ctxt->input->end))
+ SKIP(3);
}
}
xmlParsePI(ctxt);
}
}
+
+ /*
+ * detect requirement to exit there and act accordingly
+ * and avoid having instate overriden later on
+ */
+ if (ctxt->instate == XML_PARSER_EOF)
+ return;
+
/*
* This is only for internal subset. On external entities,
* the replacement is done before parsing stage
/*
* The XML REC instructs us to stop parsing right here
*/
- ctxt->instate = XML_PARSER_EOF;
+ xmlHaltParser(ctxt);
return;
}
}
* far more secure as the parser will only process data coming from
* the document entity by default.
*/
- if ((ent->checked == 0) &&
+ if (((ent->checked == 0) ||
+ ((ent->children == NULL) && (ctxt->options & XML_PARSE_NOENT))) &&
((ent->etype != XML_EXTERNAL_GENERAL_PARSED_ENTITY) ||
(ctxt->options & (XML_PARSE_NOENT | XML_PARSE_DTDVALID)))) {
unsigned long oldnbent = ctxt->nbentities;
(ret != XML_WAR_UNDECLARED_ENTITY)) {
xmlFatalErrMsgStr(ctxt, XML_ERR_UNDECLARED_ENTITY,
"Entity '%s' failed to parse\n", ent->name);
+ xmlParserEntityCheck(ctxt, 0, ent, 0);
} else if (list != NULL) {
xmlFreeNodeList(list);
list = NULL;
/*
* We are copying here, make sure there is no abuse
*/
- ctxt->sizeentcopy += ent->length;
+ ctxt->sizeentcopy += ent->length + 5;
if (xmlParserEntityCheck(ctxt, 0, ent, ctxt->sizeentcopy))
return;
/*
* We are copying here, make sure there is no abuse
*/
- ctxt->sizeentcopy += ent->length;
+ ctxt->sizeentcopy += ent->length + 5;
if (xmlParserEntityCheck(ctxt, 0, ent, ctxt->sizeentcopy))
return;
ctxt->sax->reference(ctxt->userData, name);
}
}
+ xmlParserEntityCheck(ctxt, 0, ent, 0);
ctxt->valid = 0;
}
else if ((ctxt->instate == XML_PARSER_ATTRIBUTE_VALUE) &&
(ent != NULL) &&
(ent->etype != XML_INTERNAL_PREDEFINED_ENTITY)) {
- if ((ent->checked & 1) || ((ent->checked == 0) &&
- (ent->content != NULL) &&(xmlStrchr(ent->content, '<')))) {
+ if (((ent->checked & 1) || (ent->checked == 0)) &&
+ (ent->content != NULL) && (xmlStrchr(ent->content, '<'))) {
xmlFatalErrMsgStr(ctxt, XML_ERR_LT_IN_ATTRIBUTE,
"'<' in entity '%s' is not allowed in attributes values\n", name);
}
/*
- * Predefined entites override any extra definition
+ * Predefined entities override any extra definition
*/
if ((ctxt->options & XML_PARSE_OLDSAX) == 0) {
ent = xmlGetPredefinedEntity(name);
"Entity '%s' not defined\n",
name);
}
+ xmlParserEntityCheck(ctxt, 0, ent, 0);
/* TODO ? check regressions ctxt->valid = 0; */
}
name, NULL);
ctxt->valid = 0;
}
+ xmlParserEntityCheck(ctxt, 0, NULL, 0);
} else {
/*
* Internal checking in case the entity quest barfed
* The XML REC instructs us to stop parsing
* right here
*/
- ctxt->instate = XML_PARSER_EOF;
+ xmlHaltParser(ctxt);
return;
}
}
name, NULL);
ctxt->valid = 0;
}
+ xmlParserEntityCheck(ctxt, 0, NULL, 0);
} else {
/*
* Internal checking in case the entity quest barfed
*/
if (RAW != '>') {
xmlFatalErr(ctxt, XML_ERR_DOCTYPE_NOT_FINISHED, NULL);
+ return;
}
NEXT;
}
* @prefix: the prefix to lookup
*
* Lookup the namespace name for the @prefix (which ca be NULL)
- * The prefix must come from the @ctxt->dict dictionnary
+ * The prefix must come from the @ctxt->dict dictionary
*
* Returns the namespace name or NULL if not bound
*/
xmlChar limit = 0;
const xmlChar *in = NULL, *start, *end, *last;
xmlChar *ret = NULL;
+ int line, col;
GROW;
in = (xmlChar *) CUR_PTR;
+ line = ctxt->input->line;
+ col = ctxt->input->col;
if (*in != '"' && *in != '\'') {
xmlFatalErr(ctxt, XML_ERR_ATTRIBUTE_NOT_STARTED, NULL);
return (NULL);
* pure ASCII.
*/
limit = *in++;
+ col++;
end = ctxt->input->end;
start = in;
if (in >= end) {
while ((in < end) && (*in != limit) &&
((*in == 0x20) || (*in == 0x9) ||
(*in == 0xA) || (*in == 0xD))) {
+ if (*in == 0xA) {
+ line++; col = 1;
+ } else {
+ col++;
+ }
in++;
start = in;
if (in >= end) {
}
while ((in < end) && (*in != limit) && (*in >= 0x20) &&
(*in <= 0x7f) && (*in != '&') && (*in != '<')) {
+ col++;
if ((*in++ == 0x20) && (*in == 0x20)) break;
if (in >= end) {
const xmlChar *oldbase = ctxt->input->base;
while ((in < end) && (*in != limit) &&
((*in == 0x20) || (*in == 0x9) ||
(*in == 0xA) || (*in == 0xD))) {
+ if (*in == 0xA) {
+ line++, col = 1;
+ } else {
+ col++;
+ }
in++;
if (in >= end) {
const xmlChar *oldbase = ctxt->input->base;
while ((in < end) && (*in != limit) && (*in >= 0x20) &&
(*in <= 0x7f) && (*in != '&') && (*in != '<')) {
in++;
+ col++;
if (in >= end) {
const xmlChar *oldbase = ctxt->input->base;
GROW;
if (*in != limit) goto need_complex;
}
in++;
+ col++;
if (len != NULL) {
*len = last - start;
ret = (xmlChar *) start;
ret = xmlStrndup(start, last - start);
}
CUR_PTR = in;
+ ctxt->input->line = line;
+ ctxt->input->col = col;
if (alloc) *alloc = 0;
return ret;
need_complex:
const xmlChar **atts = ctxt->atts;
int maxatts = ctxt->maxatts;
int nratts, nbatts, nbdef;
- int i, j, nbNs, attval, oldline, oldcol;
+ int i, j, nbNs, attval, oldline, oldcol, inputNr;
const xmlChar *base;
unsigned long cur;
int nsNr = ctxt->nsNr;
SHRINK;
base = ctxt->input->base;
cur = ctxt->input->cur - ctxt->input->base;
+ inputNr = ctxt->inputNr;
oldline = ctxt->input->line;
oldcol = ctxt->input->col;
nbatts = 0;
*/
SKIP_BLANKS;
GROW;
- if (ctxt->input->base != base) goto base_changed;
+ if ((ctxt->input->base != base) || (inputNr != ctxt->inputNr))
+ goto base_changed;
while (((RAW != '>') &&
((RAW != '/') || (NXT(1) != '>')) &&
attname = xmlParseAttribute2(ctxt, prefix, localname,
&aprefix, &attvalue, &len, &alloc);
- if (ctxt->input->base != base) {
+ if ((ctxt->input->base != base) || (inputNr != ctxt->inputNr)) {
if ((attvalue != NULL) && (alloc != 0))
xmlFree(attvalue);
attvalue = NULL;
const xmlChar *URL = xmlDictLookup(ctxt->dict, attvalue, len);
xmlURIPtr uri;
+ if (URL == NULL) {
+ xmlErrMemory(ctxt, "dictionary allocation failure");
+ if ((attvalue != NULL) && (alloc != 0))
+ xmlFree(attvalue);
+ return(NULL);
+ }
if (*URL != 0) {
uri = xmlParseURI((const char *) URL);
if (uri == NULL) {
else
if (nsPush(ctxt, NULL, URL) > 0) nbNs++;
skip_default_ns:
- if (alloc != 0) xmlFree(attvalue);
+ if ((attvalue != NULL) && (alloc != 0)) {
+ xmlFree(attvalue);
+ attvalue = NULL;
+ }
+ if ((RAW == '>') || (((RAW == '/') && (NXT(1) == '>'))))
+ break;
+ if (!IS_BLANK_CH(RAW)) {
+ xmlFatalErrMsg(ctxt, XML_ERR_SPACE_REQUIRED,
+ "attributes construct error\n");
+ break;
+ }
SKIP_BLANKS;
+ if ((ctxt->input->base != base) || (inputNr != ctxt->inputNr))
+ goto base_changed;
continue;
}
if (aprefix == ctxt->str_xmlns) {
else
if (nsPush(ctxt, attname, URL) > 0) nbNs++;
skip_ns:
- if (alloc != 0) xmlFree(attvalue);
+ if ((attvalue != NULL) && (alloc != 0)) {
+ xmlFree(attvalue);
+ attvalue = NULL;
+ }
+ if ((RAW == '>') || (((RAW == '/') && (NXT(1) == '>'))))
+ break;
+ if (!IS_BLANK_CH(RAW)) {
+ xmlFatalErrMsg(ctxt, XML_ERR_SPACE_REQUIRED,
+ "attributes construct error\n");
+ break;
+ }
SKIP_BLANKS;
- if (ctxt->input->base != base) goto base_changed;
+ if ((ctxt->input->base != base) || (inputNr != ctxt->inputNr))
+ goto base_changed;
continue;
}
GROW
if (ctxt->instate == XML_PARSER_EOF)
break;
- if (ctxt->input->base != base) goto base_changed;
+ if ((ctxt->input->base != base) || (inputNr != ctxt->inputNr))
+ goto base_changed;
if ((RAW == '>') || (((RAW == '/') && (NXT(1) == '>'))))
break;
if (!IS_BLANK_CH(RAW)) {
break;
}
GROW;
- if (ctxt->input->base != base) goto base_changed;
+ if ((ctxt->input->base != base) || (inputNr != ctxt->inputNr))
+ goto base_changed;
}
/*
if ((ctxt->attallocs[j] != 0) && (atts[i] != NULL))
xmlFree((xmlChar *) atts[i]);
}
+
+ /*
+ * We can't switch from one entity to another in the middle
+ * of a start tag
+ */
+ if (inputNr != ctxt->inputNr) {
+ xmlFatalErrMsg(ctxt, XML_ERR_ENTITY_BOUNDARY,
+ "Start tag doesn't start and stop in the same entity\n");
+ return(NULL);
+ }
+
ctxt->input->cur = ctxt->input->base + cur;
ctxt->input->line = oldline;
ctxt->input->col = oldcol;
xmlParseEndTag2(xmlParserCtxtPtr ctxt, const xmlChar *prefix,
const xmlChar *URI, int line, int nsNr, int tlen) {
const xmlChar *name;
+ size_t curLength;
GROW;
if ((RAW != '<') || (NXT(1) != '/')) {
}
SKIP(2);
- if ((tlen > 0) && (xmlStrncmp(ctxt->input->cur, ctxt->name, tlen) == 0)) {
- if (ctxt->input->cur[tlen] == '>') {
+ curLength = ctxt->input->end - ctxt->input->cur;
+ if ((tlen > 0) && (curLength >= (size_t)tlen) &&
+ (xmlStrncmp(ctxt->input->cur, ctxt->name, tlen) == 0)) {
+ if ((curLength >= (size_t)(tlen + 1)) &&
+ (ctxt->input->cur[tlen] == '>')) {
ctxt->input->cur += tlen + 1;
+ ctxt->input->col += tlen + 1;
goto done;
}
ctxt->input->cur += tlen;
+ ctxt->input->col += tlen;
name = (xmlChar*)1;
} else {
if (prefix == NULL)
if ((cons == ctxt->input->consumed) && (test == CUR_PTR)) {
xmlFatalErr(ctxt, XML_ERR_INTERNAL_ERROR,
"detected an error in element content\n");
- ctxt->instate = XML_PARSER_EOF;
+ xmlHaltParser(ctxt);
break;
}
}
xmlFatalErrMsgInt(ctxt, XML_ERR_INTERNAL_ERROR,
"Excessive depth in document: %d use XML_PARSE_HUGE option\n",
xmlParserMaxDepth);
- ctxt->instate = XML_PARSER_EOF;
+ xmlHaltParser(ctxt);
return;
}
encoding = xmlParseEncName(ctxt);
if (RAW != '"') {
xmlFatalErr(ctxt, XML_ERR_STRING_NOT_CLOSED, NULL);
+ xmlFree((xmlChar *) encoding);
+ return(NULL);
} else
NEXT;
} else if (RAW == '\''){
encoding = xmlParseEncName(ctxt);
if (RAW != '\'') {
xmlFatalErr(ctxt, XML_ERR_STRING_NOT_CLOSED, NULL);
+ xmlFree((xmlChar *) encoding);
+ return(NULL);
} else
NEXT;
} else {
/*
* Non standard parsing, allowing the user to ignore encoding
*/
- if (ctxt->options & XML_PARSE_IGNORE_ENC)
- return(encoding);
+ if (ctxt->options & XML_PARSE_IGNORE_ENC) {
+ xmlFree((xmlChar *) encoding);
+ return(NULL);
+ }
/*
* UTF-16 encoding stwich has already taken place at this stage,
handler = xmlFindCharEncodingHandler((const char *) encoding);
if (handler != NULL) {
- xmlSwitchToEncoding(ctxt, handler);
+ if (xmlSwitchToEncoding(ctxt, handler) < 0) {
+ /* failed to convert */
+ ctxt->errNo = XML_ERR_UNSUPPORTED_ENCODING;
+ return(NULL);
+ }
} else {
xmlFatalErrMsgStr(ctxt, XML_ERR_UNSUPPORTED_ENCODING,
"Unsupported encoding %s\n", encoding);
xmlFatalErrMsg(ctxt, XML_ERR_SPACE_REQUIRED, "Blank needed here\n");
}
xmlParseEncodingDecl(ctxt);
- if (ctxt->errNo == XML_ERR_UNSUPPORTED_ENCODING) {
+ if ((ctxt->errNo == XML_ERR_UNSUPPORTED_ENCODING) ||
+ (ctxt->instate == XML_PARSER_EOF)) {
/*
* The XML REC instructs us to stop parsing right here
*/
if (CUR == 0) {
xmlFatalErr(ctxt, XML_ERR_DOCUMENT_EMPTY, NULL);
+ return(-1);
}
/*
* Note that we will switch encoding on the fly.
*/
xmlParseXMLDecl(ctxt);
- if (ctxt->errNo == XML_ERR_UNSUPPORTED_ENCODING) {
+ if ((ctxt->errNo == XML_ERR_UNSUPPORTED_ENCODING) ||
+ (ctxt->instate == XML_PARSER_EOF)) {
/*
* The XML REC instructs us to stop parsing right here
*/
ctxt->sax->startDocument(ctxt->userData);
if (ctxt->instate == XML_PARSER_EOF)
return(-1);
+ if ((ctxt->myDoc != NULL) && (ctxt->input != NULL) &&
+ (ctxt->input->buf != NULL) && (ctxt->input->buf->compressed >= 0)) {
+ ctxt->myDoc->compression = ctxt->input->buf->compressed;
+ }
/*
* The Misc part of the Prolog
}
/**
* xmlCheckCdataPush:
- * @cur: pointer to the bock of characters
+ * @cur: pointer to the block of characters
* @len: length of the block in bytes
+ * @complete: 1 if complete CDATA block is passed in, 0 if partial block
*
* Check that the block of characters is okay as SCdata content [20]
*
* UTF-8 error occured otherwise
*/
static int
-xmlCheckCdataPush(const xmlChar *utf, int len) {
+xmlCheckCdataPush(const xmlChar *utf, int len, int complete) {
int ix;
unsigned char c;
int codepoint;
else
return(-ix);
} else if ((c & 0xe0) == 0xc0) {/* 2-byte code, starts with 110 */
- if (ix + 2 > len) return(ix);
+ if (ix + 2 > len) return(complete ? -ix : ix);
if ((utf[ix+1] & 0xc0 ) != 0x80)
return(-ix);
codepoint = (utf[ix] & 0x1f) << 6;
return(-ix);
ix += 2;
} else if ((c & 0xf0) == 0xe0) {/* 3-byte code, starts with 1110 */
- if (ix + 3 > len) return(ix);
+ if (ix + 3 > len) return(complete ? -ix : ix);
if (((utf[ix+1] & 0xc0) != 0x80) ||
((utf[ix+2] & 0xc0) != 0x80))
return(-ix);
return(-ix);
ix += 3;
} else if ((c & 0xf8) == 0xf0) {/* 4-byte code, starts with 11110 */
- if (ix + 4 > len) return(ix);
+ if (ix + 4 > len) return(complete ? -ix : ix);
if (((utf[ix+1] & 0xc0) != 0x80) ||
((utf[ix+2] & 0xc0) != 0x80) ||
((utf[ix+3] & 0xc0) != 0x80))
ctxt->sax->setDocumentLocator(ctxt->userData,
&xmlDefaultSAXLocator);
xmlFatalErr(ctxt, XML_ERR_DOCUMENT_EMPTY, NULL);
- ctxt->instate = XML_PARSER_EOF;
+ xmlHaltParser(ctxt);
#ifdef DEBUG_PUSH
xmlGenericError(xmlGenericErrorContext,
"PP: entering EOF\n");
* The XML REC instructs us to stop parsing right
* here
*/
- ctxt->instate = XML_PARSER_EOF;
+ xmlHaltParser(ctxt);
return(0);
}
ctxt->standalone = ctxt->input->standalone;
cur = ctxt->input->cur[0];
if (cur != '<') {
xmlFatalErr(ctxt, XML_ERR_DOCUMENT_EMPTY, NULL);
- ctxt->instate = XML_PARSER_EOF;
+ xmlHaltParser(ctxt);
if ((ctxt->sax) && (ctxt->sax->endDocument != NULL))
ctxt->sax->endDocument(ctxt->userData);
goto done;
goto done;
if (name == NULL) {
spacePop(ctxt);
- ctxt->instate = XML_PARSER_EOF;
+ xmlHaltParser(ctxt);
if ((ctxt->sax) && (ctxt->sax->endDocument != NULL))
ctxt->sax->endDocument(ctxt->userData);
goto done;
if ((cons == ctxt->input->consumed) && (test == CUR_PTR)) {
xmlFatalErr(ctxt, XML_ERR_INTERNAL_ERROR,
"detected an error in element content\n");
- ctxt->instate = XML_PARSER_EOF;
+ xmlHaltParser(ctxt);
break;
}
break;
int tmp;
tmp = xmlCheckCdataPush(ctxt->input->cur,
- XML_PARSER_BIG_BUFFER_SIZE);
+ XML_PARSER_BIG_BUFFER_SIZE, 0);
if (tmp < 0) {
tmp = -tmp;
ctxt->input->cur += tmp;
} else {
int tmp;
- tmp = xmlCheckCdataPush(ctxt->input->cur, base);
+ tmp = xmlCheckCdataPush(ctxt->input->cur, base, 1);
if ((tmp < 0) || (tmp != base)) {
tmp = -tmp;
ctxt->input->cur += tmp;
goto done;
} else {
xmlFatalErr(ctxt, XML_ERR_DOCUMENT_END, NULL);
- ctxt->instate = XML_PARSER_EOF;
+ xmlHaltParser(ctxt);
#ifdef DEBUG_PUSH
xmlGenericError(xmlGenericErrorContext,
"PP: entering EOF\n");
res = xmlParserInputBufferPush(ctxt->input->buf, size, chunk);
if (res < 0) {
ctxt->errNo = XML_PARSER_EOF;
- ctxt->disableSAX = 1;
+ xmlHaltParser(ctxt);
return (XML_PARSER_EOF);
}
xmlBufSetInputBaseCur(ctxt->input->buf->buffer, ctxt->input, base, cur);
((ctxt->input->cur - ctxt->input->base) > XML_MAX_LOOKUP_LIMIT)) &&
((ctxt->options & XML_PARSE_HUGE) == 0)) {
xmlFatalErr(ctxt, XML_ERR_INTERNAL_ERROR, "Huge input lookup");
- ctxt->instate = XML_PARSER_EOF;
+ xmlHaltParser(ctxt);
}
if ((ctxt->errNo != XML_ERR_OK) && (ctxt->disableSAX == 1))
return(ctxt->errNo);
#endif /* LIBXML_PUSH_ENABLED */
/**
- * xmlStopParser:
+ * xmlHaltParser:
* @ctxt: an XML parser context
*
- * Blocks further parser processing
+ * Blocks further parser processing don't override error
+ * for internal use
*/
-void
-xmlStopParser(xmlParserCtxtPtr ctxt) {
+static void
+xmlHaltParser(xmlParserCtxtPtr ctxt) {
if (ctxt == NULL)
return;
ctxt->instate = XML_PARSER_EOF;
- ctxt->errNo = XML_ERR_USER_STOP;
ctxt->disableSAX = 1;
if (ctxt->input != NULL) {
+ /*
+ * in case there was a specific allocation deallocate before
+ * overriding base
+ */
+ if (ctxt->input->free != NULL) {
+ ctxt->input->free((xmlChar *) ctxt->input->base);
+ ctxt->input->free = NULL;
+ }
ctxt->input->cur = BAD_CAST"";
ctxt->input->base = ctxt->input->cur;
}
}
/**
+ * xmlStopParser:
+ * @ctxt: an XML parser context
+ *
+ * Blocks further parser processing
+ */
+void
+xmlStopParser(xmlParserCtxtPtr ctxt) {
+ if (ctxt == NULL)
+ return;
+ xmlHaltParser(ctxt);
+ ctxt->errNo = XML_ERR_USER_STOP;
+}
+
+/**
* xmlCreateIOParserCtxt:
* @sax: a SAX handler
* @user_data: The user data returned on SAX callbacks
return(NULL);
}
+ /* We are loading a DTD */
+ ctxt->options |= XML_PARSE_DTDLOAD;
+
/*
* Set-up the SAX context
*/
return(NULL);
}
+ /* We are loading a DTD */
+ ctxt->options |= XML_PARSE_DTDLOAD;
+
/*
* Set-up the SAX context
*/
/*
* Also record the size of the entity parsed
*/
- if (ctxt->input != NULL) {
+ if (ctxt->input != NULL && oldctxt != NULL) {
oldctxt->sizeentities += ctxt->input->consumed;
oldctxt->sizeentities += (ctxt->input->cur - ctxt->input->base);
}
if (sax != NULL)
ctxt->sax = oldsax;
- oldctxt->node_seq.maximum = ctxt->node_seq.maximum;
- oldctxt->node_seq.length = ctxt->node_seq.length;
- oldctxt->node_seq.buffer = ctxt->node_seq.buffer;
+ if (oldctxt != NULL) {
+ oldctxt->node_seq.maximum = ctxt->node_seq.maximum;
+ oldctxt->node_seq.length = ctxt->node_seq.length;
+ oldctxt->node_seq.buffer = ctxt->node_seq.buffer;
+ }
ctxt->node_seq.maximum = 0;
ctxt->node_seq.length = 0;
ctxt->node_seq.buffer = NULL;
xmlFree((xmlChar *) ctxt->encoding);
ctxt->encoding = xmlStrdup((const xmlChar *) doc->encoding);
- hdlr = xmlFindCharEncodingHandler(doc->encoding);
+ hdlr = xmlFindCharEncodingHandler((const char *) doc->encoding);
if (hdlr != NULL) {
xmlSwitchToEncoding(ctxt, hdlr);
} else {
xmlCtxtUseOptionsInternal(ctxt, options, NULL);
xmlDetectSAX2(ctxt);
ctxt->myDoc = doc;
+ /* parsing in context, i.e. as within existing content */
+ ctxt->instate = XML_PARSER_CONTENT;
fake = xmlNewComment(NULL);
if (fake == NULL) {
}
cur = cur->parent;
}
- ctxt->instate = XML_PARSER_CONTENT;
}
if ((ctxt->validate) || (ctxt->replaceEntities != 0)) {
xmlSchemaCleanupTypes();
xmlRelaxNGCleanupTypes();
#endif
- xmlCleanupGlobals();
xmlResetLastError();
+ xmlCleanupGlobals();
xmlCleanupThreads(); /* must be last if called not from the main thread */
xmlCleanupMemory();
xmlParserInitialized = 0;
* DICT_FREE:
* @str: a string
*
- * Free a string if it is not owned by the "dict" dictionnary in the
+ * Free a string if it is not owned by the "dict" dictionary in the
* current scope
*/
#define DICT_FREE(str) \
if (cur == NULL)
return (NULL);
+ xmlInitParser();
ctxt = xmlCreateDocParserCtxt(cur);
if (ctxt == NULL)
{
xmlParserCtxtPtr ctxt;
+ xmlInitParser();
ctxt = xmlCreateURLParserCtxt(filename, options);
if (ctxt == NULL)
return (NULL);
{
xmlParserCtxtPtr ctxt;
+ xmlInitParser();
ctxt = xmlCreateMemoryParserCtxt(buffer, size);
if (ctxt == NULL)
return (NULL);
if (fd < 0)
return (NULL);
+ xmlInitParser();
input = xmlParserInputBufferCreateFd(fd, XML_CHAR_ENCODING_NONE);
if (input == NULL)
if (ioread == NULL)
return (NULL);
+ xmlInitParser();
input = xmlParserInputBufferCreateIO(ioread, ioclose, ioctx,
XML_CHAR_ENCODING_NONE);
return (NULL);
if (ctxt == NULL)
return (NULL);
+ xmlInitParser();
xmlCtxtReset(ctxt);
return (NULL);
if (ctxt == NULL)
return (NULL);
+ xmlInitParser();
xmlCtxtReset(ctxt);
return (NULL);
if (buffer == NULL)
return (NULL);
+ xmlInitParser();
xmlCtxtReset(ctxt);
return (NULL);
if (ctxt == NULL)
return (NULL);
+ xmlInitParser();
xmlCtxtReset(ctxt);
return (NULL);
if (ctxt == NULL)
return (NULL);
+ xmlInitParser();
xmlCtxtReset(ctxt);