blob: 363e3d70563f4f125c1e57e6f955fc44b4deab36 [file] [log] [blame]
/*
* parser.c : an XML 1.0 non-verifying parser
*
* See Copyright for the status of this software.
*
* Daniel.Veillard@w3.org
*/
#ifdef WIN32
#include "win32config.h"
#else
#include "config.h"
#endif
#include <stdio.h>
#include <string.h> /* for memset() only */
#ifdef HAVE_CTYPE_H
#include <ctype.h>
#endif
#ifdef HAVE_STDLIB_H
#include <stdlib.h>
#endif
#ifdef HAVE_SYS_STAT_H
#include <sys/stat.h>
#endif
#ifdef HAVE_FCNTL_H
#include <fcntl.h>
#endif
#ifdef HAVE_UNISTD_H
#include <unistd.h>
#endif
#ifdef HAVE_ZLIB_H
#include <zlib.h>
#endif
#include "xmlmemory.h"
#include "tree.h"
#include "parser.h"
#include "entities.h"
#include "encoding.h"
#include "valid.h"
#include "parserInternals.h"
#include "xmlIO.h"
#include "xml-error.h"
#include "uri.h"
#define XML_PARSER_BIG_BUFFER_SIZE 1000
#define XML_PARSER_BUFFER_SIZE 100
const char *xmlParserVersion = LIBXML_VERSION;
static int xmlUseNewParserDefault = 0;
/* a few of the old parser 1.8.11 entry points needed */
int xmlOldParseChunk (xmlParserCtxtPtr ctxt, const char *chunk, int size,
int terminate);
int xmlOldParseDocument(xmlParserCtxtPtr ctxt);
void xmlOldParseExternalSubset(xmlParserCtxtPtr ctxt,
const xmlChar *ExternalID, const xmlChar *SystemID);
/*
* List of XML prefixed PI allowed by W3C specs
*/
const char *xmlW3CPIs[] = {
"xml-stylesheet",
NULL
};
/*
* xmlCheckVersion:
* @version: the include version number
*
* check the compiled lib version against the include one.
* This can warn or immediately kill the application
*/
void
xmlCheckVersion(int version) {
int myversion = LIBXML_VERSION_NUMBER;
if ((myversion / 10000) != (version / 10000)) {
fprintf(stderr,
"Fatal: program compiled against libxml %d using libxml %d\n",
(version / 10000), (myversion / 10000));
exit(1);
}
if ((myversion / 100) < (version / 100)) {
fprintf(stderr,
"Warning: program compiled against libxml %d using older %d\n",
(version / 100), (myversion / 100));
}
}
/************************************************************************
* *
* Input handling functions for progressive parsing *
* *
************************************************************************/
/* #define DEBUG_INPUT */
/* #define DEBUG_STACK */
/* #define DEBUG_PUSH */
#define INPUT_CHUNK 250
/* we need to keep enough input to show errors in context */
#define LINE_LEN 80
#ifdef DEBUG_INPUT
#define CHECK_BUFFER(in) check_buffer(in)
void check_buffer(xmlParserInputPtr in) {
if (in->base != in->buf->buffer->content) {
fprintf(stderr, "xmlParserInput: base mismatch problem\n");
}
if (in->cur < in->base) {
fprintf(stderr, "xmlParserInput: cur < base problem\n");
}
if (in->cur > in->base + in->buf->buffer->use) {
fprintf(stderr, "xmlParserInput: cur > base + use problem\n");
}
fprintf(stderr,"buffer %x : content %x, cur %d, use %d, size %d\n",
(int) in, (int) in->buf->buffer->content, in->cur - in->base,
in->buf->buffer->use, in->buf->buffer->size);
}
#else
#define CHECK_BUFFER(in)
#endif
/**
* xmlParserInputRead:
* @in: an XML parser input
* @len: an indicative size for the lookahead
*
* This function refresh the input for the parser. It doesn't try to
* preserve pointers to the input buffer, and discard already read data
*
* Returns the number of xmlChars read, or -1 in case of error, 0 indicate the
* end of this entity
*/
int
xmlParserInputRead(xmlParserInputPtr in, int len) {
int ret;
int used;
int index;
#ifdef DEBUG_INPUT
fprintf(stderr, "Read\n");
#endif
if (in->buf == NULL) return(-1);
if (in->base == NULL) return(-1);
if (in->cur == NULL) return(-1);
if (in->buf->buffer == NULL) return(-1);
CHECK_BUFFER(in);
used = in->cur - in->buf->buffer->content;
ret = xmlBufferShrink(in->buf->buffer, used);
if (ret > 0) {
in->cur -= ret;
in->consumed += ret;
}
ret = xmlParserInputBufferRead(in->buf, len);
if (in->base != in->buf->buffer->content) {
/*
* the buffer has been realloced
*/
index = in->cur - in->base;
in->base = in->buf->buffer->content;
in->cur = &in->buf->buffer->content[index];
}
in->end = &in->buf->buffer->content[in->buf->buffer->use];
CHECK_BUFFER(in);
return(ret);
}
/**
* xmlParserInputGrow:
* @in: an XML parser input
* @len: an indicative size for the lookahead
*
* This function increase the input for the parser. It tries to
* preserve pointers to the input buffer, and keep already read data
*
* Returns the number of xmlChars read, or -1 in case of error, 0 indicate the
* end of this entity
*/
int
xmlParserInputGrow(xmlParserInputPtr in, int len) {
int ret;
int index;
#ifdef DEBUG_INPUT
fprintf(stderr, "Grow\n");
#endif
if (in->buf == NULL) return(-1);
if (in->base == NULL) return(-1);
if (in->cur == NULL) return(-1);
if (in->buf->buffer == NULL) return(-1);
CHECK_BUFFER(in);
index = in->cur - in->base;
if (in->buf->buffer->use > index + INPUT_CHUNK) {
CHECK_BUFFER(in);
return(0);
}
if ((in->buf->httpIO != NULL) || (in->buf->ftpIO != NULL) ||
(in->buf->file != NULL) ||
#ifdef HAVE_ZLIB_H
(in->buf->gzfile != NULL) ||
#endif
(in->buf->fd >= 0))
ret = xmlParserInputBufferGrow(in->buf, len);
else
return(0);
/*
* NOTE : in->base may be a "dandling" i.e. freed pointer in this
* block, but we use it really as an integer to do some
* pointer arithmetic. Insure will raise it as a bug but in
* that specific case, that's not !
*/
if (in->base != in->buf->buffer->content) {
/*
* the buffer has been realloced
*/
index = in->cur - in->base;
in->base = in->buf->buffer->content;
in->cur = &in->buf->buffer->content[index];
}
in->end = &in->buf->buffer->content[in->buf->buffer->use];
CHECK_BUFFER(in);
return(ret);
}
/**
* xmlParserInputShrink:
* @in: an XML parser input
*
* This function removes used input for the parser.
*/
void
xmlParserInputShrink(xmlParserInputPtr in) {
int used;
int ret;
int index;
#ifdef DEBUG_INPUT
fprintf(stderr, "Shrink\n");
#endif
if (in->buf == NULL) return;
if (in->base == NULL) return;
if (in->cur == NULL) return;
if (in->buf->buffer == NULL) return;
CHECK_BUFFER(in);
used = in->cur - in->buf->buffer->content;
if (used > INPUT_CHUNK) {
ret = xmlBufferShrink(in->buf->buffer, used - LINE_LEN);
if (ret > 0) {
in->cur -= ret;
in->consumed += ret;
}
in->end = &in->buf->buffer->content[in->buf->buffer->use];
}
CHECK_BUFFER(in);
if (in->buf->buffer->use > INPUT_CHUNK) {
return;
}
xmlParserInputBufferRead(in->buf, 2 * INPUT_CHUNK);
if (in->base != in->buf->buffer->content) {
/*
* the buffer has been realloced
*/
index = in->cur - in->base;
in->base = in->buf->buffer->content;
in->cur = &in->buf->buffer->content[index];
}
in->end = &in->buf->buffer->content[in->buf->buffer->use];
CHECK_BUFFER(in);
}
/* Needed from 2.3.5 */
void xmlSetListDoc(xmlNodePtr list, xmlDocPtr doc);
/**
* xmlSetTreeDoc:
* @tree: the top element
* @doc: the document
*
* update all nodes under the tree to point to the right document
*/
void
xmlSetTreeDoc(xmlNodePtr tree, xmlDocPtr doc) {
if (tree == NULL)
return;
if (tree->doc != doc) {
if (tree->childs != NULL)
xmlSetListDoc(tree->childs, doc);
tree->doc = doc;
}
}
/**
* xmlSetListDoc:
* @tree: the first element
* @doc: the document
*
* update all nodes in the list to point to the right document
*/
void
xmlSetListDoc(xmlNodePtr list, xmlDocPtr doc) {
xmlNodePtr cur;
if (list == NULL)
return;
cur = list;
while (cur != NULL) {
if (cur->doc != doc)
xmlSetTreeDoc(cur, doc);
cur = cur->next;
}
}
/**
* xmlAddChildList:
* @parent: the parent node
* @cur: the first node in the list
*
* Add a list of node at the end of the child list of the parent
* merging adjacent TEXT nodes (@cur may be freed)
*
* Returns the last child or NULL in case of error.
*/
xmlNodePtr
xmlAddChildList(xmlNodePtr parent, xmlNodePtr cur) {
xmlNodePtr prev;
if (parent == NULL) {
#ifdef DEBUG_TREE
xmlGenericError(xmlGenericErrorContext,
"xmlAddChild : parent == NULL\n");
#endif
return(NULL);
}
if (cur == NULL) {
#ifdef DEBUG_TREE
xmlGenericError(xmlGenericErrorContext,
"xmlAddChild : child == NULL\n");
#endif
return(NULL);
}
if ((cur->doc != NULL) && (parent->doc != NULL) &&
(cur->doc != parent->doc)) {
#ifdef DEBUG_TREE
xmlGenericError(xmlGenericErrorContext,
"Elements moved to a different document\n");
#endif
}
/*
* add the first element at the end of the children list.
*/
if (parent->childs == NULL) {
parent->childs = cur;
} else {
/*
* If cur and parent->last both are TEXT nodes, then merge them.
*/
if ((cur->type == XML_TEXT_NODE) &&
(parent->last->type == XML_TEXT_NODE) &&
(cur->name == parent->last->name)) {
#ifndef XML_USE_BUFFER_CONTENT
xmlNodeAddContent(parent->last, cur->content);
#else
xmlNodeAddContent(parent->last, xmlBufferContent(cur->content));
#endif
/*
* if it's the only child, nothing more to be done.
*/
if (cur->next == NULL) {
xmlFreeNode(cur);
return(parent->last);
}
prev = cur;
cur = cur->next;
xmlFreeNode(prev);
}
prev = parent->last;
prev->next = cur;
cur->prev = prev;
}
while (cur->next != NULL) {
cur->parent = parent;
if (cur->doc != parent->doc) {
xmlSetTreeDoc(cur, parent->doc);
}
cur = cur->next;
}
cur->parent = parent;
cur->doc = parent->doc; /* the parent may not be linked to a doc ! */
parent->last = cur;
return(cur);
}
/************************************************************************
* *
* Parser stacks related functions and macros *
* *
************************************************************************/
/* 2.3.5 portatibily */
#define xmlGenericErrorContext ctxt->userData
#define xmlGenericError \
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) \
ctxt->sax->error
int xmlSubstituteEntitiesDefaultValue = 0;
int xmlDoValidityCheckingDefaultValue = 0;
int xmlKeepBlanksDefaultValue = 0;
xmlEntityPtr xmlParseStringEntityRef(xmlParserCtxtPtr ctxt,
const xmlChar ** str);
/*
* Generic function for accessing stacks in the Parser Context
*/
#define PUSH_AND_POP(scope, type, name) \
scope int name##Push(xmlParserCtxtPtr ctxt, type value) { \
if (ctxt->name##Nr >= ctxt->name##Max) { \
ctxt->name##Max *= 2; \
ctxt->name##Tab = (void *) xmlRealloc(ctxt->name##Tab, \
ctxt->name##Max * sizeof(ctxt->name##Tab[0])); \
if (ctxt->name##Tab == NULL) { \
fprintf(stderr, "realloc failed !\n"); \
return(0); \
} \
} \
ctxt->name##Tab[ctxt->name##Nr] = value; \
ctxt->name = value; \
return(ctxt->name##Nr++); \
} \
scope type name##Pop(xmlParserCtxtPtr ctxt) { \
type ret; \
if (ctxt->name##Nr <= 0) return(0); \
ctxt->name##Nr--; \
if (ctxt->name##Nr > 0) \
ctxt->name = ctxt->name##Tab[ctxt->name##Nr - 1]; \
else \
ctxt->name = NULL; \
ret = ctxt->name##Tab[ctxt->name##Nr]; \
ctxt->name##Tab[ctxt->name##Nr] = 0; \
return(ret); \
} \
PUSH_AND_POP(extern, xmlParserInputPtr, input)
PUSH_AND_POP(extern, xmlNodePtr, node)
PUSH_AND_POP(extern, xmlChar*, name)
int spacePush(xmlParserCtxtPtr ctxt, int val) {
if (ctxt->spaceTab == NULL) {
ctxt->spaceTab = (int *) xmlMalloc(10 * sizeof(int));
ctxt->spaceNr = 1;
ctxt->spaceMax = 10;
ctxt->spaceTab[0] = -1;
ctxt->space = &ctxt->spaceTab[0];
if (ctxt->spaceTab == NULL) {
xmlGenericError(xmlGenericErrorContext,
"malloc failed !\n");
return(0);
}
}
if (ctxt->spaceNr >= ctxt->spaceMax) {
ctxt->spaceMax *= 2;
ctxt->spaceTab = (int *) xmlRealloc(ctxt->spaceTab,
ctxt->spaceMax * sizeof(ctxt->spaceTab[0]));
if (ctxt->spaceTab == NULL) {
xmlGenericError(xmlGenericErrorContext,
"realloc failed !\n");
return(0);
}
}
ctxt->spaceTab[ctxt->spaceNr] = val;
ctxt->space = &ctxt->spaceTab[ctxt->spaceNr];
return(ctxt->spaceNr++);
}
int spacePop(xmlParserCtxtPtr ctxt) {
int ret;
if (ctxt->spaceNr <= 0) return(0);
ctxt->spaceNr--;
if (ctxt->spaceNr > 0)
ctxt->space = &ctxt->spaceTab[ctxt->spaceNr - 1];
else
ctxt->space = NULL;
ret = ctxt->spaceTab[ctxt->spaceNr];
ctxt->spaceTab[ctxt->spaceNr] = -1;
return(ret);
}
/************************************************************************
* *
* Macros grabbed from 2.3.5 *
* *
************************************************************************/
void xmlParserHandlePEReference(xmlParserCtxtPtr ctxt);
xmlEntityPtr xmlParseStringPEReference(xmlParserCtxtPtr ctxt, const xmlChar **str);
void xmlParseTextDecl(xmlParserCtxtPtr ctxt);
void xmlNextChar(xmlParserCtxtPtr ctxt);
/*
* Macros for accessing the content. Those should be used only by the parser,
* and not exported.
*
* Dirty macros, i.e. one often need to make assumption on the context to
* use them
*
* CUR_PTR return the current pointer to the xmlChar to be parsed.
* To be used with extreme caution since operations consuming
* characters may move the input buffer to a different location !
* CUR returns the current xmlChar value, i.e. a 8 bit value if compiled
* This should be used internally by the parser
* only to compare to ASCII values otherwise it would break when
* running with UTF-8 encoding.
* RAW same as CUR but in the input buffer, bypass any token
* extraction that may have been done
* NXT(n) returns the n'th next xmlChar. Same as CUR is should be used only
* to compare on ASCII based substring.
* SKIP(n) Skip n xmlChar, and must also be used only to skip ASCII defined
* strings within the parser.
*
* Clean macros, not dependent of an ASCII context, expect UTF-8 encoding
*
* NEXT Skip to the next character, this does the proper decoding
* in UTF-8 mode. It also pop-up unfinished entities on the fly.
* NEXTL(l) Skip l xmlChars in the input buffer
* CUR_CHAR(l) returns the current unicode character (int), set l
* to the number of xmlChars used for the encoding [0-5].
* CUR_SCHAR same but operate on a string instead of the context
* COPY_BUF copy the current unicode char to the target buffer, increment
* the index
* GROW, SHRINK handling of input buffers
*/
#define RAW (ctxt->token ? -1 : (*ctxt->input->cur))
#define CUR (ctxt->token ? ctxt->token : (*ctxt->input->cur))
#define NXT(val) ctxt->input->cur[(val)]
#define CUR_PTR ctxt->input->cur
#define SKIP(val) do { \
ctxt->nbChars += (val),ctxt->input->cur += (val); \
if (*ctxt->input->cur == '%') xmlParserHandlePEReference(ctxt); \
if ((*ctxt->input->cur == 0) && \
(xmlParserInputGrow(ctxt->input, INPUT_CHUNK) <= 0)) \
xmlPopInput(ctxt); \
} while (0)
#define SHRINK if (ctxt->input->cur - ctxt->input->base > INPUT_CHUNK) {\
xmlParserInputShrink(ctxt->input); \
if ((*ctxt->input->cur == 0) && \
(xmlParserInputGrow(ctxt->input, INPUT_CHUNK) <= 0)) \
xmlPopInput(ctxt); \
}
#define GROW if (ctxt->input->end - ctxt->input->cur < INPUT_CHUNK) { \
xmlParserInputGrow(ctxt->input, INPUT_CHUNK); \
if ((*ctxt->input->cur == 0) && \
(xmlParserInputGrow(ctxt->input, INPUT_CHUNK) <= 0)) \
xmlPopInput(ctxt); \
}
#define SKIP_BLANKS xmlSkipBlankChars(ctxt)
#define NEXT xmlNextChar(ctxt)
#define NEXT1 { \
ctxt->input->cur++; \
ctxt->nbChars++; \
if (*ctxt->input->cur == 0) \
xmlParserInputGrow(ctxt->input, INPUT_CHUNK); \
}
#define NEXTL(l) do { \
if (*(ctxt->input->cur) == '\n') { \
ctxt->input->line++; ctxt->input->col = 1; \
} else ctxt->input->col++; \
ctxt->token = 0; ctxt->input->cur += l; \
if (*ctxt->input->cur == '%') xmlParserHandlePEReference(ctxt); \
} while (0)
#define CUR_CHAR(l) xmlCurrentChar(ctxt, &l)
#define CUR_SCHAR(s, l) xmlStringCurrentChar(ctxt, s, &l)
#define COPY_BUF(l,b,i,v) \
i += xmlCopyChar(l,&b[i],v)
/**
* xmlSkipBlankChars:
* @ctxt: the XML parser context
*
* skip all blanks character found at that point in the input streams.
* It pops up finished entities in the process if allowable at that point.
*
* Returns the number of space chars skipped
*/
int
xmlSkipBlankChars(xmlParserCtxtPtr ctxt) {
int cur, res = 0;
/*
* It's Okay to use CUR/NEXT here since all the blanks are on
* the ASCII range.
*/
do {
cur = CUR;
while (IS_BLANK(cur)) { /* CHECKED tstblanks.xml */
NEXT;
cur = CUR;
res++;
}
while ((cur == 0) && (ctxt->inputNr > 1) &&
(ctxt->instate != XML_PARSER_COMMENT)) {
xmlPopInput(ctxt);
cur = CUR;
}
/*
* Need to handle support of entities branching here
*/
if (*ctxt->input->cur == '%') xmlParserHandlePEReference(ctxt);
/* DEPR if (*ctxt->input->cur == '&') xmlParserHandleReference(ctxt); */
} while (IS_BLANK(cur)); /* CHECKED tstblanks.xml */
return(res);
}
static int xmlParserInitialized = 0;
/**
* xmlInitParser:
*
* Initialization function for the XML parser.
* This is not reentrant. Call once before processing in case of
* use in multithreaded programs.
*/
void
xmlInitParser(void) {
if (xmlParserInitialized) return;
/* make this user selectable */
xmlUseNewParserDefault = 0;
if (getenv("LIBXML_USE_NEW_PARSER"))
xmlUseNewParserDefault = 1;
else if (getenv("LIBXML_USE_OLD_PARSER"))
xmlUseNewParserDefault = 0;
xmlInitCharEncodingHandlers();
xmlInitializePredefinedEntities();
xmlDefaultSAXHandlerInit();
#ifdef LIBXML_HTML_ENABLED
htmlInitAutoClose();
htmlDefaultSAXHandlerInit();
#endif
#ifdef LIBXML_XPATH_ENABLED
xmlXPathInit();
#endif
xmlParserInitialized = 1;
}
/************************************************************************
* *
* UTF8 character input and related functions *
* *
************************************************************************/
/**
* xmlNextChar:
* @ctxt: the XML parser context
*
* Skip to the next char input char.
*/
void
xmlNextChar(xmlParserCtxtPtr ctxt) {
if (ctxt->instate == XML_PARSER_EOF) {
ctxt->token = -1;
return;
}
/*
* 2.11 End-of-Line Handling
* the literal two-character sequence "#xD#xA" or a standalone
* literal #xD, an XML processor must pass to the application
* the single character #xA.
*/
if (ctxt->token != 0) ctxt->token = 0;
else if (ctxt->charset == XML_CHAR_ENCODING_UTF8) {
if ((*ctxt->input->cur == 0) &&
(xmlParserInputGrow(ctxt->input, INPUT_CHUNK) <= 0) &&
(ctxt->instate != XML_PARSER_COMMENT)) {
/*
* If we are at the end of the current entity and
* the context allows it, we pop consumed entities
* automatically.
* the auto closing should be blocked in other cases
*/
xmlPopInput(ctxt);
} else {
if (*(ctxt->input->cur) == '\n') {
ctxt->input->line++; ctxt->input->col = 1;
} else ctxt->input->col++;
if (ctxt->charset == XML_CHAR_ENCODING_UTF8) {
/*
* We are supposed to handle UTF8, check it's valid
* From rfc2044: encoding of the Unicode values on UTF-8:
*
* UCS-4 range (hex.) UTF-8 octet sequence (binary)
* 0000 0000-0000 007F 0xxxxxxx
* 0000 0080-0000 07FF 110xxxxx 10xxxxxx
* 0000 0800-0000 FFFF 1110xxxx 10xxxxxx 10xxxxxx
*
* Check for the 0x110000 limit too
*/
const unsigned char *cur = ctxt->input->cur;
unsigned char c;
c = *cur;
if (c & 0x80) {
if (cur[1] == 0)
xmlParserInputGrow(ctxt->input, INPUT_CHUNK);
if ((cur[1] & 0xc0) != 0x80)
goto encoding_error;
if ((c & 0xe0) == 0xe0) {
unsigned int val;
if (cur[2] == 0)
xmlParserInputGrow(ctxt->input, INPUT_CHUNK);
if ((cur[2] & 0xc0) != 0x80)
goto encoding_error;
if ((c & 0xf0) == 0xf0) {
if (cur[3] == 0)
xmlParserInputGrow(ctxt->input, INPUT_CHUNK);
if (((c & 0xf8) != 0xf0) ||
((cur[3] & 0xc0) != 0x80))
goto encoding_error;
/* 4-byte code */
ctxt->input->cur += 4;
val = (cur[0] & 0x7) << 18;
val |= (cur[1] & 0x3f) << 12;
val |= (cur[2] & 0x3f) << 6;
val |= cur[3] & 0x3f;
} else {
/* 3-byte code */
ctxt->input->cur += 3;
val = (cur[0] & 0xf) << 12;
val |= (cur[1] & 0x3f) << 6;
val |= cur[2] & 0x3f;
}
if (((val > 0xd7ff) && (val < 0xe000)) ||
((val > 0xfffd) && (val < 0x10000)) ||
(val >= 0x110000)) {
if ((ctxt->sax != NULL) &&
(ctxt->sax->error != NULL))
ctxt->sax->error(ctxt->userData,
"Char 0x%X out of allowed range\n", val);
goto encoding_error;
}
} else
/* 2-byte code */
ctxt->input->cur += 2;
} else
/* 1-byte code */
ctxt->input->cur++;
} else {
/*
* Assume it's a fixed lenght encoding (1) with
* a compatibke encoding for the ASCII set, since
* XML constructs only use < 128 chars
*/
ctxt->input->cur++;
}
ctxt->nbChars++;
if (*ctxt->input->cur == 0)
xmlParserInputGrow(ctxt->input, INPUT_CHUNK);
}
} else {
ctxt->input->cur++;
ctxt->nbChars++;
if (*ctxt->input->cur == 0)
xmlParserInputGrow(ctxt->input, INPUT_CHUNK);
}
if ((*ctxt->input->cur == '%') && (!ctxt->html))
xmlParserHandlePEReference(ctxt);
if ((*ctxt->input->cur == 0) &&
(xmlParserInputGrow(ctxt->input, INPUT_CHUNK) <= 0))
xmlPopInput(ctxt);
return;
encoding_error:
/*
* If we detect an UTF8 error that probably mean that the
* input encoding didn't get properly advertized in the
* declaration header. Report the error and switch the encoding
* to ISO-Latin-1 (if you don't like this policy, just declare the
* encoding !)
*/
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) {
ctxt->sax->error(ctxt->userData,
"Input is not proper UTF-8, indicate encoding !\n");
ctxt->sax->error(ctxt->userData, "Bytes: 0x%02X 0x%02X 0x%02X 0x%02X\n",
ctxt->input->cur[0], ctxt->input->cur[1],
ctxt->input->cur[2], ctxt->input->cur[3]);
}
ctxt->charset = XML_CHAR_ENCODING_8859_1;
ctxt->input->cur++;
return;
}
/**
* xmlCurrentChar:
* @ctxt: the XML parser context
* @len: pointer to the length of the char read
*
* The current char value, if using UTF-8 this may actaully span multiple
* bytes in the input buffer. Implement the end of line normalization:
* 2.11 End-of-Line Handling
* Wherever an external parsed entity or the literal entity value
* of an internal parsed entity contains either the literal two-character
* sequence "#xD#xA" or a standalone literal #xD, an XML processor
* must pass to the application the single character #xA.
* This behavior can conveniently be produced by normalizing all
* line breaks to #xA on input, before parsing.)
*
* Returns the current char value and its lenght
*/
int
xmlCurrentChar(xmlParserCtxtPtr ctxt, int *len) {
if (ctxt->instate == XML_PARSER_EOF)
return(0);
if (ctxt->token != 0) {
*len = 0;
return(ctxt->token);
}
if ((*ctxt->input->cur >= 0x20) && (*ctxt->input->cur <= 0x7F)) {
*len = 1;
return((int) *ctxt->input->cur);
}
if (ctxt->charset == XML_CHAR_ENCODING_UTF8) {
/*
* We are supposed to handle UTF8, check it's valid
* From rfc2044: encoding of the Unicode values on UTF-8:
*
* UCS-4 range (hex.) UTF-8 octet sequence (binary)
* 0000 0000-0000 007F 0xxxxxxx
* 0000 0080-0000 07FF 110xxxxx 10xxxxxx
* 0000 0800-0000 FFFF 1110xxxx 10xxxxxx 10xxxxxx
*
* Check for the 0x110000 limit too
*/
const unsigned char *cur = ctxt->input->cur;
unsigned char c;
unsigned int val;
c = *cur;
if (c & 0x80) {
if (cur[1] == 0)
xmlParserInputGrow(ctxt->input, INPUT_CHUNK);
if ((cur[1] & 0xc0) != 0x80)
goto encoding_error;
if ((c & 0xe0) == 0xe0) {
if (cur[2] == 0)
xmlParserInputGrow(ctxt->input, INPUT_CHUNK);
if ((cur[2] & 0xc0) != 0x80)
goto encoding_error;
if ((c & 0xf0) == 0xf0) {
if (cur[3] == 0)
xmlParserInputGrow(ctxt->input, INPUT_CHUNK);
if (((c & 0xf8) != 0xf0) ||
((cur[3] & 0xc0) != 0x80))
goto encoding_error;
/* 4-byte code */
*len = 4;
val = (cur[0] & 0x7) << 18;
val |= (cur[1] & 0x3f) << 12;
val |= (cur[2] & 0x3f) << 6;
val |= cur[3] & 0x3f;
} else {
/* 3-byte code */
*len = 3;
val = (cur[0] & 0xf) << 12;
val |= (cur[1] & 0x3f) << 6;
val |= cur[2] & 0x3f;
}
} else {
/* 2-byte code */
*len = 2;
val = (cur[0] & 0x1f) << 6;
val |= cur[1] & 0x3f;
}
if (!IS_CHAR(val)) {
if ((ctxt->sax != NULL) &&
(ctxt->sax->error != NULL))
ctxt->sax->error(ctxt->userData,
"Char 0x%X out of allowed range\n", val);
goto encoding_error;
}
return(val);
} else {
/* 1-byte code */
*len = 1;
if (*ctxt->input->cur == 0xD) {
if (ctxt->input->cur[1] == 0xA) {
ctxt->nbChars++;
ctxt->input->cur++;
}
return(0xA);
}
return((int) *ctxt->input->cur);
}
}
/*
* Assume it's a fixed lenght encoding (1) with
* a compatibke encoding for the ASCII set, since
* XML constructs only use < 128 chars
*/
*len = 1;
if (*ctxt->input->cur == 0xD) {
if (ctxt->input->cur[1] == 0xA) {
ctxt->nbChars++;
ctxt->input->cur++;
}
return(0xA);
}
return((int) * (unsigned char *)ctxt->input->cur);
encoding_error:
/*
* If we detect an UTF8 error that probably mean that the
* input encoding didn't get properly advertized in the
* declaration header. Report the error and switch the encoding
* to ISO-Latin-1 (if you don't like this policy, just declare the
* encoding !)
*/
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) {
ctxt->sax->error(ctxt->userData,
"Input is not proper UTF-8, indicate encoding !\n");
ctxt->sax->error(ctxt->userData, "Bytes: 0x%02X 0x%02X 0x%02X 0x%02X\n",
ctxt->input->cur[0], ctxt->input->cur[1],
ctxt->input->cur[2], ctxt->input->cur[3]);
}
ctxt->charset = XML_CHAR_ENCODING_8859_1;
*len = 1;
return((int) *ctxt->input->cur);
}
/**
* xmlStringCurrentChar:
* @ctxt: the XML parser context
* @cur: pointer to the beginning of the char
* @len: pointer to the length of the char read
*
* The current char value, if using UTF-8 this may actaully span multiple
* bytes in the input buffer.
*
* Returns the current char value and its lenght
*/
int
xmlStringCurrentChar(xmlParserCtxtPtr ctxt, const xmlChar *cur, int *len) {
if (ctxt->charset == XML_CHAR_ENCODING_UTF8) {
/*
* We are supposed to handle UTF8, check it's valid
* From rfc2044: encoding of the Unicode values on UTF-8:
*
* UCS-4 range (hex.) UTF-8 octet sequence (binary)
* 0000 0000-0000 007F 0xxxxxxx
* 0000 0080-0000 07FF 110xxxxx 10xxxxxx
* 0000 0800-0000 FFFF 1110xxxx 10xxxxxx 10xxxxxx
*
* Check for the 0x110000 limit too
*/
unsigned char c;
unsigned int val;
c = *cur;
if (c & 0x80) {
if ((cur[1] & 0xc0) != 0x80)
goto encoding_error;
if ((c & 0xe0) == 0xe0) {
if ((cur[2] & 0xc0) != 0x80)
goto encoding_error;
if ((c & 0xf0) == 0xf0) {
if (((c & 0xf8) != 0xf0) ||
((cur[3] & 0xc0) != 0x80))
goto encoding_error;
/* 4-byte code */
*len = 4;
val = (cur[0] & 0x7) << 18;
val |= (cur[1] & 0x3f) << 12;
val |= (cur[2] & 0x3f) << 6;
val |= cur[3] & 0x3f;
} else {
/* 3-byte code */
*len = 3;
val = (cur[0] & 0xf) << 12;
val |= (cur[1] & 0x3f) << 6;
val |= cur[2] & 0x3f;
}
} else {
/* 2-byte code */
*len = 2;
val = (cur[0] & 0x1f) << 6;
val |= cur[2] & 0x3f;
}
if (!IS_CHAR(val)) {
if ((ctxt->sax != NULL) &&
(ctxt->sax->error != NULL))
ctxt->sax->error(ctxt->userData,
"Char 0x%X out of allowed range\n", val);
goto encoding_error;
}
return(val);
} else {
/* 1-byte code */
*len = 1;
return((int) *cur);
}
}
/*
* Assume it's a fixed lenght encoding (1) with
* a compatibke encoding for the ASCII set, since
* XML constructs only use < 128 chars
*/
*len = 1;
return((int) *cur);
encoding_error:
/*
* If we detect an UTF8 error that probably mean that the
* input encoding didn't get properly advertized in the
* declaration header. Report the error and switch the encoding
* to ISO-Latin-1 (if you don't like this policy, just declare the
* encoding !)
*/
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) {
ctxt->sax->error(ctxt->userData,
"Input is not proper UTF-8, indicate encoding !\n");
ctxt->sax->error(ctxt->userData, "Bytes: 0x%02X 0x%02X 0x%02X 0x%02X\n",
ctxt->input->cur[0], ctxt->input->cur[1],
ctxt->input->cur[2], ctxt->input->cur[3]);
}
ctxt->charset = XML_CHAR_ENCODING_8859_1;
*len = 1;
return((int) *cur);
}
/**
* xmlCopyCharMultiByte:
* @out: pointer to an arry of xmlChar
* @val: the char value
*
* append the char value in the array
*
* Returns the number of xmlChar written
*/
int
xmlCopyCharMultiByte(xmlChar *out, int val) {
/*
* We are supposed to handle UTF8, check it's valid
* From rfc2044: encoding of the Unicode values on UTF-8:
*
* UCS-4 range (hex.) UTF-8 octet sequence (binary)
* 0000 0000-0000 007F 0xxxxxxx
* 0000 0080-0000 07FF 110xxxxx 10xxxxxx
* 0000 0800-0000 FFFF 1110xxxx 10xxxxxx 10xxxxxx
*/
if (val >= 0x80) {
xmlChar *savedout = out;
int bits;
if (val < 0x800) { *out++= (val >> 6) | 0xC0; bits= 0; }
else if (val < 0x10000) { *out++= (val >> 12) | 0xE0; bits= 6;}
else if (val < 0x110000) { *out++= (val >> 18) | 0xF0; bits= 12; }
else {
fprintf(stderr,
"Internal error, xmlCopyChar 0x%X out of bound\n",
val);
return(0);
}
for ( ; bits >= 0; bits-= 6)
*out++= ((val >> bits) & 0x3F) | 0x80 ;
return (out - savedout);
}
*out = (xmlChar) val;
return 1;
}
/**
* xmlCopyChar:
* @len: Ignored, compatibility
* @out: pointer to an arry of xmlChar
* @val: the char value
*
* append the char value in the array
*
* Returns the number of xmlChar written
*/
int
xmlCopyChar(int len, xmlChar *out, int val) {
/* the len parameter is ignored */
if (val >= 0x80) {
return(xmlCopyCharMultiByte (out, val));
}
*out = (xmlChar) val;
return 1;
}
/************************************************************************
* *
* Commodity functions to handle entities processing *
* *
************************************************************************/
/**
* xmlFreeInputStream:
* @input: an xmlParserInputPtr
*
* Free up an input stream.
*/
void
xmlFreeInputStream(xmlParserInputPtr input) {
if (input == NULL) return;
if (input->filename != NULL) xmlFree((char *) input->filename);
if (input->directory != NULL) xmlFree((char *) input->directory);
if ((input->free != NULL) && (input->base != NULL))
input->free((xmlChar *) input->base);
if (input->buf != NULL)
xmlFreeParserInputBuffer(input->buf);
/* 2.3.5 */
if (input->encoding != NULL) xmlFree((char *) input->encoding);
if (input->version != NULL) xmlFree((char *) input->version);
memset(input, -1, sizeof(xmlParserInput));
xmlFree(input);
}
/**
* xmlNewInputStream:
* @ctxt: an XML parser context
*
* Create a new input stream structure
* Returns the new input stream or NULL
*/
xmlParserInputPtr
xmlNewInputStream(xmlParserCtxtPtr ctxt) {
xmlParserInputPtr input;
input = (xmlParserInputPtr) xmlMalloc(sizeof(xmlParserInput));
if (input == NULL) {
ctxt->errNo = XML_ERR_NO_MEMORY;
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt->userData,
"malloc: couldn't allocate a new input stream\n");
ctxt->errNo = XML_ERR_NO_MEMORY;
return(NULL);
}
input->filename = NULL;
input->directory = NULL;
input->base = NULL;
input->cur = NULL;
input->buf = NULL;
input->line = 1;
input->col = 1;
input->buf = NULL;
input->free = NULL;
input->consumed = 0;
input->length = 0;
/* 2.3.5 */
input->encoding = NULL;
input->version = NULL;
return(input);
}
/**
* xmlNewEntityInputStream:
* @ctxt: an XML parser context
* @entity: an Entity pointer
*
* Create a new input stream based on an xmlEntityPtr
*
* Returns the new input stream or NULL
*/
xmlParserInputPtr
xmlNewEntityInputStream(xmlParserCtxtPtr ctxt, xmlEntityPtr entity) {
xmlParserInputPtr input;
if (entity == NULL) {
ctxt->errNo = XML_ERR_INTERNAL_ERROR;
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt->userData,
"internal: xmlNewEntityInputStream entity = NULL\n");
ctxt->errNo = XML_ERR_INTERNAL_ERROR;
return(NULL);
}
if (entity->content == NULL) {
switch (entity->type) {
case XML_EXTERNAL_GENERAL_UNPARSED_ENTITY:
ctxt->errNo = XML_ERR_UNPARSED_ENTITY;
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt->userData,
"xmlNewEntityInputStream unparsed entity !\n");
break;
case XML_EXTERNAL_GENERAL_PARSED_ENTITY:
case XML_EXTERNAL_PARAMETER_ENTITY:
return(xmlLoadExternalEntity((char *) entity->SystemID,
(char *) entity->ExternalID, ctxt));
case XML_INTERNAL_GENERAL_ENTITY:
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt->userData,
"Internal entity %s without content !\n", entity->name);
break;
case XML_INTERNAL_PARAMETER_ENTITY:
ctxt->errNo = XML_ERR_INTERNAL_ERROR;
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt->userData,
"Internal parameter entity %s without content !\n", entity->name);
break;
case XML_INTERNAL_PREDEFINED_ENTITY:
ctxt->errNo = XML_ERR_INTERNAL_ERROR;
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt->userData,
"Predefined entity %s without content !\n", entity->name);
break;
}
return(NULL);
}
input = xmlNewInputStream(ctxt);
if (input == NULL) {
return(NULL);
}
input->filename = (char *) entity->SystemID; /* TODO !!! char <- xmlChar */
input->base = entity->content;
input->cur = entity->content;
input->length = entity->length;
input->end = &entity->content[input->length];
return(input);
}
/**
* xmlNewStringInputStream:
* @ctxt: an XML parser context
* @buffer: an memory buffer
*
* Create a new input stream based on a memory buffer.
* Returns the new input stream
*/
xmlParserInputPtr
xmlNewStringInputStream(xmlParserCtxtPtr ctxt, const xmlChar *buffer) {
xmlParserInputPtr input;
if (buffer == NULL) {
ctxt->errNo = XML_ERR_INTERNAL_ERROR;
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt->userData,
"internal: xmlNewStringInputStream string = NULL\n");
return(NULL);
}
input = xmlNewInputStream(ctxt);
if (input == NULL) {
return(NULL);
}
input->base = buffer;
input->cur = buffer;
input->length = xmlStrlen(buffer);
input->end = &buffer[input->length];
return(input);
}
/**
* xmlNewInputFromFile:
* @ctxt: an XML parser context
* @filename: the filename to use as entity
*
* Create a new input stream based on a file.
*
* Returns the new input stream or NULL in case of error
*/
xmlParserInputPtr
xmlNewInputFromFile(xmlParserCtxtPtr ctxt, const char *filename) {
xmlParserInputBufferPtr buf;
xmlParserInputPtr inputStream;
char *directory = NULL;
if (ctxt == NULL) return(NULL);
buf = xmlParserInputBufferCreateFilename(filename, XML_CHAR_ENCODING_NONE);
if (buf == NULL) {
char name[XML_PARSER_BIG_BUFFER_SIZE];
if ((ctxt->input != NULL) && (ctxt->input->directory != NULL)) {
#ifdef WIN32
sprintf(name, "%s\\%s", ctxt->input->directory, filename);
#else
sprintf(name, "%s/%s", ctxt->input->directory, filename);
#endif
buf = xmlParserInputBufferCreateFilename(name,
XML_CHAR_ENCODING_NONE);
if (buf != NULL)
directory = xmlParserGetDirectory(name);
}
if ((buf == NULL) && (ctxt->directory != NULL)) {
#ifdef WIN32
sprintf(name, "%s\\%s", ctxt->directory, filename);
#else
sprintf(name, "%s/%s", ctxt->directory, filename);
#endif
buf = xmlParserInputBufferCreateFilename(name,
XML_CHAR_ENCODING_NONE);
if (buf != NULL)
directory = xmlParserGetDirectory(name);
}
if (buf == NULL)
return(NULL);
}
if (directory == NULL)
directory = xmlParserGetDirectory(filename);
inputStream = xmlNewInputStream(ctxt);
if (inputStream == NULL) {
if (directory != NULL) xmlFree((char *) directory);
return(NULL);
}
inputStream->filename = xmlMemStrdup(filename);
inputStream->directory = directory;
inputStream->buf = buf;
inputStream->base = inputStream->buf->buffer->content;
inputStream->cur = inputStream->buf->buffer->content;
inputStream->end = &inputStream->base[inputStream->buf->buffer->use];
if ((ctxt->directory == NULL) && (directory != NULL))
ctxt->directory = (char *) xmlStrdup((const xmlChar *) directory);
return(inputStream);
}
/************************************************************************
* *
* Commodity functions to handle parser contexts *
* *
************************************************************************/
/**
* xmlInitParserCtxt:
* @ctxt: an XML parser context
*
* Initialize a parser context
*/
void
xmlInitParserCtxt(xmlParserCtxtPtr ctxt)
{
xmlSAXHandler *sax;
sax = (xmlSAXHandler *) xmlMalloc(sizeof(xmlSAXHandler));
if (sax == NULL) {
fprintf(stderr, "xmlInitParserCtxt: out of memory\n");
}
/* Allocate the Input stack */
ctxt->inputTab = (xmlParserInputPtr *) xmlMalloc(5 * sizeof(xmlParserInputPtr));
ctxt->inputNr = 0;
ctxt->inputMax = 5;
ctxt->input = NULL;
ctxt->version = NULL;
ctxt->encoding = NULL;
ctxt->standalone = -1;
ctxt->hasExternalSubset = 0;
ctxt->hasPErefs = 0;
ctxt->html = 0;
ctxt->external = 0;
ctxt->instate = XML_PARSER_START;
ctxt->token = 0;
ctxt->directory = NULL;
/* Allocate the Node stack */
ctxt->nodeTab = (xmlNodePtr *) xmlMalloc(10 * sizeof(xmlNodePtr));
ctxt->nodeNr = 0;
ctxt->nodeMax = 10;
ctxt->node = NULL;
/* Allocate the Name stack */
ctxt->nameTab = (xmlChar **) xmlMalloc(10 * sizeof(xmlChar *));
ctxt->nameNr = 0;
ctxt->nameMax = 10;
ctxt->name = NULL;
if (sax == NULL) ctxt->sax = &xmlDefaultSAXHandler;
else {
ctxt->sax = sax;
memcpy(sax, &xmlDefaultSAXHandler, sizeof(xmlSAXHandler));
}
ctxt->userData = ctxt;
ctxt->myDoc = NULL;
ctxt->wellFormed = 1;
ctxt->valid = 1;
ctxt->validate = xmlDoValidityCheckingDefaultValue;
ctxt->keepBlanks = xmlKeepBlanksDefaultValue;
ctxt->vctxt.userData = ctxt;
if (ctxt->validate) {
ctxt->vctxt.error = xmlParserValidityError;
ctxt->vctxt.warning = xmlParserValidityWarning;
} else {
ctxt->vctxt.error = NULL;
ctxt->vctxt.warning = NULL;
}
ctxt->replaceEntities = xmlSubstituteEntitiesDefaultValue;
ctxt->record_info = 0;
ctxt->nbChars = 0;
ctxt->checkIndex = 0;
ctxt->errNo = XML_ERR_OK;
xmlInitNodeInfoSeq(&ctxt->node_seq);
/* 2.3.5 */
ctxt->charset = XML_CHAR_ENCODING_UTF8;
ctxt->depth = 0;
ctxt->spaceTab = NULL;
ctxt->intSubName = NULL;
ctxt->extSubURI = NULL;
ctxt->extSubSystem = NULL;
}
/**
* xmlFreeParserCtxt:
* @ctxt: an XML parser context
*
* Free all the memory used by a parser context. However the parsed
* document in ctxt->myDoc is not freed.
*/
void
xmlFreeParserCtxt(xmlParserCtxtPtr ctxt)
{
xmlParserInputPtr input;
xmlChar *oldname;
if (ctxt == NULL) return;
while ((input = inputPop(ctxt)) != NULL) {
xmlFreeInputStream(input);
}
while ((oldname = namePop(ctxt)) != NULL) {
xmlFree(oldname);
}
if (ctxt->nameTab != NULL) xmlFree(ctxt->nameTab);
if (ctxt->nodeTab != NULL) xmlFree(ctxt->nodeTab);
if (ctxt->inputTab != NULL) xmlFree(ctxt->inputTab);
if (ctxt->version != NULL) xmlFree((char *) ctxt->version);
if (ctxt->encoding != NULL) xmlFree((char *) ctxt->encoding);
if ((ctxt->sax != NULL) && (ctxt->sax != &xmlDefaultSAXHandler))
xmlFree(ctxt->sax);
if (ctxt->directory != NULL) xmlFree((char *) ctxt->directory);
/* 2.3.5 */
if (ctxt->spaceTab != NULL) xmlFree(ctxt->spaceTab);
if (ctxt->intSubName != NULL) xmlFree(ctxt->intSubName);
if (ctxt->extSubURI != NULL) xmlFree(ctxt->extSubURI);
if (ctxt->extSubSystem != NULL) xmlFree(ctxt->extSubSystem);
xmlFree(ctxt);
}
/**
* xmlNewParserCtxt:
*
* Allocate and initialize a new parser context.
*
* Returns the xmlParserCtxtPtr or NULL
*/
xmlParserCtxtPtr
xmlNewParserCtxt()
{
xmlParserCtxtPtr ctxt;
ctxt = (xmlParserCtxtPtr) xmlMalloc(sizeof(xmlParserCtxt));
if (ctxt == NULL) {
fprintf(stderr, "xmlNewParserCtxt : cannot allocate context\n");
perror("malloc");
return(NULL);
}
memset(ctxt, 0, sizeof(xmlParserCtxt));
xmlInitParserCtxt(ctxt);
return(ctxt);
}
/**
* xmlClearParserCtxt:
* @ctxt: an XML parser context
*
* Clear (release owned resources) and reinitialize a parser context
*/
void
xmlClearParserCtxt(xmlParserCtxtPtr ctxt)
{
xmlClearNodeInfoSeq(&ctxt->node_seq);
xmlInitParserCtxt(ctxt);
}
/************************************************************************
* *
* Commodity functions to handle entities *
* Imported from 2.3.5 *
* *
************************************************************************/
int xmlParserDebugEntities = 0;
/**
* xmlPopInput:
* @ctxt: an XML parser context
*
* xmlPopInput: the current input pointed by ctxt->input came to an end
* pop it and return the next char.
*
* Returns the current xmlChar in the parser context
*/
xmlChar
xmlPopInput(xmlParserCtxtPtr ctxt) {
if (ctxt->inputNr == 1) return(0); /* End of main Input */
if (xmlParserDebugEntities)
xmlGenericError(xmlGenericErrorContext,
"Popping input %d\n", ctxt->inputNr);
xmlFreeInputStream(inputPop(ctxt));
if ((*ctxt->input->cur == 0) &&
(xmlParserInputGrow(ctxt->input, INPUT_CHUNK) <= 0))
return(xmlPopInput(ctxt));
return(CUR);
}
/**
* xmlPushInput:
* @ctxt: an XML parser context
* @input: an XML parser input fragment (entity, XML fragment ...).
*
* xmlPushInput: switch to a new input stream which is stacked on top
* of the previous one(s).
*/
void
xmlPushInput(xmlParserCtxtPtr ctxt, xmlParserInputPtr input) {
if (input == NULL) return;
if (xmlParserDebugEntities) {
if ((ctxt->input != NULL) && (ctxt->input->filename))
xmlGenericError(xmlGenericErrorContext,
"%s(%d): ", ctxt->input->filename,
ctxt->input->line);
xmlGenericError(xmlGenericErrorContext,
"Pushing input %d : %.30s\n", ctxt->inputNr+1, input->cur);
}
inputPush(ctxt, input);
GROW;
}
/**
* xmlParseCharRef:
* @ctxt: an XML parser context
*
* parse Reference declarations
*
* [66] CharRef ::= '&#' [0-9]+ ';' |
* '&#x' [0-9a-fA-F]+ ';'
*
* [ WFC: Legal Character ]
* Characters referred to using character references must match the
* production for Char.
*
* Returns the value parsed (as an int), 0 in case of error
*/
int
xmlParseCharRef(xmlParserCtxtPtr ctxt) {
unsigned int val = 0;
int count = 0;
if (ctxt->token != 0) {
val = ctxt->token;
ctxt->token = 0;
return(val);
}
/*
* Using RAW/CUR/NEXT is okay since we are working on ASCII range here
*/
if ((RAW == '&') && (NXT(1) == '#') &&
(NXT(2) == 'x')) {
SKIP(3);
GROW;
while (RAW != ';') { /* loop blocked by count */
if ((RAW >= '0') && (RAW <= '9') && (count < 20))
val = val * 16 + (CUR - '0');
else if ((RAW >= 'a') && (RAW <= 'f') && (count < 20))
val = val * 16 + (CUR - 'a') + 10;
else if ((RAW >= 'A') && (RAW <= 'F') && (count < 20))
val = val * 16 + (CUR - 'A') + 10;
else {
ctxt->errNo = XML_ERR_INVALID_HEX_CHARREF;
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt->userData,
"xmlParseCharRef: invalid hexadecimal value\n");
ctxt->wellFormed = 0;
ctxt->disableSAX = 1;
val = 0;
break;
}
NEXT;
count++;
}
if (RAW == ';') {
/* on purpose to avoid reentrancy problems with NEXT and SKIP */
ctxt->nbChars ++;
ctxt->input->cur++;
}
} else if ((RAW == '&') && (NXT(1) == '#')) {
SKIP(2);
GROW;
while (RAW != ';') { /* loop blocked by count */
if ((RAW >= '0') && (RAW <= '9') && (count < 20))
val = val * 10 + (CUR - '0');
else {
ctxt->errNo = XML_ERR_INVALID_DEC_CHARREF;
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt->userData,
"xmlParseCharRef: invalid decimal value\n");
ctxt->wellFormed = 0;
ctxt->disableSAX = 1;
val = 0;
break;
}
NEXT;
count++;
}
if (RAW == ';') {
/* on purpose to avoid reentrancy problems with NEXT and SKIP */
ctxt->nbChars ++;
ctxt->input->cur++;
}
} else {
ctxt->errNo = XML_ERR_INVALID_CHARREF;
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt->userData,
"xmlParseCharRef: invalid value\n");
ctxt->wellFormed = 0;
ctxt->disableSAX = 1;
}
/*
* [ WFC: Legal Character ]
* Characters referred to using character references must match the
* production for Char.
*/
if (IS_CHAR(val)) {
return(val);
} else {
ctxt->errNo = XML_ERR_INVALID_CHAR;
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt->userData, "CharRef: invalid xmlChar value %d\n",
val);
ctxt->wellFormed = 0;
ctxt->disableSAX = 1;
}
return(0);
}
/**
* xmlParseStringCharRef:
* @ctxt: an XML parser context
* @str: a pointer to an index in the string
*
* parse Reference declarations, variant parsing from a string rather
* than an an input flow.
*
* [66] CharRef ::= '&#' [0-9]+ ';' |
* '&#x' [0-9a-fA-F]+ ';'
*
* [ WFC: Legal Character ]
* Characters referred to using character references must match the
* production for Char.
*
* Returns the value parsed (as an int), 0 in case of error, str will be
* updated to the current value of the index
*/
int
xmlParseStringCharRef(xmlParserCtxtPtr ctxt, const xmlChar **str) {
const xmlChar *ptr;
xmlChar cur;
int val = 0;
if ((str == NULL) || (*str == NULL)) return(0);
ptr = *str;
cur = *ptr;
if ((cur == '&') && (ptr[1] == '#') && (ptr[2] == 'x')) {
ptr += 3;
cur = *ptr;
while (cur != ';') { /* Non input consuming loop */
if ((cur >= '0') && (cur <= '9'))
val = val * 16 + (cur - '0');
else if ((cur >= 'a') && (cur <= 'f'))
val = val * 16 + (cur - 'a') + 10;
else if ((cur >= 'A') && (cur <= 'F'))
val = val * 16 + (cur - 'A') + 10;
else {
ctxt->errNo = XML_ERR_INVALID_HEX_CHARREF;
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt->userData,
"xmlParseStringCharRef: invalid hexadecimal value\n");
ctxt->wellFormed = 0;
ctxt->disableSAX = 1;
val = 0;
break;
}
ptr++;
cur = *ptr;
}
if (cur == ';')
ptr++;
} else if ((cur == '&') && (ptr[1] == '#')){
ptr += 2;
cur = *ptr;
while (cur != ';') { /* Non input consuming loops */
if ((cur >= '0') && (cur <= '9'))
val = val * 10 + (cur - '0');
else {
ctxt->errNo = XML_ERR_INVALID_DEC_CHARREF;
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt->userData,
"xmlParseStringCharRef: invalid decimal value\n");
ctxt->wellFormed = 0;
ctxt->disableSAX = 1;
val = 0;
break;
}
ptr++;
cur = *ptr;
}
if (cur == ';')
ptr++;
} else {
ctxt->errNo = XML_ERR_INVALID_CHARREF;
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt->userData,
"xmlParseCharRef: invalid value\n");
ctxt->wellFormed = 0;
ctxt->disableSAX = 1;
return(0);
}
*str = ptr;
/*
* [ WFC: Legal Character ]
* Characters referred to using character references must match the
* production for Char.
*/
if (IS_CHAR(val)) {
return(val);
} else {
ctxt->errNo = XML_ERR_INVALID_CHAR;
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt->userData,
"CharRef: invalid xmlChar value %d\n", val);
ctxt->wellFormed = 0;
ctxt->disableSAX = 1;
}
return(0);
}
/**
* xmlParserHandlePEReference:
* @ctxt: the parser context
*
* [69] PEReference ::= '%' Name ';'
*
* [ WFC: No Recursion ]
* A parsed entity must not contain a recursive
* reference to itself, either directly or indirectly.
*
* [ WFC: Entity Declared ]
* In a document without any DTD, a document with only an internal DTD
* subset which contains no parameter entity references, or a document
* with "standalone='yes'", ... ... The declaration of a parameter
* entity must precede any reference to it...
*
* [ VC: Entity Declared ]
* In a document with an external subset or external parameter entities
* with "standalone='no'", ... ... The declaration of a parameter entity
* must precede any reference to it...
*
* [ WFC: In DTD ]
* Parameter-entity references may only appear in the DTD.
* NOTE: misleading but this is handled.
*
* A PEReference may have been detected in the current input stream
* the handling is done accordingly to
* http://www.w3.org/TR/REC-xml#entproc
* i.e.
* - Included in literal in entity values
* - Included as Paraemeter Entity reference within DTDs
*/
void
xmlParserHandlePEReference(xmlParserCtxtPtr ctxt) {
xmlChar *name;
xmlEntityPtr entity = NULL;
xmlParserInputPtr input;
if (ctxt->token != 0) {
return;
}
if (RAW != '%') return;
switch(ctxt->instate) {
case XML_PARSER_CDATA_SECTION:
return;
case XML_PARSER_COMMENT:
return;
case XML_PARSER_START_TAG:
return;
case XML_PARSER_END_TAG:
return;
case XML_PARSER_EOF:
ctxt->errNo = XML_ERR_PEREF_AT_EOF;
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt->userData, "PEReference at EOF\n");
ctxt->wellFormed = 0;
ctxt->disableSAX = 1;
return;
case XML_PARSER_PROLOG:
case XML_PARSER_START:
case XML_PARSER_MISC:
ctxt->errNo = XML_ERR_PEREF_IN_PROLOG;
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt->userData, "PEReference in prolog!\n");
ctxt->wellFormed = 0;
ctxt->disableSAX = 1;
return;
case XML_PARSER_ENTITY_DECL:
case XML_PARSER_CONTENT:
case XML_PARSER_ATTRIBUTE_VALUE:
case XML_PARSER_PI:
case XML_PARSER_SYSTEM_LITERAL:
/* we just ignore it there */
return;
case XML_PARSER_EPILOG:
ctxt->errNo = XML_ERR_PEREF_IN_EPILOG;
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt->userData, "PEReference in epilog!\n");
ctxt->wellFormed = 0;
ctxt->disableSAX = 1;
return;
case XML_PARSER_ENTITY_VALUE:
/*
* NOTE: in the case of entity values, we don't do the
* substitution here since we need the literal
* entity value to be able to save the internal
* subset of the document.
* This will be handled by xmlStringDecodeEntities
*/
return;
case XML_PARSER_DTD:
/*
* [WFC: Well-Formedness Constraint: PEs in Internal Subset]
* In the internal DTD subset, parameter-entity references
* can occur only where markup declarations can occur, not
* within markup declarations.
* In that case this is handled in xmlParseMarkupDecl
*/
if ((ctxt->external == 0) && (ctxt->inputNr == 1))
return;
break;
case XML_PARSER_IGNORE:
return;
}
NEXT;
name = xmlParseName(ctxt);
if (xmlParserDebugEntities)
xmlGenericError(xmlGenericErrorContext,
"PE Reference: %s\n", name);
if (name == NULL) {
ctxt->errNo = XML_ERR_PEREF_NO_NAME;
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt->userData, "xmlHandlePEReference: no name\n");
ctxt->wellFormed = 0;
ctxt->disableSAX = 1;
} else {
if (RAW == ';') {
NEXT;
if ((ctxt->sax != NULL) && (ctxt->sax->getParameterEntity != NULL))
entity = ctxt->sax->getParameterEntity(ctxt->userData, name);
if (entity == NULL) {
/*
* [ WFC: Entity Declared ]
* In a document without any DTD, a document with only an
* internal DTD subset which contains no parameter entity
* references, or a document with "standalone='yes'", ...
* ... The declaration of a parameter entity must precede
* any reference to it...
*/
if ((ctxt->standalone == 1) ||
((ctxt->hasExternalSubset == 0) &&
(ctxt->hasPErefs == 0))) {
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt->userData,
"PEReference: %%%s; not found\n", name);
ctxt->wellFormed = 0;
ctxt->disableSAX = 1;
} else {
/*
* [ VC: Entity Declared ]
* In a document with an external subset or external
* parameter entities with "standalone='no'", ...
* ... The declaration of a parameter entity must precede
* any reference to it...
*/
if ((!ctxt->disableSAX) &&
(ctxt->validate) && (ctxt->vctxt.error != NULL)) {
ctxt->vctxt.error(ctxt->vctxt.userData,
"PEReference: %%%s; not found\n", name);
} else if ((!ctxt->disableSAX) &&
(ctxt->sax != NULL) && (ctxt->sax->warning != NULL))
ctxt->sax->warning(ctxt->userData,
"PEReference: %%%s; not found\n", name);
ctxt->valid = 0;
}
} else {
if ((entity->type == XML_INTERNAL_PARAMETER_ENTITY) ||
(entity->type == XML_EXTERNAL_PARAMETER_ENTITY)) {
/*
* handle the extra spaces added before and after
* c.f. http://www.w3.org/TR/REC-xml#as-PE
* this is done independantly.
*/
input = xmlNewEntityInputStream(ctxt, entity);
xmlPushInput(ctxt, input);
if ((entity->type == XML_EXTERNAL_PARAMETER_ENTITY) &&
(RAW == '<') && (NXT(1) == '?') &&
(NXT(2) == 'x') && (NXT(3) == 'm') &&
(NXT(4) == 'l') && (IS_BLANK(NXT(5)))) {
xmlParseTextDecl(ctxt);
}
if (ctxt->token == 0)
ctxt->token = ' ';
} else {
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt->userData,
"xmlHandlePEReference: %s is not a parameter entity\n",
name);
ctxt->wellFormed = 0;
ctxt->disableSAX = 1;
}
}
} else {
ctxt->errNo = XML_ERR_PEREF_SEMICOL_MISSING;
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt->userData,
"xmlHandlePEReference: expecting ';'\n");
ctxt->wellFormed = 0;
ctxt->disableSAX = 1;
}
xmlFree(name);
}
}
/*
* Macro used to grow the current buffer.
*/
#define growBuffer(buffer) { \
buffer##_size *= 2; \
buffer = (xmlChar *) \
xmlRealloc(buffer, buffer##_size * sizeof(xmlChar)); \
if (buffer == NULL) { \
perror("realloc failed"); \
return(NULL); \
} \
}
/**
* xmlDecodeEntities:
* @ctxt: the parser context
* @what: combination of XML_SUBSTITUTE_REF and XML_SUBSTITUTE_PEREF
* @len: the len to decode (in bytes !), -1 for no size limit
* @end: an end marker xmlChar, 0 if none
* @end2: an end marker xmlChar, 0 if none
* @end3: an end marker xmlChar, 0 if none
*
* This function is deprecated, we now always process entities content
* through xmlStringDecodeEntities
*
* TODO: remove it in next major release.
*
* [67] Reference ::= EntityRef | CharRef
*
* [69] PEReference ::= '%' Name ';'
*
* Returns A newly allocated string with the substitution done. The caller
* must deallocate it !
*/
xmlChar *
xmlDecodeEntities(xmlParserCtxtPtr ctxt, int len, int what,
xmlChar end, xmlChar end2, xmlChar end3) {
static int deprecated = 0;
if (!deprecated) {
xmlGenericError(xmlGenericErrorContext,
"xmlDecodeEntities() deprecated function reached\n");
deprecated = 1;
}
return(NULL);
}
/**
* xmlStringDecodeEntities:
* @ctxt: the parser context
* @str: the input string
* @what: combination of XML_SUBSTITUTE_REF and XML_SUBSTITUTE_PEREF
* @end: an end marker xmlChar, 0 if none
* @end2: an end marker xmlChar, 0 if none
* @end3: an end marker xmlChar, 0 if none
*
* Takes a entity string content and process to do the adequate subtitutions.
*
* [67] Reference ::= EntityRef | CharRef
*
* [69] PEReference ::= '%' Name ';'
*
* Returns A newly allocated string with the substitution done. The caller
* must deallocate it !
*/
xmlChar *
xmlStringDecodeEntities(xmlParserCtxtPtr ctxt, const xmlChar *str, int what,
xmlChar end, xmlChar end2, xmlChar end3) {
xmlChar *buffer = NULL;
int buffer_size = 0;
xmlChar *current = NULL;
xmlEntityPtr ent;
int c,l;
int nbchars = 0;
if (str == NULL)
return(NULL);
if (ctxt->depth > 40) {
ctxt->errNo = XML_ERR_ENTITY_LOOP;
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt->userData,
"Detected entity reference loop\n");
ctxt->wellFormed = 0;
ctxt->disableSAX = 1;
return(NULL);
}
/*
* allocate a translation buffer.
*/
buffer_size = XML_PARSER_BIG_BUFFER_SIZE;
buffer = (xmlChar *) xmlMalloc(buffer_size * sizeof(xmlChar));
if (buffer == NULL) {
perror("xmlDecodeEntities: malloc failed");
return(NULL);
}
/*
* Ok loop until we reach one of the ending char or a size limit.
* we are operating on already parsed values.
*/
c = CUR_SCHAR(str, l);
while ((c != 0) && (c != end) && /* non input consuming loop */
(c != end2) && (c != end3)) {
if (c == 0) break;
if ((c == '&') && (str[1] == '#')) {
int val = xmlParseStringCharRef(ctxt, &str);
if (val != 0) {
COPY_BUF(0,buffer,nbchars,val);
}
} else if ((c == '&') && (what & XML_SUBSTITUTE_REF)) {
if (xmlParserDebugEntities)
xmlGenericError(xmlGenericErrorContext,
"String decoding Entity Reference: %.30s\n",
str);
ent = xmlParseStringEntityRef(ctxt, &str);
if ((ent != NULL) &&
(ent->type == XML_INTERNAL_PREDEFINED_ENTITY)) {
if (ent->content != NULL) {
COPY_BUF(0,buffer,nbchars,ent->content[0]);
} else {
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt->userData,
"internal error entity has no content\n");
}
} else if ((ent != NULL) && (ent->content != NULL)) {
xmlChar *rep;
ctxt->depth++;
rep = xmlStringDecodeEntities(ctxt, ent->content, what,
0, 0, 0);
ctxt->depth--;
if (rep != NULL) {
current = rep;
while (*current != 0) { /* non input consuming loop */
buffer[nbchars++] = *current++;
if (nbchars >
buffer_size - XML_PARSER_BUFFER_SIZE) {
growBuffer(buffer);
}
}
xmlFree(rep);
}
} else if (ent != NULL) {
int i = xmlStrlen(ent->name);
const xmlChar *cur = ent->name;
buffer[nbchars++] = '&';
if (nbchars > buffer_size - i - XML_PARSER_BUFFER_SIZE) {
growBuffer(buffer);
}
for (;i > 0;i--)
buffer[nbchars++] = *cur++;
buffer[nbchars++] = ';';
}
} else if (c == '%' && (what & XML_SUBSTITUTE_PEREF)) {
if (xmlParserDebugEntities)
xmlGenericError(xmlGenericErrorContext,
"String decoding PE Reference: %.30s\n", str);
ent = xmlParseStringPEReference(ctxt, &str);
if (ent != NULL) {
xmlChar *rep;
ctxt->depth++;
rep = xmlStringDecodeEntities(ctxt, ent->content, what,
0, 0, 0);
ctxt->depth--;
if (rep != NULL) {
current = rep;
while (*current != 0) { /* non input consuming loop */
buffer[nbchars++] = *current++;
if (nbchars >
buffer_size - XML_PARSER_BUFFER_SIZE) {
growBuffer(buffer);
}
}
xmlFree(rep);
}
}
} else {
COPY_BUF(l,buffer,nbchars,c);
str += l;
if (nbchars > buffer_size - XML_PARSER_BUFFER_SIZE) {
growBuffer(buffer);
}
}
c = CUR_SCHAR(str, l);
}
buffer[nbchars++] = 0;
return(buffer);
}
/************************************************************************
* *
* Commodity functions to handle encodings *
* *
************************************************************************/
/**
* xmlSwitchEncoding:
* @ctxt: the parser context
* @enc: the encoding value (number)
*
* change the input functions when discovering the character encoding
* of a given entity.
*/
void
xmlSwitchEncoding(xmlParserCtxtPtr ctxt, xmlCharEncoding enc)
{
switch (enc) {
case XML_CHAR_ENCODING_ERROR:
ctxt->errNo = XML_ERR_UNKNOWN_ENCODING;
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt->userData, "encoding unknown\n");
ctxt->wellFormed = 0;
break;
case XML_CHAR_ENCODING_NONE:
/* let's assume it's UTF-8 without the XML decl */
return;
case XML_CHAR_ENCODING_UTF8:
/* default encoding, no conversion should be needed */
return;
case XML_CHAR_ENCODING_UTF16LE:
ctxt->errNo = XML_ERR_UNSUPPORTED_ENCODING;
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt->userData,
"char encoding UTF16 little endian not supported\n");
break;
case XML_CHAR_ENCODING_UTF16BE:
ctxt->errNo = XML_ERR_UNSUPPORTED_ENCODING;
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt->userData,
"char encoding UTF16 big endian not supported\n");
break;
case XML_CHAR_ENCODING_UCS4LE:
ctxt->errNo = XML_ERR_UNSUPPORTED_ENCODING;
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt->userData,
"char encoding USC4 little endian not supported\n");
break;
case XML_CHAR_ENCODING_UCS4BE:
ctxt->errNo = XML_ERR_UNSUPPORTED_ENCODING;
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt->userData,
"char encoding USC4 big endian not supported\n");
break;
case XML_CHAR_ENCODING_EBCDIC:
ctxt->errNo = XML_ERR_UNSUPPORTED_ENCODING;
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt->userData,
"char encoding EBCDIC not supported\n");
break;
case XML_CHAR_ENCODING_UCS4_2143:
ctxt->errNo = XML_ERR_UNSUPPORTED_ENCODING;
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt->userData,
"char encoding UCS4 2143 not supported\n");
break;
case XML_CHAR_ENCODING_UCS4_3412:
ctxt->errNo = XML_ERR_UNSUPPORTED_ENCODING;
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt->userData,
"char encoding UCS4 3412 not supported\n");
break;
case XML_CHAR_ENCODING_UCS2:
ctxt->errNo = XML_ERR_UNSUPPORTED_ENCODING;
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt->userData,
"char encoding UCS2 not supported\n");
break;
case XML_CHAR_ENCODING_8859_1:
ctxt->errNo = XML_ERR_UNSUPPORTED_ENCODING;
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt->userData,
"char encoding ISO_8859_1 ISO Latin 1 not supported\n");
break;
case XML_CHAR_ENCODING_8859_2:
ctxt->errNo = XML_ERR_UNSUPPORTED_ENCODING;
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt->userData,
"char encoding ISO_8859_2 ISO Latin 2 not supported\n");
break;
case XML_CHAR_ENCODING_8859_3:
ctxt->errNo = XML_ERR_UNSUPPORTED_ENCODING;
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt->userData,
"char encoding ISO_8859_3 not supported\n");
break;
case XML_CHAR_ENCODING_8859_4:
ctxt->errNo = XML_ERR_UNSUPPORTED_ENCODING;
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt->userData,
"char encoding ISO_8859_4 not supported\n");
break;
case XML_CHAR_ENCODING_8859_5:
ctxt->errNo = XML_ERR_UNSUPPORTED_ENCODING;
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt->userData,
"char encoding ISO_8859_5 not supported\n");
break;
case XML_CHAR_ENCODING_8859_6:
ctxt->errNo = XML_ERR_UNSUPPORTED_ENCODING;
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt->userData,
"char encoding ISO_8859_6 not supported\n");
break;
case XML_CHAR_ENCODING_8859_7:
ctxt->errNo = XML_ERR_UNSUPPORTED_ENCODING;
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt->userData,
"char encoding ISO_8859_7 not supported\n");
break;
case XML_CHAR_ENCODING_8859_8:
ctxt->errNo = XML_ERR_UNSUPPORTED_ENCODING;
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt->userData,
"char encoding ISO_8859_8 not supported\n");
break;
case XML_CHAR_ENCODING_8859_9:
ctxt->errNo = XML_ERR_UNSUPPORTED_ENCODING;
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt->userData,
"char encoding ISO_8859_9 not supported\n");
break;
case XML_CHAR_ENCODING_2022_JP:
ctxt->errNo = XML_ERR_UNSUPPORTED_ENCODING;
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt->userData,
"char encoding ISO-2022-JPnot supported\n");
break;
case XML_CHAR_ENCODING_SHIFT_JIS:
ctxt->errNo = XML_ERR_UNSUPPORTED_ENCODING;
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt->userData,
"char encoding Shift_JISnot supported\n");
break;
case XML_CHAR_ENCODING_EUC_JP:
ctxt->errNo = XML_ERR_UNSUPPORTED_ENCODING;
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt->userData,
"char encoding EUC-JPnot supported\n");
break;
}
}
/************************************************************************
* *
* Commodity functions to handle xmlChars *
* *
************************************************************************/
/**
* xmlStrndup:
* @cur: the input xmlChar *
* @len: the len of @cur
*
* a strndup for array of xmlChar's
*
* Returns a new xmlChar * or NULL
*/
xmlChar *
xmlStrndup(const xmlChar *cur, int len) {
xmlChar *ret;
if ((cur == NULL) || (len < 0)) return(NULL);
ret = xmlMalloc((len + 1) * sizeof(xmlChar));
if (ret == NULL) {
fprintf(stderr, "malloc of %ld byte failed\n",
(len + 1) * (long)sizeof(xmlChar));
return(NULL);
}
memcpy(ret, cur, len * sizeof(xmlChar));
ret[len] = 0;
return(ret);
}
/**
* xmlStrdup:
* @cur: the input xmlChar *
*
* a strdup for array of xmlChar's
*
* Returns a new xmlChar * or NULL
*/
xmlChar *
xmlStrdup(const xmlChar *cur) {
const xmlChar *p = cur;
if (cur == NULL) return(NULL);
while (IS_CHAR(*p)) p++;
return(xmlStrndup(cur, p - cur));
}
/**
* xmlCharStrndup:
* @cur: the input char *
* @len: the len of @cur
*
* a strndup for char's to xmlChar's
*
* Returns a new xmlChar * or NULL
*/
xmlChar *
xmlCharStrndup(const char *cur, int len) {
int i;
xmlChar *ret;
if ((cur == NULL) || (len < 0)) return(NULL);
ret = xmlMalloc((len + 1) * sizeof(xmlChar));
if (ret == NULL) {
fprintf(stderr, "malloc of %ld byte failed\n",
(len + 1) * (long)sizeof(xmlChar));
return(NULL);
}
for (i = 0;i < len;i++)
ret[i] = (xmlChar) cur[i];
ret[len] = 0;
return(ret);
}
/**
* xmlCharStrdup:
* @cur: the input char *
* @len: the len of @cur
*
* a strdup for char's to xmlChar's
*
* Returns a new xmlChar * or NULL
*/
xmlChar *
xmlCharStrdup(const char *cur) {
const char *p = cur;
if (cur == NULL) return(NULL);
while (*p != '\0') p++;
return(xmlCharStrndup(cur, p - cur));
}
/**
* xmlStrcmp:
* @str1: the first xmlChar *
* @str2: the second xmlChar *
*
* a strcmp for xmlChar's
*
* Returns the integer result of the comparison
*/
int
xmlStrcmp(const xmlChar *str1, const xmlChar *str2) {
register int tmp;
if ((str1 == NULL) && (str2 == NULL)) return(0);
if (str1 == NULL) return(-1);
if (str2 == NULL) return(1);
do {
tmp = *str1++ - *str2++;
if (tmp != 0) return(tmp);
} while ((*str1 != 0) && (*str2 != 0));
return (*str1 - *str2);
}
/* Added from 2.3.5 */
/**
* xmlStrEqual:
* @str1: the first xmlChar *
* @str2: the second xmlChar *
*
* Check if both string are equal of have same content
* Should be a bit more readable and faster than xmlStrEqual()
*
* Returns 1 if they are equal, 0 if they are different
*/
int
xmlStrEqual(const xmlChar *str1, const xmlChar *str2) {
if (str1 == str2) return(1);
if (str1 == NULL) return(0);
if (str2 == NULL) return(0);
do {
if (*str1++ != *str2) return(0);
} while (*str2++);
return(1);
}
/**
* xmlStrncmp:
* @str1: the first xmlChar *
* @str2: the second xmlChar *
* @len: the max comparison length
*
* a strncmp for xmlChar's
*
* Returns the integer result of the comparison
*/
int
xmlStrncmp(const xmlChar *str1, const xmlChar *str2, int len) {
register int tmp;
if (len <= 0) return(0);
if ((str1 == NULL) && (str2 == NULL)) return(0);
if (str1 == NULL) return(-1);
if (str2 == NULL) return(1);
do {
tmp = *str1++ - *str2++;
if (tmp != 0) return(tmp);
len--;
if (len <= 0) return(0);
} while ((*str1 != 0) && (*str2 != 0));
return (*str1 - *str2);
}
/**
* xmlStrchr:
* @str: the xmlChar * array
* @val: the xmlChar to search
*
* a strchr for xmlChar's
*
* Returns the xmlChar * for the first occurence or NULL.
*/
const xmlChar *
xmlStrchr(const xmlChar *str, xmlChar val) {
if (str == NULL) return(NULL);
while (*str != 0) {
if (*str == val) return((xmlChar *) str);
str++;
}
return(NULL);
}
/**
* xmlStrstr:
* @str: the xmlChar * array (haystack)
* @val: the xmlChar to search (needle)
*
* a strstr for xmlChar's
*
* Returns the xmlChar * for the first occurence or NULL.
*/
const xmlChar *
xmlStrstr(const xmlChar *str, xmlChar *val) {
int n;
if (str == NULL) return(NULL);
if (val == NULL) return(NULL);
n = xmlStrlen(val);
if (n == 0) return(str);
while (*str != 0) {
if (*str == *val) {
if (!xmlStrncmp(str, val, n)) return((const xmlChar *) str);
}
str++;
}
return(NULL);
}
/**
* xmlStrsub:
* @str: the xmlChar * array (haystack)
* @start: the index of the first char (zero based)
* @len: the length of the substring
*
* Extract a substring of a given string
*
* Returns the xmlChar * for the first occurence or NULL.
*/
xmlChar *
xmlStrsub(const xmlChar *str, int start, int len) {
int i;
if (str == NULL) return(NULL);
if (start < 0) return(NULL);
if (len < 0) return(NULL);
for (i = 0;i < start;i++) {
if (*str == 0) return(NULL);
str++;
}
if (*str == 0) return(NULL);
return(xmlStrndup(str, len));
}
/**
* xmlStrlen:
* @str: the xmlChar * array
*
* length of a xmlChar's string
*
* Returns the number of xmlChar contained in the ARRAY.
*/
int
xmlStrlen(const xmlChar *str) {
int len = 0;
if (str == NULL) return(0);
while (*str != 0) {
str++;
len++;
}
return(len);
}
/**
* xmlStrncat:
* @cur: the original xmlChar * array
* @add: the xmlChar * array added
* @len: the length of @add
*
* a strncat for array of xmlChar's
*
* Returns a new xmlChar * containing the concatenated string.
*/
xmlChar *
xmlStrncat(xmlChar *cur, const xmlChar *add, int len) {
int size;
xmlChar *ret;
if ((add == NULL) || (len == 0))
return(cur);
if (cur == NULL)
return(xmlStrndup(add, len));
size = xmlStrlen(cur);
ret = xmlRealloc(cur, (size + len + 1) * sizeof(xmlChar));
if (ret == NULL) {
fprintf(stderr, "xmlStrncat: realloc of %ld byte failed\n",
(size + len + 1) * (long)sizeof(xmlChar));
return(cur);
}
memcpy(&ret[size], add, len * sizeof(xmlChar));
ret[size + len] = 0;
return(ret);
}
/**
* xmlStrcat:
* @cur: the original xmlChar * array
* @add: the xmlChar * array added
*
* a strcat for array of xmlChar's
*
* Returns a new xmlChar * containing the concatenated string.
*/
xmlChar *
xmlStrcat(xmlChar *cur, const xmlChar *add) {
const xmlChar *p = add;
if (add == NULL) return(cur);
if (cur == NULL)
return(xmlStrdup(add));
while (IS_CHAR(*p)) p++;
return(xmlStrncat(cur, add, p - add));
}
/************************************************************************
* *
* Commodity functions, cleanup needed ? *
* *
************************************************************************/
/**
* areBlanks:
* @ctxt: an XML parser context
* @str: a xmlChar *
* @len: the size of @str
*
* Is this a sequence of blank chars that one can ignore ?
*
* Returns 1 if ignorable 0 otherwise.
*/
static int areBlanks(xmlParserCtxtPtr ctxt, const xmlChar *str, int len) {
int i, ret;
xmlNodePtr lastChild;
/*
* Check that the string is made of blanks
*/
for (i = 0;i < len;i++)
if (!(IS_BLANK(str[i]))) return(0);
/*
* Look if the element is mixed content in the Dtd if available
*/
if (ctxt->myDoc != NULL) {
ret = xmlIsMixedElement(ctxt->myDoc, ctxt->node->name);
if (ret == 0) return(1);
if (ret == 1) return(0);
}
/*
* Do we allow an heuristic on white space
*/
if (ctxt->keepBlanks)
return(0);
if (CUR != '<') return(0);
if (ctxt->node == NULL) return(0);
if ((ctxt->node->childs == NULL) &&
(CUR == '<') && (NXT(1) == '/')) return(0);
lastChild = xmlGetLastChild(ctxt->node);
if (lastChild == NULL) {
if (ctxt->node->content != NULL) return(0);
} else if (xmlNodeIsText(lastChild))
return(0);
else if ((ctxt->node->childs != NULL) &&
(xmlNodeIsText(ctxt->node->childs)))
return(0);
return(1);
}
/**
* xmlHandleEntity:
* @ctxt: an XML parser context
* @entity: an XML entity pointer.
*
* Default handling of defined entities, when should we define a new input
* stream ? When do we just handle that as a set of chars ?
*
* OBSOLETE: to be removed at some point.
*/
void
xmlHandleEntity(xmlParserCtxtPtr ctxt, xmlEntityPtr entity) {
int len;
xmlParserInputPtr input;
if (entity->content == NULL) {
ctxt->errNo = XML_ERR_INTERNAL_ERROR;
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt->userData, "xmlHandleEntity %s: content == NULL\n",
entity->name);
ctxt->wellFormed = 0;
return;
}
len = xmlStrlen(entity->content);
if (len <= 2) goto handle_as_char;
/*
* Redefine its content as an input stream.
*/
input = xmlNewEntityInputStream(ctxt, entity);
xmlPushInput(ctxt, input);
return;
handle_as_char:
/*
* Just handle the content as a set of chars.
*/
if ((ctxt->sax != NULL) && (ctxt->sax->characters != NULL))
ctxt->sax->characters(ctxt->userData, entity->content, len);
}
/*
* Forward definition for recusive behaviour.
*/
void xmlParsePEReference(xmlParserCtxtPtr ctxt);
void xmlParseReference(xmlParserCtxtPtr ctxt);
/************************************************************************
* *
* Extra stuff for namespace support *
* Relates to http://www.w3.org/TR/WD-xml-names *
* *
************************************************************************/
/**
* xmlNamespaceParseNCName:
* @ctxt: an XML parser context
*
* parse an XML namespace name.
*
* [NS 3] NCName ::= (Letter | '_') (NCNameChar)*
*
* [NS 4] NCNameChar ::= Letter | Digit | '.' | '-' | '_' |
* CombiningChar | Extender
*
* Returns the namespace name or NULL
*/
xmlChar *
xmlNamespaceParseNCName(xmlParserCtxtPtr ctxt) {
xmlChar buf[XML_MAX_NAMELEN];
int len = 0;
if (!IS_LETTER(CUR) && (CUR != '_')) return(NULL);
while ((IS_LETTER(CUR)) || (IS_DIGIT(CUR)) ||
(CUR == '.') || (CUR == '-') ||
(CUR == '_') ||
(IS_COMBINING(CUR)) ||
(IS_EXTENDER(CUR))) {
buf[len++] = CUR;
NEXT;
if (len >= XML_MAX_NAMELEN) {
fprintf(stderr,
"xmlNamespaceParseNCName: reached XML_MAX_NAMELEN limit\n");
while ((IS_LETTER(CUR)) || (IS_DIGIT(CUR)) ||
(CUR == '.') || (CUR == '-') ||
(CUR == '_') ||
(IS_COMBINING(CUR)) ||
(IS_EXTENDER(CUR)))
NEXT;
break;
}
}
return(xmlStrndup(buf, len));
}
/**
* xmlNamespaceParseQName:
* @ctxt: an XML parser context
* @prefix: a xmlChar **
*
* parse an XML qualified name
*
* [NS 5] QName ::= (Prefix ':')? LocalPart
*
* [NS 6] Prefix ::= NCName
*
* [NS 7] LocalPart ::= NCName
*
* Returns the local part, and prefix is updated
* to get the Prefix if any.
*/
xmlChar *
xmlNamespaceParseQName(xmlParserCtxtPtr ctxt, xmlChar **prefix) {
xmlChar *ret = NULL;
*prefix = NULL;
ret = xmlNamespaceParseNCName(ctxt);
if (CUR == ':') {
*prefix = ret;
NEXT;
ret = xmlNamespaceParseNCName(ctxt);
}
return(ret);
}
/**
* xmlSplitQName:
* @name: an XML parser context
* @prefix: a xmlChar **
*
* parse an XML qualified name string
*
* [NS 5] QName ::= (Prefix ':')? LocalPart
*
* [NS 6] Prefix ::= NCName
*
* [NS 7] LocalPart ::= NCName
*
* Returns the local part, and prefix is updated
* to get the Prefix if any.
*/
xmlChar *
xmlSplitQName(const xmlChar *name, xmlChar **prefix) {
xmlChar buf[XML_MAX_NAMELEN + 5];
xmlChar *buffer = NULL;
int len = 0;
int max = XML_MAX_NAMELEN;
xmlChar *ret = NULL;
const xmlChar *cur = name;
int c;
*prefix = NULL;
#ifndef XML_XML_NAMESPACE
/* xml: prefix is not really a namespace */
if ((cur[0] == 'x') && (cur[1] == 'm') &&
(cur[2] == 'l') && (cur[3] == ':'))
return(xmlStrdup(name));
#endif
/* nasty but valid */
if (cur[0] == ':')
return(xmlStrdup(name));
c = *cur++;
while ((c != 0) && (c != ':') && (len < max)) { /* tested bigname.xml */
buf[len++] = c;
c = *cur++;
}
if (len >= max) {
/*
* Okay someone managed to make a huge name, so he's ready to pay
* for the processing speed.
*/
max = len * 2;
buffer = (xmlChar *) xmlMalloc(max * sizeof(xmlChar));
if (buffer == NULL) {
fprintf(stderr, "xmlSplitQName: out of memory\n");
return(NULL);
}
memcpy(buffer, buf, len);
while ((c != 0) && (c != ':')) { /* tested bigname.xml */
if (len + 10 > max) {
max *= 2;
buffer = (xmlChar *) xmlRealloc(buffer,
max * sizeof(xmlChar));
if (buffer == NULL) {
fprintf(stderr, "xmlSplitQName: out of memory\n");
return(NULL);
}
}
buffer[len++] = c;
c = *cur++;
}
buffer[len] = 0;
}
if (buffer == NULL)
ret = xmlStrndup(buf, len);
else {
ret = buffer;
buffer = NULL;
max = XML_MAX_NAMELEN;
}
if (c == ':') {
c = *cur++;
if (c == 0) return(ret);
*prefix = ret;
len = 0;
while ((c != 0) && (len < max)) { /* tested bigname2.xml */
buf[len++] = c;
c = *cur++;
}
if (len >= max) {
/*
* Okay someone managed to make a huge name, so he's ready to pay
* for the processing speed.
*/
max = len * 2;
buffer = (xmlChar *) xmlMalloc(max * sizeof(xmlChar));
if (buffer == NULL) {
fprintf(stderr, "xmlSplitQName: out of memory\n");
return(NULL);
}
memcpy(buffer, buf, len);
while (c != 0) { /* tested bigname2.xml */
if (len + 10 > max) {
max *= 2;
buffer = (xmlChar *) xmlRealloc(buffer,
max * sizeof(xmlChar));
if (buffer == NULL) {
fprintf(stderr, "xmlSplitQName: out of memory\n");
return(NULL);
}
}
buffer[len++] = c;
c = *cur++;
}
buffer[len] = 0;
}
if (buffer == NULL)
ret = xmlStrndup(buf, len);
else {
ret = buffer;
}
}
return(ret);
#if 0
xmlChar *ret = NULL;
const xmlChar *q;
const xmlChar *cur = name;