blob: b603c7208b27f010a6c4efcee371fb55015d48ad [file] [log] [blame]
/*
* xmlreader.c: implements the xmlTextReader streaming node API
*
* NOTE:
* XmlTextReader.Normalization Property won't be supported, since
* it makes the parser non compliant to the XML recommendation
*
* See Copyright for the status of this software.
*
* daniel@veillard.com
*/
/*
* TODOs:
* - XML Schemas validation
*/
#define IN_LIBXML
#include "libxml.h"
#ifdef LIBXML_READER_ENABLED
#include <string.h> /* for memset() only ! */
#include <stdarg.h>
#ifdef HAVE_CTYPE_H
#include <ctype.h>
#endif
#ifdef HAVE_STDLIB_H
#include <stdlib.h>
#endif
#include <libxml/xmlmemory.h>
#include <libxml/xmlIO.h>
#include <libxml/xmlreader.h>
#include <libxml/parserInternals.h>
#ifdef LIBXML_SCHEMAS_ENABLED
#include <libxml/relaxng.h>
#include <libxml/xmlschemas.h>
#endif
#include <libxml/uri.h>
#ifdef LIBXML_XINCLUDE_ENABLED
#include <libxml/xinclude.h>
#endif
#ifdef LIBXML_PATTERN_ENABLED
#include <libxml/pattern.h>
#endif
#include "buf.h"
#define MAX_ERR_MSG_SIZE 64000
/*
* The following VA_COPY was coded following an example in
* the Samba project. It may not be sufficient for some
* esoteric implementations of va_list but (hopefully) will
* be sufficient for libxml2.
*/
#ifndef VA_COPY
#ifdef HAVE_VA_COPY
#define VA_COPY(dest, src) va_copy(dest, src)
#else
#ifdef HAVE___VA_COPY
#define VA_COPY(dest,src) __va_copy(dest, src)
#else
#ifndef VA_LIST_IS_ARRAY
#define VA_COPY(dest,src) (dest) = (src)
#else
#include <string.h>
#define VA_COPY(dest,src) memcpy((char *)(dest),(char *)(src),sizeof(va_list))
#endif
#endif
#endif
#endif
/* #define DEBUG_CALLBACKS */
/* #define DEBUG_READER */
/**
* TODO:
*
* macro to flag unimplemented blocks
*/
#define TODO \
xmlGenericError(xmlGenericErrorContext, \
"Unimplemented block at %s:%d\n", \
__FILE__, __LINE__);
#ifdef DEBUG_READER
#define DUMP_READER xmlTextReaderDebug(reader);
#else
#define DUMP_READER
#endif
#define CHUNK_SIZE 512
/************************************************************************
* *
* The parser: maps the Text Reader API on top of the existing *
* parsing routines building a tree *
* *
************************************************************************/
#define XML_TEXTREADER_INPUT 1
#define XML_TEXTREADER_CTXT 2
typedef enum {
XML_TEXTREADER_NONE = -1,
XML_TEXTREADER_START= 0,
XML_TEXTREADER_ELEMENT= 1,
XML_TEXTREADER_END= 2,
XML_TEXTREADER_EMPTY= 3,
XML_TEXTREADER_BACKTRACK= 4,
XML_TEXTREADER_DONE= 5,
XML_TEXTREADER_ERROR= 6
} xmlTextReaderState;
typedef enum {
XML_TEXTREADER_NOT_VALIDATE = 0,
XML_TEXTREADER_VALIDATE_DTD = 1,
XML_TEXTREADER_VALIDATE_RNG = 2,
XML_TEXTREADER_VALIDATE_XSD = 4
} xmlTextReaderValidate;
struct _xmlTextReader {
int mode; /* the parsing mode */
xmlDocPtr doc; /* when walking an existing doc */
xmlTextReaderValidate validate;/* is there any validation */
int allocs; /* what structure were deallocated */
xmlTextReaderState state;
xmlParserCtxtPtr ctxt; /* the parser context */
xmlSAXHandlerPtr sax; /* the parser SAX callbacks */
xmlParserInputBufferPtr input; /* the input */
startElementSAXFunc startElement;/* initial SAX callbacks */
endElementSAXFunc endElement; /* idem */
startElementNsSAX2Func startElementNs;/* idem */
endElementNsSAX2Func endElementNs; /* idem */
charactersSAXFunc characters;
cdataBlockSAXFunc cdataBlock;
unsigned int base; /* base of the segment in the input */
unsigned int cur; /* current position in the input */
xmlNodePtr node; /* current node */
xmlNodePtr curnode;/* current attribute node */
int depth; /* depth of the current node */
xmlNodePtr faketext;/* fake xmlNs chld */
int preserve;/* preserve the resulting document */
xmlBufPtr buffer; /* used to return const xmlChar * */
xmlDictPtr dict; /* the context dictionary */
/* entity stack when traversing entities content */
xmlNodePtr ent; /* Current Entity Ref Node */
int entNr; /* Depth of the entities stack */
int entMax; /* Max depth of the entities stack */
xmlNodePtr *entTab; /* array of entities */
/* error handling */
xmlTextReaderErrorFunc errorFunc; /* callback function */
void *errorFuncArg; /* callback function user argument */
#ifdef LIBXML_SCHEMAS_ENABLED
/* Handling of RelaxNG validation */
xmlRelaxNGPtr rngSchemas; /* The Relax NG schemas */
xmlRelaxNGValidCtxtPtr rngValidCtxt;/* The Relax NG validation context */
int rngPreserveCtxt; /* 1 if the context was provided by the user */
int rngValidErrors;/* The number of errors detected */
xmlNodePtr rngFullNode; /* the node if RNG not progressive */
/* Handling of Schemas validation */
xmlSchemaPtr xsdSchemas; /* The Schemas schemas */
xmlSchemaValidCtxtPtr xsdValidCtxt;/* The Schemas validation context */
int xsdPreserveCtxt; /* 1 if the context was provided by the user */
int xsdValidErrors;/* The number of errors detected */
xmlSchemaSAXPlugPtr xsdPlug; /* the schemas plug in SAX pipeline */
#endif
#ifdef LIBXML_XINCLUDE_ENABLED
/* Handling of XInclude processing */
int xinclude; /* is xinclude asked for */
const xmlChar * xinclude_name; /* the xinclude name from dict */
xmlXIncludeCtxtPtr xincctxt; /* the xinclude context */
int in_xinclude; /* counts for xinclude */
#endif
#ifdef LIBXML_PATTERN_ENABLED
int patternNr; /* number of preserve patterns */
int patternMax; /* max preserve patterns */
xmlPatternPtr *patternTab; /* array of preserve patterns */
#endif
int preserves; /* level of preserves */
int parserFlags; /* the set of options set */
/* Structured error handling */
xmlStructuredErrorFunc sErrorFunc; /* callback function */
};
#define NODE_IS_EMPTY 0x1
#define NODE_IS_PRESERVED 0x2
#define NODE_IS_SPRESERVED 0x4
/**
* CONSTSTR:
*
* Macro used to return an interned string
*/
#define CONSTSTR(str) xmlDictLookup(reader->dict, (str), -1)
#define CONSTQSTR(p, str) xmlDictQLookup(reader->dict, (p), (str))
static int xmlTextReaderReadTree(xmlTextReaderPtr reader);
static int xmlTextReaderNextTree(xmlTextReaderPtr reader);
/************************************************************************
* *
* Our own version of the freeing routines as we recycle nodes *
* *
************************************************************************/
/**
* DICT_FREE:
* @str: a string
*
* Free a string if it is not owned by the "dict" dictionary in the
* current scope
*/
#define DICT_FREE(str) \
if ((str) && ((!dict) || \
(xmlDictOwns(dict, (const xmlChar *)(str)) == 0))) \
xmlFree((char *)(str));
static void xmlTextReaderFreeNode(xmlTextReaderPtr reader, xmlNodePtr cur);
static void xmlTextReaderFreeNodeList(xmlTextReaderPtr reader, xmlNodePtr cur);
/**
* xmlFreeID:
* @not: A id
*
* Deallocate the memory used by an id definition
*/
static void
xmlFreeID(xmlIDPtr id) {
xmlDictPtr dict = NULL;
if (id == NULL) return;
if (id->doc != NULL)
dict = id->doc->dict;
if (id->value != NULL)
DICT_FREE(id->value)
if (id->name != NULL)
DICT_FREE(id->name)
xmlFree(id);
}
/**
* xmlTextReaderRemoveID:
* @doc: the document
* @attr: the attribute
*
* Remove the given attribute from the ID table maintained internally.
*
* Returns -1 if the lookup failed and 0 otherwise
*/
static int
xmlTextReaderRemoveID(xmlDocPtr doc, xmlAttrPtr attr) {
xmlIDTablePtr table;
xmlIDPtr id;
xmlChar *ID;
if (doc == NULL) return(-1);
if (attr == NULL) return(-1);
table = (xmlIDTablePtr) doc->ids;
if (table == NULL)
return(-1);
ID = xmlNodeListGetString(doc, attr->children, 1);
if (ID == NULL)
return(-1);
id = xmlHashLookup(table, ID);
xmlFree(ID);
if (id == NULL || id->attr != attr) {
return(-1);
}
id->name = attr->name;
attr->name = NULL;
id->attr = NULL;
return(0);
}
/**
* xmlTextReaderFreeProp:
* @reader: the xmlTextReaderPtr used
* @cur: the node
*
* Free a node.
*/
static void
xmlTextReaderFreeProp(xmlTextReaderPtr reader, xmlAttrPtr cur) {
xmlDictPtr dict;
if ((reader != NULL) && (reader->ctxt != NULL))
dict = reader->ctxt->dict;
else
dict = NULL;
if (cur == NULL) return;
if ((__xmlRegisterCallbacks) && (xmlDeregisterNodeDefaultValue))
xmlDeregisterNodeDefaultValue((xmlNodePtr) cur);
/* Check for ID removal -> leading to invalid references ! */
if ((cur->parent != NULL) && (cur->parent->doc != NULL) &&
((cur->parent->doc->intSubset != NULL) ||
(cur->parent->doc->extSubset != NULL))) {
if (xmlIsID(cur->parent->doc, cur->parent, cur))
xmlTextReaderRemoveID(cur->parent->doc, cur);
}
if (cur->children != NULL)
xmlTextReaderFreeNodeList(reader, cur->children);
DICT_FREE(cur->name);
if ((reader != NULL) && (reader->ctxt != NULL) &&
(reader->ctxt->freeAttrsNr < 100)) {
cur->next = reader->ctxt->freeAttrs;
reader->ctxt->freeAttrs = cur;
reader->ctxt->freeAttrsNr++;
} else {
xmlFree(cur);
}
}
/**
* xmlTextReaderFreePropList:
* @reader: the xmlTextReaderPtr used
* @cur: the first property in the list
*
* Free a property and all its siblings, all the children are freed too.
*/
static void
xmlTextReaderFreePropList(xmlTextReaderPtr reader, xmlAttrPtr cur) {
xmlAttrPtr next;
while (cur != NULL) {
next = cur->next;
xmlTextReaderFreeProp(reader, cur);
cur = next;
}
}
/**
* xmlTextReaderFreeNodeList:
* @reader: the xmlTextReaderPtr used
* @cur: the first node in the list
*
* Free a node and all its siblings, this is a recursive behaviour, all
* the children are freed too.
*/
static void
xmlTextReaderFreeNodeList(xmlTextReaderPtr reader, xmlNodePtr cur) {
xmlNodePtr next;
xmlNodePtr parent;
xmlDictPtr dict;
size_t depth = 0;
if ((reader != NULL) && (reader->ctxt != NULL))
dict = reader->ctxt->dict;
else
dict = NULL;
if (cur == NULL) return;
if (cur->type == XML_NAMESPACE_DECL) {
xmlFreeNsList((xmlNsPtr) cur);
return;
}
if ((cur->type == XML_DOCUMENT_NODE) ||
(cur->type == XML_HTML_DOCUMENT_NODE)) {
xmlFreeDoc((xmlDocPtr) cur);
return;
}
while (1) {
while ((cur->type != XML_DTD_NODE) &&
(cur->type != XML_ENTITY_REF_NODE) &&
(cur->children != NULL) &&
(cur->children->parent == cur)) {
cur = cur->children;
depth += 1;
}
next = cur->next;
parent = cur->parent;
/* unroll to speed up freeing the document */
if (cur->type != XML_DTD_NODE) {
if ((__xmlRegisterCallbacks) && (xmlDeregisterNodeDefaultValue))
xmlDeregisterNodeDefaultValue(cur);
if (((cur->type == XML_ELEMENT_NODE) ||
(cur->type == XML_XINCLUDE_START) ||
(cur->type == XML_XINCLUDE_END)) &&
(cur->properties != NULL))
xmlTextReaderFreePropList(reader, cur->properties);
if ((cur->content != (xmlChar *) &(cur->properties)) &&
(cur->type != XML_ELEMENT_NODE) &&
(cur->type != XML_XINCLUDE_START) &&
(cur->type != XML_XINCLUDE_END) &&
(cur->type != XML_ENTITY_REF_NODE)) {
DICT_FREE(cur->content);
}
if (((cur->type == XML_ELEMENT_NODE) ||
(cur->type == XML_XINCLUDE_START) ||
(cur->type == XML_XINCLUDE_END)) &&
(cur->nsDef != NULL))
xmlFreeNsList(cur->nsDef);
/*
* we don't free element names here they are interned now
*/
if ((cur->type != XML_TEXT_NODE) &&
(cur->type != XML_COMMENT_NODE))
DICT_FREE(cur->name);
if (((cur->type == XML_ELEMENT_NODE) ||
(cur->type == XML_TEXT_NODE)) &&
(reader != NULL) && (reader->ctxt != NULL) &&
(reader->ctxt->freeElemsNr < 100)) {
cur->next = reader->ctxt->freeElems;
reader->ctxt->freeElems = cur;
reader->ctxt->freeElemsNr++;
} else {
xmlFree(cur);
}
}
if (next != NULL) {
cur = next;
} else {
if ((depth == 0) || (parent == NULL))
break;
depth -= 1;
cur = parent;
cur->children = NULL;
}
}
}
/**
* xmlTextReaderFreeNode:
* @reader: the xmlTextReaderPtr used
* @cur: the node
*
* Free a node, this is a recursive behaviour, all the children are freed too.
* This doesn't unlink the child from the list, use xmlUnlinkNode() first.
*/
static void
xmlTextReaderFreeNode(xmlTextReaderPtr reader, xmlNodePtr cur) {
xmlDictPtr dict;
if ((reader != NULL) && (reader->ctxt != NULL))
dict = reader->ctxt->dict;
else
dict = NULL;
if (cur->type == XML_DTD_NODE) {
xmlFreeDtd((xmlDtdPtr) cur);
return;
}
if (cur->type == XML_NAMESPACE_DECL) {
xmlFreeNs((xmlNsPtr) cur);
return;
}
if (cur->type == XML_ATTRIBUTE_NODE) {
xmlTextReaderFreeProp(reader, (xmlAttrPtr) cur);
return;
}
if ((cur->children != NULL) &&
(cur->type != XML_ENTITY_REF_NODE)) {
if (cur->children->parent == cur)
xmlTextReaderFreeNodeList(reader, cur->children);
cur->children = NULL;
}
if ((__xmlRegisterCallbacks) && (xmlDeregisterNodeDefaultValue))
xmlDeregisterNodeDefaultValue(cur);
if (((cur->type == XML_ELEMENT_NODE) ||
(cur->type == XML_XINCLUDE_START) ||
(cur->type == XML_XINCLUDE_END)) &&
(cur->properties != NULL))
xmlTextReaderFreePropList(reader, cur->properties);
if ((cur->content != (xmlChar *) &(cur->properties)) &&
(cur->type != XML_ELEMENT_NODE) &&
(cur->type != XML_XINCLUDE_START) &&
(cur->type != XML_XINCLUDE_END) &&
(cur->type != XML_ENTITY_REF_NODE)) {
DICT_FREE(cur->content);
}
if (((cur->type == XML_ELEMENT_NODE) ||
(cur->type == XML_XINCLUDE_START) ||
(cur->type == XML_XINCLUDE_END)) &&
(cur->nsDef != NULL))
xmlFreeNsList(cur->nsDef);
/*
* we don't free names here they are interned now
*/
if ((cur->type != XML_TEXT_NODE) &&
(cur->type != XML_COMMENT_NODE))
DICT_FREE(cur->name);
if (((cur->type == XML_ELEMENT_NODE) ||
(cur->type == XML_TEXT_NODE)) &&
(reader != NULL) && (reader->ctxt != NULL) &&
(reader->ctxt->freeElemsNr < 100)) {
cur->next = reader->ctxt->freeElems;
reader->ctxt->freeElems = cur;
reader->ctxt->freeElemsNr++;
} else {
xmlFree(cur);
}
}
static void
xmlTextReaderFreeIDTableEntry(void *id, const xmlChar *name ATTRIBUTE_UNUSED) {
xmlFreeID((xmlIDPtr) id);
}
/**
* xmlTextReaderFreeIDTable:
* @table: An id table
*
* Deallocate the memory used by an ID hash table.
*/
static void
xmlTextReaderFreeIDTable(xmlIDTablePtr table) {
xmlHashFree(table, xmlTextReaderFreeIDTableEntry);
}
/**
* xmlTextReaderFreeDoc:
* @reader: the xmlTextReaderPtr used
* @cur: pointer to the document
*
* Free up all the structures used by a document, tree included.
*/
static void
xmlTextReaderFreeDoc(xmlTextReaderPtr reader, xmlDocPtr cur) {
xmlDtdPtr extSubset, intSubset;
if (cur == NULL) return;
if ((__xmlRegisterCallbacks) && (xmlDeregisterNodeDefaultValue))
xmlDeregisterNodeDefaultValue((xmlNodePtr) cur);
/*
* Do this before freeing the children list to avoid ID lookups
*/
if (cur->ids != NULL) xmlTextReaderFreeIDTable((xmlIDTablePtr) cur->ids);
cur->ids = NULL;
if (cur->refs != NULL) xmlFreeRefTable((xmlRefTablePtr) cur->refs);
cur->refs = NULL;
extSubset = cur->extSubset;
intSubset = cur->intSubset;
if (intSubset == extSubset)
extSubset = NULL;
if (extSubset != NULL) {
xmlUnlinkNode((xmlNodePtr) cur->extSubset);
cur->extSubset = NULL;
xmlFreeDtd(extSubset);
}
if (intSubset != NULL) {
xmlUnlinkNode((xmlNodePtr) cur->intSubset);
cur->intSubset = NULL;
xmlFreeDtd(intSubset);
}
if (cur->children != NULL) xmlTextReaderFreeNodeList(reader, cur->children);
if (cur->version != NULL) xmlFree((char *) cur->version);
if (cur->name != NULL) xmlFree((char *) cur->name);
if (cur->encoding != NULL) xmlFree((char *) cur->encoding);
if (cur->oldNs != NULL) xmlFreeNsList(cur->oldNs);
if (cur->URL != NULL) xmlFree((char *) cur->URL);
if (cur->dict != NULL) xmlDictFree(cur->dict);
xmlFree(cur);
}
/************************************************************************
* *
* The reader core parser *
* *
************************************************************************/
#ifdef DEBUG_READER
static void
xmlTextReaderDebug(xmlTextReaderPtr reader) {
if ((reader == NULL) || (reader->ctxt == NULL)) {
fprintf(stderr, "xmlTextReader NULL\n");
return;
}
fprintf(stderr, "xmlTextReader: state %d depth %d ",
reader->state, reader->depth);
if (reader->node == NULL) {
fprintf(stderr, "node = NULL\n");
} else {
fprintf(stderr, "node %s\n", reader->node->name);
}
fprintf(stderr, " input: base %d, cur %d, depth %d: ",
reader->base, reader->cur, reader->ctxt->nodeNr);
if (reader->input->buffer == NULL) {
fprintf(stderr, "buffer is NULL\n");
} else {
#ifdef LIBXML_DEBUG_ENABLED
xmlDebugDumpString(stderr,
&reader->input->buffer->content[reader->cur]);
#endif
fprintf(stderr, "\n");
}
}
#endif
/**
* xmlTextReaderEntPush:
* @reader: the xmlTextReaderPtr used
* @value: the entity reference node
*
* Pushes a new entity reference node on top of the entities stack
*
* Returns 0 in case of error, the index in the stack otherwise
*/
static int
xmlTextReaderEntPush(xmlTextReaderPtr reader, xmlNodePtr value)
{
if (reader->entMax <= 0) {
reader->entMax = 10;
reader->entTab = (xmlNodePtr *) xmlMalloc(reader->entMax *
sizeof(reader->entTab[0]));
if (reader->entTab == NULL) {
xmlGenericError(xmlGenericErrorContext, "xmlMalloc failed !\n");
return (0);
}
}
if (reader->entNr >= reader->entMax) {
reader->entMax *= 2;
reader->entTab =
(xmlNodePtr *) xmlRealloc(reader->entTab,
reader->entMax *
sizeof(reader->entTab[0]));
if (reader->entTab == NULL) {
xmlGenericError(xmlGenericErrorContext, "xmlRealloc failed !\n");
return (0);
}
}
reader->entTab[reader->entNr] = value;
reader->ent = value;
return (reader->entNr++);
}
/**
* xmlTextReaderEntPop:
* @reader: the xmlTextReaderPtr used
*
* Pops the top element entity from the entities stack
*
* Returns the entity just removed
*/
static xmlNodePtr
xmlTextReaderEntPop(xmlTextReaderPtr reader)
{
xmlNodePtr ret;
if (reader->entNr <= 0)
return (NULL);
reader->entNr--;
if (reader->entNr > 0)
reader->ent = reader->entTab[reader->entNr - 1];
else
reader->ent = NULL;
ret = reader->entTab[reader->entNr];
reader->entTab[reader->entNr] = NULL;
return (ret);
}
/**
* xmlTextReaderStartElement:
* @ctx: the user data (XML parser context)
* @fullname: The element name, including namespace prefix
* @atts: An array of name/value attributes pairs, NULL terminated
*
* called when an opening tag has been processed.
*/
static void
xmlTextReaderStartElement(void *ctx, const xmlChar *fullname,
const xmlChar **atts) {
xmlParserCtxtPtr ctxt = (xmlParserCtxtPtr) ctx;
xmlTextReaderPtr reader = ctxt->_private;
#ifdef DEBUG_CALLBACKS
printf("xmlTextReaderStartElement(%s)\n", fullname);
#endif
if ((reader != NULL) && (reader->startElement != NULL)) {
reader->startElement(ctx, fullname, atts);
if ((ctxt->node != NULL) && (ctxt->input != NULL) &&
(ctxt->input->cur != NULL) && (ctxt->input->cur[0] == '/') &&
(ctxt->input->cur[1] == '>'))
ctxt->node->extra = NODE_IS_EMPTY;
}
if (reader != NULL)
reader->state = XML_TEXTREADER_ELEMENT;
}
/**
* xmlTextReaderEndElement:
* @ctx: the user data (XML parser context)
* @fullname: The element name, including namespace prefix
*
* called when an ending tag has been processed.
*/
static void
xmlTextReaderEndElement(void *ctx, const xmlChar *fullname) {
xmlParserCtxtPtr ctxt = (xmlParserCtxtPtr) ctx;
xmlTextReaderPtr reader = ctxt->_private;
#ifdef DEBUG_CALLBACKS
printf("xmlTextReaderEndElement(%s)\n", fullname);
#endif
if ((reader != NULL) && (reader->endElement != NULL)) {
reader->endElement(ctx, fullname);
}
}
/**
* xmlTextReaderStartElementNs:
* @ctx: the user data (XML parser context)
* @localname: the local name of the element
* @prefix: the element namespace prefix if available
* @URI: the element namespace name if available
* @nb_namespaces: number of namespace definitions on that node
* @namespaces: pointer to the array of prefix/URI pairs namespace definitions
* @nb_attributes: the number of attributes on that node
* nb_defaulted: the number of defaulted attributes.
* @attributes: pointer to the array of (localname/prefix/URI/value/end)
* attribute values.
*
* called when an opening tag has been processed.
*/
static void
xmlTextReaderStartElementNs(void *ctx,
const xmlChar *localname,
const xmlChar *prefix,
const xmlChar *URI,
int nb_namespaces,
const xmlChar **namespaces,
int nb_attributes,
int nb_defaulted,
const xmlChar **attributes)
{
xmlParserCtxtPtr ctxt = (xmlParserCtxtPtr) ctx;
xmlTextReaderPtr reader = ctxt->_private;
#ifdef DEBUG_CALLBACKS
printf("xmlTextReaderStartElementNs(%s)\n", localname);
#endif
if ((reader != NULL) && (reader->startElementNs != NULL)) {
reader->startElementNs(ctx, localname, prefix, URI, nb_namespaces,
namespaces, nb_attributes, nb_defaulted,
attributes);
if ((ctxt->node != NULL) && (ctxt->input != NULL) &&
(ctxt->input->cur != NULL) && (ctxt->input->cur[0] == '/') &&
(ctxt->input->cur[1] == '>'))
ctxt->node->extra = NODE_IS_EMPTY;
}
if (reader != NULL)
reader->state = XML_TEXTREADER_ELEMENT;
}
/**
* xmlTextReaderEndElementNs:
* @ctx: the user data (XML parser context)
* @localname: the local name of the element
* @prefix: the element namespace prefix if available
* @URI: the element namespace name if available
*
* called when an ending tag has been processed.
*/
static void
xmlTextReaderEndElementNs(void *ctx,
const xmlChar * localname,
const xmlChar * prefix,
const xmlChar * URI)
{
xmlParserCtxtPtr ctxt = (xmlParserCtxtPtr) ctx;
xmlTextReaderPtr reader = ctxt->_private;
#ifdef DEBUG_CALLBACKS
printf("xmlTextReaderEndElementNs(%s)\n", localname);
#endif
if ((reader != NULL) && (reader->endElementNs != NULL)) {
reader->endElementNs(ctx, localname, prefix, URI);
}
}
/**
* xmlTextReaderCharacters:
* @ctx: the user data (XML parser context)
* @ch: a xmlChar string
* @len: the number of xmlChar
*
* receiving some chars from the parser.
*/
static void
xmlTextReaderCharacters(void *ctx, const xmlChar *ch, int len)
{
xmlParserCtxtPtr ctxt = (xmlParserCtxtPtr) ctx;
xmlTextReaderPtr reader = ctxt->_private;
#ifdef DEBUG_CALLBACKS
printf("xmlTextReaderCharacters()\n");
#endif
if ((reader != NULL) && (reader->characters != NULL)) {
reader->characters(ctx, ch, len);
}
}
/**
* xmlTextReaderCDataBlock:
* @ctx: the user data (XML parser context)
* @value: The pcdata content
* @len: the block length
*
* called when a pcdata block has been parsed
*/
static void
xmlTextReaderCDataBlock(void *ctx, const xmlChar *ch, int len)
{
xmlParserCtxtPtr ctxt = (xmlParserCtxtPtr) ctx;
xmlTextReaderPtr reader = ctxt->_private;
#ifdef DEBUG_CALLBACKS
printf("xmlTextReaderCDataBlock()\n");
#endif
if ((reader != NULL) && (reader->cdataBlock != NULL)) {
reader->cdataBlock(ctx, ch, len);
}
}
/**
* xmlTextReaderPushData:
* @reader: the xmlTextReaderPtr used
*
* Push data down the progressive parser until a significant callback
* got raised.
*
* Returns -1 in case of failure, 0 otherwise
*/
static int
xmlTextReaderPushData(xmlTextReaderPtr reader) {
xmlBufPtr inbuf;
int val, s;
xmlTextReaderState oldstate;
int alloc;
if ((reader->input == NULL) || (reader->input->buffer == NULL))
return(-1);
oldstate = reader->state;
reader->state = XML_TEXTREADER_NONE;
inbuf = reader->input->buffer;
alloc = xmlBufGetAllocationScheme(inbuf);
while (reader->state == XML_TEXTREADER_NONE) {
if (xmlBufUse(inbuf) < reader->cur + CHUNK_SIZE) {
/*
* Refill the buffer unless we are at the end of the stream
*/
if (reader->mode != XML_TEXTREADER_MODE_EOF) {
val = xmlParserInputBufferRead(reader->input, 4096);
if ((val == 0) &&
(alloc == XML_BUFFER_ALLOC_IMMUTABLE)) {
if (xmlBufUse(inbuf) == reader->cur) {
reader->mode = XML_TEXTREADER_MODE_EOF;
reader->state = oldstate;
}
} else if (val < 0) {
reader->mode = XML_TEXTREADER_MODE_EOF;
reader->state = oldstate;
if ((oldstate != XML_TEXTREADER_START) ||
(reader->ctxt->myDoc != NULL))
return(val);
} else if (val == 0) {
/* mark the end of the stream and process the remains */
reader->mode = XML_TEXTREADER_MODE_EOF;
break;
}
} else
break;
}
/*
* parse by block of CHUNK_SIZE bytes, various tests show that
* it's the best tradeoff at least on a 1.2GH Duron
*/
if (xmlBufUse(inbuf) >= reader->cur + CHUNK_SIZE) {
val = xmlParseChunk(reader->ctxt,
(const char *) xmlBufContent(inbuf) + reader->cur,
CHUNK_SIZE, 0);
reader->cur += CHUNK_SIZE;
if (val != 0)
reader->ctxt->wellFormed = 0;
if (reader->ctxt->wellFormed == 0)
break;
} else {
s = xmlBufUse(inbuf) - reader->cur;
val = xmlParseChunk(reader->ctxt,
(const char *) xmlBufContent(inbuf) + reader->cur,
s, 0);
reader->cur += s;
if (val != 0)
reader->ctxt->wellFormed = 0;
break;
}
}
/*
* Discard the consumed input when needed and possible
*/
if (reader->mode == XML_TEXTREADER_MODE_INTERACTIVE) {
if (alloc != XML_BUFFER_ALLOC_IMMUTABLE) {
if ((reader->cur >= 4096) &&
(xmlBufUse(inbuf) - reader->cur <= CHUNK_SIZE)) {
val = xmlBufShrink(inbuf, reader->cur);
if (val >= 0) {
reader->cur -= val;
}
}
}
}
/*
* At the end of the stream signal that the work is done to the Push
* parser.
*/
else if (reader->mode == XML_TEXTREADER_MODE_EOF) {
if (reader->state != XML_TEXTREADER_DONE) {
s = xmlBufUse(inbuf) - reader->cur;
val = xmlParseChunk(reader->ctxt,
(const char *) xmlBufContent(inbuf) + reader->cur,
s, 1);
reader->cur = xmlBufUse(inbuf);
reader->state = XML_TEXTREADER_DONE;
if (val != 0) {
if (reader->ctxt->wellFormed)
reader->ctxt->wellFormed = 0;
else
return(-1);
}
}
}
reader->state = oldstate;
if (reader->ctxt->wellFormed == 0) {
reader->mode = XML_TEXTREADER_MODE_EOF;
return(-1);
}
return(0);
}
#ifdef LIBXML_REGEXP_ENABLED
/**
* xmlTextReaderValidatePush:
* @reader: the xmlTextReaderPtr used
*
* Push the current node for validation
*/
static void
xmlTextReaderValidatePush(xmlTextReaderPtr reader ATTRIBUTE_UNUSED) {
xmlNodePtr node = reader->node;
#ifdef LIBXML_VALID_ENABLED
if ((reader->validate == XML_TEXTREADER_VALIDATE_DTD) &&
(reader->ctxt != NULL) && (reader->ctxt->validate == 1)) {
if ((node->ns == NULL) || (node->ns->prefix == NULL)) {
reader->ctxt->valid &= xmlValidatePushElement(&reader->ctxt->vctxt,
reader->ctxt->myDoc, node, node->name);
} else {
/* TODO use the BuildQName interface */
xmlChar *qname;
qname = xmlStrdup(node->ns->prefix);
qname = xmlStrcat(qname, BAD_CAST ":");
qname = xmlStrcat(qname, node->name);
reader->ctxt->valid &= xmlValidatePushElement(&reader->ctxt->vctxt,
reader->ctxt->myDoc, node, qname);
if (qname != NULL)
xmlFree(qname);
}
}
#endif /* LIBXML_VALID_ENABLED */
#ifdef LIBXML_SCHEMAS_ENABLED
if ((reader->validate == XML_TEXTREADER_VALIDATE_RNG) &&
(reader->rngValidCtxt != NULL)) {
int ret;
if (reader->rngFullNode != NULL) return;
ret = xmlRelaxNGValidatePushElement(reader->rngValidCtxt,
reader->ctxt->myDoc,
node);
if (ret == 0) {
/*
* this element requires a full tree
*/
node = xmlTextReaderExpand(reader);
if (node == NULL) {
ret = -1;
} else {
ret = xmlRelaxNGValidateFullElement(reader->rngValidCtxt,
reader->ctxt->myDoc,
node);
reader->rngFullNode = node;
}
}
if (ret != 1)
reader->rngValidErrors++;
}
#endif
}
/**
* xmlTextReaderValidateCData:
* @reader: the xmlTextReaderPtr used
* @data: pointer to the CData
* @len: length of the CData block in bytes.
*
* Push some CData for validation
*/
static void
xmlTextReaderValidateCData(xmlTextReaderPtr reader,
const xmlChar *data, int len) {
#ifdef LIBXML_VALID_ENABLED
if ((reader->validate == XML_TEXTREADER_VALIDATE_DTD) &&
(reader->ctxt != NULL) && (reader->ctxt->validate == 1)) {
reader->ctxt->valid &= xmlValidatePushCData(&reader->ctxt->vctxt,
data, len);
}
#endif /* LIBXML_VALID_ENABLED */
#ifdef LIBXML_SCHEMAS_ENABLED
if ((reader->validate == XML_TEXTREADER_VALIDATE_RNG) &&
(reader->rngValidCtxt != NULL)) {
int ret;
if (reader->rngFullNode != NULL) return;
ret = xmlRelaxNGValidatePushCData(reader->rngValidCtxt, data, len);
if (ret != 1)
reader->rngValidErrors++;
}
#endif
}
/**
* xmlTextReaderValidatePop:
* @reader: the xmlTextReaderPtr used
*
* Pop the current node from validation
*/
static void
xmlTextReaderValidatePop(xmlTextReaderPtr reader) {
xmlNodePtr node = reader->node;
#ifdef LIBXML_VALID_ENABLED
if ((reader->validate == XML_TEXTREADER_VALIDATE_DTD) &&
(reader->ctxt != NULL) && (reader->ctxt->validate == 1)) {
if ((node->ns == NULL) || (node->ns->prefix == NULL)) {
reader->ctxt->valid &= xmlValidatePopElement(&reader->ctxt->vctxt,
reader->ctxt->myDoc, node, node->name);
} else {
/* TODO use the BuildQName interface */
xmlChar *qname;
qname = xmlStrdup(node->ns->prefix);
qname = xmlStrcat(qname, BAD_CAST ":");
qname = xmlStrcat(qname, node->name);
reader->ctxt->valid &= xmlValidatePopElement(&reader->ctxt->vctxt,
reader->ctxt->myDoc, node, qname);
if (qname != NULL)
xmlFree(qname);
}
}
#endif /* LIBXML_VALID_ENABLED */
#ifdef LIBXML_SCHEMAS_ENABLED
if ((reader->validate == XML_TEXTREADER_VALIDATE_RNG) &&
(reader->rngValidCtxt != NULL)) {
int ret;
if (reader->rngFullNode != NULL) {
if (node == reader->rngFullNode)
reader->rngFullNode = NULL;
return;
}
ret = xmlRelaxNGValidatePopElement(reader->rngValidCtxt,
reader->ctxt->myDoc,
node);
if (ret != 1)
reader->rngValidErrors++;
}
#endif
}
/**
* xmlTextReaderValidateEntity:
* @reader: the xmlTextReaderPtr used
*
* Handle the validation when an entity reference is encountered and
* entity substitution is not activated. As a result the parser interface
* must walk through the entity and do the validation calls
*/
static void
xmlTextReaderValidateEntity(xmlTextReaderPtr reader) {
xmlNodePtr oldnode = reader->node;
xmlNodePtr node = reader->node;
do {
if (node->type == XML_ENTITY_REF_NODE) {
if ((node->children != NULL) &&
(node->children->type == XML_ENTITY_DECL) &&
(node->children->children != NULL)) {
xmlTextReaderEntPush(reader, node);
node = node->children->children;
continue;
} else {
/*
* The error has probably been raised already.
*/
if (node == oldnode)
break;
goto skip_children;
}
#ifdef LIBXML_REGEXP_ENABLED
} else if (node->type == XML_ELEMENT_NODE) {
reader->node = node;
xmlTextReaderValidatePush(reader);
} else if ((node->type == XML_TEXT_NODE) ||
(node->type == XML_CDATA_SECTION_NODE)) {
xmlTextReaderValidateCData(reader, node->content,
xmlStrlen(node->content));
#endif
}
/*
* go to next node
*/
if (node->children != NULL) {
node = node->children;
continue;
} else if (node->type == XML_ELEMENT_NODE) {
xmlTextReaderValidatePop(reader);
}
skip_children:
if (node->next != NULL) {
node = node->next;
continue;
}
do {
node = node->parent;
if (node->type == XML_ELEMENT_NODE) {
xmlNodePtr tmp;
if (reader->entNr == 0) {
while ((tmp = node->last) != NULL) {
if ((tmp->extra & NODE_IS_PRESERVED) == 0) {
xmlUnlinkNode(tmp);
xmlTextReaderFreeNode(reader, tmp);
} else
break;
}
}
reader->node = node;
xmlTextReaderValidatePop(reader);
}
if ((node->type == XML_ENTITY_DECL) &&
(reader->ent != NULL) && (reader->ent->children == node)) {
node = xmlTextReaderEntPop(reader);
}
if (node == oldnode)
break;
if (node->next != NULL) {
node = node->next;
break;
}
} while ((node != NULL) && (node != oldnode));
} while ((node != NULL) && (node != oldnode));
reader->node = oldnode;
}
#endif /* LIBXML_REGEXP_ENABLED */
/**
* xmlTextReaderGetSuccessor:
* @cur: the current node
*
* Get the successor of a node if available.
*
* Returns the successor node or NULL
*/
static xmlNodePtr
xmlTextReaderGetSuccessor(xmlNodePtr cur) {
if (cur == NULL) return(NULL) ; /* ERROR */
if (cur->next != NULL) return(cur->next) ;
do {
cur = cur->parent;
if (cur == NULL) break;
if (cur->next != NULL) return(cur->next);
} while (cur != NULL);
return(cur);
}
/**
* xmlTextReaderDoExpand:
* @reader: the xmlTextReaderPtr used
*
* Makes sure that the current node is fully read as well as all its
* descendant. It means the full DOM subtree must be available at the
* end of the call.
*
* Returns 1 if the node was expanded successfully, 0 if there is no more
* nodes to read, or -1 in case of error
*/
static int
xmlTextReaderDoExpand(xmlTextReaderPtr reader) {
int val;
if ((reader == NULL) || (reader->node == NULL) || (reader->ctxt == NULL))
return(-1);
do {
if (reader->ctxt->instate == XML_PARSER_EOF) return(1);
if (xmlTextReaderGetSuccessor(reader->node) != NULL)
return(1);
if (reader->ctxt->nodeNr < reader->depth)
return(1);
if (reader->mode == XML_TEXTREADER_MODE_EOF)
return(1);
val = xmlTextReaderPushData(reader);
if (val < 0){
reader->mode = XML_TEXTREADER_MODE_ERROR;
return(-1);
}
} while(reader->mode != XML_TEXTREADER_MODE_EOF);
return(1);
}
/**
* xmlTextReaderCollectSiblings:
* @node: the first child
*
* Traverse depth-first through all sibling nodes and their children
* nodes and concatenate their content. This is an auxiliary function
* to xmlTextReaderReadString.
*
* Returns a string containing the content, or NULL in case of error.
*/
static xmlChar *
xmlTextReaderCollectSiblings(xmlNodePtr node)
{
xmlBufferPtr buffer;
xmlChar *ret;
if ((node == NULL) || (node->type == XML_NAMESPACE_DECL))
return(NULL);
buffer = xmlBufferCreate();
if (buffer == NULL)
return NULL;
for ( ; node != NULL; node = node->next) {
switch (node->type) {
case XML_TEXT_NODE:
case XML_CDATA_SECTION_NODE:
xmlBufferCat(buffer, node->content);
break;
case XML_ELEMENT_NODE: {
xmlChar *tmp;
tmp = xmlTextReaderCollectSiblings(node->children);
xmlBufferCat(buffer, tmp);
xmlFree(tmp);
break;
}
default:
break;
}
}
ret = buffer->content;
buffer->content = NULL;
xmlBufferFree(buffer);
return(ret);
}
/**
* xmlTextReaderRead:
* @reader: the xmlTextReaderPtr used
*
* Moves the position of the current instance to the next node in
* the stream, exposing its properties.
*
* Returns 1 if the node was read successfully, 0 if there is no more
* nodes to read, or -1 in case of error
*/
int
xmlTextReaderRead(xmlTextReaderPtr reader) {
int val, olddepth = 0;
xmlTextReaderState oldstate = XML_TEXTREADER_START;
xmlNodePtr oldnode = NULL;
if (reader == NULL)
return(-1);
reader->curnode = NULL;
if (reader->doc != NULL)
return(xmlTextReaderReadTree(reader));
if (reader->ctxt == NULL)
return(-1);
#ifdef DEBUG_READER
fprintf(stderr, "\nREAD ");
DUMP_READER
#endif
if (reader->mode == XML_TEXTREADER_MODE_INITIAL) {
reader->mode = XML_TEXTREADER_MODE_INTERACTIVE;
/*
* Initial state
*/
do {
val = xmlTextReaderPushData(reader);
if (val < 0){
reader->mode = XML_TEXTREADER_MODE_ERROR;
reader->state = XML_TEXTREADER_ERROR;
return(-1);
}
} while ((reader->ctxt->node == NULL) &&
((reader->mode != XML_TEXTREADER_MODE_EOF) &&
(reader->state != XML_TEXTREADER_DONE)));
if (reader->ctxt->node == NULL) {
if (reader->ctxt->myDoc != NULL) {
reader->node = reader->ctxt->myDoc->children;
}
if (reader->node == NULL){
reader->mode = XML_TEXTREADER_MODE_ERROR;
reader->state = XML_TEXTREADER_ERROR;
return(-1);
}
reader->state = XML_TEXTREADER_ELEMENT;
} else {
if (reader->ctxt->myDoc != NULL) {
reader->node = reader->ctxt->myDoc->children;
}
if (reader->node == NULL)
reader->node = reader->ctxt->nodeTab[0];
reader->state = XML_TEXTREADER_ELEMENT;
}
reader->depth = 0;
reader->ctxt->parseMode = XML_PARSE_READER;
goto node_found;
}
oldstate = reader->state;
olddepth = reader->ctxt->nodeNr;
oldnode = reader->node;
get_next_node:
if (reader->node == NULL) {
if (reader->mode == XML_TEXTREADER_MODE_EOF)
return(0);
else
return(-1);
}
/*
* If we are not backtracking on ancestors or examined nodes,
* that the parser didn't finished or that we aren't at the end
* of stream, continue processing.
*/
while ((reader->node != NULL) && (reader->node->next == NULL) &&
(reader->ctxt->nodeNr == olddepth) &&
((oldstate == XML_TEXTREADER_BACKTRACK) ||
(reader->node->children == NULL) ||
(reader->node->type == XML_ENTITY_REF_NODE) ||
((reader->node->children != NULL) &&
(reader->node->children->type == XML_TEXT_NODE) &&
(reader->node->children->next == NULL)) ||
(reader->node->type == XML_DTD_NODE) ||
(reader->node->type == XML_DOCUMENT_NODE) ||
(reader->node->type == XML_HTML_DOCUMENT_NODE)) &&
((reader->ctxt->node == NULL) ||
(reader->ctxt->node == reader->node) ||
(reader->ctxt->node == reader->node->parent)) &&
(reader->ctxt->instate != XML_PARSER_EOF)) {
val = xmlTextReaderPushData(reader);
if (val < 0){
reader->mode = XML_TEXTREADER_MODE_ERROR;
reader->state = XML_TEXTREADER_ERROR;
return(-1);
}
if (reader->node == NULL)
goto node_end;
}
if (oldstate != XML_TEXTREADER_BACKTRACK) {
if ((reader->node->children != NULL) &&
(reader->node->type != XML_ENTITY_REF_NODE) &&
(reader->node->type != XML_XINCLUDE_START) &&
(reader->node->type != XML_DTD_NODE)) {
reader->node = reader->node->children;
reader->depth++;
reader->state = XML_TEXTREADER_ELEMENT;
goto node_found;
}
}
if (reader->node->next != NULL) {
if ((oldstate == XML_TEXTREADER_ELEMENT) &&
(reader->node->type == XML_ELEMENT_NODE) &&
(reader->node->children == NULL) &&
((reader->node->extra & NODE_IS_EMPTY) == 0)
#ifdef LIBXML_XINCLUDE_ENABLED
&& (reader->in_xinclude <= 0)
#endif
) {
reader->state = XML_TEXTREADER_END;
goto node_found;
}
#ifdef LIBXML_REGEXP_ENABLED
if ((reader->validate) &&
(reader->node->type == XML_ELEMENT_NODE))
xmlTextReaderValidatePop(reader);
#endif /* LIBXML_REGEXP_ENABLED */
if ((reader->preserves > 0) &&
(reader->node->extra & NODE_IS_SPRESERVED))
reader->preserves--;
reader->node = reader->node->next;
reader->state = XML_TEXTREADER_ELEMENT;
/*
* Cleanup of the old node
*/
if ((reader->preserves == 0) &&
#ifdef LIBXML_XINCLUDE_ENABLED
(reader->in_xinclude == 0) &&
#endif
(reader->entNr == 0) &&
(reader->node->prev != NULL) &&
(reader->node->prev->type != XML_DTD_NODE)) {
xmlNodePtr tmp = reader->node->prev;
if ((tmp->extra & NODE_IS_PRESERVED) == 0) {
xmlUnlinkNode(tmp);
xmlTextReaderFreeNode(reader, tmp);
}
}
goto node_found;
}
if ((oldstate == XML_TEXTREADER_ELEMENT) &&
(reader->node->type == XML_ELEMENT_NODE) &&
(reader->node->children == NULL) &&
((reader->node->extra & NODE_IS_EMPTY) == 0)) {;
reader->state = XML_TEXTREADER_END;
goto node_found;
}
#ifdef LIBXML_REGEXP_ENABLED
if ((reader->validate != XML_TEXTREADER_NOT_VALIDATE) && (reader->node->type == XML_ELEMENT_NODE))
xmlTextReaderValidatePop(reader);
#endif /* LIBXML_REGEXP_ENABLED */
if ((reader->preserves > 0) &&
(reader->node->extra & NODE_IS_SPRESERVED))
reader->preserves--;
reader->node = reader->node->parent;
if ((reader->node == NULL) ||
(reader->node->type == XML_DOCUMENT_NODE) ||
#ifdef LIBXML_DOCB_ENABLED
(reader->node->type == XML_DOCB_DOCUMENT_NODE) ||
#endif
(reader->node->type == XML_HTML_DOCUMENT_NODE)) {
if (reader->mode != XML_TEXTREADER_MODE_EOF) {
val = xmlParseChunk(reader->ctxt, "", 0, 1);
reader->state = XML_TEXTREADER_DONE;
if (val != 0)
return(-1);
}
reader->node = NULL;
reader->depth = -1;
/*
* Cleanup of the old node
*/
if ((oldnode != NULL) && (reader->preserves == 0) &&
#ifdef LIBXML_XINCLUDE_ENABLED
(reader->in_xinclude == 0) &&
#endif
(reader->entNr == 0) &&
(oldnode->type != XML_DTD_NODE) &&
((oldnode->extra & NODE_IS_PRESERVED) == 0)) {
xmlUnlinkNode(oldnode);
xmlTextReaderFreeNode(reader, oldnode);
}
goto node_end;
}
if ((reader->preserves == 0) &&
#ifdef LIBXML_XINCLUDE_ENABLED
(reader->in_xinclude == 0) &&
#endif
(reader->entNr == 0) &&
(reader->node->last != NULL) &&
((reader->node->last->extra & NODE_IS_PRESERVED) == 0)) {
xmlNodePtr tmp = reader->node->last;
xmlUnlinkNode(tmp);
xmlTextReaderFreeNode(reader, tmp);
}
reader->depth--;
reader->state = XML_TEXTREADER_BACKTRACK;
node_found:
DUMP_READER
/*
* If we are in the middle of a piece of CDATA make sure it's finished
*/
if ((reader->node != NULL) &&
(reader->node->next == NULL) &&
((reader->node->type == XML_TEXT_NODE) ||
(reader->node->type == XML_CDATA_SECTION_NODE))) {
if (xmlTextReaderExpand(reader) == NULL)
return -1;
}
#ifdef LIBXML_XINCLUDE_ENABLED
/*
* Handle XInclude if asked for
*/
if ((reader->xinclude) && (reader->node != NULL) &&
(reader->node->type == XML_ELEMENT_NODE) &&
(reader->node->ns != NULL) &&
((xmlStrEqual(reader->node->ns->href, XINCLUDE_NS)) ||
(xmlStrEqual(reader->node->ns->href, XINCLUDE_OLD_NS)))) {
if (reader->xincctxt == NULL) {
reader->xincctxt = xmlXIncludeNewContext(reader->ctxt->myDoc);
xmlXIncludeSetFlags(reader->xincctxt,
reader->parserFlags & (~XML_PARSE_NOXINCNODE));
}
/*
* expand that node and process it
*/
if (xmlTextReaderExpand(reader) == NULL)
return -1;
xmlXIncludeProcessNode(reader->xincctxt, reader->node);
}
if ((reader->node != NULL) && (reader->node->type == XML_XINCLUDE_START)) {
reader->in_xinclude++;
goto get_next_node;
}
if ((reader->node != NULL) && (reader->node->type == XML_XINCLUDE_END)) {
reader->in_xinclude--;
goto get_next_node;
}
#endif
/*
* Handle entities enter and exit when in entity replacement mode
*/
if ((reader->node != NULL) &&
(reader->node->type == XML_ENTITY_REF_NODE) &&
(reader->ctxt != NULL) && (reader->ctxt->replaceEntities == 1)) {
if ((reader->node->children != NULL) &&
(reader->node->children->type == XML_ENTITY_DECL) &&
(reader->node->children->children != NULL)) {
xmlTextReaderEntPush(reader, reader->node);
reader->node = reader->node->children->children;
}
#ifdef LIBXML_REGEXP_ENABLED
} else if ((reader->node != NULL) &&
(reader->node->type == XML_ENTITY_REF_NODE) &&
(reader->ctxt != NULL) && (reader->validate)) {
xmlTextReaderValidateEntity(reader);
#endif /* LIBXML_REGEXP_ENABLED */
}
if ((reader->node != NULL) &&
(reader->node->type == XML_ENTITY_DECL) &&
(reader->ent != NULL) && (reader->ent->children == reader->node)) {
reader->node = xmlTextReaderEntPop(reader);
reader->depth++;
goto get_next_node;
}
#ifdef LIBXML_REGEXP_ENABLED
if ((reader->validate != XML_TEXTREADER_NOT_VALIDATE) && (reader->node != NULL)) {
xmlNodePtr node = reader->node;
if ((node->type == XML_ELEMENT_NODE) &&
((reader->state != XML_TEXTREADER_END) &&
(reader->state != XML_TEXTREADER_BACKTRACK))) {
xmlTextReaderValidatePush(reader);
} else if ((node->type == XML_TEXT_NODE) ||
(node->type == XML_CDATA_SECTION_NODE)) {
xmlTextReaderValidateCData(reader, node->content,
xmlStrlen(node->content));
}
}
#endif /* LIBXML_REGEXP_ENABLED */
#ifdef LIBXML_PATTERN_ENABLED
if ((reader->patternNr > 0) && (reader->state != XML_TEXTREADER_END) &&
(reader->state != XML_TEXTREADER_BACKTRACK)) {
int i;
for (i = 0;i < reader->patternNr;i++) {
if (xmlPatternMatch(reader->patternTab[i], reader->node) == 1) {
xmlTextReaderPreserve(reader);
break;
}
}
}
#endif /* LIBXML_PATTERN_ENABLED */
#ifdef LIBXML_SCHEMAS_ENABLED
if ((reader->validate == XML_TEXTREADER_VALIDATE_XSD) &&
(reader->xsdValidErrors == 0) &&
(reader->xsdValidCtxt != NULL)) {
reader->xsdValidErrors = !xmlSchemaIsValid(reader->xsdValidCtxt);
}
#endif /* LIBXML_PATTERN_ENABLED */
return(1);
node_end:
reader->state = XML_TEXTREADER_DONE;
return(0);
}
/**
* xmlTextReaderReadState:
* @reader: the xmlTextReaderPtr used
*
* Gets the read state of the reader.
*
* Returns the state value, or -1 in case of error
*/
int
xmlTextReaderReadState(xmlTextReaderPtr reader) {
if (reader == NULL)
return(-1);
return(reader->mode);
}
/**
* xmlTextReaderExpand:
* @reader: the xmlTextReaderPtr used
*
* Reads the contents of the current node and the full subtree. It then makes
* the subtree available until the next xmlTextReaderRead() call
*
* Returns a node pointer valid until the next xmlTextReaderRead() call
* or NULL in case of error.
*/
xmlNodePtr
xmlTextReaderExpand(xmlTextReaderPtr reader) {
if ((reader == NULL) || (reader->node == NULL))
return(NULL);
if (reader->doc != NULL)
return(reader->node);
if (reader->ctxt == NULL)
return(NULL);
if (xmlTextReaderDoExpand(reader) < 0)
return(NULL);
return(reader->node);
}
/**
* xmlTextReaderNext:
* @reader: the xmlTextReaderPtr used
*
* Skip to the node following the current one in document order while
* avoiding the subtree if any.
*
* Returns 1 if the node was read successfully, 0 if there is no more
* nodes to read, or -1 in case of error
*/
int
xmlTextReaderNext(xmlTextReaderPtr reader) {
int ret;
xmlNodePtr cur;
if (reader == NULL)
return(-1);
if (reader->doc != NULL)
return(xmlTextReaderNextTree(reader));
cur = reader->node;
if ((cur == NULL) || (cur->type != XML_ELEMENT_NODE))
return(xmlTextReaderRead(reader));
if (reader->state == XML_TEXTREADER_END || reader->state == XML_TEXTREADER_BACKTRACK)
return(xmlTextReaderRead(reader));
if (cur->extra & NODE_IS_EMPTY)
return(xmlTextReaderRead(reader));
do {
ret = xmlTextReaderRead(reader);
if (ret != 1)
return(ret);
} while (reader->node != cur);
return(xmlTextReaderRead(reader));
}
#ifdef LIBXML_WRITER_ENABLED
/**
* xmlTextReaderReadInnerXml:
* @reader: the xmlTextReaderPtr used
*
* Reads the contents of the current node, including child nodes and markup.
*
* Returns a string containing the XML content, or NULL if the current node
* is neither an element nor attribute, or has no child nodes. The
* string must be deallocated by the caller.
*/
xmlChar *
xmlTextReaderReadInnerXml(xmlTextReaderPtr reader ATTRIBUTE_UNUSED)
{
xmlChar *resbuf;
xmlNodePtr node, cur_node;
xmlBufferPtr buff, buff2;
xmlDocPtr doc;
if (xmlTextReaderExpand(reader) == NULL) {
return NULL;
}
doc = reader->node->doc;
buff = xmlBufferCreate();
if (buff == NULL)
return NULL;
for (cur_node = reader->node->children; cur_node != NULL;
cur_node = cur_node->next) {
/* XXX: Why is the node copied? */
node = xmlDocCopyNode(cur_node, doc, 1);
buff2 = xmlBufferCreate();
if (xmlNodeDump(buff2, doc, node, 0, 0) == -1) {
xmlFreeNode(node);
xmlBufferFree(buff2);
xmlBufferFree(buff);
return NULL;
}
xmlBufferCat(buff, buff2->content);
xmlFreeNode(node);
xmlBufferFree(buff2);
}
resbuf = buff->content;
buff->content = NULL;
xmlBufferFree(buff);
return resbuf;
}
#endif
#ifdef LIBXML_WRITER_ENABLED
/**
* xmlTextReaderReadOuterXml:
* @reader: the xmlTextReaderPtr used
*
* Reads the contents of the current node, including child nodes and markup.
*
* Returns a string containing the node and any XML content, or NULL if the
* current node cannot be serialized. The string must be deallocated
* by the caller.
*/
xmlChar *
xmlTextReaderReadOuterXml(xmlTextReaderPtr reader ATTRIBUTE_UNUSED)
{
xmlChar *resbuf;
xmlNodePtr node;
xmlBufferPtr buff;
xmlDocPtr doc;
if (xmlTextReaderExpand(reader) == NULL) {
return NULL;
}
node = reader->node;
doc = node->doc;
/* XXX: Why is the node copied? */
if (node->type == XML_DTD_NODE) {
node = (xmlNodePtr) xmlCopyDtd((xmlDtdPtr) node);
} else {
node = xmlDocCopyNode(node, doc, 1);
}
buff = xmlBufferCreate();
if (xmlNodeDump(buff, doc, node, 0, 0) == -1) {
xmlFreeNode(node);
xmlBufferFree(buff);
return NULL;
}
resbuf = buff->content;
buff->content = NULL;
xmlFreeNode(node);
xmlBufferFree(buff);
return resbuf;
}
#endif
/**
* xmlTextReaderReadString:
* @reader: the xmlTextReaderPtr used
*
* Reads the contents of an element or a text node as a string.
*
* Returns a string containing the contents of the Element or Text node,
* or NULL if the reader is positioned on any other type of node.
* The string must be deallocated by the caller.
*/
xmlChar *
xmlTextReaderReadString(xmlTextReaderPtr reader)
{
xmlNodePtr node;
if ((reader == NULL) || (reader->node == NULL))
return(NULL);
node = (reader->curnode != NULL) ? reader->curnode : reader->node;
switch (node->type) {
case XML_TEXT_NODE:
if (node->content != NULL)
return(xmlStrdup(node->content));
break;
case XML_ELEMENT_NODE:
if (xmlTextReaderDoExpand(reader) != -1) {
return xmlTextReaderCollectSiblings(node->children);
}
break;
case XML_ATTRIBUTE_NODE:
TODO
break;
default:
break;
}
return(NULL);
}
#if 0
/**
* xmlTextReaderReadBase64:
* @reader: the xmlTextReaderPtr used
* @array: a byte array to store the content.
* @offset: the zero-based index into array where the method should
* begin to write.
* @len: the number of bytes to write.
*
* Reads and decodes the Base64 encoded contents of an element and
* stores the result in a byte buffer.
*
* Returns the number of bytes written to array, or zero if the current
* instance is not positioned on an element or -1 in case of error.
*/
int
xmlTextReaderReadBase64(xmlTextReaderPtr reader,
unsigned char *array ATTRIBUTE_UNUSED,
int offset ATTRIBUTE_UNUSED,
int len ATTRIBUTE_UNUSED) {
if ((reader == NULL) || (reader->ctxt == NULL))
return(-1);
if (reader->ctxt->wellFormed != 1)
return(-1);
if ((reader->node == NULL) || (reader->node->type == XML_ELEMENT_NODE))
return(0);
TODO
return(0);
}
/**
* xmlTextReaderReadBinHex:
* @reader: the xmlTextReaderPtr used
* @array: a byte array to store the content.
* @offset: the zero-based index into array where the method should
* begin to write.
* @len: the number of bytes to write.
*
* Reads and decodes the BinHex encoded contents of an element and
* stores the result in a byte buffer.
*
* Returns the number of bytes written to array, or zero if the current
* instance is not positioned on an element or -1 in case of error.
*/
int
xmlTextReaderReadBinHex(xmlTextReaderPtr reader,
unsigned char *array ATTRIBUTE_UNUSED,
int offset ATTRIBUTE_UNUSED,
int len ATTRIBUTE_UNUSED) {
if ((reader == NULL) || (reader->ctxt == NULL))
return(-1);
if (reader->ctxt->wellFormed != 1)
return(-1);
if ((reader->node == NULL) || (reader->node->type == XML_ELEMENT_NODE))
return(0);
TODO
return(0);
}
#endif
/************************************************************************
* *
* Operating on a preparsed tree *
* *
************************************************************************/
static int
xmlTextReaderNextTree(xmlTextReaderPtr reader)
{
if (reader == NULL)
return(-1);
if (reader->state == XML_TEXTREADER_END)
return(0);
if (reader->node == NULL) {
if (reader->doc->children == NULL) {
reader->state = XML_TEXTREADER_END;
return(0);
}
reader->node = reader->doc->children;
reader->state = XML_TEXTREADER_START;
return(1);
}
if (reader->state != XML_TEXTREADER_BACKTRACK) {
/* Here removed traversal to child, because we want to skip the subtree,
replace with traversal to sibling to skip subtree */
if (reader->node->next != 0) {
/* Move to sibling if present,skipping sub-tree */
reader->node = reader->node->next;
reader->state = XML_TEXTREADER_START;
return(1);
}
/* if reader->node->next is NULL mean no subtree for current node,
so need to move to sibling of parent node if present */
reader->state = XML_TEXTREADER_BACKTRACK;
/* This will move to parent if present */
xmlTextReaderRead(reader);
}
if (reader->node->next != 0) {
reader->node = reader->node->next;
reader->state = XML_TEXTREADER_START;
return(1);
}
if (reader->node->parent != 0) {
if (reader->node->parent->type == XML_DOCUMENT_NODE) {
reader->state = XML_TEXTREADER_END;
return(0);
}
reader->node = reader->node->parent;
reader->depth--;
reader->state = XML_TEXTREADER_BACKTRACK;
/* Repeat process to move to sibling of parent node if present */
xmlTextReaderNextTree(reader);
}
reader->state = XML_TEXTREADER_END;
return(1);
}
/**
* xmlTextReaderReadTree:
* @reader: the xmlTextReaderPtr used
*
* Moves the position of the current instance to the next node in
* the stream, exposing its properties.
*
* Returns 1 if the node was read successfully, 0 if there is no more
* nodes to read, or -1 in case of error
*/
static int
xmlTextReaderReadTree(xmlTextReaderPtr reader) {
if (reader->state == XML_TEXTREADER_END)
return(0);
next_node:
if (reader->node == NULL) {
if (reader->doc->children == NULL) {
reader->state = XML_TEXTREADER_END;
return(0);
}
reader->node = reader->doc->children;
reader->state = XML_TEXTREADER_START;
goto found_node;
}
if ((reader->state != XML_TEXTREADER_BACKTRACK) &&
(reader->node->type != XML_DTD_NODE) &&
(reader->node->type != XML_XINCLUDE_START) &&
(reader->node->type != XML_ENTITY_REF_NODE)) {
if (reader->node->children != NULL) {
reader->node = reader->node->children;
reader->depth++;
reader->state = XML_TEXTREADER_START;
goto found_node;
}
if (reader->node->type == XML_ATTRIBUTE_NODE) {
reader->state = XML_TEXTREADER_BACKTRACK;
goto found_node;
}
}
if (reader->node->next != NULL) {
reader->node = reader->node->next;
reader->state = XML_TEXTREADER_START;
goto found_node;
}
if (reader->node->parent != NULL) {
if ((reader->node->parent->type == XML_DOCUMENT_NODE) ||
(reader->node->parent->type == XML_HTML_DOCUMENT_NODE)) {
reader->state = XML_TEXTREADER_END;
return(0);
}
reader->node = reader->node->parent;
reader->depth--;
reader->state = XML_TEXTREADER_BACKTRACK;
goto found_node;
}
reader->state = XML_TEXTREADER_END;
found_node:
if ((reader->node->type == XML_XINCLUDE_START) ||
(reader->node->type == XML_XINCLUDE_END))
goto next_node;
return(1);
}
/**
* xmlTextReaderNextSibling:
* @reader: the xmlTextReaderPtr used
*
* Skip to the node following the current one in document order while
* avoiding the subtree if any.
* Currently implemented only for Readers built on a document
*
* Returns 1 if the node was read successfully, 0 if there is no more
* nodes to read, or -1 in case of error
*/
int
xmlTextReaderNextSibling(xmlTextReaderPtr reader) {
if (reader == NULL)
return(-1);
if (reader->doc == NULL) {
/* TODO */
return(-1);
}
if (reader->state == XML_TEXTREADER_END)
return(0);
if (reader->node == NULL)
return(xmlTextReaderNextTree(reader));
if (reader->node->next != NULL) {
reader->node = reader->node->next;
reader->state = XML_TEXTREADER_START;
return(1);
}
return(0);
}
/************************************************************************
* *
* Constructor and destructors *
* *
************************************************************************/
/**
* xmlNewTextReader:
* @input: the xmlParserInputBufferPtr used to read data
* @URI: the URI information for the source if available
*
* Create an xmlTextReader structure fed with @input
*
* Returns the new xmlTextReaderPtr or NULL in case of error
*/
xmlTextReaderPtr
xmlNewTextReader(xmlParserInputBufferPtr input, const char *URI) {
xmlTextReaderPtr ret;
if (input == NULL)
return(NULL);
ret = xmlMalloc(sizeof(xmlTextReader));
if (ret == NULL) {
xmlGenericError(xmlGenericErrorContext,
"xmlNewTextReader : malloc failed\n");
return(NULL);
}
memset(ret, 0, sizeof(xmlTextReader));
ret->doc = NULL;
ret->entTab = NULL;
ret->entMax = 0;
ret->entNr = 0;
ret->input = input;
ret->buffer = xmlBufCreateSize(100);
if (ret->buffer == NULL) {
xmlFree(ret);
xmlGenericError(xmlGenericErrorContext,
"xmlNewTextReader : malloc failed\n");
return(NULL);
}
/* no operation on a reader should require a huge buffer */
xmlBufSetAllocationScheme(ret->buffer,
XML_BUFFER_ALLOC_BOUNDED);
ret->sax = (xmlSAXHandler *) xmlMalloc(sizeof(xmlSAXHandler));
if (ret->sax == NULL) {
xmlBufFree(ret->buffer);
xmlFree(ret);
xmlGenericError(xmlGenericErrorContext,
"xmlNewTextReader : malloc failed\n");
return(NULL);
}
xmlSAXVersion(ret->sax, 2);
ret->startElement = ret->sax->startElement;
ret->sax->startElement = xmlTextReaderStartElement;
ret->endElement = ret->sax->endElement;
ret->sax->endElement = xmlTextReaderEndElement;
#ifdef LIBXML_SAX1_ENABLED
if (ret->sax->initialized == XML_SAX2_MAGIC) {
#endif /* LIBXML_SAX1_ENABLED */
ret->startElementNs = ret->sax->startElementNs;
ret->sax->startElementNs = xmlTextReaderStartElementNs;
ret->endElementNs = ret->sax->endElementNs;
ret->sax->endElementNs = xmlTextReaderEndElementNs;
#ifdef LIBXML_SAX1_ENABLED
} else {
ret->startElementNs = NULL;
ret->endElementNs = NULL;
}
#endif /* LIBXML_SAX1_ENABLED */
ret->characters = ret->sax->characters;
ret->sax->characters = xmlTextReaderCharacters;
ret->sax->ignorableWhitespace = xmlTextReaderCharacters;
ret->cdataBlock = ret->sax->cdataBlock;
ret->sax->cdataBlock = xmlTextReaderCDataBlock;
ret->mode = XML_TEXTREADER_MODE_INITIAL;
ret->node = NULL;
ret->curnode = NULL;
if (xmlBufUse(ret->input->buffer) < 4) {
xmlParserInputBufferRead(input, 4);
}
if (xmlBufUse(ret->input->buffer) >= 4) {
ret->ctxt = xmlCreatePushParserCtxt(ret->sax, NULL,
(const char *) xmlBufContent(ret->input->buffer),
4, URI);
ret->base = 0;
ret->cur = 4;
} else {
ret->ctxt = xmlCreatePushParserCtxt(ret->sax, NULL, NULL, 0, URI);
ret->base = 0;
ret->cur = 0;
}
if (ret->ctxt == NULL) {
xmlGenericError(xmlGenericErrorContext,
"xmlNewTextReader : malloc failed\n");
xmlBufFree(ret->buffer);
xmlFree(ret->sax);
xmlFree(ret);
return(NULL);
}
ret->ctxt->parseMode = XML_PARSE_READER;
ret->ctxt->_private = ret;
ret->ctxt->linenumbers = 1;
ret->ctxt->dictNames = 1;
ret->allocs = XML_TEXTREADER_CTXT;
/*
* use the parser dictionary to allocate all elements and attributes names
*/
ret->ctxt->docdict = 1;
ret->dict = ret->ctxt->dict;
#ifdef LIBXML_XINCLUDE_ENABLED
ret->xinclude = 0;
#endif
#ifdef LIBXML_PATTERN_ENABLED
ret->patternMax = 0;
ret->patternTab = NULL;
#endif
return(ret);
}
/**
* xmlNewTextReaderFilename:
* @URI: the URI of the resource to process
*
* Create an xmlTextReader structure fed with the resource at @URI
*
* Returns the new xmlTextReaderPtr or NULL in case of error
*/
xmlTextReaderPtr
xmlNewTextReaderFilename(const char *URI) {
xmlParserInputBufferPtr input;
xmlTextReaderPtr ret;
char *directory = NULL;
input = xmlParserInputBufferCreateFilename(URI, XML_CHAR_ENCODING_NONE);
if (input == NULL)
return(NULL);
ret = xmlNewTextReader(input, URI);
if (ret == NULL) {
xmlFreeParserInputBuffer(input);
return(NULL);
}
ret->allocs |= XML_TEXTREADER_INPUT;
if (ret->ctxt->directory == NULL)
directory = xmlParserGetDirectory(URI);
if ((ret->ctxt->directory == NULL) && (directory != NULL))
ret->ctxt->directory = (char *) xmlStrdup((xmlChar *) directory);
if (directory != NULL)
xmlFree(directory);
return(ret);
}
/**
* xmlFreeTextReader:
* @reader: the xmlTextReaderPtr
*
* Deallocate all the resources associated to the reader
*/
void
xmlFreeTextReader(xmlTextReaderPtr reader) {
if (reader == NULL)
return;
#ifdef LIBXML_SCHEMAS_ENABLED
if (reader->rngSchemas != NULL) {
xmlRelaxNGFree(reader->rngSchemas);
reader->rngSchemas = NULL;
}
if (reader->rngValidCtxt != NULL) {
if (! reader->rngPreserveCtxt)
xmlRelaxNGFreeValidCtxt(reader->rngValidCtxt);
reader->rngValidCtxt = NULL;
}
if (reader->xsdPlug != NULL) {
xmlSchemaSAXUnplug(reader->xsdPlug);
reader->xsdPlug = NULL;
}
if (reader->xsdValidCtxt != NULL) {
if (! reader->xsdPreserveCtxt)
xmlSchemaFreeValidCtxt(reader->xsdValidCtxt);
reader->xsdValidCtxt = NULL;
}
if (reader->xsdSchemas != NULL) {
xmlSchemaFree(reader->xsdSchemas);
reader->xsdSchemas = NULL;
}
#endif
#ifdef LIBXML_XINCLUDE_ENABLED
if (reader->xincctxt != NULL)
xmlXIncludeFreeContext(reader->xincctxt);
#endif
#ifdef LIBXML_PATTERN_ENABLED
if (reader->patternTab != NULL) {
int i;
for (i = 0;i < reader->patternNr;i++) {
if (reader->patternTab[i] != NULL)
xmlFreePattern(reader->patternTab[i]);
}
xmlFree(reader->patternTab);
}
#endif
if (reader->faketext != NULL) {
xmlFreeNode(reader->faketext);
}
if (reader->ctxt != NULL) {
if (reader->dict == reader->ctxt->dict)
reader->dict = NULL;
if ((reader->ctxt->vctxt.vstateTab != NULL) &&
(reader->ctxt->vctxt.vstateMax > 0)){
#ifdef LIBXML_REGEXP_ENABLED
while (reader->ctxt->vctxt.vstateNr > 0)
xmlValidatePopElement(&reader->ctxt->vctxt, NULL, NULL, NULL);
#endif /* LIBXML_REGEXP_ENABLED */
xmlFree(reader->ctxt->vctxt.vstateTab);
reader->ctxt->vctxt.vstateTab = NULL;
reader->ctxt->vctxt.vstateMax = 0;
}
if (reader->ctxt->myDoc != NULL) {
if (reader->preserve == 0)
xmlTextReaderFreeDoc(reader, reader->ctxt->myDoc);
reader->ctxt->myDoc = NULL;
}
if (reader->allocs & XML_TEXTREADER_CTXT)
xmlFreeParserCtxt(reader->ctxt);
}
if (reader->sax != NULL)
xmlFree(reader->sax);
if ((reader->input != NULL) && (reader->allocs & XML_TEXTREADER_INPUT))
xmlFreeParserInputBuffer(reader->input);
if (reader->buffer != NULL)
xmlBufFree(reader->buffer);
if (reader->entTab != NULL)
xmlFree(reader->entTab);
if (reader->dict != NULL)
xmlDictFree(reader->dict);
xmlFree(reader);
}
/************************************************************************
* *
* Methods for XmlTextReader *
* *
************************************************************************/
/**
* xmlTextReaderClose:
* @reader: the xmlTextReaderPtr used
*
* This method releases any resources allocated by the current instance
* changes the state to Closed and close any underlying input.
*
* Returns 0 or -1 in case of error
*/
int
xmlTextReaderClose(xmlTextReaderPtr reader) {
if (reader == NULL)
return(-1);
reader->node = NULL;
reader->curnode = NULL;
reader->mode = XML_TEXTREADER_MODE_CLOSED;
if (reader->ctxt != NULL) {
xmlStopParser(reader->ctxt);
if (reader->ctxt->myDoc != NULL) {
if (reader->preserve == 0)
xmlTextReaderFreeDoc(reader, reader->ctxt->myDoc);
reader->ctxt->myDoc = NULL;
}
}
if ((reader->input != NULL) && (reader->allocs & XML_TEXTREADER_INPUT)) {
xmlFreeParserInputBuffer(reader->input);
reader->allocs -= XML_TEXTREADER_INPUT;
}
return(0);
}
/**
* xmlTextReaderGetAttributeNo:
* @reader: the xmlTextReaderPtr used
* @no: the zero-based index of the attribute relative to the containing element
*
* Provides the value of the attribute with the specified index relative
* to the containing element.
*
* Returns a string containing the value of the specified attribute, or NULL
* in case of error. The string must be deallocated by the caller.
*/
xmlChar *
xmlTextReaderGetAttributeNo(xmlTextReaderPtr reader, int no) {
xmlChar *ret;
int i;
xmlAttrPtr cur;
xmlNsPtr ns;
if (reader == NULL)
return(NULL);
if (reader->node == NULL)
return(NULL);
if (reader->curnode != NULL)
return(NULL);
/* TODO: handle the xmlDecl */
if (reader->node->type != XML_ELEMENT_NODE)
return(NULL);
ns = reader->node->nsDef;
for (i = 0;(i < no) && (ns != NULL);i++) {
ns = ns->next;
}
if (ns != NULL)
return(xmlStrdup(ns->href));
cur = reader->node->properties;
if (cur == NULL)
return(NULL);
for (;i < no;i++) {
cur = cur->next;
if (cur == NULL)
return(NULL);
}
/* TODO walk the DTD if present */
ret = xmlNodeListGetString(reader->node->doc, cur->children, 1);
if (ret == NULL) return(xmlStrdup((xmlChar *)""));
return(ret);
}
/**
* xmlTextReaderGetAttribute:
* @reader: the xmlTextReaderPtr used
* @name: the qualified name of the attribute.
*
* Provides the value of the attribute with the specified qualified name.
*
* Returns a string containing the value of the specified attribute, or NULL
* in case of error. The string must be deallocated by the caller.
*/
xmlChar *
xmlTextReaderGetAttribute(xmlTextReaderPtr reader, const xmlChar *name) {
xmlChar *prefix = NULL;
xmlChar *localname;
xmlNsPtr ns;
xmlChar *ret = NULL;
if ((reader == NULL) || (name == NULL))
return(NULL);
if (reader->node == NULL)
return(NULL);
if (reader->curnode != NULL)
return(NULL);
/* TODO: handle the xmlDecl */
if (reader->node->type != XML_ELEMENT_NODE)
return(NULL);
localname = xmlSplitQName2(name, &prefix);
if (localname == NULL) {
/*
* Namespace default decl
*/
if (xmlStrEqual(name, BAD_CAST "xmlns")) {
ns = reader->node->nsDef;
while (ns != NULL) {
if (ns->prefix == NULL) {
return(xmlStrdup(ns->href));
}
ns = ns->next;
}
return NULL;
}
return(xmlGetNoNsProp(reader->node, name));
}
/*
* Namespace default decl
*/
if (xmlStrEqual(prefix, BAD_CAST "xmlns")) {
ns = reader->node->nsDef;
while (ns != NULL) {
if ((ns->prefix != NULL) && (xmlStrEqual(ns->prefix, localname))) {
ret = xmlStrdup(ns->href);
break;
}
ns = ns->next;
}
} else {
ns = xmlSearchNs(reader->node->doc, reader->node, prefix);
if (ns != NULL)
ret = xmlGetNsProp(reader->node, localname, ns->href);
}
xmlFree(localname);
if (prefix != NULL)
xmlFree(prefix);
return(ret);
}
/**
* xmlTextReaderGetAttributeNs:
* @reader: the xmlTextReaderPtr used
* @localName: the local name of the attribute.
* @namespaceURI: the namespace URI of the attribute.
*
* Provides the value of the specified attribute
*
* Returns a string containing the value of the specified attribute, or NULL
* in case of error. The string must be deallocated by the caller.
*/
xmlChar *
xmlTextReaderGetAttributeNs(xmlTextReaderPtr reader, const xmlChar *localName,
const xmlChar *namespaceURI) {
xmlChar *prefix = NULL;
xmlNsPtr ns;
if ((reader == NULL) || (localName == NULL))
return(NULL);
if (reader->node == NULL)
return(NULL);
if (reader->curnode != NULL)
return(NULL);
/* TODO: handle the xmlDecl */
if (reader->node->type != XML_ELEMENT_NODE)
return(NULL);
if (xmlStrEqual(namespaceURI, BAD_CAST "http://www.w3.org/2000/xmlns/")) {
if (! xmlStrEqual(localName, BAD_CAST "xmlns")) {
prefix = BAD_CAST localName;
}
ns = reader->node->nsDef;
while (ns != NULL) {
if ((prefix == NULL && ns->prefix == NULL) ||
((ns->prefix != NULL) && (xmlStrEqual(ns->prefix, localName)))) {
return xmlStrdup(ns->href);
}
ns = ns->next;
}
return NULL;
}
return(xmlGetNsProp(reader->node, localName, namespaceURI));
}
/**
* xmlTextReaderGetRemainder:
* @reader: the xmlTextReaderPtr used
*
* Method to get the remainder of the buffered XML. this method stops the
* parser, set its state to End Of File and return the input stream with
* what is left that the parser did not use.
*
* The implementation is not good, the parser certainly progressed past
* what's left in reader->input, and there is an allocation problem. Best
* would be to rewrite it differently.
*
* Returns the xmlParserInputBufferPtr attached to the XML or NULL
* in case of error.
*/
xmlParserInputBufferPtr
xmlTextReaderGetRemainder(xmlTextReaderPtr reader) {
xmlParserInputBufferPtr ret = NULL;
if (reader == NULL)
return(NULL);
if (reader->node == NULL)
return(NULL);
reader->node = NULL;
reader->curnode = NULL;
reader->mode = XML_TEXTREADER_MODE_EOF;
if (reader->ctxt != NULL) {
xmlStopParser(reader->ctxt);
if (reader->ctxt->myDoc != NULL) {
if (reader->preserve == 0)
xmlTextReaderFreeDoc(reader, reader->ctxt->myDoc);
reader->ctxt->myDoc = NULL;
}
}
if (reader->allocs & XML_TEXTREADER_INPUT) {
ret = reader->input;
reader->input = NULL;
reader->allocs -= XML_TEXTREADER_INPUT;
} else {
/*
* Hum, one may need to duplicate the data structure because
* without reference counting the input may be freed twice:
* - by the layer which allocated it.
* - by the layer to which would have been returned to.
*/
TODO
return(NULL);
}
return(ret);
}
/**
* xmlTextReaderLookupNamespace:
* @reader: the xmlTextReaderPtr used
* @prefix: the prefix whose namespace URI is to be resolved. To return
* the default namespace, specify NULL
*
* Resolves a namespace prefix in the scope of the current element.
*
* Returns a string containing the namespace URI to which the prefix maps
* or NULL in case of error. The string must be deallocated by the caller.
*/
xmlChar *
xmlTextReaderLookupNamespace(xmlTextReaderPtr reader, const xmlChar *prefix) {
xmlNsPtr ns;
if (reader == NULL)
return(NULL);
if (reader->node == NULL)
return(NULL);
ns = xmlSearchNs(reader->node->doc, reader->node, prefix);
if (ns == NULL)
return(NULL);
return(xmlStrdup(ns->href));
}
/**
* xmlTextReaderMoveToAttributeNo:
* @reader: the xmlTextReaderPtr used
* @no: the zero-based index of the attribute relative to the containing
* element.
*
* Moves the position of the current instance to the attribute with
* the specified index relative to the containing element.
*
* Returns 1 in case of success, -1 in case of error, 0 if not found
*/
int
xmlTextReaderMoveToAttributeNo(xmlTextReaderPtr reader, int no) {
int i;
xmlAttrPtr cur;
xmlNsPtr ns;
if (reader == NULL)
return(-1);
if (reader->node == NULL)
return(-1);
/* TODO: handle the xmlDecl */
if (reader->node->type != XML_ELEMENT_NODE)
return(-1);
reader->curnode = NULL;
ns = reader->node->nsDef;
for (i = 0;(i < no) && (ns != NULL);i++) {
ns = ns->next;
}
if (ns != NULL) {
reader->curnode = (xmlNodePtr) ns;
return(1);
}
cur = reader->node->properties;
if (cur == NULL)
return(0);
for (;i < no;i++) {
cur = cur->next;
if (cur == NULL)
return(0);
}
/* TODO walk the DTD if present */
reader->curnode = (xmlNodePtr) cur;
return(1);
}
/**
* xmlTextReaderMoveToAttribute:
* @reader: the xmlTextReaderPtr used
* @name: the qualified name of the attribute.
*
* Moves the position of the current instance to the attribute with
* the specified qualified name.
*
* Returns 1 in case of success, -1 in case of error, 0 if not found
*/
int
xmlTextReaderMoveToAttribute(xmlTextReaderPtr reader, const xmlChar *name) {
xmlChar *prefix = NULL;
xmlChar *localname;
xmlNsPtr ns;
xmlAttrPtr prop;
if ((reader == NULL) || (name == NULL))
return(-1);
if (reader->node == NULL)
return(-1);
/* TODO: handle the xmlDecl */
if (reader->node->type != XML_ELEMENT_NODE)
return(0);
localname = xmlSplitQName2(name, &prefix);
if (localname == NULL) {
/*
* Namespace default decl
*/
if (xmlStrEqual(name, BAD_CAST "xmlns")) {
ns = reader->node->nsDef;
while (ns != NULL) {
if (ns->prefix == NULL) {
reader->curnode = (xmlNodePtr) ns;
return(1);
}
ns = ns->next;
}
return(0);
}
prop = reader->node->properties;
while (prop != NULL) {
/*
* One need to have
* - same attribute names
* - and the attribute carrying that namespace
*/
if ((xmlStrEqual(prop->name, name)) &&
((prop->ns == NULL) || (prop->ns->prefix == NULL))) {
reader->curnode = (xmlNodePtr) prop;
return(1);
}
prop = prop->next;
}
return(0);
}
/*
* Namespace default decl
*/
if (xmlStrEqual(prefix, BAD_CAST "xmlns")) {
ns = reader->node->nsDef;
while (ns != NULL) {
if ((ns->prefix != NULL) && (xmlStrEqual(ns->prefix, localname))) {
reader->curnode = (xmlNodePtr) ns;
goto found;
}
ns = ns->next;
}
goto not_found;
}
prop = reader->node->properties;
while (prop != NULL) {
/*
* One need to have
* - same attribute names
* - and the attribute carrying that namespace
*/
if ((xmlStrEqual(prop->name, localname)) &&
(prop->ns != NULL) && (xmlStrEqual(prop->ns->prefix, prefix))) {
reader->curnode = (xmlNodePtr) prop;
goto found;
}
prop = prop->next;
}
not_found:
if (localname != NULL)
xmlFree(localname);
if (prefix != NULL)
xmlFree(prefix);
return(0);
found:
if (localname != NULL)
xmlFree(localname);
if (prefix != NULL)
xmlFree(prefix);
return(1);
}
/**
* xmlTextReaderMoveToAttributeNs:
* @reader: the xmlTextReaderPtr used
* @localName: the local name of the attribute.
* @namespaceURI: the namespace URI of the attribute.
*
* Moves the position of the current instance to the attribute with the
* specified local name and namespace URI.
*
* Returns 1 in case of success, -1 in case of error, 0 if not found
*/
int
xmlTextReaderMoveToAttributeNs(xmlTextReaderPtr reader,
const xmlChar *localName, const xmlChar *namespaceURI) {
xmlAttrPtr prop;
xmlNodePtr node;
xmlNsPtr ns;
xmlChar *prefix = NULL;
if ((reader == NULL) || (localName == NULL) || (namespaceURI == NULL))
return(-1);
if (reader->node == NULL)
return(-1);
if (reader->node->type != XML_ELEMENT_NODE)
return(0);
node = reader->node;
if (xmlStrEqual(namespaceURI, BAD_CAST "http://www.w3.org/2000/xmlns/")) {
if (! xmlStrEqual(localName, BAD_CAST "xmlns")) {
prefix = BAD_CAST localName;
}
ns = reader->node->nsDef;
while (ns != NULL) {
if ((prefix == NULL && ns->prefix == NULL) ||
((ns->prefix != NULL) && (xmlStrEqual(ns->prefix, localName)))) {
reader->curnode = (xmlNodePtr) ns;
return(1);
}
ns = ns->next;
}
return(0);
}
prop = node->properties;
while (prop != NULL) {
/*
* One need to have
* - same attribute names
* - and the attribute carrying that namespace
*/
if (xmlStrEqual(prop->name, localName) &&
((prop->ns != NULL) &&
(xmlStrEqual(prop->ns->href, namespaceURI)))) {
reader->curnode = (xmlNodePtr) prop;
return(1);
}
prop = prop->next;
}
return(0);
}
/**
* xmlTextReaderMoveToFirstAttribute:
* @reader: the xmlTextReaderPtr used
*
* Moves the position of the current instance to the first attribute
* associated with the current node.
*
* Returns 1 in case of success, -1 in case of error, 0 if not found
*/
int
xmlTextReaderMoveToFirstAttribute(xmlTextReaderPtr reader) {
if (reader == NULL)
return(-1);
if (reader->node == NULL)
return(-1);
if (reader->node->type != XML_ELEMENT_NODE)
return(0);
if (reader->node->nsDef != NULL) {
reader->curnode = (xmlNodePtr) reader->node->nsDef;
return(1);
}
if (reader->node->properties != NULL) {
reader->curnode = (xmlNodePtr) reader->node->properties;
return(1);
}
return(0);
}
/**
* xmlTextReaderMoveToNextAttribute:
* @reader: the xmlTextReaderPtr used
*
* Moves the position of the current instance to the next attribute
* associated with the current node.
*
* Returns 1 in case of success, -1 in case of error, 0 if not found
*/
int
xmlTextReaderMoveToNextAttribute(xmlTextReaderPtr reader) {
if (reader == NULL)
return(-1);
if (reader->node == NULL)
return(-1);
if (reader->node->type != XML_ELEMENT_NODE)
return(0);
if (reader->curnode == NULL)
return(xmlTextReaderMoveToFirstAttribute(reader));
if (reader->curnode->type == XML_NAMESPACE_DECL) {
xmlNsPtr ns = (xmlNsPtr) reader->curnode;
if (ns->next != NULL) {
reader->curnode = (xmlNodePtr) ns->next;
return(1);
}
if (reader->node->properties != NULL) {
reader->curnode = (xmlNodePtr) reader->node->properties;
return(1);
}
return(0);
} else if ((reader->curnode->type == XML_ATTRIBUTE_NODE) &&
(reader->curnode->next != NULL)) {
reader->curnode = reader->curnode->next;
return(1);
}
return(0);
}
/**
* xmlTextReaderMoveToElement:
* @reader: the xmlTextReaderPtr used
*
* Moves the position of the current instance to the node that
* contains the current Attribute node.
*
* Returns 1 in case of success, -1 in case of error, 0 if not moved
*/
int
xmlTextReaderMoveToElement(xmlTextReaderPtr reader) {
if (reader == NULL)
return(-1);
if (reader->node == NULL)
return(-1);
if (reader->node->type != XML_ELEMENT_NODE)
return(0);
if (reader->curnode != NULL) {
reader->curnode = NULL;
return(1);
}
return(0);
}
/**
* xmlTextReaderReadAttributeValue:
* @reader: the xmlTextReaderPtr used
*
* Parses an attribute value into one or more Text and EntityReference nodes.
*
* Returns 1 in case of success, 0 if the reader was not positioned on an
* attribute node or all the attribute values have been read, or -1
* in case of error.
*/
int
xmlTextReaderReadAttributeValue(xmlTextReaderPtr reader) {
if (reader == NULL)
return(-1);
if (reader->node == NULL)
return(-1);
if (reader->curnode == NULL)
return(0);
if (reader->curnode->type == XML_ATTRIBUTE_NODE) {
if (reader->curnode->children == NULL)
return(0);
reader->curnode = reader->curnode->children;
} else if (reader->curnode->type == XML_NAMESPACE_DECL) {
xmlNsPtr ns = (xmlNsPtr) reader->curnode;
if (reader->faketext == NULL) {
reader->faketext = xmlNewDocText(reader->node->doc,
ns->href);
} else {
if ((reader->faketext->content != NULL) &&
(reader->faketext->content !=
(xmlChar *) &(reader->faketext->properties)))
xmlFree(reader->faketext->content);
reader->faketext->content = xmlStrdup(ns->href);
}
reader->curnode = reader->faketext;
} else {
if (reader->curnode->next == NULL)
return(0);
reader->curnode = reader->curnode->next;
}
return(1);
}
/**
* xmlTextReaderConstEncoding:
* @reader: the xmlTextReaderPtr used
*
* Determine the encoding of the document being read.
*
* Returns a string containing the encoding of the document or NULL in
* case of error. The string is deallocated with the reader.
*/
const xmlChar *
xmlTextReaderConstEncoding(xmlTextReaderPtr reader) {
xmlDocPtr doc = NULL;
if (reader == NULL)
return(NULL);
if (reader->doc != NULL)
doc = reader->doc;
else if (reader->ctxt != NULL)
doc = reader->ctxt->myDoc;
if (doc == NULL)
return(NULL);
if (doc->encoding == NULL)
return(NULL);
else
return(CONSTSTR(doc->encoding));
}
/************************************************************************
* *
* Access API to the current node *
* *
************************************************************************/
/**
* xmlTextReaderAttributeCount:
* @reader: the xmlTextReaderPtr used
*
* Provides the number of attributes of the current node
*
* Returns 0 i no attributes, -1 in case of error or the attribute count
*/
int
xmlTextReaderAttributeCount(xmlTextReaderPtr reader) {
int ret;
xmlAttrPtr attr;
xmlNsPtr ns;
xmlNodePtr node;
if (reader == NULL)
return(-1);
if (reader->node == NULL)
return(0);
if (reader->curnode != NULL)
node = reader->curnode;
else
node = reader->node;
if (node->type != XML_ELEMENT_NODE)
return(0);
if ((reader->state == XML_TEXTREADER_END) ||
(reader->state == XML_TEXTREADER_BACKTRACK))
return(0);
ret = 0;
attr = node->properties;
while (attr != NULL) {
ret++;
attr = attr->next;
}
ns = node->nsDef;
while (ns != NULL) {
ret++;
ns = ns->next;
}
return(ret);
}
/**
* xmlTextReaderNodeType:
* @reader: the xmlTextReaderPtr used
*
* Get the node type of the current node
* Reference:
* http://www.gnu.org/software/dotgnu/pnetlib-doc/System/Xml/XmlNodeType.html
*
* Returns the xmlReaderTypes of the current node or -1 in case of error
*/
int
xmlTextReaderNodeType(xmlTextReaderPtr reader) {
xmlNodePtr node;
if (reader == NULL)
return(-1);
if (reader->node == NULL)
return(XML_READER_TYPE_NONE);
if (reader->curnode != NULL)
node = reader->curnode;
else
node = reader->node;
switch (node->type) {
case XML_ELEMENT_NODE:
if ((reader->state == XML_TEXTREADER_END) ||
(reader->state == XML_TEXTREADER_BACKTRACK))
return(XML_READER_TYPE_END_ELEMENT);
return(XML_READER_TYPE_ELEMENT);
case XML_NAMESPACE_DECL:
case XML_ATTRIBUTE_NODE:
return(XML_READER_TYPE_ATTRIBUTE);
case XML_TEXT_NODE:
if (xmlIsBlankNode(reader->node)) {
if (xmlNodeGetSpacePreserve(reader->node))
return(XML_READER_TYPE_SIGNIFICANT_WHITESPACE