📄 tree.c
字号:
/*
* tree.c : implementation of access function for an XML tree.
*
* References:
* XHTML 1.0 W3C REC: http://www.w3.org/TR/2002/REC-xhtml1-20020801/
*
* See Copyright for the status of this software.
*
* daniel@veillard.com
*
*/
#define IN_LIBXML
#include "libxml.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_ZLIB_H
#include <zlib.h>
#endif
#include <libxml/xmlmemory.h>
#include <libxml/tree.h>
#include <libxml/parser.h>
#include <libxml/uri.h>
#include <libxml/entities.h>
#include <libxml/valid.h>
#include <libxml/xmlerror.h>
#include <libxml/parserInternals.h>
#include <libxml/globals.h>
#ifdef LIBXML_HTML_ENABLED
#include <libxml/HTMLtree.h>
#endif
#ifdef LIBXML_DEBUG_ENABLED
#include <libxml/debugXML.h>
#endif
int __xmlRegisterCallbacks = 0;
xmlNsPtr xmlNewReconciliedNs(xmlDocPtr doc, xmlNodePtr tree, xmlNsPtr ns);
/************************************************************************
* *
* Tree memory error handler *
* *
************************************************************************/
/**
* xmlTreeErrMemory:
* @extra: extra informations
*
* Handle an out of memory condition
*/
static void
xmlTreeErrMemory(const char *extra)
{
__xmlSimpleError(XML_FROM_TREE, XML_ERR_NO_MEMORY, NULL, NULL, extra);
}
/**
* xmlTreeErr:
* @code: the error number
* @extra: extra informations
*
* Handle an out of memory condition
*/
static void
xmlTreeErr(int code, xmlNodePtr node, const char *extra)
{
const char *msg = NULL;
switch(code) {
case XML_TREE_INVALID_HEX:
msg = "invalid hexadecimal character value\n";
break;
case XML_TREE_INVALID_DEC:
msg = "invalid decimal character value\n";
break;
case XML_TREE_UNTERMINATED_ENTITY:
msg = "unterminated entity reference %15s\n";
break;
default:
msg = "unexpected error number\n";
}
__xmlSimpleError(XML_FROM_TREE, code, node, msg, extra);
}
/************************************************************************
* *
* A few static variables and macros *
* *
************************************************************************/
/* #undef xmlStringText */
const xmlChar xmlStringText[] = { 't', 'e', 'x', 't', 0 };
/* #undef xmlStringTextNoenc */
const xmlChar xmlStringTextNoenc[] =
{ 't', 'e', 'x', 't', 'n', 'o', 'e', 'n', 'c', 0 };
/* #undef xmlStringComment */
const xmlChar xmlStringComment[] = { 'c', 'o', 'm', 'm', 'e', 'n', 't', 0 };
static int xmlCompressMode = 0;
static int xmlCheckDTD = 1;
#define UPDATE_LAST_CHILD_AND_PARENT(n) if ((n) != NULL) { \
xmlNodePtr ulccur = (n)->children; \
if (ulccur == NULL) { \
(n)->last = NULL; \
} else { \
while (ulccur->next != NULL) { \
ulccur->parent = (n); \
ulccur = ulccur->next; \
} \
ulccur->parent = (n); \
(n)->last = ulccur; \
}}
/* #define DEBUG_BUFFER */
/* #define DEBUG_TREE */
/************************************************************************
* *
* Functions to move to entities.c once the *
* API freeze is smoothen and they can be made public. *
* *
************************************************************************/
#include <libxml/hash.h>
#ifdef LIBXML_TREE_ENABLED
/**
* xmlGetEntityFromDtd:
* @dtd: A pointer to the DTD to search
* @name: The entity name
*
* Do an entity lookup in the DTD entity hash table and
* return the corresponding entity, if found.
*
* Returns A pointer to the entity structure or NULL if not found.
*/
static xmlEntityPtr
xmlGetEntityFromDtd(xmlDtdPtr dtd, const xmlChar *name) {
xmlEntitiesTablePtr table;
if((dtd != NULL) && (dtd->entities != NULL)) {
table = (xmlEntitiesTablePtr) dtd->entities;
return((xmlEntityPtr) xmlHashLookup(table, name));
/* return(xmlGetEntityFromTable(table, name)); */
}
return(NULL);
}
/**
* xmlGetParameterEntityFromDtd:
* @dtd: A pointer to the DTD to search
* @name: The entity name
*
* Do an entity lookup in the DTD pararmeter entity hash table and
* return the corresponding entity, if found.
*
* Returns A pointer to the entity structure or NULL if not found.
*/
static xmlEntityPtr
xmlGetParameterEntityFromDtd(xmlDtdPtr dtd, const xmlChar *name) {
xmlEntitiesTablePtr table;
if ((dtd != NULL) && (dtd->pentities != NULL)) {
table = (xmlEntitiesTablePtr) dtd->pentities;
return((xmlEntityPtr) xmlHashLookup(table, name));
/* return(xmlGetEntityFromTable(table, name)); */
}
return(NULL);
}
#endif /* LIBXML_TREE_ENABLED */
/************************************************************************
* *
* QName handling helper *
* *
************************************************************************/
/**
* xmlBuildQName:
* @ncname: the Name
* @prefix: the prefix
* @memory: preallocated memory
* @len: preallocated memory length
*
* Builds the QName @prefix:@ncname in @memory if there is enough space
* and prefix is not NULL nor empty, otherwise allocate a new string.
* If prefix is NULL or empty it returns ncname.
*
* Returns the new string which must be freed by the caller if different from
* @memory and @ncname or NULL in case of error
*/
xmlChar *
xmlBuildQName(const xmlChar *ncname, const xmlChar *prefix,
xmlChar *memory, int len) {
int lenn, lenp;
xmlChar *ret;
if (ncname == NULL) return(NULL);
if (prefix == NULL) return((xmlChar *) ncname);
lenn = strlen((char *) ncname);
lenp = strlen((char *) prefix);
if ((memory == NULL) || (len < lenn + lenp + 2)) {
ret = (xmlChar *) xmlMallocAtomic(lenn + lenp + 2);
if (ret == NULL) {
xmlTreeErrMemory("building QName");
return(NULL);
}
} else {
ret = memory;
}
memcpy(&ret[0], prefix, lenp);
ret[lenp] = ':';
memcpy(&ret[lenp + 1], ncname, lenn);
ret[lenn + lenp + 1] = 0;
return(ret);
}
/**
* xmlSplitQName2:
* @name: the full QName
* @prefix: a xmlChar **
*
* parse an XML qualified name string
*
* [NS 5] QName ::= (Prefix ':')? LocalPart
*
* [NS 6] Prefix ::= NCName
*
* [NS 7] LocalPart ::= NCName
*
* Returns NULL if not a QName, otherwise the local part, and prefix
* is updated to get the Prefix if any.
*/
xmlChar *
xmlSplitQName2(const xmlChar *name, xmlChar **prefix) {
int len = 0;
xmlChar *ret = NULL;
if (prefix == NULL) return(NULL);
*prefix = NULL;
if (name == NULL) return(NULL);
#ifndef XML_XML_NAMESPACE
/* xml: prefix is not really a namespace */
if ((name[0] == 'x') && (name[1] == 'm') &&
(name[2] == 'l') && (name[3] == ':'))
return(NULL);
#endif
/* nasty but valid */
if (name[0] == ':')
return(NULL);
/*
* we are not trying to validate but just to cut, and yes it will
* work even if this is as set of UTF-8 encoded chars
*/
while ((name[len] != 0) && (name[len] != ':'))
len++;
if (name[len] == 0)
return(NULL);
*prefix = xmlStrndup(name, len);
if (*prefix == NULL) {
xmlTreeErrMemory("QName split");
return(NULL);
}
ret = xmlStrdup(&name[len + 1]);
if (ret == NULL) {
xmlTreeErrMemory("QName split");
if (*prefix != NULL) {
xmlFree(*prefix);
*prefix = NULL;
}
return(NULL);
}
return(ret);
}
/**
* xmlSplitQName3:
* @name: the full QName
* @len: an int *
*
* parse an XML qualified name string,i
*
* returns NULL if it is not a Qualified Name, otherwise, update len
* with the lenght in byte of the prefix and return a pointer
*/
const xmlChar *
xmlSplitQName3(const xmlChar *name, int *len) {
int l = 0;
if (name == NULL) return(NULL);
if (len == NULL) return(NULL);
/* nasty but valid */
if (name[0] == ':')
return(NULL);
/*
* we are not trying to validate but just to cut, and yes it will
* work even if this is as set of UTF-8 encoded chars
*/
while ((name[l] != 0) && (name[l] != ':'))
l++;
if (name[l] == 0)
return(NULL);
*len = l;
return(&name[l+1]);
}
/************************************************************************
* *
* Check Name, NCName and QName strings *
* *
************************************************************************/
#define CUR_SCHAR(s, l) xmlStringCurrentChar(NULL, s, &l)
#if defined(LIBXML_TREE_ENABLED) || defined(LIBXML_XPATH_ENABLED) || defined(LIBXML_SCHEMAS_ENABLED) || defined(LIBXML_DEBUG_ENABLED) || defined (LIBXML_HTML_ENABLED)
/**
* xmlValidateNCName:
* @value: the value to check
* @space: allow spaces in front and end of the string
*
* Check that a value conforms to the lexical space of NCName
*
* Returns 0 if this validates, a positive error code number otherwise
* and -1 in case of internal or API error.
*/
int
xmlValidateNCName(const xmlChar *value, int space) {
const xmlChar *cur = value;
int c,l;
if (value == NULL)
return(-1);
/*
* First quick algorithm for ASCII range
*/
if (space)
while (IS_BLANK_CH(*cur)) cur++;
if (((*cur >= 'a') && (*cur <= 'z')) || ((*cur >= 'A') && (*cur <= 'Z')) ||
(*cur == '_'))
cur++;
else
goto try_complex;
while (((*cur >= 'a') && (*cur <= 'z')) ||
((*cur >= 'A') && (*cur <= 'Z')) ||
((*cur >= '0') && (*cur <= '9')) ||
(*cur == '_') || (*cur == '-') || (*cur == '.'))
cur++;
if (space)
while (IS_BLANK_CH(*cur)) cur++;
if (*cur == 0)
return(0);
try_complex:
/*
* Second check for chars outside the ASCII range
*/
cur = value;
c = CUR_SCHAR(cur, l);
if (space) {
while (IS_BLANK(c)) {
cur += l;
c = CUR_SCHAR(cur, l);
}
}
if ((!IS_LETTER(c)) && (c != '_'))
return(1);
cur += l;
c = CUR_SCHAR(cur, l);
while (IS_LETTER(c) || IS_DIGIT(c) || (c == '.') ||
(c == '-') || (c == '_') || IS_COMBINING(c) ||
IS_EXTENDER(c)) {
cur += l;
c = CUR_SCHAR(cur, l);
}
if (space) {
while (IS_BLANK(c)) {
cur += l;
c = CUR_SCHAR(cur, l);
}
}
if (c != 0)
return(1);
return(0);
}
#endif
#if defined(LIBXML_TREE_ENABLED) || defined(LIBXML_SCHEMAS_ENABLED)
/**
* xmlValidateQName:
* @value: the value to check
* @space: allow spaces in front and end of the string
*
* Check that a value conforms to the lexical space of QName
*
* Returns 0 if this validates, a positive error code number otherwise
* and -1 in case of internal or API error.
*/
int
xmlValidateQName(const xmlChar *value, int space) {
const xmlChar *cur = value;
int c,l;
if (value == NULL)
return(-1);
/*
* First quick algorithm for ASCII range
*/
if (space)
while (IS_BLANK_CH(*cur)) cur++;
if (((*cur >= 'a') && (*cur <= 'z')) || ((*cur >= 'A') && (*cur <= 'Z')) ||
(*cur == '_'))
cur++;
else
goto try_complex;
while (((*cur >= 'a') && (*cur <= 'z')) ||
((*cur >= 'A') && (*cur <= 'Z')) ||
((*cur >= '0') && (*cur <= '9')) ||
(*cur == '_') || (*cur == '-') || (*cur == '.'))
cur++;
if (*cur == ':') {
cur++;
if (((*cur >= 'a') && (*cur <= 'z')) ||
((*cur >= 'A') && (*cur <= 'Z')) ||
(*cur == '_'))
cur++;
else
goto try_complex;
while (((*cur >= 'a') && (*cur <= 'z')) ||
((*cur >= 'A') && (*cur <= 'Z')) ||
((*cur >= '0') && (*cur <= '9')) ||
(*cur == '_') || (*cur == '-') || (*cur == '.'))
cur++;
}
if (space)
while (IS_BLANK_CH(*cur)) cur++;
if (*cur == 0)
return(0);
try_complex:
/*
* Second check for chars outside the ASCII range
*/
cur = value;
c = CUR_SCHAR(cur, l);
if (space) {
while (IS_BLANK(c)) {
cur += l;
c = CUR_SCHAR(cur, l);
}
}
if ((!IS_LETTER(c)) && (c != '_'))
return(1);
cur += l;
c = CUR_SCHAR(cur, l);
while (IS_LETTER(c) || IS_DIGIT(c) || (c == '.') ||
(c == '-') || (c == '_') || IS_COMBINING(c) ||
IS_EXTENDER(c)) {
cur += l;
c = CUR_SCHAR(cur, l);
}
if (c == ':') {
cur += l;
c = CUR_SCHAR(cur, l);
if ((!IS_LETTER(c)) && (c != '_'))
return(1);
cur += l;
c = CUR_SCHAR(cur, l);
while (IS_LETTER(c) || IS_DIGIT(c) || (c == '.') ||
(c == '-') || (c == '_') || IS_COMBINING(c) ||
IS_EXTENDER(c)) {
cur += l;
c = CUR_SCHAR(cur, l);
}
}
if (space) {
while (IS_BLANK(c)) {
cur += l;
c = CUR_SCHAR(cur, l);
}
}
if (c != 0)
return(1);
⌨️ 快捷键说明
复制代码
Ctrl + C
搜索代码
Ctrl + F
全屏模式
F11
切换主题
Ctrl + Shift + D
显示快捷键
?
增大字号
Ctrl + =
减小字号
Ctrl + -