Speed, conformance testing, more parsing, general improvements, Daniel.
diff --git a/ChangeLog b/ChangeLog
index 425103a..d0e3ba7 100644
--- a/ChangeLog
+++ b/ChangeLog
@@ -1,3 +1,12 @@
+Sun Jan 17 20:06:36 CET 1999 Daniel Veillard <Daniel.Veillard@w3.org>
+
+ * parser.c, tree.[ch] : more work toward conformance testing,
+ added a last element to accelerate parsing of very flat structures
+ started working on internal subset Element content declaration.
+ * valid.[ch] : first cut at adding code toward validation.
+ * previous changes had also small impact on most files, especially
+ the conformance testing using James Clark test suite.
+
Sun Jan 17 14:45:06 CET 1999 Daniel Veillard <Daniel.Veillard@w3.org>
* test/* : updated the examples, most of them were not well
diff --git a/Makefile.am b/Makefile.am
index df550e8..87a2bae 100644
--- a/Makefile.am
+++ b/Makefile.am
@@ -16,7 +16,8 @@
error.h \
parser.c \
debugXML.c \
- tree.c
+ tree.c \
+ valid.c
xmlincdir = $(includedir)/gnome-xml
xmlinc_HEADERS = \
@@ -24,7 +25,8 @@
encoding.h \
parser.h \
debugXML.h \
- tree.h
+ tree.h \
+ valid.h
DEPS = $(top_builddir)/libxml.la
LDADDS = $(top_builddir)/libxml.la @Z_LIBS@
diff --git a/SAX.c b/SAX.c
index 9b7e504..46afbb9 100644
--- a/SAX.c
+++ b/SAX.c
@@ -1,6 +1,8 @@
/*
* SAX.c : Default SAX handler to build a tree.
*
+ * See Copyright for the status of this software.
+ *
* Daniel Veillard <Daniel.Veillard@w3.org>
*/
diff --git a/debugXML.c b/debugXML.c
index bb913fd..5abec63 100644
--- a/debugXML.c
+++ b/debugXML.c
@@ -2,6 +2,8 @@
* debugXML.c : This is a set of routines used for debugging the tree
* produced by the XML parser.
*
+ * See Copyright for the status of this software.
+ *
* Daniel Veillard <Daniel.Veillard@w3.org>
*/
diff --git a/encoding.c b/encoding.c
index ce86fff..295ca39 100644
--- a/encoding.c
+++ b/encoding.c
@@ -16,8 +16,6 @@
*
* See Copyright for the status of this software.
*
- * $Id$
- *
* Daniel.Veillard@w3.org
*/
diff --git a/encoding.h b/encoding.h
index 72348c3..b5a11f9 100644
--- a/encoding.h
+++ b/encoding.h
@@ -17,8 +17,6 @@
*
* See Copyright for the status of this software.
*
- * $Id$
- *
* Daniel.Veillard@w3.org
*/
diff --git a/entities.c b/entities.c
index 30506cb..ef6aa79 100644
--- a/entities.c
+++ b/entities.c
@@ -3,7 +3,7 @@
*
* See Copyright for the status of this software.
*
- * $Id$
+ * Daniel.Veillard@w3.org
*/
#include <stdio.h>
@@ -186,14 +186,15 @@
const CHAR *ExternalID, const CHAR *SystemID, CHAR *content) {
xmlEntitiesTablePtr table;
- if (doc->dtd == NULL) {
- fprintf(stderr, "xmlAddDtdEntity: document without Dtd !\n");
+ if (doc->extSubset == NULL) {
+ fprintf(stderr,
+ "xmlAddDtdEntity: document without external subset !\n");
return;
}
- table = (xmlEntitiesTablePtr) doc->dtd->entities;
+ table = (xmlEntitiesTablePtr) doc->extSubset->entities;
if (table == NULL) {
table = xmlCreateEntitiesTable();
- doc->dtd->entities = table;
+ doc->extSubset->entities = table;
}
xmlAddEntity(table, name, type, ExternalID, SystemID, content);
}
@@ -214,12 +215,22 @@
const CHAR *ExternalID, const CHAR *SystemID, CHAR *content) {
xmlEntitiesTablePtr table;
- table = (xmlEntitiesTablePtr) doc->entities;
+ if (doc == NULL) {
+ fprintf(stderr,
+ "xmlAddDocEntity: document is NULL !\n");
+ return;
+ }
+ if (doc->intSubset == NULL) {
+ fprintf(stderr,
+ "xmlAddDtdEntity: document without internal subset !\n");
+ return;
+ }
+ table = (xmlEntitiesTablePtr) doc->intSubset->entities;
if (table == NULL) {
table = xmlCreateEntitiesTable();
- doc->entities = table;
+ doc->intSubset->entities = table;
}
- xmlAddEntity(doc->entities, name, type, ExternalID, SystemID, content);
+ xmlAddEntity(table, name, type, ExternalID, SystemID, content);
}
/**
@@ -238,8 +249,8 @@
xmlEntityPtr cur;
xmlEntitiesTablePtr table;
- if ((doc->dtd != NULL) && (doc->dtd->entities != NULL)) {
- table = (xmlEntitiesTablePtr) doc->dtd->entities;
+ if ((doc->extSubset != NULL) && (doc->extSubset->entities != NULL)) {
+ table = (xmlEntitiesTablePtr) doc->extSubset->entities;
for (i = 0;i < table->nb_entities;i++) {
cur = &table->table[i];
if (!xmlStrcmp(cur->name, name)) return(cur);
@@ -265,8 +276,8 @@
xmlEntityPtr cur;
xmlEntitiesTablePtr table;
- if (doc->entities != NULL) {
- table = (xmlEntitiesTablePtr) doc->entities;
+ if ((doc->intSubset != NULL) && (doc->intSubset->entities != NULL)) {
+ table = (xmlEntitiesTablePtr) doc->intSubset->entities;
for (i = 0;i < table->nb_entities;i++) {
cur = &table->table[i];
if (!xmlStrcmp(cur->name, name)) return(cur);
diff --git a/entities.h b/entities.h
index 0830101..86c425f 100644
--- a/entities.h
+++ b/entities.h
@@ -3,7 +3,7 @@
*
* See Copyright for the status of this software.
*
- * $Id$
+ * Daniel.Veillard@w3.org
*/
#ifndef __XML_ENTITIES_H__
diff --git a/error.c b/error.c
index a67bdbd..eb93d96 100644
--- a/error.c
+++ b/error.c
@@ -1,6 +1,8 @@
/*
* error.c: module displaying/handling XML parser errors
*
+ * See Copyright for the status of this software.
+ *
* Daniel Veillard <Daniel.Veillard@w3.org>
*/
diff --git a/include/libxml/encoding.h b/include/libxml/encoding.h
index 72348c3..b5a11f9 100644
--- a/include/libxml/encoding.h
+++ b/include/libxml/encoding.h
@@ -17,8 +17,6 @@
*
* See Copyright for the status of this software.
*
- * $Id$
- *
* Daniel.Veillard@w3.org
*/
diff --git a/include/libxml/entities.h b/include/libxml/entities.h
index 0830101..86c425f 100644
--- a/include/libxml/entities.h
+++ b/include/libxml/entities.h
@@ -3,7 +3,7 @@
*
* See Copyright for the status of this software.
*
- * $Id$
+ * Daniel.Veillard@w3.org
*/
#ifndef __XML_ENTITIES_H__
diff --git a/include/libxml/parser.h b/include/libxml/parser.h
index 4cd2ead..a06d481 100644
--- a/include/libxml/parser.h
+++ b/include/libxml/parser.h
@@ -3,7 +3,7 @@
*
* See Copyright for the status of this software.
*
- * $Id$
+ * Daniel.Veillard@w3.org
*/
#ifndef __XML_PARSER_H__
@@ -46,6 +46,7 @@
typedef struct xmlParserCtxt {
struct xmlSAXHandler *sax; /* The SAX handler */
xmlDocPtr doc; /* the document being built */
+ int wellFormed; /* is the document well formed */
/* Input stream stack */
xmlParserInputPtr input; /* Current input stream */
@@ -131,15 +132,8 @@
#include "entities.h"
/*
- * Interfaces
+ * CHAR handling
*/
-extern int xmlParseDocument(xmlParserCtxtPtr ctxt);
-extern xmlDocPtr xmlParseDoc(CHAR *cur);
-extern xmlDocPtr xmlParseMemory(char *buffer, int size);
-extern xmlDocPtr xmlParseFile(const char *filename);
-extern xmlDocPtr xmlSAXParseDoc(xmlSAXHandlerPtr sax, CHAR *cur);
-extern xmlDocPtr xmlSAXParseMemory(xmlSAXHandlerPtr sax, char *buffer, int size);
-extern xmlDocPtr xmlSAXParseFile(xmlSAXHandlerPtr sax, const char *filename);
extern CHAR *xmlStrdup(const CHAR *input);
extern CHAR *xmlStrndup(const CHAR *input, int n);
extern CHAR *xmlStrchr(const CHAR *str, CHAR val);
@@ -149,6 +143,29 @@
extern CHAR *xmlStrcat(CHAR *cur, const CHAR *add);
extern CHAR *xmlStrncat(CHAR *cur, const CHAR *add, int len);
+/*
+ * Interfaces
+ */
+extern xmlDocPtr xmlParseDoc(CHAR *cur);
+extern xmlDocPtr xmlParseMemory(char *buffer, int size);
+extern xmlDocPtr xmlParseFile(const char *filename);
+
+/*
+ * Recovery mode
+ */
+extern xmlDocPtr xmlRecoverDoc(CHAR *cur);
+extern xmlDocPtr xmlRecoverMemory(char *buffer, int size);
+extern xmlDocPtr xmlRecoverFile(const char *filename);
+
+/*
+ * Internal routines
+ */
+extern int xmlParseDocument(xmlParserCtxtPtr ctxt);
+extern xmlDocPtr xmlSAXParseDoc(xmlSAXHandlerPtr sax, CHAR *cur, int recovery);
+extern xmlDocPtr xmlSAXParseMemory(xmlSAXHandlerPtr sax, char *buffer,
+ int size, int recovery);
+extern xmlDocPtr xmlSAXParseFile(xmlSAXHandlerPtr sax, const char *filename,
+ int recovery);
extern void xmlInitParserCtxt(xmlParserCtxtPtr ctx);
extern void xmlClearParserCtxt(xmlParserCtxtPtr ctx);
extern void xmlSetupParserForBuffer(xmlParserCtxtPtr ctx, const CHAR* buffer,
diff --git a/include/libxml/tree.h b/include/libxml/tree.h
index 3921f7e..b6e26cf 100644
--- a/include/libxml/tree.h
+++ b/include/libxml/tree.h
@@ -4,7 +4,7 @@
*
* See Copyright for the status of this software.
*
- * $Id$
+ * Daniel.Veillard@w3.org
*/
#ifndef __XML_TREE_H__
@@ -15,6 +15,8 @@
extern "C" {
#endif
+#include <stdio.h>
+
/*
* The different element types carried by an XML tree
*
@@ -61,18 +63,51 @@
* TODO !!!!
*/
+#define XML_ATTRIBUTE_NONE 1
+#define XML_ATTRIBUTE_REQUIRED 2
+#define XML_ATTRIBUTE_IMPLIED 3
+#define XML_ATTRIBUTE_FIXED 4
+
+#define XML_ATTRIBUTE_STRING 1
+#define XML_ATTRIBUTE_ID 2
+#define XML_ATTRIBUTE_IDREF 3
+#define XML_ATTRIBUTE_IDREFS 4
+#define XML_ATTRIBUTE_ENTITY 5
+#define XML_ATTRIBUTE_ENTITIES 6
+#define XML_ATTRIBUTE_NMTOKEN 7
+#define XML_ATTRIBUTE_NMTOKENS 8
+#define XML_ATTRIBUTE_ENUMERATED 9
+
/*
* a DTD Element definition.
*/
+#define XML_ELEMENT_CONTENT_PCDATA 1
+#define XML_ELEMENT_CONTENT_ELEMENT 2
+#define XML_ELEMENT_CONTENT_SEQ 3
+#define XML_ELEMENT_CONTENT_OR 4
+
+#define XML_ELEMENT_CONTENT_ONCE 1
+#define XML_ELEMENT_CONTENT_OPT 2
+#define XML_ELEMENT_CONTENT_MULT 3
+#define XML_ELEMENT_CONTENT_PLUS 4
+
+typedef struct xmlElementContent {
+ int type; /* PCDATA, ELEMENT, SEQ or OR */
+ int ocur; /* ONCE, OPT, MULT or PLUS */
+ const CHAR *name; /* Element name */
+ struct xmlElementContent *c1; /* first child */
+ struct xmlElementContent *c2; /* second child */
+} xmlElementContent, *xmlElementContentPtr;
+
#define XML_ELEMENT_TYPE_EMPTY 1
#define XML_ELEMENT_TYPE_ANY 2
#define XML_ELEMENT_TYPE_MIXED 3
#define XML_ELEMENT_TYPE_ELEMENT 4
typedef struct xmlElement {
- const CHAR *name; /* Element name */
- int type; /* type (too simple, to extend ...) */
- /* TODO !!! more needed */
+ const CHAR *name; /* Element name */
+ int type; /* The type */
+ xmlElementContentPtr content; /* the allowed element content */
} xmlElement, *xmlElementPtr;
/*
@@ -132,6 +167,7 @@
struct xmlNode *next; /* next sibling link */
struct xmlNode *prev; /* previous sibling link */
struct xmlNode *childs; /* parent->childs link */
+ struct xmlNode *last; /* last child link */
struct xmlAttr *properties; /* properties list */
const CHAR *name; /* the name of the node, or the entity */
xmlNs *ns; /* pointer to the associated namespace */
@@ -153,9 +189,9 @@
const CHAR *encoding; /* encoding, if any */
int compression;/* level of zlib compression */
int standalone; /* standalone document (no external refs) */
- struct xmlDtd *dtd; /* the document DTD if available */
+ struct xmlDtd *intSubset; /* the document internal subset */
+ struct xmlDtd *extSubset; /* the document external subset */
struct xmlNs *oldNs; /* Global namespace, the old way */
- void *entities; /* Hash table for general entities if any */
struct xmlNode *root; /* the document tree */
} xmlDoc, *xmlDocPtr;
@@ -169,6 +205,8 @@
/*
* Creating/freeing new structures
*/
+extern xmlDtdPtr xmlCreateIntSubset(xmlDocPtr doc, const CHAR *name,
+ const CHAR *ExternalID, const CHAR *SystemID);
extern xmlDtdPtr xmlNewDtd(xmlDocPtr doc, const CHAR *name,
const CHAR *ExternalID, const CHAR *SystemID);
extern void xmlFreeDtd(xmlDtdPtr cur);
@@ -240,7 +278,7 @@
*/
extern xmlAttrPtr xmlSetProp(xmlNodePtr node, const CHAR *name,
const CHAR *value);
-extern const CHAR *xmlGetProp(xmlNodePtr node, const CHAR *name);
+extern CHAR *xmlGetProp(xmlNodePtr node, const CHAR *name);
extern xmlNodePtr xmlStringGetNodeList(xmlDocPtr doc, const CHAR *value);
extern xmlNodePtr xmlStringLenGetNodeList(xmlDocPtr doc, const CHAR *value,
int len);
diff --git a/include/libxml/valid.h b/include/libxml/valid.h
new file mode 100644
index 0000000..d6765ff
--- /dev/null
+++ b/include/libxml/valid.h
@@ -0,0 +1,18 @@
+/*
+ * valid.h : interface to the DTD handling and the validity checking
+ *
+ * See Copyright for the status of this software.
+ *
+ * Daniel.Veillard@w3.org
+ */
+
+
+#ifndef __XML_VALID_H__
+#define __XML_VALID_H__
+#include "tree.h"
+
+extern xmlElementPtr xmlAddElementDecl(xmlDtdPtr dtd, char *name, int type,
+ xmlElementContentPtr content);
+extern xmlElementContentPtr xmlNewElementContent(CHAR *name, int type);
+extern void xmlFreeElementContent(xmlElementContentPtr cur);
+#endif /* __XML_VALID_H__ */
diff --git a/parser.c b/parser.c
index 5464b26..a466b8b 100644
--- a/parser.c
+++ b/parser.c
@@ -3,7 +3,7 @@
*
* See Copyright for the status of this software.
*
- * $Id$
+ * Daniel.Veillard@w3.org
*/
#ifdef WIN32
@@ -30,6 +30,7 @@
#include "tree.h"
#include "parser.h"
#include "entities.h"
+#include "valid.h"
/************************************************************************
* *
@@ -182,6 +183,38 @@
return(input);
}
+/**
+ * xmlNewStringInputStream:
+ * @ctxt: an XML parser context
+ * @entity: an Entity pointer
+ *
+ * Create a new input stream based on a memory buffer.
+ * return vakues: the new input stream
+ */
+xmlParserInputPtr
+xmlNewStringInputStream(xmlParserCtxtPtr ctxt, CHAR *string) {
+ xmlParserInputPtr input;
+
+ if (string == NULL) {
+ if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
+ ctxt->sax->error(ctxt,
+ "internal: xmlNewStringInputStream string = NULL\n");
+ return(NULL);
+ }
+ input = (xmlParserInputPtr) malloc(sizeof(xmlParserInput));
+ if (input == NULL) {
+ if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
+ ctxt->sax->error(ctxt, "malloc: couldn't allocate a new input stream\n");
+ return(NULL);
+ }
+ input->filename = NULL;
+ input->base = string;
+ input->cur = string;
+ input->line = 1;
+ input->col = 1;
+ return(input);
+}
+
/*
* A few macros needed to help building the parser.
*/
@@ -198,8 +231,10 @@
* any Unicode character, excluding the surrogate blocks, FFFE, and FFFF.
*/
#define IS_CHAR(c) \
- (((c) == 0x09) || ((c) == 0x0a) || ((c) == 0x0d) || \
- (((c) >= 0x20) && ((c) != 0xFFFE) && ((c) != 0xFFFF)))
+ ((((c) == 0x09) || ((c) == 0x0a) || ((c) == 0x0d) || \
+ (((c) >= 0x20) && ((c) != 0xFFFE) && ((c) != 0xFFFF))) && \
+ (((c) <= 0xD7FF) || ((c) >= 0xE000)) && ((c) >= 0) && \
+ ((c) <= 0x10FFFF))
/*
* [3] S ::= (#x20 | #x9 | #xD | #xA)+
@@ -579,8 +614,9 @@
* any Unicode character, excluding the surrogate blocks, FFFE, and FFFF.
*/
#define IS_CHAR(c) \
- (((c) == 0x09) || ((c) == 0x0a) || ((c) == 0x0d) || ((c) >= 0x20) ||\
- ((c) == 0xa))
+ ((((c) == 0x09) || ((c) == 0x0a) || ((c) == 0x0d) || \
+ (((c) >= 0x20) && ((c) != 0xFFFE) && ((c) != 0xFFFF))) && \
+ (((c) <= 0xD7FF) || ((c) >= 0xE000)) && ((c) <= 0x10FFFF))
/*
* [85] BaseChar ::= ... long list see REC ...
@@ -935,6 +971,7 @@
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt, "xmlHandleEntity %s: content == NULL\n",
entity->name);
+ ctxt->wellFormed = 0;
return;
}
len = xmlStrlen(entity->content);
@@ -1081,6 +1118,7 @@
if (CUR != '"') {
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt, "String not closed \"%.50s\"\n", q);
+ ctxt->wellFormed = 0;
} else {
ret = xmlStrndup(q, CUR_PTR - q);
NEXT;
@@ -1092,6 +1130,7 @@
if (CUR != '\'') {
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt, "String not closed \"%.50s\"\n", q);
+ ctxt->wellFormed = 0;
} else {
ret = xmlStrndup(q, CUR_PTR - q);
NEXT;
@@ -1184,6 +1223,7 @@
if (!garbage)
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt, "xmlParseNamespace found garbage\n");
+ ctxt->wellFormed = 0;
NEXT;
}
}
@@ -1321,6 +1361,7 @@
if (!IS_CHAR(CUR)) {
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt, "Unfinished EntityValue\n");
+ ctxt->wellFormed = 0;
} else {
ret = xmlStrncat(ret, q, CUR_PTR - q);
NEXT;
@@ -1354,6 +1395,7 @@
if (!IS_CHAR(CUR)) {
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt, "Unfinished EntityValue\n");
+ ctxt->wellFormed = 0;
} else {
ret = xmlStrncat(ret, q, CUR_PTR - q);
NEXT;
@@ -1361,6 +1403,7 @@
} else {
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt, "xmlParseEntityValue \" or ' expected\n");
+ ctxt->wellFormed = 0;
}
return(ret);
@@ -1387,6 +1430,12 @@
q = CUR_PTR;
while ((IS_CHAR(CUR)) && (CUR != '"')) {
+ if (CUR == '<') {
+ if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
+ ctxt->sax->error(ctxt,
+ "Unescaped '<' not allowed in attributes values\n");
+ ctxt->wellFormed = 0;
+ }
if (CUR == '&') {
ret = xmlStrncat(ret, q, CUR_PTR - q);
cur = xmlParseReference(ctxt);
@@ -1419,6 +1468,7 @@
if (!IS_CHAR(CUR)) {
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt, "Unfinished AttValue\n");
+ ctxt->wellFormed = 0;
} else {
ret = xmlStrncat(ret, q, CUR_PTR - q);
NEXT;
@@ -1427,6 +1477,12 @@
NEXT;
q = CUR_PTR;
while ((IS_CHAR(CUR)) && (CUR != '\'')) {
+ if (CUR == '<') {
+ if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
+ ctxt->sax->error(ctxt,
+ "Unescaped '<' not allowed in attributes values\n");
+ ctxt->wellFormed = 0;
+ }
if (CUR == '&') {
ret = xmlStrncat(ret, q, CUR_PTR - q);
cur = xmlParseReference(ctxt);
@@ -1459,6 +1515,7 @@
if (!IS_CHAR(CUR)) {
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt, "Unfinished AttValue\n");
+ ctxt->wellFormed = 0;
} else {
ret = xmlStrncat(ret, q, CUR_PTR - q);
NEXT;
@@ -1466,6 +1523,7 @@
} else {
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt, "AttValue: \" or ' expected\n");
+ ctxt->wellFormed = 0;
}
return(ret);
@@ -1494,6 +1552,7 @@
if (!IS_CHAR(CUR)) {
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt, "Unfinished SystemLiteral\n");
+ ctxt->wellFormed = 0;
} else {
ret = xmlStrndup(q, CUR_PTR - q);
NEXT;
@@ -1506,6 +1565,7 @@
if (!IS_CHAR(CUR)) {
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt, "Unfinished SystemLiteral\n");
+ ctxt->wellFormed = 0;
} else {
ret = xmlStrndup(q, CUR_PTR - q);
NEXT;
@@ -1513,6 +1573,7 @@
} else {
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt, "SystemLiteral \" or ' expected\n");
+ ctxt->wellFormed = 0;
}
return(ret);
@@ -1540,6 +1601,7 @@
if (CUR != '"') {
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt, "Unfinished PubidLiteral\n");
+ ctxt->wellFormed = 0;
} else {
ret = xmlStrndup(q, CUR_PTR - q);
NEXT;
@@ -1552,6 +1614,7 @@
if (!IS_LETTER(CUR)) {
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt, "Unfinished PubidLiteral\n");
+ ctxt->wellFormed = 0;
} else {
ret = xmlStrndup(q, CUR_PTR - q);
NEXT;
@@ -1559,6 +1622,7 @@
} else {
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt, "SystemLiteral \" or ' expected\n");
+ ctxt->wellFormed = 0;
}
return(ret);
@@ -1583,9 +1647,17 @@
q = CUR_PTR;
while ((IS_CHAR(CUR)) && (CUR != '<') &&
(CUR != '&')) {
+ if ((CUR == ']') && (NXT(1) == ']') &&
+ (NXT(2) == '>')) {
+ if (cdata) break;
+ else {
+ if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
+ ctxt->sax->error(ctxt,
+ "Sequence ']]>' not allowed in content\n");
+ ctxt->wellFormed = 0;
+ }
+ }
NEXT;
- if ((cdata) && (CUR == ']') && (NXT(1) == ']') &&
- (NXT(2) == '>')) break;
}
if (q == CUR_PTR) return;
@@ -1621,28 +1693,52 @@
(NXT(2) == 'S') && (NXT(3) == 'T') &&
(NXT(4) == 'E') && (NXT(5) == 'M')) {
SKIP(6);
+ if (!IS_BLANK(CUR)) {
+ if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
+ ctxt->sax->error(ctxt,
+ "Space required after 'SYSTEM'\n");
+ ctxt->wellFormed = 0;
+ }
SKIP_BLANKS;
URI = xmlParseSystemLiteral(ctxt);
- if (URI == NULL)
+ if (URI == NULL) {
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt,
"xmlParseExternalID: SYSTEM, no URI\n");
+ ctxt->wellFormed = 0;
+ }
} else if ((CUR == 'P') && (NXT(1) == 'U') &&
(NXT(2) == 'B') && (NXT(3) == 'L') &&
(NXT(4) == 'I') && (NXT(5) == 'C')) {
SKIP(6);
+ if (!IS_BLANK(CUR)) {
+ if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
+ ctxt->sax->error(ctxt,
+ "Space required after 'PUBLIC'\n");
+ ctxt->wellFormed = 0;
+ }
SKIP_BLANKS;
*publicID = xmlParsePubidLiteral(ctxt);
- if (*publicID == NULL)
+ if (*publicID == NULL) {
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt,
"xmlParseExternalID: PUBLIC, no Public Identifier\n");
+ ctxt->wellFormed = 0;
+ }
+ if (!IS_BLANK(CUR)) {
+ if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
+ ctxt->sax->error(ctxt,
+ "Space required after the Public Identifier\n");
+ ctxt->wellFormed = 0;
+ }
SKIP_BLANKS;
URI = xmlParseSystemLiteral(ctxt);
- if (URI == NULL)
+ if (URI == NULL) {
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt,
"xmlParseExternalID: PUBLIC, no URI\n");
+ ctxt->wellFormed = 0;
+ }
}
return(URI);
}
@@ -1682,15 +1778,18 @@
while (IS_CHAR(CUR) &&
((CUR == ':') || (CUR != '>') ||
(*r != '-') || (*q != '-'))) {
- if ((*r == '-') && (*q == '-'))
+ if ((*r == '-') && (*q == '-')) {
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt,
"Comment must not contain '--' (double-hyphen)`\n");
+ ctxt->wellFormed = 0;
+ }
NEXT;r++;q++;
}
if (!IS_CHAR(CUR)) {
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt, "Comment not terminated \n<!--%.50s\n", start);
+ ctxt->wellFormed = 0;
} else {
NEXT;
if (create) {
@@ -1782,8 +1881,9 @@
NEXT;
if (!IS_CHAR(CUR)) {
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
- ctxt->sax->error(ctxt, "xmlParsePI: PI %s never end ...\n",
- target);
+ ctxt->sax->error(ctxt,
+ "xmlParsePI: PI %s never end ...\n", target);
+ ctxt->wellFormed = 0;
} else {
CHAR *data;
@@ -1809,6 +1909,8 @@
} else {
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt, "xmlParsePI : no target name\n");
+ ctxt->wellFormed = 0;
+
/********* Should we try to complete parsing the PI ???
while (IS_CHAR(CUR) &&
(CUR != '?') && (CUR != '>'))
@@ -1861,6 +1963,7 @@
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt,
"xmlParseAttributeListDecl: no name for Element\n");
+ ctxt->wellFormed = 0;
return;
}
SKIP_BLANKS;
@@ -1903,13 +2006,22 @@
if ((CUR == '<') && (NXT(1) == '!') &&
(NXT(2) == 'E') && (NXT(3) == 'N') &&
(NXT(4) == 'T') && (NXT(5) == 'I') &&
- (NXT(6) == 'T') && (NXT(7) == 'Y') &&
- (IS_BLANK(NXT(8)))) {
+ (NXT(6) == 'T') && (NXT(7) == 'Y')) {
SKIP(8);
- SKIP_BLANKS;
+ if (!IS_BLANK(CUR)) {
+ if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
+ ctxt->sax->error(ctxt, "Space required after '<!ENTITY'\n");
+ ctxt->wellFormed = 0;
+ }
+ SKIP_BLANKS;
if (CUR == '%') {
NEXT;
+ if (!IS_BLANK(CUR)) {
+ if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
+ ctxt->sax->error(ctxt, "Space required after '%'\n");
+ ctxt->wellFormed = 0;
+ }
SKIP_BLANKS;
isParameter = 1;
}
@@ -1918,8 +2030,15 @@
if (name == NULL) {
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt, "xmlParseEntityDecl: no name\n");
+ ctxt->wellFormed = 0;
return;
}
+ if (!IS_BLANK(CUR)) {
+ if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
+ ctxt->sax->error(ctxt,
+ "Space required after the entity name\n");
+ ctxt->wellFormed = 0;
+ }
SKIP_BLANKS;
/*
@@ -1949,11 +2068,23 @@
NULL, NULL, value);
} else {
URI = xmlParseExternalID(ctxt, &literal);
+ if ((CUR != '>') && (!IS_BLANK(CUR))) {
+ if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
+ ctxt->sax->error(ctxt,
+ "Space required before 'NDATA'\n");
+ ctxt->wellFormed = 0;
+ }
SKIP_BLANKS;
if ((CUR == 'N') && (NXT(1) == 'D') &&
(NXT(2) == 'A') && (NXT(3) == 'T') &&
(NXT(4) == 'A')) {
SKIP(5);
+ if (!IS_BLANK(CUR)) {
+ if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
+ ctxt->sax->error(ctxt,
+ "Space required after 'NDATA'\n");
+ ctxt->wellFormed = 0;
+ }
SKIP_BLANKS;
ndata = xmlParseName(ctxt);
xmlAddDocEntity(ctxt->doc, name,
@@ -1971,6 +2102,7 @@
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt,
"xmlParseEntityDecl: entity %s not terminated\n", name);
+ ctxt->wellFormed = 0;
} else
NEXT;
if (name != NULL) free(name);
@@ -1982,6 +2114,64 @@
}
/**
+ * xmlParseDefaultDecl:
+ * @ctxt: an XML parser context
+ * @value: Receive a possible fixed default value for the attribute
+ *
+ * Parse an attribute default declaration
+ *
+ * [60] DefaultDecl ::= '#REQUIRED' | '#IMPLIED' | (('#FIXED' S)? AttValue)
+ *
+ * returns: XML_ATTRIBUTE_NONE, XML_ATTRIBUTE_REQUIRED, XML_ATTRIBUTE_IMPLIED
+ * or XML_ATTRIBUTE_FIXED.
+ */
+
+int
+xmlParseDefaultDecl(xmlParserCtxtPtr ctxt, CHAR **value) {
+ int val;
+ CHAR *ret;
+
+ *value = NULL;
+ if ((CUR == '#') && (NXT(1) == 'R') &&
+ (NXT(2) == 'E') && (NXT(3) == 'Q') &&
+ (NXT(4) == 'U') && (NXT(5) == 'I') &&
+ (NXT(6) == 'R') && (NXT(7) == 'E') &&
+ (NXT(8) == 'D')) {
+ SKIP(9);
+ return(XML_ATTRIBUTE_REQUIRED);
+ }
+ if ((CUR == '#') && (NXT(1) == 'I') &&
+ (NXT(2) == 'M') && (NXT(3) == 'P') &&
+ (NXT(4) == 'L') && (NXT(5) == 'I') &&
+ (NXT(6) == 'E') && (NXT(7) == 'D')) {
+ SKIP(8);
+ return(XML_ATTRIBUTE_IMPLIED);
+ }
+ val = XML_ATTRIBUTE_NONE;
+ if ((CUR == '#') && (NXT(1) == 'F') &&
+ (NXT(2) == 'I') && (NXT(3) == 'X') &&
+ (NXT(4) == 'E') && (NXT(5) == 'D')) {
+ SKIP(6);
+ val = XML_ATTRIBUTE_FIXED;
+ if (!IS_BLANK(CUR)) {
+ if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
+ ctxt->sax->error(ctxt, "Space required after '#FIXED'\n");
+ ctxt->wellFormed = 0;
+ }
+ SKIP_BLANKS;
+ }
+ ret = xmlParseAttValue(ctxt);
+ if (ret == NULL) {
+ if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
+ ctxt->sax->error(ctxt,
+ "Attribute default value declaration error\n");
+ ctxt->wellFormed = 0;
+ } else
+ *value = ret;
+ return(val);
+}
+
+/**
* xmlParseEnumeratedType:
* @ctxt: an XML parser context
* @name: ???
@@ -2003,6 +2193,7 @@
/*
* TODO !!!
*/
+ fprintf(stderr, "Production [57] EnumeratedType not yet supported\n");
while ((IS_CHAR(CUR)) && (CUR != '>'))
NEXT;
}
@@ -2012,7 +2203,7 @@
* @ctxt: an XML parser context
* @name: ???
*
- * : parse the Attribute list def for an element
+ * parse the Attribute list def for an element
*
* [54] AttType ::= StringType | TokenizedType | EnumeratedType
*
@@ -2021,46 +2212,54 @@
* [56] TokenizedType ::= 'ID' | 'IDREF' | 'IDREFS' | 'ENTITY' |
* 'ENTITIES' | 'NMTOKEN' | 'NMTOKENS'
*
- * TODO: not implemented !!!
+ * Returns: the attribute type
*/
-void
+int
xmlParseAttributeType(xmlParserCtxtPtr ctxt, CHAR *name) {
/* TODO !!! */
if ((CUR == 'C') && (NXT(1) == 'D') &&
(NXT(2) == 'A') && (NXT(3) == 'T') &&
(NXT(4) == 'A')) {
SKIP(5);
+ return(XML_ATTRIBUTE_STRING);
} else if ((CUR == 'I') && (NXT(1) == 'D')) {
SKIP(2);
+ return(XML_ATTRIBUTE_ID);
} else if ((CUR == 'I') && (NXT(1) == 'D') &&
(NXT(2) == 'R') && (NXT(3) == 'E') &&
(NXT(4) == 'F')) {
SKIP(5);
+ return(XML_ATTRIBUTE_IDREF);
} else if ((CUR == 'I') && (NXT(1) == 'D') &&
(NXT(2) == 'R') && (NXT(3) == 'E') &&
(NXT(4) == 'F') && (NXT(5) == 'S')) {
SKIP(6);
+ return(XML_ATTRIBUTE_IDREFS);
} else if ((CUR == 'E') && (NXT(1) == 'N') &&
(NXT(2) == 'T') && (NXT(3) == 'I') &&
(NXT(4) == 'T') && (NXT(5) == 'Y')) {
SKIP(6);
+ return(XML_ATTRIBUTE_ENTITY);
} else if ((CUR == 'E') && (NXT(1) == 'N') &&
(NXT(2) == 'T') && (NXT(3) == 'I') &&
(NXT(4) == 'T') && (NXT(5) == 'I') &&
(NXT(6) == 'E') && (NXT(7) == 'S')) {
SKIP(8);
+ return(XML_ATTRIBUTE_ENTITIES);
} else if ((CUR == 'N') && (NXT(1) == 'M') &&
(NXT(2) == 'T') && (NXT(3) == 'O') &&
(NXT(4) == 'K') && (NXT(5) == 'E') &&
(NXT(6) == 'N')) {
SKIP(7);
+ return(XML_ATTRIBUTE_NMTOKEN);
} else if ((CUR == 'N') && (NXT(1) == 'M') &&
(NXT(2) == 'T') && (NXT(3) == 'O') &&
(NXT(4) == 'K') && (NXT(5) == 'E') &&
(NXT(6) == 'N') && (NXT(7) == 'S')) {
- } else {
- xmlParseEnumeratedType(ctxt, name);
+ return(XML_ATTRIBUTE_NMTOKENS);
}
+ xmlParseEnumeratedType(ctxt, name);
+ return(XML_ATTRIBUTE_ENUMERATED);
}
/**
@@ -2077,51 +2276,183 @@
*/
void
xmlParseAttributeListDecl(xmlParserCtxtPtr ctxt) {
- CHAR *name;
+ CHAR *elemName;
+ CHAR *attrName;
/* TODO !!! */
if ((CUR == '<') && (NXT(1) == '!') &&
(NXT(2) == 'A') && (NXT(3) == 'T') &&
(NXT(4) == 'T') && (NXT(5) == 'L') &&
(NXT(6) == 'I') && (NXT(7) == 'S') &&
- (NXT(8) == 'T') && (IS_BLANK(NXT(9)))) {
+ (NXT(8) == 'T')) {
SKIP(9);
- SKIP_BLANKS;
- name = xmlParseName(ctxt);
- if (name == NULL) {
+ if (!IS_BLANK(CUR)) {
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
- ctxt->sax->error(ctxt,
- "xmlParseAttributeListDecl: no name for Element\n");
+ ctxt->sax->error(ctxt, "Space required after '<!ATTLIST'\n");
+ ctxt->wellFormed = 0;
+ }
+ SKIP_BLANKS;
+ elemName = xmlParseName(ctxt);
+ if (elemName == NULL) {
+ if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
+ ctxt->sax->error(ctxt, "ATTLIST: no name for Element\n");
+ ctxt->wellFormed = 0;
return;
}
SKIP_BLANKS;
while (CUR != '>') {
const CHAR *check = CUR_PTR;
+ int type;
+ int def;
+ CHAR *defaultValue = NULL;
- xmlParseAttributeType(ctxt, name);
+ attrName = xmlParseName(ctxt);
+ if (attrName == NULL) {
+ if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
+ ctxt->sax->error(ctxt, "ATTLIST: no name for Attribute\n");
+ ctxt->wellFormed = 0;
+ break;
+ }
+ if (!IS_BLANK(CUR)) {
+ if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
+ ctxt->sax->error(ctxt,
+ "Space required after the attribute name\n");
+ ctxt->wellFormed = 0;
+ break;
+ }
SKIP_BLANKS;
+
+ type = xmlParseAttributeType(ctxt, attrName);
+ if (type <= 0) break;
+
+ if (!IS_BLANK(CUR)) {
+ if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
+ ctxt->sax->error(ctxt,
+ "Space required after the attribute type\n");
+ ctxt->wellFormed = 0;
+ break;
+ }
+ SKIP_BLANKS;
+
+ def = xmlParseDefaultDecl(ctxt, &defaultValue);
+ if (def <= 0) break;
+
+ if (CUR != '>') {
+ if (!IS_BLANK(CUR)) {
+ if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
+ ctxt->sax->error(ctxt,
+ "Space required after the attribute default value\n");
+ ctxt->wellFormed = 0;
+ break;
+ }
+ SKIP_BLANKS;
+ }
if (check == CUR_PTR) {
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt,
- "xmlParseAttributeListDecl: detected error\n");
+ "xmlParseAttributeListDecl: detected internal error\n");
break;
}
+ if (attrName != NULL)
+ free(attrName);
+ if (defaultValue != NULL)
+ free(defaultValue);
}
if (CUR == '>')
NEXT;
- free(name);
+ free(elemName);
}
}
/**
- * xmlParseElementContentDecl:
+ * xmlParseElementMixedContentDecl:
* @ctxt: an XML parser context
- * @name: ???
*
- * parse the declaration for an Element content
- * either Mixed or Children, the cases EMPTY and ANY being handled
+ * parse the declaration for a Mixed Element content
+ * The leading '(' and spaces have been skipped in xmlParseElementContentDecl
*
+ * [51] Mixed ::= '(' S? '#PCDATA' (S? '|' S? Name)* S? ')*' |
+ * '(' S? '#PCDATA' S? ')'
+ *
+ * returns: the list of the xmlElementContentPtr describing the element choices
+ */
+xmlElementContentPtr
+xmlParseElementMixedContentDecl(xmlParserCtxtPtr ctxt) {
+ xmlElementContentPtr ret = NULL, cur = NULL;
+ CHAR *elem = NULL;
+
+ if ((CUR == '#') && (NXT(1) == 'P') &&
+ (NXT(2) == 'C') && (NXT(3) == 'D') &&
+ (NXT(4) == 'A') && (NXT(5) == 'T') &&
+ (NXT(6) == 'A')) {
+ SKIP(7);
+ SKIP_BLANKS;
+ if ((CUR == '(') || (CUR == '|')) {
+ ret = cur = xmlNewElementContent(NULL, XML_ELEMENT_CONTENT_PCDATA);
+ if (ret == NULL) return(NULL);
+ } else {
+ if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
+ ctxt->sax->error(ctxt,
+ "xmlParseElementMixedContentDecl : '|' or ')' expected\n");
+ ctxt->wellFormed = 0;
+ return(NULL);
+ }
+ while (CUR == '|') {
+ if (elem == NULL) {
+ ret = xmlNewElementContent(NULL, XML_ELEMENT_CONTENT_OR);
+ if (ret == NULL) return(NULL);
+ ret->c1 = cur;
+ } else {
+ cur->c1 = xmlNewElementContent(elem,
+ XML_ELEMENT_CONTENT_ELEMENT);
+ cur->c2 = xmlNewElementContent(NULL, XML_ELEMENT_CONTENT_OR);
+ cur = cur->c2;
+ if (cur == NULL) return(NULL);
+ }
+ SKIP_BLANKS;
+ elem = xmlParseName(ctxt);
+ if (elem == NULL) {
+ if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
+ ctxt->sax->error(ctxt,
+ "xmlParseElementMixedContentDecl : Name expected\n");
+ ctxt->wellFormed = 0;
+ xmlFreeElementContent(cur);
+ return(NULL);
+ }
+ SKIP_BLANKS;
+ }
+ if (CUR == ')') {
+ if (elem != NULL)
+ cur->c2 = xmlNewElementContent(elem,
+ XML_ELEMENT_CONTENT_ELEMENT);
+ NEXT;
+ } else {
+ if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
+ ctxt->sax->error(ctxt,
+ "xmlParseElementMixedContentDecl : '|' or ')' expected\n");
+ ctxt->wellFormed = 0;
+ xmlFreeElementContent(ret);
+ return(NULL);
+ }
+
+ } else {
+ if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
+ ctxt->sax->error(ctxt,
+ "xmlParseElementMixedContentDecl : '#PCDATA' expected\n");
+ ctxt->wellFormed = 0;
+ }
+ return(ret);
+}
+
+/**
+ * xmlParseElementChildrenContentDecl:
+ * @ctxt: an XML parser context
+ *
+ * parse the declaration for a Mixed Element content
+ * The leading '(' and spaces have been skipped in xmlParseElementContentDecl
+ *
+ *
* [47] children ::= (choice | seq) ('?' | '*' | '+')?
*
* [48] cp ::= (Name | choice | seq) ('?' | '*' | '+')?
@@ -2130,23 +2461,217 @@
*
* [50] seq ::= '(' S? cp ( S? ',' S? cp )* S? ')'
*
- * or
+ * returns: the tree of xmlElementContentPtr describing the element
+ * hierarchy.
+ */
+xmlElementContentPtr
+xmlParseElementChildrenContentDecl(xmlParserCtxtPtr ctxt) {
+ xmlElementContentPtr ret = NULL, cur = NULL, last = NULL, op = NULL;
+ CHAR *elem;
+ CHAR type = 0;
+
+ SKIP_BLANKS;
+ if (CUR == '(') {
+ /* Recurse on first child */
+ NEXT;
+ SKIP_BLANKS;
+ cur = ret = xmlParseElementChildrenContentDecl(ctxt);
+ SKIP_BLANKS;
+ } else {
+ elem = xmlParseName(ctxt);
+ if (elem == NULL) {
+ if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
+ ctxt->sax->error(ctxt,
+ "xmlParseElementChildrenContentDecl : Name or '(' expected\n");
+ ctxt->wellFormed = 0;
+ return(NULL);
+ }
+ cur = ret = xmlNewElementContent(elem, XML_ELEMENT_CONTENT_ELEMENT);
+ if (CUR == '?') {
+ ret->ocur = XML_ELEMENT_CONTENT_OPT;
+ NEXT;
+ } else if (CUR == '*') {
+ ret->ocur = XML_ELEMENT_CONTENT_MULT;
+ NEXT;
+ } else if (CUR == '+') {
+ ret->ocur = XML_ELEMENT_CONTENT_PLUS;
+ NEXT;
+ } else {
+ ret->ocur = XML_ELEMENT_CONTENT_ONCE;
+ }
+ }
+ SKIP_BLANKS;
+ while (CUR != ')') {
+ /*
+ * Each loop we parse one separator and one element.
+ */
+ if (CUR == ',') {
+ if (type == 0) type = CUR;
+
+ /*
+ * Detect "Name | Name , Name" error
+ */
+ else if (type != CUR) {
+ if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
+ ctxt->sax->error(ctxt,
+ "xmlParseElementChildrenContentDecl : '%c' expected\n",
+ type);
+ ctxt->wellFormed = 0;
+ xmlFreeElementContent(ret);
+ return(NULL);
+ }
+
+ op = xmlNewElementContent(NULL, XML_ELEMENT_CONTENT_SEQ);
+ if (op == NULL) {
+ xmlFreeElementContent(ret);
+ return(NULL);
+ }
+ if (last == NULL) {
+ op->c1 = ret;
+ ret = cur = op;
+ } else {
+ cur->c2 = op;
+ op->c1 = last;
+ cur =op;
+ }
+ } else if (CUR == '|') {
+ if (type == 0) type = CUR;
+
+ /*
+ * Detect "Name , Name | Name" error
+ */
+ else if (type != CUR) {
+ if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
+ ctxt->sax->error(ctxt,
+ "xmlParseElementChildrenContentDecl : '%c' expected\n",
+ type);
+ ctxt->wellFormed = 0;
+ xmlFreeElementContent(ret);
+ return(NULL);
+ }
+
+ op = xmlNewElementContent(NULL, XML_ELEMENT_CONTENT_OR);
+ if (op == NULL) {
+ xmlFreeElementContent(ret);
+ return(NULL);
+ }
+ if (last == NULL) {
+ op->c1 = ret;
+ ret = cur = op;
+ } else {
+ cur->c2 = op;
+ op->c1 = last;
+ cur =op;
+ }
+ } else {
+ if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
+ ctxt->sax->error(ctxt,
+ "xmlParseElementChildrenContentDecl : ',' '|' or ')' expected\n");
+ ctxt->wellFormed = 0;
+ xmlFreeElementContent(ret);
+ return(NULL);
+ }
+ SKIP_BLANKS;
+ if (CUR == '(') {
+ /* Recurse on second child */
+ NEXT;
+ SKIP_BLANKS;
+ cur = xmlParseElementChildrenContentDecl(ctxt);
+ SKIP_BLANKS;
+ } else {
+ elem = xmlParseName(ctxt);
+ if (elem == NULL) {
+ if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
+ ctxt->sax->error(ctxt,
+ "xmlParseElementChildrenContentDecl : Name or '(' expected\n");
+ ctxt->wellFormed = 0;
+ return(NULL);
+ }
+ cur = xmlNewElementContent(elem, XML_ELEMENT_CONTENT_ELEMENT);
+ }
+ if (CUR == '?') {
+ ret->ocur = XML_ELEMENT_CONTENT_OPT;
+ NEXT;
+ } else if (CUR == '*') {
+ ret->ocur = XML_ELEMENT_CONTENT_MULT;
+ NEXT;
+ } else if (CUR == '+') {
+ ret->ocur = XML_ELEMENT_CONTENT_PLUS;
+ NEXT;
+ } else {
+ ret->ocur = XML_ELEMENT_CONTENT_ONCE;
+ }
+ SKIP_BLANKS;
+ }
+ NEXT;
+ if (CUR == '?') {
+ ret->ocur = XML_ELEMENT_CONTENT_OPT;
+ NEXT;
+ } else if (CUR == '*') {
+ ret->ocur = XML_ELEMENT_CONTENT_MULT;
+ NEXT;
+ } else if (CUR == '+') {
+ ret->ocur = XML_ELEMENT_CONTENT_PLUS;
+ NEXT;
+ } else {
+ ret->ocur = XML_ELEMENT_CONTENT_ONCE;
+ }
+ return(ret);
+}
+
+/**
+ * xmlParseElementContentDecl:
+ * @ctxt: an XML parser context
+ * @name: the name of the element being defined.
+ * @result: the Element Content pointer will be stored here if any
*
- * [51] Mixed ::= '(' S? '#PCDATA' (S? '|' S? Name)* S? ')*' |
- * '(' S? '#PCDATA' S? ')'
+ * parse the declaration for an Element content either Mixed or Children,
+ * the cases EMPTY and ANY are handled directly in xmlParseElementDecl
+ *
+ * [46] contentspec ::= 'EMPTY' | 'ANY' | Mixed | children
*
- * TODO: not implemented !!!
+ * returns: the type of element content XML_ELEMENT_TYPE_xxx
*/
-void
-xmlParseElementContentDecl(xmlParserCtxtPtr ctxt, CHAR *name) {
- /*
- * TODO This has to be parsed correctly, currently we just skip until
- * we reach the first '>'.
- * !!!
- */
- while ((IS_CHAR(CUR)) && (CUR != '>'))
- NEXT;
+int
+xmlParseElementContentDecl(xmlParserCtxtPtr ctxt, CHAR *name,
+ xmlElementContentPtr *result) {
+
+ xmlElementContentPtr tree = NULL;
+ int res;
+
+ *result = NULL;
+
+ if (CUR != '(') {
+ if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
+ ctxt->sax->error(ctxt,
+ "xmlParseElementContentDecl : '(' expected\n");
+ ctxt->wellFormed = 0;
+ return(-1);
+ }
+ NEXT;
+ SKIP_BLANKS;
+ if ((CUR == '#') && (NXT(1) == 'P') &&
+ (NXT(2) == 'C') && (NXT(3) == 'D') &&
+ (NXT(4) == 'A') && (NXT(5) == 'T') &&
+ (NXT(6) == 'A')) {
+ tree = xmlParseElementMixedContentDecl(ctxt);
+ res = XML_ELEMENT_TYPE_MIXED;
+ } else {
+ tree = xmlParseElementChildrenContentDecl(ctxt);
+ res = XML_ELEMENT_TYPE_ELEMENT;
+ }
+ SKIP_BLANKS;
+ /****************************
+ if (CUR != ')') {
+ if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
+ ctxt->sax->error(ctxt,
+ "xmlParseElementContentDecl : ')' expected\n");
+ ctxt->wellFormed = 0;
+ return(-1);
+ }
+ ****************************/
+ return(res);
}
/**
@@ -2157,26 +2682,40 @@
*
* [45] elementdecl ::= '<!ELEMENT' S Name S contentspec S? '>'
*
- * [46] contentspec ::= 'EMPTY' | 'ANY' | Mixed | children
- *
* TODO There is a check [ VC: Unique Element Type Declaration ]
*/
-void
+int
xmlParseElementDecl(xmlParserCtxtPtr ctxt) {
CHAR *name;
+ int ret = -1;
+ xmlElementContentPtr content = NULL;
if ((CUR == '<') && (NXT(1) == '!') &&
(NXT(2) == 'E') && (NXT(3) == 'L') &&
(NXT(4) == 'E') && (NXT(5) == 'M') &&
(NXT(6) == 'E') && (NXT(7) == 'N') &&
- (NXT(8) == 'T') && (IS_BLANK(NXT(9)))) {
+ (NXT(8) == 'T')) {
SKIP(9);
+ if (!IS_BLANK(CUR)) {
+ if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
+ ctxt->sax->error(ctxt,
+ "Space required after 'ELEMENT'\n");
+ ctxt->wellFormed = 0;
+ }
SKIP_BLANKS;
name = xmlParseName(ctxt);
if (name == NULL) {
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
- ctxt->sax->error(ctxt, "xmlParseElementDecl: no name for Element\n");
- return;
+ ctxt->sax->error(ctxt,
+ "xmlParseElementDecl: no name for Element\n");
+ ctxt->wellFormed = 0;
+ return(-1);
+ }
+ if (!IS_BLANK(CUR)) {
+ if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
+ ctxt->sax->error(ctxt,
+ "Space required after the element name\n");
+ ctxt->wellFormed = 0;
}
SKIP_BLANKS;
if ((CUR == 'E') && (NXT(1) == 'M') &&
@@ -2186,23 +2725,39 @@
/*
* Element must always be empty.
*/
+ ret = XML_ELEMENT_TYPE_EMPTY;
} else if ((CUR == 'A') && (NXT(1) == 'N') &&
(NXT(2) == 'Y')) {
SKIP(3);
/*
* Element is a generic container.
*/
+ ret = XML_ELEMENT_TYPE_ANY;
+ } else if (CUR == '(') {
+ ret = xmlParseElementContentDecl(ctxt, name, &content);
} else {
- xmlParseElementContentDecl(ctxt, name);
+ if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
+ ctxt->sax->error(ctxt,
+ "xmlParseElementDecl: 'EMPTY', 'ANY' or '(' expected\n");
+ ctxt->wellFormed = 0;
+ if (name != NULL) free(name);
+ return(-1);
}
SKIP_BLANKS;
if (CUR != '>') {
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt,
"xmlParseElementDecl: expected '>' at the end\n");
- } else
+ ctxt->wellFormed = 0;
+ } else {
NEXT;
+ xmlAddElementDecl(ctxt->doc->intSubset, name, ret, content);
+ }
+ if (name != NULL) {
+ free(name);
+ }
}
+ return(ret);
}
/**
@@ -2254,7 +2809,8 @@
else {
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt,
- "xmlParseCharRef: invalid hexa value\n");
+ "xmlParseCharRef: invalid hexadecimal value\n");
+ ctxt->wellFormed = 0;
val = 0;
break;
}
@@ -2271,6 +2827,7 @@
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt,
"xmlParseCharRef: invalid decimal value\n");
+ ctxt->wellFormed = 0;
val = 0;
break;
}
@@ -2281,6 +2838,7 @@
} else {
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt, "xmlParseCharRef: invalid value\n");
+ ctxt->wellFormed = 0;
}
/*
* Check the value IS_CHAR ...
@@ -2293,6 +2851,7 @@
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt, "xmlParseCharRef: invalid CHAR value %d\n",
val);
+ ctxt->wellFormed = 0;
}
return(NULL);
}
@@ -2311,6 +2870,7 @@
CHAR *ret = NULL;
const CHAR *q;
CHAR *name;
+ xmlEntityPtr ent;
xmlParserInputPtr input = NULL;
q = CUR_PTR;
@@ -2320,10 +2880,62 @@
if (name == NULL) {
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt, "xmlParseEntityRef: no name\n");
+ ctxt->wellFormed = 0;
} else {
if (CUR == ';') {
NEXT;
/*
+ * Well Formedness Constraint if:
+ * - standalone
+ * or
+ * - no external subset and no external parameter entities
+ * referenced
+ * then
+ * the entity referenced must have been declared
+ *
+ * TODO: to be double checked !!!
+ */
+ ent = xmlGetDocEntity(ctxt->doc, name);
+ if ((ctxt->doc->standalone) ||
+ ((ctxt->doc->intSubset == NULL) &&
+ (ctxt->doc->extSubset == NULL))) {
+ if (ent == NULL) {
+ if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
+ ctxt->sax->error(ctxt,
+ "Entity '%s' not defined\n", name);
+ ctxt->wellFormed = 0;
+ }
+ }
+
+ /*
+ * Well Formedness Constraint :
+ * The referenced entity must be a parsed entity.
+ */
+ if (ent != NULL) {
+ switch (ent->type) {
+ case XML_INTERNAL_PARAMETER_ENTITY:
+ case XML_EXTERNAL_PARAMETER_ENTITY:
+ if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
+ ctxt->sax->error(ctxt,
+ "Attempt to reference the parameter entity '%s'\n", name);
+ ctxt->wellFormed = 0;
+ break;
+
+ case XML_EXTERNAL_GENERAL_UNPARSED_ENTITY:
+ if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
+ ctxt->sax->error(ctxt,
+ "Attempt to reference unparsed entity '%s'\n", name);
+ ctxt->wellFormed = 0;
+ break;
+ }
+ }
+
+ /*
+ * Well Formedness Constraint :
+ * The referenced entity must not lead to recursion !
+ */
+
+ /*
* We parsed the entity reference correctly, call SAX
* interface for the proper behaviour:
* - get a new input stream
@@ -2340,7 +2952,9 @@
char cst[2] = { '&', 0 };
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
- ctxt->sax->error(ctxt, "xmlParseEntityRef: expecting ';'\n");
+ ctxt->sax->error(ctxt,
+ "xmlParseEntityRef: expecting ';'\n");
+ ctxt->wellFormed = 0;
ret = xmlStrndup(cst, 1);
ret = xmlStrcat(ret, name);
}
@@ -2363,7 +2977,14 @@
CHAR *
xmlParseReference(xmlParserCtxtPtr ctxt) {
if ((CUR == '&') && (NXT(1) == '#')) {
- return(xmlParseCharRef(ctxt));
+ CHAR *val = xmlParseCharRef(ctxt);
+ xmlParserInputPtr in;
+
+ if (val != NULL) {
+ in = xmlNewStringInputStream(ctxt, val);
+ xmlPushInput(ctxt, in);
+ }
+ return(NULL);
} else if (CUR == '&') {
return(xmlParseEntityRef(ctxt));
}
@@ -2392,6 +3013,7 @@
if (name == NULL) {
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt, "xmlParsePEReference: no name\n");
+ ctxt->wellFormed = 0;
} else {
if (CUR == ';') {
NEXT;
@@ -2399,7 +3021,7 @@
if (entity == NULL) {
if ((ctxt->sax != NULL) && (ctxt->sax->warning != NULL))
ctxt->sax->warning(ctxt,
- "xmlParsePEReference: %%%s; not found\n");
+ "xmlParsePEReference: %%%s; not found\n", name);
} else {
input = xmlNewEntityInputStream(ctxt, entity);
xmlPushInput(ctxt, input);
@@ -2408,7 +3030,9 @@
char cst[2] = { '%', 0 };
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
- ctxt->sax->error(ctxt, "xmlParsePEReference: expecting ';'\n");
+ ctxt->sax->error(ctxt,
+ "xmlParsePEReference: expecting ';'\n");
+ ctxt->wellFormed = 0;
ret = xmlStrndup(cst, 1);
ret = xmlStrcat(ret, name);
}
@@ -2449,6 +3073,7 @@
if (name == NULL) {
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt, "xmlParseDocTypeDecl : no DOCTYPE name !\n");
+ ctxt->wellFormed = 0;
}
SKIP_BLANKS;
@@ -2459,7 +3084,7 @@
URI = xmlParseExternalID(ctxt, &ExternalID);
SKIP_BLANKS;
- dtd = xmlNewDtd(ctxt->doc, name, ExternalID, URI);
+ dtd = xmlCreateIntSubset(ctxt->doc, name, ExternalID, URI);
/*
* Is there any DTD definition ?
@@ -2482,6 +3107,7 @@
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt,
"xmlParseDocTypeDecl: error detected in Markup declaration\n");
+ ctxt->wellFormed = 0;
break;
}
}
@@ -2494,6 +3120,7 @@
if (CUR != '>') {
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt, "DOCTYPE unproperly terminated\n");
+ ctxt->wellFormed = 0;
/* We shouldn't try to resynchronize ... */
}
NEXT;
@@ -2527,7 +3154,7 @@
*/
xmlAttrPtr xmlParseAttribute(xmlParserCtxtPtr ctxt, xmlNodePtr node) {
- CHAR *name;
+ CHAR *name, *val;
CHAR *ns;
CHAR *value = NULL;
xmlAttrPtr ret;
@@ -2536,6 +3163,7 @@
if (name == NULL) {
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt, "error parsing attribute name\n");
+ ctxt->wellFormed = 0;
return(NULL);
}
@@ -2549,8 +3177,9 @@
value = xmlParseAttValue(ctxt);
} else {
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
- ctxt->sax->error(ctxt, "Specification mandate value for attribute %s\n",
- name);
+ ctxt->sax->error(ctxt,
+ "Specification mandate value for attribute %s\n", name);
+ ctxt->wellFormed = 0;
}
/*
@@ -2579,9 +3208,20 @@
return(NULL);
}
- ret = xmlNewProp(ctxt->node, name, NULL);
- if (ret != NULL)
- ret->val = xmlStringGetNodeList(ctxt->doc, value);
+ /*
+ * Well formedness requires at most one declaration of an attribute
+ */
+ if ((val = xmlGetProp(ctxt->node, name)) != NULL) {
+ free(val);
+ if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
+ ctxt->sax->error(ctxt, "Attribute %s redefined\n", name);
+ ctxt->wellFormed = 0;
+ ret = NULL;
+ } else {
+ ret = xmlNewProp(ctxt->node, name, NULL);
+ if (ret != NULL)
+ ret->val = xmlStringGetNodeList(ctxt->doc, value);
+ }
if (ns != NULL)
free(ns);
@@ -2621,7 +3261,13 @@
NEXT;
name = xmlNamespaceParseQName(ctxt, &namespace);
- if (name == NULL) return(NULL);
+ if (name == NULL) {
+ if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
+ ctxt->sax->error(ctxt,
+ "xmlParseStartTag: invalid element name\n");
+ ctxt->wellFormed = 0;
+ return(NULL);
+ }
/*
* Note : the namespace resolution is deferred until the end of the
@@ -2659,6 +3305,7 @@
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt,
"xmlParseStartTag: problem parsing attributes\n");
+ ctxt->wellFormed = 0;
break;
}
}
@@ -2719,6 +3366,7 @@
if ((CUR != '<') || (NXT(1) != '/')) {
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt, "xmlParseEndTag: '</' not found\n");
+ ctxt->wellFormed = 0;
return;
}
SKIP(2);
@@ -2742,6 +3390,7 @@
if ((!IS_CHAR(CUR)) || (CUR != '>')) {
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt, "End tag : expected '>'\n");
+ ctxt->wellFormed = 0;
} else
NEXT;
@@ -2778,12 +3427,14 @@
if (!IS_CHAR(CUR)) {
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt, "CData section not finished\n%.50s\n", base);
+ ctxt->wellFormed = 0;
return;
}
r = NEXT;
if (!IS_CHAR(CUR)) {
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt, "CData section not finished\n%.50s\n", base);
+ ctxt->wellFormed = 0;
return;
}
s = NEXT;
@@ -2794,6 +3445,7 @@
if (!IS_CHAR(CUR)) {
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt, "CData section not finished\n%.50s\n", base);
+ ctxt->wellFormed = 0;
return;
}
@@ -2802,9 +3454,9 @@
*/
if (ctxt->sax != NULL) {
if (areBlanks(ctxt, base, CUR_PTR - base))
- ctxt->sax->ignorableWhitespace(ctxt, base, 0, CUR_PTR - base);
+ ctxt->sax->ignorableWhitespace(ctxt, base, 0, (CUR_PTR - base) - 2);
else
- ctxt->sax->characters(ctxt, base, 0, CUR_PTR - base);
+ ctxt->sax->characters(ctxt, base, 0, (CUR_PTR - base) - 2);
}
}
@@ -2891,7 +3543,9 @@
if (test == CUR_PTR) {
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
- ctxt->sax->error(ctxt, "detected an error in element content\n");
+ ctxt->sax->error(ctxt,
+ "detected an error in element content\n");
+ ctxt->wellFormed = 0;
break;
}
}
@@ -2946,6 +3600,7 @@
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt, "Couldn't find end of Start Tag\n%.30s\n",
openTag);
+ ctxt->wellFormed = 0;
/*
* end of parsing of this node.
@@ -2963,6 +3618,7 @@
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt,
"Premature end of data in tag %.30s\n", openTag);
+ ctxt->wellFormed = 0;
/*
* end of parsing of this node.
@@ -2985,15 +3641,18 @@
ctxt->sax->error(ctxt,
"Start and End tags don't use the same namespace\n%.30s\n%.30s\n",
openTag, endTag);
+ ctxt->wellFormed = 0;
}
if (endTag == NULL ) {
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt, "The End tag has no name\n%.30s\n", openTag);
+ ctxt->wellFormed = 0;
} else if (xmlStrcmp(ret->name, endTag)) {
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt,
"Start and End tags don't use the same name\n%.30s\n%.30s\n",
openTag, endTag);
+ ctxt->wellFormed = 0;
}
/*
* SAX: End of Tag
@@ -3071,6 +3730,7 @@
if (CUR != '=') {
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt, "xmlParseVersionInfo : expected '='\n");
+ ctxt->wellFormed = 0;
return(NULL);
}
NEXT;
@@ -3082,6 +3742,7 @@
if (CUR != '"') {
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt, "String not closed\n%.50s\n", q);
+ ctxt->wellFormed = 0;
} else
NEXT;
} else if (CUR == '\''){
@@ -3091,11 +3752,14 @@
if (CUR != '\'') {
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt, "String not closed\n%.50s\n", q);
+ ctxt->wellFormed = 0;
} else
NEXT;
} else {
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
- ctxt->sax->error(ctxt, "xmlParseVersionInfo : expected ' or \"\n");
+ ctxt->sax->error(ctxt,
+ "xmlParseVersionInfo : expected ' or \"\n");
+ ctxt->wellFormed = 0;
}
}
return(version);
@@ -3128,6 +3792,7 @@
} else {
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt, "Invalid XML encoding name\n");
+ ctxt->wellFormed = 0;
}
return(ret);
}
@@ -3160,6 +3825,7 @@
if (CUR != '=') {
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt, "xmlParseEncodingDecl : expected '='\n");
+ ctxt->wellFormed = 0;
return(NULL);
}
NEXT;
@@ -3171,6 +3837,7 @@
if (CUR != '"') {
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt, "String not closed\n%.50s\n", q);
+ ctxt->wellFormed = 0;
} else
NEXT;
} else if (CUR == '\''){
@@ -3180,11 +3847,14 @@
if (CUR != '\'') {
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt, "String not closed\n%.50s\n", q);
+ ctxt->wellFormed = 0;
} else
NEXT;
} else if (CUR == '"'){
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
- ctxt->sax->error(ctxt, "xmlParseEncodingDecl : expected ' or \"\n");
+ ctxt->sax->error(ctxt,
+ "xmlParseEncodingDecl : expected ' or \"\n");
+ ctxt->wellFormed = 0;
}
}
return(encoding);
@@ -3214,7 +3884,9 @@
SKIP(10);
if (CUR != '=') {
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
- ctxt->sax->error(ctxt, "XML standalone declaration : expected '='\n");
+ ctxt->sax->error(ctxt,
+ "XML standalone declaration : expected '='\n");
+ ctxt->wellFormed = 0;
return(standalone);
}
NEXT;
@@ -3231,10 +3903,12 @@
} else {
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt, "standalone accepts only 'yes' or 'no'\n");
+ ctxt->wellFormed = 0;
}
if (CUR != '\'') {
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt, "String not closed\n");
+ ctxt->wellFormed = 0;
} else
NEXT;
} else if (CUR == '"'){
@@ -3248,16 +3922,20 @@
SKIP(3);
} else {
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
- ctxt->sax->error(ctxt, "standalone accepts only 'yes' or 'no'\n");
+ ctxt->sax->error(ctxt,
+ "standalone accepts only 'yes' or 'no'\n");
+ ctxt->wellFormed = 0;
}
if (CUR != '"') {
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt, "String not closed\n");
+ ctxt->wellFormed = 0;
} else
NEXT;
} else {
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt, "Standalone value not found\n");
+ ctxt->wellFormed = 0;
}
}
return(standalone);
@@ -3281,6 +3959,11 @@
*/
SKIP(5);
+ if (!IS_BLANK(CUR)) {
+ if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
+ ctxt->sax->error(ctxt, "Blank needed after '<?xml'\n");
+ ctxt->wellFormed = 0;
+ }
SKIP_BLANKS;
/*
@@ -3295,11 +3978,30 @@
/*
* We may have the encoding declaration
*/
+ if (!IS_BLANK(CUR)) {
+ if ((CUR == '?') && (NXT(1) == '>')) {
+ SKIP(2);
+ return;
+ }
+ if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
+ ctxt->sax->error(ctxt, "Blank needed here\n");
+ ctxt->wellFormed = 0;
+ }
ctxt->doc->encoding = xmlParseEncodingDecl(ctxt);
/*
* We may have the standalone status.
*/
+ if ((ctxt->doc->encoding != NULL) && (!IS_BLANK(CUR))) {
+ if ((CUR == '?') && (NXT(1) == '>')) {
+ SKIP(2);
+ return;
+ }
+ if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
+ ctxt->sax->error(ctxt, "Blank needed here\n");
+ ctxt->wellFormed = 0;
+ }
+ SKIP_BLANKS;
ctxt->doc->standalone = xmlParseSDDecl(ctxt);
SKIP_BLANKS;
@@ -3309,10 +4011,12 @@
/* Deprecated old WD ... */
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt, "XML declaration must end-up with '?>'\n");
+ ctxt->wellFormed = 0;
NEXT;
} else {
if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
ctxt->sax->error(ctxt, "parsing XML declaration: '?>' expected\n");
+ ctxt->wellFormed = 0;
MOVETO_ENDTAG(CUR_PTR);
NEXT;
}
@@ -3377,7 +4081,19 @@
/*
* Wipe out everything which is before the first '<'
*/
- SKIP_BLANKS;
+ if (IS_BLANK(CUR)) {
+ if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
+ ctxt->sax->error(ctxt,
+ "Extra spaces at the beginning of the document are not allowed\n");
+ ctxt->wellFormed = 0;
+ SKIP_BLANKS;
+ }
+
+ if (CUR == 0) {
+ if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
+ ctxt->sax->error(ctxt, "Document is empty\n");
+ ctxt->wellFormed = 0;
+ }
/*
* Check for the XMLDecl in the Prolog.
@@ -3434,11 +4150,19 @@
*/
xmlParseMisc(ctxt);
+ if (CUR != 0) {
+ if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
+ ctxt->sax->error(ctxt,
+ "Extra content at the end of the document\n");
+ ctxt->wellFormed = 0;
+ }
+
/*
* SAX: end of the document processing.
*/
if (ctxt->sax)
ctxt->sax->endDocument(ctxt);
+ if (! ctxt->wellFormed) return(-1);
return(0);
}
@@ -3446,6 +4170,8 @@
* xmlSAXParseDoc :
* @sax: the SAX handler block
* @cur: a pointer to an array of CHAR
+ * @recovery: work in recovery mode, i.e. tries to read no Well Formed
+ * documents
*
* parse an XML in-memory document and build a tree.
* It use the given SAX function block to handle the parsing callback.
@@ -3454,7 +4180,7 @@
* return values: the resulting document tree
*/
-xmlDocPtr xmlSAXParseDoc(xmlSAXHandlerPtr sax, CHAR *cur) {
+xmlDocPtr xmlSAXParseDoc(xmlSAXHandlerPtr sax, CHAR *cur, int recovery) {
xmlDocPtr ret;
xmlParserCtxtPtr ctxt;
xmlParserInputPtr input;
@@ -3485,7 +4211,12 @@
xmlParseDocument(ctxt);
- ret = ctxt->doc;
+ if ((ctxt->wellFormed) || recovery) ret = ctxt->doc;
+ else {
+ ret = NULL;
+ xmlFreeDoc(ctxt->doc);
+ ctxt->doc = NULL;
+ }
free(ctxt->nodeTab);
free(ctxt->inputTab);
if (input->filename != NULL)
@@ -3506,13 +4237,29 @@
*/
xmlDocPtr xmlParseDoc(CHAR *cur) {
- return(xmlSAXParseDoc(NULL, cur));
+ return(xmlSAXParseDoc(NULL, cur, 0));
+}
+
+/**
+ * xmlRecoverDoc :
+ * @cur: a pointer to an array of CHAR
+ *
+ * parse an XML in-memory document and build a tree.
+ * In the case the document is not Well Formed, a tree is built anyway
+ *
+ * return values: the resulting document tree
+ */
+
+xmlDocPtr xmlRecoverDoc(CHAR *cur) {
+ return(xmlSAXParseDoc(NULL, cur, 1));
}
/**
* xmlSAXParseFile :
* @sax: the SAX handler block
* @filename: the filename
+ * @recovery: work in recovery mode, i.e. tries to read no Well Formed
+ * documents
*
* parse an XML file and build a tree. Automatic support for ZLIB/Compress
* compressed document is provided by default if found at compile-time.
@@ -3522,7 +4269,8 @@
* return values: the resulting document tree
*/
-xmlDocPtr xmlSAXParseFile(xmlSAXHandlerPtr sax, const char *filename) {
+xmlDocPtr xmlSAXParseFile(xmlSAXHandlerPtr sax, const char *filename,
+ int recovery) {
xmlDocPtr ret;
#ifdef HAVE_ZLIB_H
gzFile input;
@@ -3624,7 +4372,12 @@
xmlParseDocument(ctxt);
- ret = ctxt->doc;
+ if ((ctxt->wellFormed) || recovery) ret = ctxt->doc;
+ else {
+ ret = NULL;
+ xmlFreeDoc(ctxt->doc);
+ ctxt->doc = NULL;
+ }
free(buffer);
free(ctxt->nodeTab);
free(ctxt->inputTab);
@@ -3647,7 +4400,22 @@
*/
xmlDocPtr xmlParseFile(const char *filename) {
- return(xmlSAXParseFile(NULL, filename));
+ return(xmlSAXParseFile(NULL, filename, 0));
+}
+
+/**
+ * xmlRecoverFile :
+ * @filename: the filename
+ *
+ * parse an XML file and build a tree. Automatic support for ZLIB/Compress
+ * compressed document is provided by default if found at compile-time.
+ * In the case the document is not Well Formed, a tree is built anyway
+ *
+ * return values: the resulting document tree
+ */
+
+xmlDocPtr xmlRecoverFile(const char *filename) {
+ return(xmlSAXParseFile(NULL, filename, 1));
}
/**
@@ -3655,6 +4423,8 @@
* @sax: the SAX handler block
* @cur: an pointer to a char array
* @size: the siwe of the array
+ * @recovery: work in recovery mode, i.e. tries to read no Well Formed
+ * documents
*
* parse an XML in-memory block and use the given SAX function block
* to handle the parsing callback. If sax is NULL, fallback to the default
@@ -3665,7 +4435,8 @@
* return values: the resulting document tree
*/
-xmlDocPtr xmlSAXParseMemory(xmlSAXHandlerPtr sax, char *buffer, int size) {
+xmlDocPtr xmlSAXParseMemory(xmlSAXHandlerPtr sax, char *buffer, int size,
+ int recovery) {
xmlDocPtr ret;
xmlParserCtxtPtr ctxt;
xmlParserInputPtr input;
@@ -3702,7 +4473,12 @@
xmlParseDocument(ctxt);
- ret = ctxt->doc;
+ if ((ctxt->wellFormed) || recovery) ret = ctxt->doc;
+ else {
+ ret = NULL;
+ xmlFreeDoc(ctxt->doc);
+ ctxt->doc = NULL;
+ }
free(ctxt->nodeTab);
free(ctxt->inputTab);
if (input->filename != NULL)
@@ -3724,7 +4500,22 @@
*/
xmlDocPtr xmlParseMemory(char *buffer, int size) {
- return(xmlSAXParseMemory(NULL, buffer, size));
+ return(xmlSAXParseMemory(NULL, buffer, size, 0));
+}
+
+/**
+ * xmlRecoverMemory :
+ * @cur: an pointer to a char array
+ * @size: the size of the array
+ *
+ * parse an XML in-memory block and build a tree.
+ * In the case the document is not Well Formed, a tree is built anyway
+ *
+ * return values: the resulting document tree
+ */
+
+xmlDocPtr xmlRecoverMemory(char *buffer, int size) {
+ return(xmlSAXParseMemory(NULL, buffer, size, 1));
}
/**
@@ -3751,6 +4542,7 @@
ctxt->sax = &xmlDefaultSAXHandler;
ctxt->doc = NULL;
+ ctxt->wellFormed = 1;
ctxt->record_info = 0;
xmlInitNodeInfoSeq(&ctxt->node_seq);
}
diff --git a/parser.h b/parser.h
index 4cd2ead..a06d481 100644
--- a/parser.h
+++ b/parser.h
@@ -3,7 +3,7 @@
*
* See Copyright for the status of this software.
*
- * $Id$
+ * Daniel.Veillard@w3.org
*/
#ifndef __XML_PARSER_H__
@@ -46,6 +46,7 @@
typedef struct xmlParserCtxt {
struct xmlSAXHandler *sax; /* The SAX handler */
xmlDocPtr doc; /* the document being built */
+ int wellFormed; /* is the document well formed */
/* Input stream stack */
xmlParserInputPtr input; /* Current input stream */
@@ -131,15 +132,8 @@
#include "entities.h"
/*
- * Interfaces
+ * CHAR handling
*/
-extern int xmlParseDocument(xmlParserCtxtPtr ctxt);
-extern xmlDocPtr xmlParseDoc(CHAR *cur);
-extern xmlDocPtr xmlParseMemory(char *buffer, int size);
-extern xmlDocPtr xmlParseFile(const char *filename);
-extern xmlDocPtr xmlSAXParseDoc(xmlSAXHandlerPtr sax, CHAR *cur);
-extern xmlDocPtr xmlSAXParseMemory(xmlSAXHandlerPtr sax, char *buffer, int size);
-extern xmlDocPtr xmlSAXParseFile(xmlSAXHandlerPtr sax, const char *filename);
extern CHAR *xmlStrdup(const CHAR *input);
extern CHAR *xmlStrndup(const CHAR *input, int n);
extern CHAR *xmlStrchr(const CHAR *str, CHAR val);
@@ -149,6 +143,29 @@
extern CHAR *xmlStrcat(CHAR *cur, const CHAR *add);
extern CHAR *xmlStrncat(CHAR *cur, const CHAR *add, int len);
+/*
+ * Interfaces
+ */
+extern xmlDocPtr xmlParseDoc(CHAR *cur);
+extern xmlDocPtr xmlParseMemory(char *buffer, int size);
+extern xmlDocPtr xmlParseFile(const char *filename);
+
+/*
+ * Recovery mode
+ */
+extern xmlDocPtr xmlRecoverDoc(CHAR *cur);
+extern xmlDocPtr xmlRecoverMemory(char *buffer, int size);
+extern xmlDocPtr xmlRecoverFile(const char *filename);
+
+/*
+ * Internal routines
+ */
+extern int xmlParseDocument(xmlParserCtxtPtr ctxt);
+extern xmlDocPtr xmlSAXParseDoc(xmlSAXHandlerPtr sax, CHAR *cur, int recovery);
+extern xmlDocPtr xmlSAXParseMemory(xmlSAXHandlerPtr sax, char *buffer,
+ int size, int recovery);
+extern xmlDocPtr xmlSAXParseFile(xmlSAXHandlerPtr sax, const char *filename,
+ int recovery);
extern void xmlInitParserCtxt(xmlParserCtxtPtr ctx);
extern void xmlClearParserCtxt(xmlParserCtxtPtr ctx);
extern void xmlSetupParserForBuffer(xmlParserCtxtPtr ctx, const CHAR* buffer,
diff --git a/tester.c b/tester.c
index 6830917..945464e 100644
--- a/tester.c
+++ b/tester.c
@@ -3,7 +3,7 @@
*
* See Copyright for the status of this software.
*
- * $Id$
+ * Daniel.Veillard@w3.org
*/
#ifdef WIN32
@@ -32,13 +32,14 @@
static int debug = 0;
static int copy = 0;
+static int recovery = 0;
/*
* Note: there is a couple of errors introduced on purpose.
*/
static CHAR buffer[] =
"\n\
-<?xml version=\"1.0\">\n\
+<?xml version=\"1.0\"?>\n\
<?xml:namespace ns = \"http://www.ietf.org/standards/dav/\" prefix = \"D\"?>\n\
<?xml:namespace ns = \"http://www.w3.com/standards/z39.50/\" prefix = \"Z\"?>\n\
<D:propertyupdate>\n\
@@ -109,7 +110,10 @@
/*
* build an XML tree from a string;
*/
- doc = xmlParseFile(filename);
+ if (recovery)
+ doc = xmlRecoverFile(filename);
+ else
+ doc = xmlParseFile(filename);
/*
* test intermediate copy if needed.
@@ -140,7 +144,10 @@
/*
* build an XML tree from a string;
*/
- doc = xmlParseDoc(buf);
+ if (recovery)
+ doc = xmlRecoverDoc(buf);
+ else
+ doc = xmlParseDoc(buf);
/*
* test intermediate copy if needed.
@@ -174,6 +181,9 @@
debug++;
else if ((!strcmp(argv[i], "-copy")) || (!strcmp(argv[i], "--copy")))
copy++;
+ else if ((!strcmp(argv[i], "-recover")) ||
+ (!strcmp(argv[i], "--recover")))
+ recovery++;
}
for (i = 1; i < argc ; i++) {
if (argv[i][0] != '-') {
diff --git a/tree.c b/tree.c
index e1dbd29..4d643d8 100644
--- a/tree.c
+++ b/tree.c
@@ -3,9 +3,9 @@
*
* See Copyright for the status of this software.
*
- * $Id$
- *
* TODO Cleanup the Dump mechanism.
+ *
+ * Daniel.Veillard@w3.org
*/
#include "config.h"
@@ -27,6 +27,15 @@
static int xmlCompressMode = 0;
+#define UPDATE_LAST_CHILD(n) if ((n) != NULL) { \
+ xmlNodePtr ulccur = (n)->childs; \
+ if (ulccur == NULL) { \
+ (n)->last = NULL; \
+ } else { \
+ while (ulccur->next != NULL) ulccur = ulccur->next; \
+ (n)->last = ulccur; \
+} }
+
/************************************************************************
* *
* Allocation and deallocation of basic structures *
@@ -234,9 +243,10 @@
const CHAR *ExternalID, const CHAR *SystemID) {
xmlDtdPtr cur;
- if ((doc != NULL) && (doc->dtd != NULL)) {
+ if ((doc != NULL) && (doc->extSubset != NULL)) {
fprintf(stderr, "xmlNewDtd(%s): document %s already have a DTD %s\n",
- /* !!! */ (char *) name, doc->name, /* !!! */ (char *)doc->dtd->name);
+ /* !!! */ (char *) name, doc->name,
+ /* !!! */ (char *)doc->extSubset->name);
}
/*
@@ -263,7 +273,58 @@
cur->elements = NULL;
cur->entities = NULL;
if (doc != NULL)
- doc->dtd = cur;
+ doc->extSubset = cur;
+
+ return(cur);
+}
+
+/**
+ * xmlCreateIntSubset:
+ * @doc: the document pointer
+ * @name: the DTD name
+ * @ExternalID: the external ID
+ * @SystemID: the system ID
+ *
+ * Creatte the internal subset of a document
+ * return values: a pointer to the new DTD structure
+ */
+xmlDtdPtr
+xmlCreateIntSubset(xmlDocPtr doc, const CHAR *name,
+ const CHAR *ExternalID, const CHAR *SystemID) {
+ xmlDtdPtr cur;
+
+ if ((doc != NULL) && (doc->intSubset != NULL)) {
+ fprintf(stderr,
+ "xmlCreateIntSubset(): document %s already have an internal subset\n",
+ doc->name);
+ return(NULL);
+ }
+
+ /*
+ * Allocate a new DTD and fill the fields.
+ */
+ cur = (xmlDtdPtr) malloc(sizeof(xmlDtd));
+ if (cur == NULL) {
+ fprintf(stderr, "xmlNewDtd : malloc failed\n");
+ return(NULL);
+ }
+
+ if (name != NULL)
+ cur->name = xmlStrdup(name);
+ else
+ cur->name = NULL;
+ if (ExternalID != NULL)
+ cur->ExternalID = xmlStrdup(ExternalID);
+ else
+ cur->ExternalID = NULL;
+ if (SystemID != NULL)
+ cur->SystemID = xmlStrdup(SystemID);
+ else
+ cur->SystemID = NULL;
+ cur->elements = NULL;
+ cur->entities = NULL;
+ if (doc != NULL)
+ doc->intSubset = cur;
return(cur);
}
@@ -319,10 +380,10 @@
cur->version = xmlStrdup(version);
cur->name = NULL;
cur->root = NULL;
- cur->dtd = NULL;
+ cur->intSubset = NULL;
+ cur->extSubset = NULL;
cur->oldNs = NULL;
cur->encoding = NULL;
- cur->entities = NULL;
cur->standalone = -1;
cur->compression = xmlCompressMode;
#ifndef WITHOUT_CORBA
@@ -342,17 +403,18 @@
void
xmlFreeDoc(xmlDocPtr cur) {
if (cur == NULL) {
+#ifdef DEBUG_TREE
fprintf(stderr, "xmlFreeDoc : document == NULL\n");
+#endif
return;
}
free((char *) cur->version);
if (cur->name != NULL) free((char *) cur->name);
if (cur->encoding != NULL) free((char *) cur->encoding);
if (cur->root != NULL) xmlFreeNode(cur->root);
- if (cur->dtd != NULL) xmlFreeDtd(cur->dtd);
+ if (cur->intSubset != NULL) xmlFreeDtd(cur->intSubset);
+ if (cur->extSubset != NULL) xmlFreeDtd(cur->extSubset);
if (cur->oldNs != NULL) xmlFreeNsList(cur->oldNs);
- if (cur->entities != NULL)
- xmlFreeEntitiesTable((xmlEntitiesTablePtr) cur->entities);
memset(cur, -1, sizeof(xmlDoc));
free(cur);
}
@@ -803,6 +865,7 @@
cur->next = NULL;
cur->prev = NULL;
cur->childs = NULL;
+ cur->last = NULL;
cur->properties = NULL;
cur->name = xmlStrdup(name);
cur->ns = ns;
@@ -834,8 +897,10 @@
cur = xmlNewNode(ns, name);
if (cur != NULL) {
cur->doc = doc;
- if (content != NULL)
+ if (content != NULL) {
cur->childs = xmlStringGetNodeList(doc, content);
+ UPDATE_LAST_CHILD(cur);
+ }
}
return(cur);
}
@@ -867,6 +932,7 @@
cur->next = NULL;
cur->prev = NULL;
cur->childs = NULL;
+ cur->last = NULL;
cur->properties = NULL;
cur->type = XML_TEXT_NODE;
cur->name = xmlStrdup(xmlStringText);
@@ -907,6 +973,7 @@
cur->next = NULL;
cur->prev = NULL;
cur->childs = NULL;
+ cur->last = NULL;
cur->properties = NULL;
if (name[0] == '&') {
int len;
@@ -973,6 +1040,7 @@
cur->prev = NULL;
cur->next = NULL;
cur->childs = NULL;
+ cur->last = NULL;
cur->properties = NULL;
cur->type = XML_TEXT_NODE;
cur->name = xmlStrdup(xmlStringText);
@@ -1030,6 +1098,7 @@
cur->prev = NULL;
cur->next = NULL;
cur->childs = NULL;
+ cur->last = NULL;
cur->properties = NULL;
cur->type = XML_COMMENT_NODE;
cur->name = xmlStrdup(xmlStringText);
@@ -1104,11 +1173,12 @@
cur->doc = parent->doc;
if (parent->childs == NULL) {
parent->childs = cur;
+ parent->last = cur;
} else {
- prev = parent->childs;
- while (prev->next != NULL) prev = prev->next;
+ prev = parent->last;
prev->next = cur;
cur->prev = prev;
+ parent->last = cur;
}
return(cur);
@@ -1146,6 +1216,7 @@
*/
cur->parent = parent;
cur->doc = parent->doc; /* the parent may not be linked to a doc ! */
+
/*
* Handle the case where parent->content != NULL, in that case it will
* create a intermediate TEXT node.
@@ -1159,17 +1230,19 @@
if (text->next != NULL)
text->next->prev = text;
parent->childs = text;
+ UPDATE_LAST_CHILD(parent);
free(parent->content);
parent->content = NULL;
}
}
if (parent->childs == NULL) {
parent->childs = cur;
+ parent->last = cur;
} else {
- prev = parent->childs;
- while (prev->next != NULL) prev = prev->next;
+ prev = parent->last;
prev->next = cur;
cur->prev = prev;
+ parent->last = cur;
}
return(cur);
@@ -1184,23 +1257,11 @@
*/
xmlNodePtr
xmlGetLastChild(xmlNodePtr parent) {
- xmlNodePtr last;
-
if (parent == NULL) {
fprintf(stderr, "xmlGetLastChild : parent == NULL\n");
return(NULL);
}
-
- /*
- * add the new element at the end of the childs list.
- */
- if (parent->childs == NULL) {
- return(NULL);
- } else {
- last = parent->childs;
- while (last->next != NULL) last = last->next;
- }
- return(last);
+ return(parent->last);
}
/**
@@ -1264,6 +1325,8 @@
}
if ((cur->parent != NULL) && (cur->parent->childs == cur))
cur->parent->childs = cur->next;
+ if ((cur->parent != NULL) && (cur->parent->last == cur))
+ cur->parent->last = cur->prev;
if (cur->next != NULL)
cur->next->prev = cur->prev;
if (cur->prev != NULL)
@@ -1419,6 +1482,7 @@
ret->next = NULL;
ret->prev = NULL;
ret->childs = NULL;
+ ret->last = NULL;
ret->properties = NULL;
if (node->name != NULL)
ret->name = xmlStrdup(node->name);
@@ -1469,6 +1533,7 @@
}
if (node->childs != NULL)
ret->childs = xmlStaticCopyNodeList(node->childs, doc, ret);
+ UPDATE_LAST_CHILD(ret);
return(ret);
}
@@ -1601,11 +1666,8 @@
ret->standalone = doc->standalone;
if (!recursive) return(ret);
- if (doc->dtd != NULL)
- ret->dtd = xmlCopyDtd(doc->dtd);
- if (doc->entities != NULL)
- ret->entities = (void *) xmlCopyEntitiesTable(
- (xmlEntitiesTablePtr) doc->entities);
+ if (doc->intSubset != NULL)
+ ret->intSubset = xmlCopyDtd(doc->intSubset);
if (doc->oldNs != NULL)
ret->oldNs = xmlCopyNamespaceList(doc->oldNs);
if (doc->root != NULL)
@@ -1677,6 +1739,7 @@
}
if (cur->childs != NULL) xmlFreeNode(cur->childs);
cur->childs = xmlStringGetNodeList(cur->doc, content);
+ UPDATE_LAST_CHILD(cur);
break;
case XML_ATTRIBUTE_NODE:
break;
@@ -1688,6 +1751,7 @@
case XML_COMMENT_NODE:
if (cur->content != NULL) free(cur->content);
if (cur->childs != NULL) xmlFreeNode(cur->childs);
+ cur->last = cur->childs = NULL;
if (content != NULL)
cur->content = xmlStrdup(content);
else
@@ -1723,6 +1787,7 @@
}
if (cur->childs != NULL) xmlFreeNode(cur->childs);
cur->childs = xmlStringLenGetNodeList(cur->doc, content, len);
+ UPDATE_LAST_CHILD(cur);
break;
case XML_ATTRIBUTE_NODE:
break;
@@ -1734,6 +1799,7 @@
case XML_COMMENT_NODE:
if (cur->content != NULL) free(cur->content);
if (cur->childs != NULL) xmlFreeNode(cur->childs);
+ cur->childs = cur->last = NULL;
if (content != NULL)
cur->content = xmlStrndup(content, len);
else
@@ -1743,6 +1809,8 @@
break;
case XML_NOTATION_NODE:
if (cur->content != NULL) free(cur->content);
+ if (cur->childs != NULL) xmlFreeNode(cur->childs);
+ cur->childs = cur->last = NULL;
if (content != NULL)
cur->content = xmlStrndup(content, len);
else
@@ -1772,24 +1840,22 @@
xmlNodePtr last = NULL, new;
if (cur->childs != NULL) {
- last = cur->childs;
- while (last->next != NULL) last = last->next;
+ last = cur->last;
} else {
if (cur->content != NULL) {
cur->childs = xmlStringGetNodeList(cur->doc, cur->content);
+ UPDATE_LAST_CHILD(cur);
free(cur->content);
cur->content = NULL;
- if (cur->childs != NULL) {
- last = cur->childs;
- while (last->next != NULL) last = last->next;
- }
+ last = cur->last;
}
}
- new = xmlStringLenGetNodeList(cur->doc, content, len);
+ new = xmlNewTextLen(content, len);
if (new != NULL) {
xmlAddChild(cur, new);
- if ((last != NULL) && (last->next == new))
+ if ((last != NULL) && (last->next == new)) {
xmlTextMerge(last, new);
+ }
}
break;
}
@@ -1936,19 +2002,14 @@
*
* Search and get the value of an attribute associated to a node
* This does the entity substitution.
- *
* return values: the attribute value or NULL if not found.
*/
-const CHAR *xmlGetProp(xmlNodePtr node, const CHAR *name) {
+CHAR *xmlGetProp(xmlNodePtr node, const CHAR *name) {
xmlAttrPtr prop = node->properties;
while (prop != NULL) {
- if (!xmlStrcmp(prop->name, name)) {
- if (prop->val != NULL)
- return(xmlNodeListGetString(node->doc, prop->val, 1));
- else
- return(xmlStrndup("", 1));
- }
+ if (!xmlStrcmp(prop->name, name))
+ return(xmlNodeListGetString(node->doc, prop->val, 1));
prop = prop->next;
}
return(NULL);
@@ -2188,10 +2249,10 @@
*/
static void
xmlDtdDump(xmlDocPtr doc) {
- xmlDtdPtr cur = doc->dtd;
+ xmlDtdPtr cur = doc->intSubset;
if (cur == NULL) {
- fprintf(stderr, "xmlDtdDump : DTD == NULL\n");
+ fprintf(stderr, "xmlDtdDump : no internal subset\n");
return;
}
xmlBufferWriteChar("<!DOCTYPE ");
@@ -2207,15 +2268,13 @@
xmlBufferWriteCHAR(cur->SystemID);
xmlBufferWriteChar("\"");
}
- if ((cur->entities == NULL) && (doc->entities == NULL)) {
+ if (cur->entities == NULL) {
xmlBufferWriteChar(">\n");
return;
}
xmlBufferWriteChar(" [\n");
if (cur->entities != NULL)
xmlDumpEntitiesTable((xmlEntitiesTablePtr) cur->entities);
- if (doc->entities != NULL)
- xmlDumpEntitiesTable((xmlEntitiesTablePtr) doc->entities);
xmlBufferWriteChar("]");
/* TODO !!! a lot more things to dump ... */
@@ -2239,13 +2298,13 @@
}
xmlBufferWriteChar(" ");
xmlBufferWriteCHAR(cur->name);
- xmlBufferWriteChar("=\"");
value = xmlNodeListGetString(doc, cur->val, 0);
if (value) {
+ xmlBufferWriteChar("=\"");
xmlBufferWriteCHAR(value);
+ xmlBufferWriteChar("\"");
free(value);
}
- xmlBufferWriteChar("\"");
}
/**
@@ -2401,7 +2460,7 @@
break;
}
xmlBufferWriteChar("?>\n");
- if ((cur->dtd != NULL) || (cur->entities != NULL))
+ if (cur->intSubset != NULL)
xmlDtdDump(cur);
if (cur->root != NULL) {
/* global namespace definitions, the old way */
@@ -2425,7 +2484,9 @@
void
xmlDocDumpMemory(xmlDocPtr cur, CHAR**mem, int *size) {
if (cur == NULL) {
- fprintf(stderr, "xmlDocDump : document == NULL\n");
+#ifdef DEBUG_TREE
+ fprintf(stderr, "xmlDocDumpMemory : document == NULL\n");
+#endif
*mem = NULL;
*size = 0;
return;
@@ -2501,7 +2562,9 @@
void
xmlDocDump(FILE *f, xmlDocPtr cur) {
if (cur == NULL) {
+#ifdef DEBUG_TREE
fprintf(stderr, "xmlDocDump : document == NULL\n");
+#endif
return;
}
buffer_index = 0;
diff --git a/tree.h b/tree.h
index 3921f7e..b6e26cf 100644
--- a/tree.h
+++ b/tree.h
@@ -4,7 +4,7 @@
*
* See Copyright for the status of this software.
*
- * $Id$
+ * Daniel.Veillard@w3.org
*/
#ifndef __XML_TREE_H__
@@ -15,6 +15,8 @@
extern "C" {
#endif
+#include <stdio.h>
+
/*
* The different element types carried by an XML tree
*
@@ -61,18 +63,51 @@
* TODO !!!!
*/
+#define XML_ATTRIBUTE_NONE 1
+#define XML_ATTRIBUTE_REQUIRED 2
+#define XML_ATTRIBUTE_IMPLIED 3
+#define XML_ATTRIBUTE_FIXED 4
+
+#define XML_ATTRIBUTE_STRING 1
+#define XML_ATTRIBUTE_ID 2
+#define XML_ATTRIBUTE_IDREF 3
+#define XML_ATTRIBUTE_IDREFS 4
+#define XML_ATTRIBUTE_ENTITY 5
+#define XML_ATTRIBUTE_ENTITIES 6
+#define XML_ATTRIBUTE_NMTOKEN 7
+#define XML_ATTRIBUTE_NMTOKENS 8
+#define XML_ATTRIBUTE_ENUMERATED 9
+
/*
* a DTD Element definition.
*/
+#define XML_ELEMENT_CONTENT_PCDATA 1
+#define XML_ELEMENT_CONTENT_ELEMENT 2
+#define XML_ELEMENT_CONTENT_SEQ 3
+#define XML_ELEMENT_CONTENT_OR 4
+
+#define XML_ELEMENT_CONTENT_ONCE 1
+#define XML_ELEMENT_CONTENT_OPT 2
+#define XML_ELEMENT_CONTENT_MULT 3
+#define XML_ELEMENT_CONTENT_PLUS 4
+
+typedef struct xmlElementContent {
+ int type; /* PCDATA, ELEMENT, SEQ or OR */
+ int ocur; /* ONCE, OPT, MULT or PLUS */
+ const CHAR *name; /* Element name */
+ struct xmlElementContent *c1; /* first child */
+ struct xmlElementContent *c2; /* second child */
+} xmlElementContent, *xmlElementContentPtr;
+
#define XML_ELEMENT_TYPE_EMPTY 1
#define XML_ELEMENT_TYPE_ANY 2
#define XML_ELEMENT_TYPE_MIXED 3
#define XML_ELEMENT_TYPE_ELEMENT 4
typedef struct xmlElement {
- const CHAR *name; /* Element name */
- int type; /* type (too simple, to extend ...) */
- /* TODO !!! more needed */
+ const CHAR *name; /* Element name */
+ int type; /* The type */
+ xmlElementContentPtr content; /* the allowed element content */
} xmlElement, *xmlElementPtr;
/*
@@ -132,6 +167,7 @@
struct xmlNode *next; /* next sibling link */
struct xmlNode *prev; /* previous sibling link */
struct xmlNode *childs; /* parent->childs link */
+ struct xmlNode *last; /* last child link */
struct xmlAttr *properties; /* properties list */
const CHAR *name; /* the name of the node, or the entity */
xmlNs *ns; /* pointer to the associated namespace */
@@ -153,9 +189,9 @@
const CHAR *encoding; /* encoding, if any */
int compression;/* level of zlib compression */
int standalone; /* standalone document (no external refs) */
- struct xmlDtd *dtd; /* the document DTD if available */
+ struct xmlDtd *intSubset; /* the document internal subset */
+ struct xmlDtd *extSubset; /* the document external subset */
struct xmlNs *oldNs; /* Global namespace, the old way */
- void *entities; /* Hash table for general entities if any */
struct xmlNode *root; /* the document tree */
} xmlDoc, *xmlDocPtr;
@@ -169,6 +205,8 @@
/*
* Creating/freeing new structures
*/
+extern xmlDtdPtr xmlCreateIntSubset(xmlDocPtr doc, const CHAR *name,
+ const CHAR *ExternalID, const CHAR *SystemID);
extern xmlDtdPtr xmlNewDtd(xmlDocPtr doc, const CHAR *name,
const CHAR *ExternalID, const CHAR *SystemID);
extern void xmlFreeDtd(xmlDtdPtr cur);
@@ -240,7 +278,7 @@
*/
extern xmlAttrPtr xmlSetProp(xmlNodePtr node, const CHAR *name,
const CHAR *value);
-extern const CHAR *xmlGetProp(xmlNodePtr node, const CHAR *name);
+extern CHAR *xmlGetProp(xmlNodePtr node, const CHAR *name);
extern xmlNodePtr xmlStringGetNodeList(xmlDocPtr doc, const CHAR *value);
extern xmlNodePtr xmlStringLenGetNodeList(xmlDocPtr doc, const CHAR *value,
int len);
diff --git a/valid.c b/valid.c
new file mode 100644
index 0000000..215e463
--- /dev/null
+++ b/valid.c
@@ -0,0 +1,162 @@
+/*
+ * valid.c : part of the code use to do the DTD handling and the validity
+ * checking
+ *
+ * See Copyright for the status of this software.
+ *
+ * Daniel.Veillard@w3.org
+ */
+
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include "valid.h"
+#include "parser.h"
+
+/****************************************************************
+ * *
+ * Util functions for data allocation/deallocation *
+ * *
+ ****************************************************************/
+
+/**
+ * xmlNewElementContent:
+ * @name: the subelement name or NULL
+ * @type: the type of element content decl
+ *
+ * Allocate an element content structure.
+ *
+ * return values: NULL if not, othervise the new element content structure
+ */
+xmlElementContentPtr
+xmlNewElementContent(CHAR *name, int type) {
+ xmlElementContentPtr ret;
+
+ switch(type) {
+ case XML_ELEMENT_CONTENT_ELEMENT:
+ if (name == NULL) {
+ fprintf(stderr, "xmlNewElementContent : name == NULL !\n");
+ }
+ break;
+ case XML_ELEMENT_CONTENT_PCDATA:
+ case XML_ELEMENT_CONTENT_SEQ:
+ case XML_ELEMENT_CONTENT_OR:
+ if (name != NULL) {
+ fprintf(stderr, "xmlNewElementContent : name != NULL !\n");
+ }
+ break;
+ default:
+ fprintf(stderr, "xmlNewElementContent: unknown type %d\n", type);
+ exit(1);
+ }
+ ret = (xmlElementContentPtr) malloc(sizeof(xmlElementContent));
+ if (ret == NULL) {
+ fprintf(stderr, "xmlNewElementContent : out of memory!\n");
+ return(NULL);
+ }
+ ret->type = type;
+ ret->ocur = XML_ELEMENT_CONTENT_ONCE;
+ ret->name = xmlStrdup(name);
+ return(ret);
+}
+
+/**
+ * xmlNewElementContent:
+ * @name: the subelement name or NULL
+ * @type: the type of element content decl
+ *
+ * Free an element content structure. This is a recursive call !
+ */
+void
+xmlFreeElementContent(xmlElementContentPtr cur) {
+/* TODO !!! */
+}
+
+/****************************************************************
+ * *
+ * Registration of DTD declarations *
+ * *
+ ****************************************************************/
+
+
+/**
+ * xmlAddElementDecl:
+ * @name: the entity name
+ *
+ * Register a new element declaration
+ *
+ * return values: NULL if not, othervise the entity
+ */
+xmlElementPtr
+xmlAddElementDecl(xmlDtdPtr dtd, char *name, int type,
+ xmlElementContentPtr content) {
+ xmlElementPtr ret;
+
+ if (dtd == NULL) {
+ fprintf(stderr, "xmlAddElementDecl: dtd == NULL\n");
+ return(NULL);
+ }
+ if (name == NULL) {
+ fprintf(stderr, "xmlAddElementDecl: name == NULL\n");
+ return(NULL);
+ }
+ switch (type) {
+ case XML_ELEMENT_TYPE_EMPTY:
+ if (content != NULL) {
+ fprintf(stderr,
+ "xmlAddElementDecl: content != NULL for EMPTY\n");
+ return(NULL);
+ }
+ break;
+ case XML_ELEMENT_TYPE_ANY:
+ if (content != NULL) {
+ fprintf(stderr,
+ "xmlAddElementDecl: content != NULL for ANY\n");
+ return(NULL);
+ }
+ break;
+ case XML_ELEMENT_TYPE_MIXED:
+ if (content == NULL) {
+ fprintf(stderr,
+ "xmlAddElementDecl: content == NULL for MIXED\n");
+ return(NULL);
+ }
+ break;
+ case XML_ELEMENT_TYPE_ELEMENT:
+ if (content == NULL) {
+ fprintf(stderr,
+ "xmlAddElementDecl: content == NULL for ELEMENT\n");
+ return(NULL);
+ }
+ break;
+ default:
+ fprintf(stderr, "xmlAddElementDecl: unknown type %d\n", type);
+ return(NULL);
+ }
+
+ /*
+ * Validity Check:
+ * Search the DTD for previous declarations of the ELEMENT
+ */
+ /* TODO */
+
+ /*
+ * Create and fill the structure.
+ */
+ ret = (xmlElementPtr) malloc(sizeof(xmlElement));
+ if (ret == NULL) {
+ fprintf(stderr, "xmlAddElementDecl: out of memory\n");
+ return(NULL);
+ }
+ ret->type = type;
+ ret->name = xmlStrdup(name);
+ ret->content = content;
+
+ /*
+ * Insert the structure in the DTD Element table
+ */
+ /* TODO */
+
+ return(ret);
+}
+
diff --git a/valid.h b/valid.h
new file mode 100644
index 0000000..d6765ff
--- /dev/null
+++ b/valid.h
@@ -0,0 +1,18 @@
+/*
+ * valid.h : interface to the DTD handling and the validity checking
+ *
+ * See Copyright for the status of this software.
+ *
+ * Daniel.Veillard@w3.org
+ */
+
+
+#ifndef __XML_VALID_H__
+#define __XML_VALID_H__
+#include "tree.h"
+
+extern xmlElementPtr xmlAddElementDecl(xmlDtdPtr dtd, char *name, int type,
+ xmlElementContentPtr content);
+extern xmlElementContentPtr xmlNewElementContent(CHAR *name, int type);
+extern void xmlFreeElementContent(xmlElementContentPtr cur);
+#endif /* __XML_VALID_H__ */