Finally commiting work done on the plane, major cleanup,

spread some serious anti bitrot all over the place:
- parserInternals.c parserInternals.h parser.c Makefile.am:
  created a new module parserInternals.c, moved most of the
  code shared by the various parsers there, as well as
  deprecated  code from parser.c. More cleanup of parser.c
- uri.c: fixed a problem when URI is NULL
- valid.c: speedup when looking for an attribute declaration
Daniel

ChangeLog

@@ -1,3 +1,12 @@
+Wed Sep 13 22:03:18 CEST 2000 Daniel Veillard <Daniel.Veillard@w3.org>
+
+	* parserInternals.c parserInternals.h parser.c Makefile.am:
+	  created a new module parserInternals.c, moved most of the
+	  code shared by the various parsers there, as well as
+	  deprecated  code from parser.c. More cleanup of parser.c
+	* uri.c: fixed a problem when URI is NULL
+	* valid.c: speedup when looking for an attribute declaration
+
 Sun Sep 10 17:53:48 CEST 2000 Daniel Veillard <Daniel.Veillard@w3.org>
 
 	* uri.c tree.c SAX.c parser.c entities.c debugXML.c: finished

HTMLparser.c

@@ -48,7 +48,6 @@
 #include "xml-error.h"
 
 #define HTML_MAX_NAMELEN 1000
-#define INPUT_CHUNK     50
 #define HTML_PARSER_BIG_BUFFER_SIZE 1000
 #define HTML_PARSER_BUFFER_SIZE 100
 

Makefile.am

@@ -20,6 +20,7 @@ libxml_la_SOURCES = \
 		entities.c \
 		encoding.c \
 		error.c \
+		parserInternals.c \
 		parser.c \
 		tree.c \
 		xmlIO.c \

include/libxml/parser.h

@@ -371,6 +371,7 @@ xmlDocPtr	xmlRecoverFile		(const char *filename);
  * Less common routines and SAX interfaces
  */
 int		xmlParseDocument	(xmlParserCtxtPtr ctxt);
+int		xmlParseExtParsedEnt	(xmlParserCtxtPtr ctxt);
 xmlDocPtr	xmlSAXParseDoc		(xmlSAXHandlerPtr sax,
 					 xmlChar *cur,
 					 int recovery);
@@ -388,6 +389,9 @@ xmlDocPtr	xmlSAXParseMemory	(xmlSAXHandlerPtr sax,
 xmlDocPtr	xmlSAXParseFile		(xmlSAXHandlerPtr sax,
 					 const char *filename,
 					 int recovery);
+xmlDocPtr	xmlSAXParseEntity	(xmlSAXHandlerPtr sax,
+					 const char *filename);
+xmlDocPtr	xmlParseEntity		(const char *filename);
 xmlDtdPtr	xmlParseDTD		(const xmlChar *ExternalID,
 					 const xmlChar *SystemID);
 xmlDtdPtr	xmlSAXParseDTD		(xmlSAXHandlerPtr sax,

include/libxml/parserInternals.h

@@ -15,8 +15,18 @@
 extern "C" {
 #endif
 
+ /* 
+  * Identifiers can be longer, but this will be more costly
+  * at runtime.
+  */
 #define XML_MAX_NAMELEN 100
 
+/*
+ * The parser tries to always have that amount of input ready
+ * one of the point is providing context when reporting errors
+ */
+#define INPUT_CHUNK	250
+
 /************************************************************************
  *									*
  * UNICODE version of the macros.      					*
@@ -87,6 +97,18 @@ extern "C" {
 #define MOVETO_STARTTAG(p)						\
     while ((*p) && (*(p) != '<')) (p)++
 
+/**
+ * Global vaiables affecting the default parser behaviour.
+ */
+
+extern int xmlParserDebugEntities;
+extern int xmlGetWarningsDefaultValue;
+extern int xmlParserDebugEntities;
+extern int xmlSubstituteEntitiesDefaultValue;
+extern int xmlDoValidityCheckingDefaultValue;
+extern int xmlPedanticParserDefaultValue;
+extern int xmlKeepBlanksDefaultValue;
+
 /*
  * Function to finish teh work of the macros where needed
  */
@@ -101,12 +123,6 @@ int			xmlIsExtender	(int c);
 int			xmlIsCombining	(int c);
 int			xmlIsChar	(int c);
 
-/**
- * Not for the faint of heart
- */
-
-extern int xmlParserDebugEntities;
-
 /**
  * Parser context
  */
@@ -141,6 +157,7 @@ xmlChar			xmlPopInput		(xmlParserCtxtPtr ctxt);
 void			xmlFreeInputStream	(xmlParserInputPtr input);
 xmlParserInputPtr	xmlNewInputFromFile	(xmlParserCtxtPtr ctxt,
 						 const char *filename);
+xmlParserInputPtr	xmlNewInputStream	(xmlParserCtxtPtr ctxt);
 
 /**
  * Namespaces.
@@ -248,6 +265,18 @@ int			inputPush		(xmlParserCtxtPtr ctxt,
 						 xmlParserInputPtr value);
 xmlParserInputPtr	inputPop		(xmlParserCtxtPtr ctxt);
 
+/*
+ * other comodities shared between parser.c and parserInternals
+ */
+int			xmlSkipBlankChars	(xmlParserCtxtPtr ctxt);
+int			xmlStringCurrentChar	(xmlParserCtxtPtr ctxt,
+						 const xmlChar *cur,
+						 int *len);
+void			xmlParserHandlePEReference(xmlParserCtxtPtr ctxt);
+void			xmlParserHandleReference(xmlParserCtxtPtr ctxt);
+xmlChar                *namePop			(xmlParserCtxtPtr ctxt);
+int			xmlCheckLanguageID	(const xmlChar *lang);
+
 /*
  * Really core function shared with HTML parser
  */

parser.h

@@ -371,6 +371,7 @@ xmlDocPtr	xmlRecoverFile		(const char *filename);
  * Less common routines and SAX interfaces
  */
 int		xmlParseDocument	(xmlParserCtxtPtr ctxt);
+int		xmlParseExtParsedEnt	(xmlParserCtxtPtr ctxt);
 xmlDocPtr	xmlSAXParseDoc		(xmlSAXHandlerPtr sax,
 					 xmlChar *cur,
 					 int recovery);
@@ -388,6 +389,9 @@ xmlDocPtr	xmlSAXParseMemory	(xmlSAXHandlerPtr sax,
 xmlDocPtr	xmlSAXParseFile		(xmlSAXHandlerPtr sax,
 					 const char *filename,
 					 int recovery);
+xmlDocPtr	xmlSAXParseEntity	(xmlSAXHandlerPtr sax,
+					 const char *filename);
+xmlDocPtr	xmlParseEntity		(const char *filename);
 xmlDtdPtr	xmlParseDTD		(const xmlChar *ExternalID,
 					 const xmlChar *SystemID);
 xmlDtdPtr	xmlSAXParseDTD		(xmlSAXHandlerPtr sax,

parserInternals.h

@@ -15,8 +15,18 @@
 extern "C" {
 #endif
 
+ /* 
+  * Identifiers can be longer, but this will be more costly
+  * at runtime.
+  */
 #define XML_MAX_NAMELEN 100
 
+/*
+ * The parser tries to always have that amount of input ready
+ * one of the point is providing context when reporting errors
+ */
+#define INPUT_CHUNK	250
+
 /************************************************************************
  *									*
  * UNICODE version of the macros.      					*
@@ -87,6 +97,18 @@ extern "C" {
 #define MOVETO_STARTTAG(p)						\
     while ((*p) && (*(p) != '<')) (p)++
 
+/**
+ * Global vaiables affecting the default parser behaviour.
+ */
+
+extern int xmlParserDebugEntities;
+extern int xmlGetWarningsDefaultValue;
+extern int xmlParserDebugEntities;
+extern int xmlSubstituteEntitiesDefaultValue;
+extern int xmlDoValidityCheckingDefaultValue;
+extern int xmlPedanticParserDefaultValue;
+extern int xmlKeepBlanksDefaultValue;
+
 /*
  * Function to finish teh work of the macros where needed
  */
@@ -101,12 +123,6 @@ int			xmlIsExtender	(int c);
 int			xmlIsCombining	(int c);
 int			xmlIsChar	(int c);
 
-/**
- * Not for the faint of heart
- */
-
-extern int xmlParserDebugEntities;
-
 /**
  * Parser context
  */
@@ -141,6 +157,7 @@ xmlChar			xmlPopInput		(xmlParserCtxtPtr ctxt);
 void			xmlFreeInputStream	(xmlParserInputPtr input);
 xmlParserInputPtr	xmlNewInputFromFile	(xmlParserCtxtPtr ctxt,
 						 const char *filename);
+xmlParserInputPtr	xmlNewInputStream	(xmlParserCtxtPtr ctxt);
 
 /**
  * Namespaces.
@@ -248,6 +265,18 @@ int			inputPush		(xmlParserCtxtPtr ctxt,
 						 xmlParserInputPtr value);
 xmlParserInputPtr	inputPop		(xmlParserCtxtPtr ctxt);
 
+/*
+ * other comodities shared between parser.c and parserInternals
+ */
+int			xmlSkipBlankChars	(xmlParserCtxtPtr ctxt);
+int			xmlStringCurrentChar	(xmlParserCtxtPtr ctxt,
+						 const xmlChar *cur,
+						 int *len);
+void			xmlParserHandlePEReference(xmlParserCtxtPtr ctxt);
+void			xmlParserHandleReference(xmlParserCtxtPtr ctxt);
+xmlChar                *namePop			(xmlParserCtxtPtr ctxt);
+int			xmlCheckLanguageID	(const xmlChar *lang);
+
 /*
  * Really core function shared with HTML parser
  */

uri.c

@@ -1467,6 +1467,11 @@ xmlBuildURI(const xmlChar *URI, const xmlChar *base) {
     xmlURIPtr bas = NULL;
     xmlURIPtr res = NULL;
 
+    if ((URI == NULL) && (base == NULL))
+	return(NULL);
+    if (URI == NULL)
+	return((xmlChar *) xmlMemStrdup((const char *) base));
+
     /*
      * 1) The URI reference is parsed into the potential four components and
      *    fragment identifier, as described in Section 4.3.

valid.c

@@ -2294,14 +2294,41 @@ xmlGetDtdQElementDesc(xmlDtdPtr dtd, const xmlChar *name,
 xmlAttributePtr
 xmlGetDtdAttrDesc(xmlDtdPtr dtd, const xmlChar *elem, const xmlChar *name) {
     xmlAttributeTablePtr table;
+    xmlElementTablePtr etable;
     xmlAttributePtr cur;
+    xmlElementPtr ecur;
     xmlChar *uqname = NULL, *prefix = NULL;
     int i;
 
     if (dtd == NULL) return(NULL);
     if (dtd->attributes == NULL) return(NULL);
-    table = (xmlAttributeTablePtr) dtd->attributes;
 
+    /*
+     * Faster lookup through the element table
+     */
+    etable = (xmlElementTablePtr) dtd->elements;
+    if (etable != NULL) {
+	for (i = 0;i < etable->nb_elements;i++) {
+	    ecur = etable->table[i];
+	    if (!xmlStrcmp(ecur->name, elem)) {
+		cur = ecur->attributes;
+		while (cur != NULL) {
+		    if (!xmlStrcmp(cur->name, name))
+			return(cur);
+                    cur = cur->nexth;
+		}
+		/* TODO: same accelerator for QNames !!! */
+		break;
+	    }
+	}
+    }
+    /*
+     * Miss on the element table, retry on the attribute one
+     */
+
+    table = (xmlAttributeTablePtr) dtd->attributes;
+    if (table == NULL)
+	return(NULL);
     for (i = 0;i < table->nb_attributes;i++) {
         cur = table->table[i];
 	if ((!xmlStrcmp(cur->name, name)) &&

xlink.c

@@ -164,12 +164,13 @@ xlinkIsLink	(xmlDocPtr doc, xmlNodePtr node) {
 		} else {
 		    xmlChar buf[200];
 #ifdef HAVE_SNPRINTF
-		    snprintf((char *) buf, 199, "%s:external-linkset",
+		    snprintf((char *) buf, sizeof(buf), "%s:external-linkset",
 			     (char *) xlink->prefix);
 #else
 		    sprintf((char *) buf, "%s:external-linkset",
 			    (char *) xlink->prefix);
 #endif
+                    buf[sizeof(buf) - 1] = 0;
 		    if (!xmlStrcmp(role, buf))
 			ret = XLINK_TYPE_EXTENDED_SET;