parser: Use hash tables to avoid quadratic behavior

Use a hash table to lookup namespaces by prefix. The hash table stores
an index into the namespace table. Auxiliary data for namespaces is
stored in a separate array along the main namespace table.

Use a hash table to verify attribute uniqueness. The hash table stores
an index into the attribute table.

Reuse hash value from the dictionary to avoid computing them twice.

See #346.

SAX2.c

@@ -1870,8 +1870,12 @@ xmlSAX2AttributeNs(xmlParserCtxtPtr ctxt,
     /*
      * Note: if prefix == NULL, the attribute is not in the default namespace
      */
-    if (prefix != NULL)
+    if (prefix != NULL) {
-	namespace = xmlSearchNs(ctxt->myDoc, ctxt->node, prefix);
+	namespace = xmlParserNsLookupSax(ctxt, prefix);
+	if ((namespace == NULL) && (xmlStrEqual(prefix, BAD_CAST "xml"))) {
+	    namespace = xmlSearchNs(ctxt->myDoc, ctxt->node, prefix);
+	}
+    }
 
     /*
      * allocate the node
@@ -2201,6 +2205,9 @@ xmlSAX2StartElementNs(void *ctx,
              */
 	    continue;
 	}
+
+        xmlParserNsUpdateSax(ctxt, pref, ns);
+
 #ifdef LIBXML_VALID_ENABLED
 	if ((!ctxt->html) && ctxt->validate && ctxt->wellFormed &&
 	    ctxt->myDoc && ctxt->myDoc->intSubset) {
@@ -2242,7 +2249,7 @@ xmlSAX2StartElementNs(void *ctx,
      * Note that, if prefix is NULL, this searches for the default Ns
      */
     if ((URI != NULL) && (ret->ns == NULL)) {
-        ret->ns = xmlSearchNs(ctxt->myDoc, parent, prefix);
+        ret->ns = xmlParserNsLookupSax(ctxt, prefix);
 	if ((ret->ns == NULL) && (xmlStrEqual(prefix, BAD_CAST "xml"))) {
 	    ret->ns = xmlSearchNs(ctxt->myDoc, ret, prefix);
 	}

include/libxml/parser.h

@@ -172,6 +172,8 @@ typedef enum {
 } xmlParserMode;
 
 typedef struct _xmlStartTag xmlStartTag;
+typedef struct _xmlParserNsData xmlParserNsData;
+typedef struct _xmlAttrHashBucket xmlAttrHashBucket;
 
 /**
  * xmlParserCtxt:
@@ -282,7 +284,7 @@ struct _xmlParserCtxt {
     int                nsNr;          /* the number of inherited namespaces */
     int                nsMax;         /* the size of the arrays */
     const xmlChar *   *nsTab;         /* the array of prefix/namespace name */
-    int               *attallocs;     /* which attribute were allocated */
+    unsigned          *attallocs;     /* which attribute were allocated */
     xmlStartTag       *pushTab;       /* array of data for push */
     xmlHashTablePtr    attsDefault;   /* defaulted attributes if any */
     xmlHashTablePtr    attsSpecial;   /* non-CDATA attributes if any */
@@ -319,6 +321,10 @@ struct _xmlParserCtxt {
     unsigned short     nbErrors;    /* number of errors */
     unsigned short   nbWarnings;    /* number of warnings */
     unsigned            maxAmpl;    /* maximum amplification factor */
+
+    xmlParserNsData       *nsdb;    /* namespace database */
+    unsigned        attrHashMax;    /* allocated size */
+    xmlAttrHashBucket *attrHash;    /* atttribute hash table */
 };
 
 /**

include/private/parser.h

@@ -49,4 +49,16 @@ xmlDetectEncoding(xmlParserCtxtPtr ctxt);
 XML_HIDDEN void
 xmlSetDeclaredEncoding(xmlParserCtxtPtr ctxt, xmlChar *encoding);
 
+/*
+ * These functions allow SAX handlers to attach extra data to namespaces
+ * efficiently and should be made public.
+ */
+void
+xmlParserNsFree(xmlParserNsData *nsdb);
+XML_HIDDEN int
+xmlParserNsUpdateSax(xmlParserCtxtPtr ctxt, const xmlChar *prefix,
+                     void *saxData);
+XML_HIDDEN void *
+xmlParserNsLookupSax(xmlParserCtxtPtr ctxt, const xmlChar *prefix);
+
 #endif /* XML_PARSER_H_PRIVATE__ */

parserInternals.c

@@ -2061,7 +2061,9 @@ xmlFreeParserCtxt(xmlParserCtxtPtr ctxt)
     if (ctxt->vctxt.nodeTab != NULL) xmlFree(ctxt->vctxt.nodeTab);
     if (ctxt->atts != NULL) xmlFree((xmlChar * *)ctxt->atts);
     if (ctxt->dict != NULL) xmlDictFree(ctxt->dict);
-    if (ctxt->nsTab != NULL) xmlFree((char *) ctxt->nsTab);
+    if (ctxt->nsTab != NULL) xmlFree(ctxt->nsTab);
+    if (ctxt->nsdb != NULL) xmlParserNsFree(ctxt->nsdb);
+    if (ctxt->attrHash != NULL) xmlFree(ctxt->attrHash);
     if (ctxt->pushTab != NULL) xmlFree(ctxt->pushTab);
     if (ctxt->attallocs != NULL) xmlFree(ctxt->attallocs);
     if (ctxt->attsDefault != NULL)