Merge branch 'tizen_base' of ssh://review.tizen.org:29418/platform/upstream/libxml2...

[platform/upstream/libxml2.git] / xmlregexp.c

diff --git a/xmlregexp.c b/xmlregexp.c
index 3e912ab..d255fbf 100644 (file)
--- a/xmlregexp.c
+++ b/xmlregexp.c
@@ -1544,6 +1544,7 @@ static int
 xmlFAGenerateTransitions(xmlRegParserCtxtPtr ctxt, xmlRegStatePtr from,
                         xmlRegStatePtr to, xmlRegAtomPtr atom) {
     xmlRegStatePtr end;
+    int nullable = 0;
 
     if (atom == NULL) {
        ERROR("genrate transition: atom == NULL");
@@ -1730,6 +1731,13 @@ xmlFAGenerateTransitions(xmlRegParserCtxtPtr ctxt, xmlRegStatePtr from,
     if (xmlRegAtomPush(ctxt, atom) < 0) {
        return(-1);
     }
+    if ((atom->quant == XML_REGEXP_QUANT_RANGE) &&
+        (atom->min == 0) && (atom->max > 0)) {
+       nullable = 1;
+       atom->min = 1;
+        if (atom->max == 1)
+           atom->quant = XML_REGEXP_QUANT_OPT;
+    }
     xmlRegStateAddTrans(ctxt, from, atom, to, -1, -1);
     ctxt->state = end;
     switch (atom->quant) {
@@ -1747,11 +1755,8 @@ xmlFAGenerateTransitions(xmlRegParserCtxtPtr ctxt, xmlRegStatePtr from,
            xmlRegStateAddTrans(ctxt, to, atom, to, -1, -1);
            break;
        case XML_REGEXP_QUANT_RANGE:
-#if DV_test
-           if (atom->min == 0) {
+           if (nullable)
                xmlFAGenerateEpsilonTransition(ctxt, from, to);
-           }
-#endif
            break;
        default:
            break;
@@ -2805,18 +2810,21 @@ xmlRegCheckCharacterRange(xmlRegAtomType type, int codepoint, int neg,
            break;
         case XML_REGEXP_NOTSPACE:
            neg = !neg;
+            /* Falls through. */
         case XML_REGEXP_ANYSPACE:
            ret = ((codepoint == '\n') || (codepoint == '\r') ||
                   (codepoint == '\t') || (codepoint == ' '));
            break;
         case XML_REGEXP_NOTINITNAME:
            neg = !neg;
+            /* Falls through. */
         case XML_REGEXP_INITNAME:
            ret = (IS_LETTER(codepoint) ||
                   (codepoint == '_') || (codepoint == ':'));
            break;
         case XML_REGEXP_NOTNAMECHAR:
            neg = !neg;
+            /* Falls through. */
         case XML_REGEXP_NAMECHAR:
            ret = (IS_LETTER(codepoint) || IS_DIGIT(codepoint) ||
                   (codepoint == '.') || (codepoint == '-') ||
@@ -2825,11 +2833,13 @@ xmlRegCheckCharacterRange(xmlRegAtomType type, int codepoint, int neg,
            break;
         case XML_REGEXP_NOTDECIMAL:
            neg = !neg;
+            /* Falls through. */
         case XML_REGEXP_DECIMAL:
            ret = xmlUCSIsCatNd(codepoint);
            break;
         case XML_REGEXP_REALCHAR:
            neg = !neg;
+            /* Falls through. */
         case XML_REGEXP_NOTREALCHAR:
            ret = xmlUCSIsCatP(codepoint);
            if (ret == 0)
@@ -4084,8 +4094,9 @@ rollback:
                    xmlFree(exec->errString);
                exec->errString = xmlStrdup(value);
                exec->errState = exec->state;
-               memcpy(exec->errCounts, exec->counts,
-                      exec->comp->nbCounters * sizeof(int));
+                if (exec->comp->nbCounters)
+                    memcpy(exec->errCounts, exec->counts,
+                           exec->comp->nbCounters * sizeof(int));
            }
 
            /*
@@ -4875,7 +4886,8 @@ xmlFAParseCharClassEsc(xmlRegParserCtxtPtr ctxt) {
        }
        NEXT;
        xmlFAParseCharProp(ctxt);
-       ctxt->atom->neg = 1;
+        if (ctxt->atom != NULL)
+           ctxt->atom->neg = 1;
        if (CUR != '}') {
            ERROR("Expecting '}'");
            return;
@@ -5046,17 +5058,18 @@ xmlFAParseCharRange(xmlRegParserCtxtPtr ctxt) {
                return;
        }
         len = 1;
-    } else if ((cur != 0x5B) && (cur != 0x5D)) {
+    } else if ((cur != '\0') && (cur != 0x5B) && (cur != 0x5D)) {
         end = CUR_SCHAR(ctxt->cur, len);
     } else {
        ERROR("Expecting the end of a char range");
        return;
     }
-    NEXTL(len);
+
     /* TODO check that the values are acceptable character ranges for XML */
     if (end < start) {
        ERROR("End of range is before start of range");
     } else {
+        NEXTL(len);
         xmlRegAtomAddRange(ctxt, ctxt->atom, ctxt->neg,
                           XML_REGEXP_CHARVAL, start, end, NULL);
     }
@@ -6345,7 +6358,7 @@ struct _xmlExpCtxt {
 /**
  * xmlExpNewCtxt:
  * @maxNodes:  the maximum number of nodes
- * @dict:  optional dictionnary to use internally
+ * @dict:  optional dictionary to use internally
  *
  * Creates a new context for manipulating expressions
  *
@@ -7204,7 +7217,7 @@ xmlExpStringDerive(xmlExpCtxtPtr ctxt, xmlExpNodePtr exp,
         return(NULL);
     }
     /*
-     * check the string is in the dictionnary, if yes use an interned
+     * check the string is in the dictionary, if yes use an interned
      * copy, otherwise we know it's not an acceptable input
      */
     input = xmlDictExists(ctxt->dict, str, len);