ICU-491.11.2.tar.gz

[apple/icu.git] / icuSources / tools / toolutil / xmlparser.cpp

diff --git a/icuSources/tools/toolutil/xmlparser.cpp b/icuSources/tools/toolutil/xmlparser.cpp
index 195fa30385c9f26422e5ccbffa1c045e9bdec59d..134fdcf8d0911929040cd3e49a7a5dd5353ecd88 100644 (file)
--- a/icuSources/tools/toolutil/xmlparser.cpp
+++ b/icuSources/tools/toolutil/xmlparser.cpp
@@ -1,7 +1,7 @@
 /*
 *******************************************************************************
 *
 /*
 *******************************************************************************
 *

-*   Copyright (C) 2004-2006, International Business Machines
+*   Copyright (C) 2004-2010, International Business Machines

 *   Corporation and others.  All Rights Reserved.
 *
 *******************************************************************************
 *   Corporation and others.  All Rights Reserved.
 *
 *******************************************************************************


@@ -61,27 +61,32 @@ UXMLParser::UXMLParser(UErrorCode &status) :
       //      example:  "<?xml version=1.0 encoding="utf-16" ?>
       //      This is a sloppy implementation - just look for the leading <?xml and the closing ?>
       //            allow for a possible leading BOM.
       //      example:  "<?xml version=1.0 encoding="utf-16" ?>
       //      This is a sloppy implementation - just look for the leading <?xml and the closing ?>
       //            allow for a possible leading BOM.

-      mXMLDecl(UnicodeString("(?s)\\uFEFF?<\\?xml.+?\\?>"), 0, status),
+      mXMLDecl(UnicodeString("(?s)\\uFEFF?<\\?xml.+?\\?>", -1, US_INV), 0, status),

       
       //  XML Comment   production #15
       //     example:  "<!-- whatever -->
       //       note, does not detect an illegal "--" within comments
       
       //  XML Comment   production #15
       //     example:  "<!-- whatever -->
       //       note, does not detect an illegal "--" within comments

-      mXMLComment(UnicodeString("(?s)<!--.+?-->"), 0, status),
+      mXMLComment(UnicodeString("(?s)<!--.+?-->", -1, US_INV), 0, status),

       
       //  XML Spaces
       //      production [3]
       
       //  XML Spaces
       //      production [3]

-      mXMLSP(UnicodeString(XML_SPACES "+"), 0, status),
+      mXMLSP(UnicodeString(XML_SPACES "+", -1, US_INV), 0, status),

       
       //  XML Doctype decl  production #28
       //     example   "<!DOCTYPE foo SYSTEM "somewhere" >
       
       //  XML Doctype decl  production #28
       //     example   "<!DOCTYPE foo SYSTEM "somewhere" >

+      //       or      "<!DOCTYPE foo [internal dtd]>

       //    TODO:  we don't actually parse the DOCTYPE or internal subsets.
       //           Some internal dtd subsets could confuse this simple-minded
       //    TODO:  we don't actually parse the DOCTYPE or internal subsets.
       //           Some internal dtd subsets could confuse this simple-minded

-      //           attempt at skipping over them.
-      mXMLDoctype(UnicodeString("(?s)<!DOCTYPE.+?>"), 0, status),
+      //           attempt at skipping over them, specifically, occcurences
+      //           of closeing square brackets.  These could appear in comments, 
+      //           or in parameter entity declarations, for example.
+      mXMLDoctype(UnicodeString(
+           "(?s)<!DOCTYPE.*?(>|\\[.*?\\].*?>)", -1, US_INV
+           ), 0, status),

       
       //  XML PI     production #16
       //     example   "<?target stuff?>
       
       //  XML PI     production #16
       //     example   "<?target stuff?>

-      mXMLPI(UnicodeString("(?s)<\\?.+?\\?>"), 0, status),
+      mXMLPI(UnicodeString("(?s)<\\?.+?\\?>", -1, US_INV), 0, status),

       
       //  XML Element Start   Productions #40, #41
       //          example   <foo att1='abc'  att2="d e f" >
       
       //  XML Element Start   Productions #40, #41
       //          example   <foo att1='abc'  att2="d e f" >


@@ -92,11 +97,11 @@ UXMLParser::UXMLParser(UErrorCode &status) :
                 XML_SPACES "+" XML_NAME XML_SPACES "*=" XML_SPACES "*"     // match  "ATTR_NAME = "
                 "(?:(?:\\\'[^<\\\']*?\\\')|(?:\\\"[^<\\\"]*?\\\"))"        // match  '"attribute value"'
           ")*"                                                             //   * for zero or more attributes.
                 XML_SPACES "+" XML_NAME XML_SPACES "*=" XML_SPACES "*"     // match  "ATTR_NAME = "
                 "(?:(?:\\\'[^<\\\']*?\\\')|(?:\\\"[^<\\\"]*?\\\"))"        // match  '"attribute value"'
           ")*"                                                             //   * for zero or more attributes.

-          XML_SPACES "*?>"), 0, status),                               // match " >"
+          XML_SPACES "*?>", -1, US_INV), 0, status),                               // match " >"

       
       //  XML Element End     production #42
       //     example   </foo>
       
       //  XML Element End     production #42
       //     example   </foo>

-      mXMLElemEnd (UnicodeString("</(" XML_NAME ")" XML_SPACES "*>"), 0, status),
+      mXMLElemEnd (UnicodeString("</(" XML_NAME ")" XML_SPACES "*>", -1, US_INV), 0, status),

       
       // XML Element Empty    production #44
       //     example   <foo att1="abc"   att2="d e f" />
       
       // XML Element Empty    production #44
       //     example   <foo att1="abc"   att2="d e f" />


@@ -105,11 +110,11 @@ UXMLParser::UXMLParser(UErrorCode &status) :
                 XML_SPACES "+" XML_NAME XML_SPACES "*=" XML_SPACES "*"     // match  "ATTR_NAME = "
                 "(?:(?:\\\'[^<\\\']*?\\\')|(?:\\\"[^<\\\"]*?\\\"))"        // match  '"attribute value"'
           ")*"                                                             //   * for zero or more attributes.
                 XML_SPACES "+" XML_NAME XML_SPACES "*=" XML_SPACES "*"     // match  "ATTR_NAME = "
                 "(?:(?:\\\'[^<\\\']*?\\\')|(?:\\\"[^<\\\"]*?\\\"))"        // match  '"attribute value"'
           ")*"                                                             //   * for zero or more attributes.

-          XML_SPACES "*?/>"), 0, status),                              // match " />"
+          XML_SPACES "*?/>", -1, US_INV), 0, status),                              // match " />"

       
 
       // XMLCharData.  Everything but '<'.  Note that & will be dealt with later.
       
 
       // XMLCharData.  Everything but '<'.  Note that & will be dealt with later.

-      mXMLCharData(UnicodeString("(?s)[^<]*"), 0, status),
+      mXMLCharData(UnicodeString("(?s)[^<]*", -1, US_INV), 0, status),

 
       // Attribute name = "value".  XML Productions 10, 40/41
       //  Capture group 1 is name, 
 
       // Attribute name = "value".  XML Productions 10, 40/41
       //  Capture group 1 is name, 


@@ -121,14 +126,14 @@ UXMLParser::UXMLParser(UErrorCode &status) :
       //        Here, we match a single attribute, and make its name and
       //        attribute value available to the parser code.
       mAttrValue(UnicodeString(XML_SPACES "+("  XML_NAME ")"  XML_SPACES "*=" XML_SPACES "*"
       //        Here, we match a single attribute, and make its name and
       //        attribute value available to the parser code.
       mAttrValue(UnicodeString(XML_SPACES "+("  XML_NAME ")"  XML_SPACES "*=" XML_SPACES "*"

-         "((?:\\\'[^<\\\']*?\\\')|(?:\\\"[^<\\\"]*?\\\"))"), 0, status),
+         "((?:\\\'[^<\\\']*?\\\')|(?:\\\"[^<\\\"]*?\\\"))", -1, US_INV), 0, status),

 
 
 
 

-      mAttrNormalizer(UnicodeString(XML_SPACES), 0, status),
+      mAttrNormalizer(UnicodeString(XML_SPACES, -1, US_INV), 0, status),

 
       // Match any of the new-line sequences in content.
       //   All are changed to \u000a.
 
       // Match any of the new-line sequences in content.
       //   All are changed to \u000a.

-      mNewLineNormalizer(UnicodeString("\\u000d\\u000a|\\u000d\\u0085|\\u000a|\\u000d|\\u0085|\\u2028"), 0, status),
+      mNewLineNormalizer(UnicodeString("\\u000d\\u000a|\\u000d\\u0085|\\u000a|\\u000d|\\u0085|\\u2028", -1, US_INV), 0, status),

 
       // & char references
       //   We will figure out what we've got based on which capture group has content.
 
       // & char references
       //   We will figure out what we've got based on which capture group has content.


@@ -558,7 +563,7 @@ UnicodeString
 UXMLParser::scanContent(UErrorCode &status) {
     UnicodeString  result;
     if (mXMLCharData.lookingAt(fPos, status)) {
 UXMLParser::scanContent(UErrorCode &status) {
     UnicodeString  result;
     if (mXMLCharData.lookingAt(fPos, status)) {

-        result = mXMLCharData.group(0, status);
+        result = mXMLCharData.group((int32_t)0, status);

         // Normalize the new-lines.  (Before char ref substitution)
         mNewLineNormalizer.reset(result);
         result = mNewLineNormalizer.replaceAll(fOneLF, status);
         // Normalize the new-lines.  (Before char ref substitution)
         mNewLineNormalizer.reset(result);
         result = mNewLineNormalizer.replaceAll(fOneLF, status);


@@ -617,7 +622,7 @@ UXMLParser::replaceCharRefs(UnicodeString &s, UErrorCode &status) {
             // An unrecognized &entity;  Leave it alone.
             //  TODO:  check that it really looks like an entity, and is not some
             //         random & in the text.
             // An unrecognized &entity;  Leave it alone.
             //  TODO:  check that it really looks like an entity, and is not some
             //         random & in the text.

-            replacement = mAmps.group(0, status);
+            replacement = mAmps.group((int32_t)0, status);

         }
         mAmps.appendReplacement(result, replacement, status);
     }
         }
         mAmps.appendReplacement(result, replacement, status);
     }


@@ -710,8 +715,9 @@ UXMLElement::appendText(UnicodeString &text, UBool recurse) const {
     int32_t i, count=fChildren.size();
     for(i=0; i<count; ++i) {
         node=(const UObject *)fChildren.elementAt(i);
     int32_t i, count=fChildren.size();
     for(i=0; i<count; ++i) {
         node=(const UObject *)fChildren.elementAt(i);

-        if(node->getDynamicClassID()==UnicodeString::getStaticClassID()) {
-            text.append(*(const UnicodeString *)node);
+        const UnicodeString *s=dynamic_cast<const UnicodeString *>(node);
+        if(s!=NULL) {
+            text.append(*s);

         } else if(recurse) /* must be a UXMLElement */ {
             ((const UXMLElement *)node)->appendText(text, recurse);
         }
         } else if(recurse) /* must be a UXMLElement */ {
             ((const UXMLElement *)node)->appendText(text, recurse);
         }


@@ -761,7 +767,7 @@ const UObject *
 UXMLElement::getChild(int32_t i, UXMLNodeType &type) const {
     if(0<=i && i<fChildren.size()) {
         const UObject *node=(const UObject *)fChildren.elementAt(i);
 UXMLElement::getChild(int32_t i, UXMLNodeType &type) const {
     if(0<=i && i<fChildren.size()) {
         const UObject *node=(const UObject *)fChildren.elementAt(i);

-        if(node->getDynamicClassID()==UXMLElement::getStaticClassID()) {
+        if(dynamic_cast<const UXMLElement *>(node)!=NULL) {

             type=UXML_NODE_TYPE_ELEMENT;
         } else {
             type=UXML_NODE_TYPE_STRING;
             type=UXML_NODE_TYPE_ELEMENT;
         } else {
             type=UXML_NODE_TYPE_STRING;


@@ -782,10 +788,9 @@ UXMLElement::nextChildElement(int32_t &i) const {
     int32_t count=fChildren.size();
     while(i<count) {
         node=(const UObject *)fChildren.elementAt(i++);
     int32_t count=fChildren.size();
     while(i<count) {
         node=(const UObject *)fChildren.elementAt(i++);

-        // TODO: see if ICU can use C++ instanceof instead of its own poor man's RTTI
-        // if(node instanceof UXMLElement) {
-        if(node->getDynamicClassID()==UXMLElement::getStaticClassID()) {
-            return (const UXMLElement *)node;
+        const UXMLElement *elem=dynamic_cast<const UXMLElement *>(node);
+        if(elem!=NULL) {
+            return elem;

         }
     }
     return NULL;
         }
     }
     return NULL;


@@ -804,8 +809,8 @@ UXMLElement::getChildElement(const UnicodeString &name) const {
     int32_t i, count=fChildren.size();
     for(i=0; i<count; ++i) {
         node=(const UObject *)fChildren.elementAt(i);
     int32_t i, count=fChildren.size();
     for(i=0; i<count; ++i) {
         node=(const UObject *)fChildren.elementAt(i);

-        if(node->getDynamicClassID()==UXMLElement::getStaticClassID()) {
-            const UXMLElement *elem=(const UXMLElement *)node;
+        const UXMLElement *elem=dynamic_cast<const UXMLElement *>(node);
+        if(elem!=NULL) {

             if(p==elem->fName) {
                 return elem;
             }
             if(p==elem->fName) {
                 return elem;
             }