Re: [xml] XHTML Doc serialization and meta element

[Rob Richards <rrichards ctindustries net>]

Daniel Veillard wrote:

>  Another way would be to make libxml2 not touch anything if it detects
> any http-equiv meta tag, and if it doesn't detect one try to put the
> tag serialization in the output buffer, but not in the tree itself.
> This would requires some tweaking of the serialization code but should
> not be too hard. This would be more satisfactory from an API POV since
> serialization would not modify the input tree, and would not disturb
> headers if the application is already taking care of it, while still
> adding the meta if the application is ignorant about it.
>  What do you think ?
>  
Not sure which way this is going to be handled, but I took a crack at 
doing it when serializing.
Attached is the diff of the changes I made (only meant to look at not 
meant to be commited).
It looks a bit hackish, but couldnt see any other way to do it.
Didn't touch the HTMLTree.c file at this point till its decided if this 
is to be done during serialization or one of the other possible ways.

Another thing doing it this way is previously, meta element is only 
added if dumping the document element (html).
Now, if the head element were dumped as the top node in 
xmlNodeDumpOutput, the meta element is still going to be added to the 
output as there is no way to know then first node being dumped since all 
that is avaiable is the node and the xmlSaveCtxt (probably not a big 
deal though).

Rob

Index: xmlsave.c
===================================================================
RCS file: /cvs/gnome/gnome-xml/xmlsave.c,v
retrieving revision 1.25
diff -c -r1.25 xmlsave.c
*** xmlsave.c   8 Aug 2005 14:44:11 -0000       1.25
--- xmlsave.c   24 Aug 2005 22:47:54 -0000
***************
*** 860,871 ****
--- 860,873 ----
        is_xhtml = xmlIsXHTML(dtd->SystemID, dtd->ExternalID);
        if (is_xhtml < 0) is_xhtml = 0;
      }
+ /*
      if (is_xhtml) {
        if (encoding != NULL)
            htmlSetMetaEncoding(cur, (const xmlChar *) ctxt->encoding);
        else
            htmlSetMetaEncoding(cur, BAD_CAST "UTF-8");
      }
+ */
  #endif
      if (cur->children != NULL) {
          xmlNodePtr child = cur->children;
***************
*** 1086,1092 ****
   */
  static void
  xhtmlNodeDumpOutput(xmlSaveCtxtPtr ctxt, xmlNodePtr cur) {
!     int format;
      xmlNodePtr tmp;
      xmlChar *start, *end;
      xmlOutputBufferPtr buf;
--- 1088,1094 ----
   */
  static void
  xhtmlNodeDumpOutput(xmlSaveCtxtPtr ctxt, xmlNodePtr cur) {
!     int format, addmeta = 0;
      xmlNodePtr tmp;
      xmlChar *start, *end;
      xmlOutputBufferPtr buf;
***************
*** 1210,1227 ****
      if (cur->properties != NULL)
          xhtmlAttrListDumpOutput(ctxt, cur->properties);
  
      if ((cur->type == XML_ELEMENT_NODE) && (cur->children == NULL)) {
        if (((cur->ns == NULL) || (cur->ns->prefix == NULL)) &&
!           (xhtmlIsEmpty(cur) == 1)) {
            /*
             * C.2. Empty Elements
             */
            xmlOutputBufferWrite(buf, 3, " />");
        } else {
            /*
             * C.3. Element Minimization and Empty Element Content
             */
!           xmlOutputBufferWrite(buf, 3, "></");
            if ((cur->ns != NULL) && (cur->ns->prefix != NULL)) {
                xmlOutputBufferWriteString(buf, (const char *)cur->ns->prefix);
                xmlOutputBufferWrite(buf, 1, ":");
--- 1212,1256 ----
      if (cur->properties != NULL)
          xhtmlAttrListDumpOutput(ctxt, cur->properties);
  
+       if ((cur->type == XML_ELEMENT_NODE) && 
+               (xmlStrcasecmp(cur->name, BAD_CAST"head") == 0) && 
+               (cur->parent != NULL) && 
+               (xmlStrcasecmp(cur->parent->name, BAD_CAST"html") == 0)) {
+               tmp = cur->children;
+               while (tmp != NULL) {
+                       if ((xmlStrcasecmp(tmp->name, BAD_CAST"meta") == 0) &&
+                               (xmlHasProp(tmp, BAD_CAST"http-equiv"))) {
+                               break;
+                       }
+                       tmp = tmp->next;
+               }
+               if (tmp == NULL)
+                       addmeta = 1;
+       }
+ 
      if ((cur->type == XML_ELEMENT_NODE) && (cur->children == NULL)) {
        if (((cur->ns == NULL) || (cur->ns->prefix == NULL)) &&
!           ((xhtmlIsEmpty(cur) == 1) && (addmeta == 0))) {
            /*
             * C.2. Empty Elements
             */
            xmlOutputBufferWrite(buf, 3, " />");
        } else {
+               if (addmeta == 1) {
+                       xmlOutputBufferWrite(buf, 1, ">");
+                       xmlOutputBufferWriteString(buf,
+                               "<meta http-equiv=\"Content-Type\" content=\"text/html; charset=");
+                       if (ctxt->encoding) {
+                               xmlOutputBufferWriteString(buf, ctxt->encoding);
+                       } else {
+                               xmlOutputBufferWrite(buf, 5, "UTF-8");
+                       }
+                       xmlOutputBufferWrite(buf, 3, "\" /");
+               }
            /*
             * C.3. Element Minimization and Empty Element Content
             */
!                       xmlOutputBufferWrite(buf, 3, "></");
            if ((cur->ns != NULL) && (cur->ns->prefix != NULL)) {
                xmlOutputBufferWriteString(buf, (const char *)cur->ns->prefix);
                xmlOutputBufferWrite(buf, 1, ":");
***************
*** 1232,1237 ****
--- 1261,1276 ----
        return;
      }
      xmlOutputBufferWrite(buf, 1, ">");
+       if (addmeta == 1) {
+               xmlOutputBufferWriteString(buf,
+                       "<meta http-equiv=\"Content-Type\" content=\"text/html; charset=");
+               if (ctxt->encoding) {
+                       xmlOutputBufferWriteString(buf, ctxt->encoding);
+               } else {
+                       xmlOutputBufferWrite(buf, 5, "UTF-8");
+               }
+               xmlOutputBufferWrite(buf, 4, "\" />");
+       }
      if ((cur->type != XML_ELEMENT_NODE) && (cur->content != NULL)) {
        xmlOutputBufferWriteEscape(buf, cur->content, ctxt->escape);
      }
***************
*** 1832,1837 ****
--- 1871,1877 ----
          is_xhtml = xmlIsXHTML(dtd->SystemID, dtd->ExternalID);
          if (is_xhtml < 0)
              is_xhtml = 0;
+ /*
          if ((is_xhtml) && (cur->parent == (xmlNodePtr) doc) &&
              (cur->type == XML_ELEMENT_NODE) &&
              (xmlStrEqual(cur->name, BAD_CAST "html"))) {
***************
*** 1841,1846 ****
--- 1881,1887 ----
              else
                  htmlSetMetaEncoding((htmlDocPtr) doc, BAD_CAST "UTF-8");
          }
+ */
      }
  
      if (is_xhtml)