Please review: parsing HTML 5 media elements

[Harri Porten]

Hi!

Germain had added support for the <audio> and <video> elements to the KDE 
4.1. feature plan. Not sure whether he already did some work in his git 
repo. As the freeze is coming near I went ahead and wanted to lay the 
groundwork for them. Details can be implemented later.

I'd like to ask either Allen or Germain to take a look at the necessary 
parser extension. For better readability he attached patch does not 
contain the generated dtd.h and dtd.c code. The element creation code is 
just a dummy printing out a message.

Harri.

P.S. Note to bystanders: we already have audio support in KHTML for some 
time now. But that was based on the "Audio" JS element. Equally 
non-standard as <audio> so far...
-------------- next part --------------
Index: misc/htmltags.in
===================================================================
--- misc/htmltags.in	(revision 808184)
+++ misc/htmltags.in	(working copy)
@@ -4,6 +4,7 @@
 address
 applet
 area
+audio
 b
 base
 basefont
@@ -80,6 +81,7 @@
 script
 select
 small
+source
 span
 strike
 strong
@@ -99,5 +101,6 @@
 u
 ul
 var
+video
 wbr
 xmp
Index: html/dtd.cpp
===================================================================
--- html/dtd.cpp	(revision 808184)
+++ html/dtd.cpp	(working copy)
@@ -52,6 +52,7 @@
     5, // ID_ADDRESS
     1, // ID_APPLET
     0, // ID_AREA
+    1, // ID_AUDIO
     1, // ID_B
     0, // ID_BASE
     0, // ID_BASEFONT
@@ -128,6 +129,7 @@
     1, // ID_SCRIPT
     6, // ID_SELECT
     1, // ID_SMALL
+    1, // ID_SOURCE
     1, // ID_SPAN
     1, // ID_STRIKE
     1, // ID_STRONG
@@ -147,6 +149,7 @@
     1, // ID_U
     5, // ID_UL
     1, // ID_VAR
+    1, // ID_VIDEO
     0, // ID_WBR
     5, // ID_XMP
     0, // ID_TEXT
@@ -160,6 +163,7 @@
     REQUIRED,  // ID_ADDRESS
     REQUIRED,  // ID_APPLET
     FORBIDDEN, // ID_AREA
+    REQUIRED,  // ID_AUDIO
     REQUIRED,  // ID_B
     FORBIDDEN, // ID_BASE
     FORBIDDEN, // ID_BASEFONT
@@ -236,6 +240,7 @@
     REQUIRED,  // ID_SCRIPT
     REQUIRED,  // ID_SELECT
     REQUIRED,  // ID_SMALL
+    FORBIDDEN, // ID_SOURCE
     REQUIRED,  // ID_SPAN
     REQUIRED,  // ID_STRIKE
     REQUIRED,  // ID_STRONG
@@ -255,7 +260,8 @@
     REQUIRED,  // ID_U
     REQUIRED,  // ID_UL
     REQUIRED,  // ID_VAR
-    FORBIDDEN,  // ID_WBR
+    REQUIRED,  // ID_VIDEO
+    FORBIDDEN, // ID_WBR
     REQUIRED,  // ID_XMP
     REQUIRED   // ID_TEXT
 };
@@ -316,6 +322,8 @@
     ID_NOBR,        // ?
     ID_WBR,         // ?
     ID_CANVAS,
+    ID_AUDIO,
+    ID_VIDEO,
     0
 };
 
@@ -543,6 +551,9 @@
     case ID_APPLET:
         // OBJECT: %flow | PARAM *
         return check_flow(childID, true) || childID == ID_PARAM;
+    case ID_AUDIO:
+    case ID_VIDEO:
+        return check_flow(childID, true) || childID == ID_SOURCE;
     case ID_PRE:
     case ID_XMP:
     case ID_PLAINTEXT:
Index: html/htmlparser.cpp
===================================================================
--- html/htmlparser.cpp	(revision 808184)
+++ html/htmlparser.cpp	(working copy)
@@ -1061,7 +1061,18 @@
         n = scriptElement;
         break;
     }
-
+    case ID_AUDIO:
+        qDebug("CREATING AUDIO ELEMENT");
+        n = new HTMLGenericElementImpl(document, t->tid);
+        break;
+    case ID_VIDEO:
+        qDebug("CREATING VIDEO ELEMENT");
+        n = new HTMLGenericElementImpl(document, t->tid);
+        break;
+    case ID_SOURCE:
+        qDebug("CREATING SOURCE ELEMENT");
+        n = new HTMLGenericElementImpl(document, t->tid);
+        break;
 // tables
     case ID_TABLE:
         n = new HTMLTableElementImpl(document);