<?xml version="1.0" encoding="UTF-8" standalone="yes" ?>
<!DOCTYPE bugzilla SYSTEM "https://bugs.webkit.org/page.cgi?id=bugzilla.dtd">

<bugzilla version="5.0.4.1"
          urlbase="https://bugs.webkit.org/"
          
          maintainer="admin@webkit.org"
>

    <bug>
          <bug_id>12147</bug_id>
          
          <creation_ts>2007-01-06 21:02:37 -0800</creation_ts>
          <short_desc>Linked favicons no longer render in address bar</short_desc>
          <delta_ts>2007-12-11 18:45:26 -0800</delta_ts>
          <reporter_accessible>1</reporter_accessible>
          <cclist_accessible>1</cclist_accessible>
          <classification_id>1</classification_id>
          <classification>Unclassified</classification>
          <product>WebKit</product>
          <component>New Bugs</component>
          <version>420+</version>
          <rep_platform>Mac</rep_platform>
          <op_sys>OS X 10.4</op_sys>
          <bug_status>RESOLVED</bug_status>
          <resolution>FIXED</resolution>
          
          
          <bug_file_loc>http://www.informationgift.com/ud/faviconic/testfavicon_yeslink.html</bug_file_loc>
          <status_whiteboard></status_whiteboard>
          <keywords>InRadar, Regression</keywords>
          <priority>P2</priority>
          <bug_severity>Normal</bug_severity>
          <target_milestone>---</target_milestone>
          
          
          <everconfirmed>1</everconfirmed>
          <reporter name="Lewis Francis">lewis</reporter>
          <assigned_to name="Brady Eidson">beidson</assigned_to>
          <cc>beidson</cc>
    
    <cc>ddkilzer</cc>
    
    <cc>mrowe</cc>
          

      

      

      

          <comment_sort_order>oldest_to_newest</comment_sort_order>  
          <long_desc isprivate="0" >
    <commentid>36634</commentid>
    <comment_count>0</comment_count>
    <who name="Lewis Francis">lewis</who>
    <bug_when>2007-01-06 21:02:37 -0800</bug_when>
    <thetext>Using a link tag to specify a favicon different from /favicon.ico fails to render in the address bar. At least, it does so for a multi-icon set (containing 1, 4, 8, and 32-bit icons).  Safari favors the 4-bit icon in the multi-res set. Ideally, Webkit would favor the 32-bit icon. Bug present in latest nightly as of this report.</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>36290</commentid>
    <comment_count>1</comment_count>
    <who name="Brady Eidson">beidson</who>
    <bug_when>2007-01-08 10:30:33 -0800</bug_when>
    <thetext>I think the bug here isn&apos;t that we fail to respect the linked icon, or that we &quot;prefer the 4-bit version&quot; -
I believe the bug here is that we grab frame 0 from the image - or the first indexed icon from the icon set - and its the wrong one!

I think we have a radar for this but it&apos;s been low priority as we haven&apos;t seen this reported in the wild yet.

Lemme see if I can find the radar and/or file one.</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>36293</commentid>
    <comment_count>2</comment_count>
    <who name="Brady Eidson">beidson</who>
    <bug_when>2007-01-08 10:37:52 -0800</bug_when>
    <thetext>&lt;rdar://problem/4680377&gt;

I will bump this up a notch on the priority list.

But a warning!  The algorithm won&apos;t be &quot;choose the deepest icon&quot; - it will be &quot;choose the deepest icon of the correct size or, if none is found, choose the deepest icon of the incorrect size&quot; as that is what we did in the previous WebIconDatabase</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>36240</commentid>
    <comment_count>3</comment_count>
    <who name="Brady Eidson">beidson</who>
    <bug_when>2007-01-08 10:38:13 -0800</bug_when>
    <thetext>&lt;rdar://problem/4680377&gt;

I will bump this up a notch on the priority list.

But a warning!  The algorithm won&apos;t be &quot;choose the deepest icon&quot; - it will be &quot;choose the deepest icon of the correct size or, if none is found, choose the deepest icon of the incorrect size&quot; as that is what we did in the previous WebIconDatabase</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>36241</commentid>
    <comment_count>4</comment_count>
    <who name="Brady Eidson">beidson</who>
    <bug_when>2007-01-08 10:38:51 -0800</bug_when>
    <thetext>Sorry bout the double post - no clue how that happened...  lame!</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>36245</commentid>
    <comment_count>5</comment_count>
    <who name="David Kilzer (:ddkilzer)">ddkilzer</who>
    <bug_when>2007-01-08 10:46:19 -0800</bug_when>
    <thetext>*** Bug 12002 has been marked as a duplicate of this bug. ***</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>36246</commentid>
    <comment_count>6</comment_count>
    <who name="David Kilzer (:ddkilzer)">ddkilzer</who>
    <bug_when>2007-01-08 10:51:46 -0800</bug_when>
    <thetext>(In reply to comment #5)
&gt; *** Bug 12002 has been marked as a duplicate of this bug. ***

Interesting note from Bug 12002 Comment #0:

Code in WebKit/DOM/WebDOMOperations.m for DOMHTMLLinkElement leads me to
believe that this icon is expected to be loaded (or was loaded at one time)
since saving a WebArchive uses this code path and will try to save the icon to
the WebArchive file (and fail).

It&apos;s too bad Bug 11882 hasn&apos;t landed yet, or you could write a webarchive test for this.

</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>63977</commentid>
    <comment_count>7</comment_count>
    <who name="Lewis Francis">lewis</who>
    <bug_when>2007-12-11 17:51:44 -0800</bug_when>
    <thetext>Both of the issues in this report (sorry about that) appear to be resolved; work as expected for me in both Safari 3 and today&apos;s Webkit on Leopard.</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>63981</commentid>
    <comment_count>8</comment_count>
    <who name="David Kilzer (:ddkilzer)">ddkilzer</who>
    <bug_when>2007-12-11 18:45:26 -0800</bug_when>
    <thetext>Marking as RESOLVED/FIXED per Comment #7 and Radar.

</thetext>
  </long_desc>
      
      

    </bug>

</bugzilla>