<?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>34262</bug_id>
          
          <creation_ts>2010-01-28 06:35:17 -0800</creation_ts>
          <short_desc>Accept XHTML MP content type as XHTML content</short_desc>
          <delta_ts>2010-05-13 15:07:21 -0700</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>528+ (Nightly build)</version>
          <rep_platform>S60 Hardware</rep_platform>
          <op_sys>S60 3rd edition</op_sys>
          <bug_status>CLOSED</bug_status>
          <resolution>FIXED</resolution>
          
          
          <bug_file_loc></bug_file_loc>
          <status_whiteboard></status_whiteboard>
          <keywords>Qt</keywords>
          <priority>P2</priority>
          <bug_severity>Normal</bug_severity>
          <target_milestone>---</target_milestone>
          
          <blocked>34751</blocked>
    
    <blocked>35784</blocked>
          <everconfirmed>1</everconfirmed>
          <reporter>jtkbugz</reporter>
          <assigned_to name="Laszlo Gombos">laszlo.gombos</assigned_to>
          <cc>charles.wei</cc>
    
    <cc>commit-queue</cc>
    
    <cc>darin</cc>
    
    <cc>ddkilzer</cc>
    
    <cc>eric</cc>
    
    <cc>hausmann</cc>
    
    <cc>jukka.jokiniva</cc>
    
    <cc>kenneth</cc>
    
    <cc>koivisto</cc>
    
    <cc>koshuin</cc>
    
    <cc>laszlo.gombos</cc>
    
    <cc>mjs</cc>
    
    <cc>staikos</cc>
    
    <cc>timothy.lu</cc>
    
    <cc>tonikitoo</cc>
    
    <cc>webkit.review.bot</cc>
    
    <cc>zalan</cc>
          

      

      

      

          <comment_sort_order>oldest_to_newest</comment_sort_order>  
          <long_desc isprivate="0" >
    <commentid>185460</commentid>
    <comment_count>0</comment_count>
    <who name="">jtkbugz</who>
    <bug_when>2010-01-28 06:35:17 -0800</bug_when>
    <thetext>STEPS TO REPRODUCE:
1. Open the browser.
2. Type the URL &quot;www.ovi.com&quot; in the URL/search bar.
3. Press the &quot;Go&quot; button
ACTUAL RESULTS:
Progress Indiacation shows loaded and the button&apos;s state changes to &quot;Reload&quot;. But the page content is not loaded

EXPECTED RESULTS:
Page content should load.

NOTE: The same behavior observed in ANOMALY browser</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>185476</commentid>
    <comment_count>1</comment_count>
    <who name="Simon Hausmann">hausmann</who>
    <bug_when>2010-01-28 07:36:55 -0800</bug_when>
    <thetext>I assume this is on S60. Can you reproduce this in the simulator with a backtrace?</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>188970</commentid>
    <comment_count>2</comment_count>
    <who name="Jukka Jokiniva">jukka.jokiniva</who>
    <bug_when>2010-02-09 05:01:47 -0800</bug_when>
    <thetext>I tested this with N5800 &amp; Anomaly. I works with 4.6.0, but not with 4.6.1.</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>191852</commentid>
    <comment_count>3</comment_count>
    <who name="Janne Koskinen">koshuin</who>
    <bug_when>2010-02-19 03:28:00 -0800</bug_when>
    <thetext>Ok. tested this with 4.6.2 as well. tried several https websites and none of them respond. This can be reproed on emulator. Seems that it just immediately stops loading page with SSL.</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>191894</commentid>
    <comment_count>4</comment_count>
    <who name="Janne Koskinen">koshuin</who>
    <bug_when>2010-02-19 06:32:31 -0800</bug_when>
    <thetext>Ok now found the real reason. forget https and friends nothign to do with that, blaming https was just unfortunate coincidence :)

This is issue is coming from user-agent string change https://bugs.webkit.org/show_bug.cgi?id=31961. after the change our client is now considered mobile client and server redirects to mobile page. Is the client that is wrong or the server that is wrong you&apos;ll be the judge:

GET /ui/appmanager/m/d?ar=EUROPE&amp;co=EUROPE&amp;la=en&amp;te=NonNokia&amp;_nfpb=true&amp;_pageLabel=CATEGORY_SERVICES&amp;_nfls=false&amp;cdncc=FI&amp; HTTP/1.1
User-Agent: Mozilla/5.0 (SymbianOS/9.4; U; Series60/5.0; en-GB) AppleWebKit/532.4 (KHTML, like Gecko) Anomaly/0.0.0 Safari/532.4
Accept: application/xml,application/xhtml+xml,text/html;q=0.9,text/plain;q=0.8,image/png,*/*;q=0.5
Referer: http://www.ovi.com/
Connection: Keep-Alive
Accept-Encoding: gzip
Accept-Language: en-GB,*
Host: mobile.nokia.mobi

HTTP/1.1 200 OK
Server: &quot;Nokia&quot;
Content-Length: 7130
Content-Type: application/vnd.wap.xhtml+xml
X-Powered-By: Servlet/2.5 JSP/2.1
Cache-Control: max-age=104
Date: Fri, 19 Feb 2010 13:08:28 GMT
Connection: keep-alive

As seen from above we don&apos;t accept vnd.wap.xhtml+xml but server insists it upon us. Also notice that for some reason we are NonNokia client on 5800 express music, so something is missing from the agent-string.

Now we don&apos;t support rendering of wap pages, or we would but we don&apos;t have it flagged on...
so the task is to have:

ENABLE_WML = 1
ENABLE_WCSS = 1

and assure it works on wap pages.

I have wireshark dumps on 4.6 vs 4.6.2 and can send them on request.</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>192450</commentid>
    <comment_count>5</comment_count>
    <who name="Laszlo Gombos">laszlo.gombos</who>
    <bug_when>2010-02-22 06:37:57 -0800</bug_when>
    <thetext>As far as I know QtWebKit does nto plan to productize WML support, so server should change.</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>192497</commentid>
    <comment_count>6</comment_count>
    <who name="Simon Hausmann">hausmann</who>
    <bug_when>2010-02-22 09:59:21 -0800</bug_when>
    <thetext>(In reply to comment #5)
&gt; As far as I know QtWebKit does nto plan to productize WML support, so server
&gt; should change.

Great analysis Janne!

If it&apos;s just ovi.com, then I agree and the server should be fixed. In the worst case of it affecting other sites sending us WML, too, then I think we may have to change the user agent.

Zalan, can you help us with this to get ovi.com fixed?</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>192510</commentid>
    <comment_count>7</comment_count>
    <who name="alan">zalan</who>
    <bug_when>2010-02-22 10:23:50 -0800</bug_when>
    <thetext>(In reply to comment #6)
&gt; (In reply to comment #5)
&gt; &gt; As far as I know QtWebKit does nto plan to productize WML support, so server
&gt; &gt; should change.
&gt; 
&gt; Great analysis Janne!
&gt; 
&gt; If it&apos;s just ovi.com, then I agree and the server should be fixed. In the worst
&gt; case of it affecting other sites sending us WML, too, then I think we may have
&gt; to change the user agent.
&gt; 
&gt; Zalan, can you help us with this to get ovi.com fixed?
Sure, I&apos;ll forward it to some guys, they might be able to help...about the string, as seen here (http://www.zytrax.com/tech/web/mobile_ids.html), we do (or used to) include the phone model too in the UA string. I&apos;m sure the server scans for the &quot;Nokia&quot; str and redirects the browser accordingly. -not saying it is right. 

Here is an example of a production UA string:
Mozilla/5.0 (SymbianOS/9.4; U; Series60/5.0 Nokia5800d-1b/20.2.014; Profile/MIDP-2.1 Configuration/CLDC-1.1 ) AppleWebKit/413 (KHTML, like Gecko) Safari/413</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>192781</commentid>
    <comment_count>8</comment_count>
    <who name="Laszlo Gombos">laszlo.gombos</who>
    <bug_when>2010-02-22 21:40:42 -0800</bug_when>
    <thetext>I would think that the &quot;532.4&quot; version info should be the primary importance for the server to determine the client WebKit capabilities (and not the device model number as the same device model can have different browser and WebKit versions installed).</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>192787</commentid>
    <comment_count>9</comment_count>
    <who name="Petri Ojala">peojala11</who>
    <bug_when>2010-02-22 22:12:09 -0800</bug_when>
    <thetext>I think www.ovi.com is not alone. 

I tested some web pages and at least &quot;www.facebook.com&quot; is not loading.</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>192823</commentid>
    <comment_count>10</comment_count>
    <who name="Petri Ojala">peojala11</who>
    <bug_when>2010-02-23 01:11:24 -0800</bug_when>
    <thetext>www.facebook.com tries to redirect to m.facebook.com. 
And its Content-Type: application/vnd.wap.xhtml+xml; charset=UTF-8</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>192895</commentid>
    <comment_count>11</comment_count>
    <who name="alan">zalan</who>
    <bug_when>2010-02-23 05:30:22 -0800</bug_when>
    <thetext>(In reply to comment #8)
&gt; I would think that the &quot;532.4&quot; version info should be the primary importance
&gt; for the server to determine the client WebKit capabilities (and not the device
&gt; model number as the same device model can have different browser and WebKit
&gt; versions installed).

In an ideal world, it would be the Accept header and not the UA string. 
In a short run, enabling WML would fix the issue, but it is far from being the optimal solution for users and for Nokia in general. Is there any way to talk to some of the major content providers and let them know about the capabilities (or actually the lack of capabilities -missing wml) of the upcoming s60 browser?</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>192910</commentid>
    <comment_count>12</comment_count>
    <who name="Janne Koskinen">koshuin</who>
    <bug_when>2010-02-23 06:20:03 -0800</bug_when>
    <thetext>(In reply to comment #8)
&gt; I would think that the &quot;532.4&quot; version info should be the primary importance
&gt; for the server to determine the client WebKit capabilities (and not the device
&gt; model number as the same device model can have different browser and WebKit
&gt; versions installed).

Webkit version doesn&apos;t define what features it has enabled. We could have the same build with or without WML support. Server shouldn&apos;t be worried about webkit version. It is the browser that is built upon it that will define the features that it will support.</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>193488</commentid>
    <comment_count>13</comment_count>
    <who name="Laszlo Gombos">laszlo.gombos</who>
    <bug_when>2010-02-24 13:32:41 -0800</bug_when>
    <thetext>First, the returned content does not seemed to be WML, it seems to be XHTML-MP.

application/vnd.wap.xhtml+xml is the mime type for XHTML-MP.

We do not have XHTML-MP enabled for QtWebKit at the moment. I had a failed attempt to do so on the trunk 4 month ago  - https://bugs.webkit.org/show_bug.cgi?id=30270.

I will test if turning on XHTML-MP would help. Enabling XHTML-MP is has a lot less impact that it would be to enable full WML support.</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>194988</commentid>
    <comment_count>14</comment_count>
    <who name="Petri Ojala">peojala11</who>
    <bug_when>2010-03-02 02:08:23 -0800</bug_when>
    <thetext>I did test it too. I defined ENABLE_XHTMLMP flag as enabled and now web pages are displayed correctly (ovi.com and facebook.com).

So, I think we should enable this feature for Symbian platform (at least)?</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>195895</commentid>
    <comment_count>15</comment_count>
      <attachid>49986</attachid>
    <who name="Petri Ojala">peojala11</who>
    <bug_when>2010-03-03 22:21:09 -0800</bug_when>
    <thetext>Created attachment 49986
Patch

Patch to turn on XHTML-MP support for Symbian platform</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>195897</commentid>
    <comment_count>16</comment_count>
    <who name="WebKit Review Bot">webkit.review.bot</who>
    <bug_when>2010-03-03 22:24:39 -0800</bug_when>
    <thetext>Attachment 49986 did not pass style-queue:

Failed to run &quot;WebKitTools/Scripts/check-webkit-style&quot; exit_code: 1
WebCore/ChangeLog:9:  Line contains tab character.  [whitespace/tab] [5]
Total errors found: 1 in 2 files


If any of these errors are false positives, please file a bug against check-webkit-style.</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>195899</commentid>
    <comment_count>17</comment_count>
      <attachid>49987</attachid>
    <who name="Petri Ojala">peojala11</who>
    <bug_when>2010-03-03 22:29:48 -0800</bug_when>
    <thetext>Created attachment 49987
Patch 2

Style issue fixed</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>195941</commentid>
    <comment_count>18</comment_count>
    <who name="Simon Hausmann">hausmann</who>
    <bug_when>2010-03-04 01:20:17 -0800</bug_when>
    <thetext>I&apos;m not sure we should enable a feature that produces quite a few different code paths in WebKit just for Symbian.

Let me ask differently: What real world _feature_ of the XHTML-MP profile do we really _need_ need? What feature in this profile makes ovi.com &quot;better&quot; over a version of the same web site with &quot;desktop&quot; profile?</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>196341</commentid>
    <comment_count>19</comment_count>
      <attachid>49987</attachid>
    <who name="Laszlo Gombos">laszlo.gombos</who>
    <bug_when>2010-03-04 21:10:40 -0800</bug_when>
    <thetext>Comment on attachment 49987
Patch 2

I agree with Simon. If we end-up concluding to turn on XHTML-MP we should do it for all QtWebKit ports, not just for Symbian. Marking the patch for r-.</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>196345</commentid>
    <comment_count>20</comment_count>
      <attachid>50086</attachid>
    <who name="Laszlo Gombos">laszlo.gombos</who>
    <bug_when>2010-03-04 21:26:27 -0800</bug_when>
    <thetext>Created attachment 50086
Patch3 - enable the mime type

It seems that it might be enough to enable processing the &quot;application/vnd.wap.xhtml+xml&quot; MIME type response (it this point) no need to enable full XHTML-MP build flag. I would love to get rid of the XHTMLMP guard all together rather than adding the PLATFORM(QT) guard, but I&apos;m afraid that will need more discussion with other ports.

Tested on Linux QtWebKit build by faking the user-agent. ovi.com and facebook.com loaded OK. 

I consider this the next step, we should continue to looking into what is the correct user-agent for Symbian.</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>196375</commentid>
    <comment_count>21</comment_count>
    <who name="alan">zalan</who>
    <bug_when>2010-03-05 00:23:22 -0800</bug_when>
    <thetext>&gt; Let me ask differently: What real world _feature_ of the XHTML-MP profile do we
&gt; really _need_ need? What feature in this profile makes ovi.com &quot;better&quot; over a
&gt; version of the same web site with &quot;desktop&quot; profile?
Apparently nothing, as this content has not much to do with mobile profile. It&apos;s only the http header that says so. The content (.html) defines itself as xhtml in the meta tag

&lt;meta http-equiv=&quot;content-type&quot; content=&quot;application/xhtml+xml; charset=UTF-8&quot;/&gt;

So it is nothing more, than just a light/mobile version of a web service with a misconfigured http server (or whoever adds the content type).

&gt;I&apos;m not sure we should enable a feature that produces quite a few different code &gt;paths in WebKit just for Symbian
Coming from the other end; what would be the benefit of enabling it for all the platforms? Would you enable WML for desktop? (i know we dont have it anymore, but let&apos;s just assume we do for a sec) It is pretty much the same in the context of targeted platforms (resource constrained devices). -mp is just yet another iteration of WML in some sense (both were initiated by OMA)

...and by grepping for the changes that mp involves, there is really only a few instances where it deviates from the baseline. nothing major.</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>196505</commentid>
    <comment_count>22</comment_count>
    <who name="Tor Arne Vestbø">vestbo</who>
    <bug_when>2010-03-05 09:39:42 -0800</bug_when>
    <thetext>
Please follow the QtWebKit bug reporting guidelines when reporting bugs.

See http://trac.webkit.org/wiki/QtWebKitBugs

Specifically:

  - The &apos;QtWebKit&apos; component should be used for bugs/features in the public QtWebKit API layer, not to signify that the bug is specific to the Qt port of WebKit

    http://trac.webkit.org/wiki/QtWebKitBugs#Component</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>196536</commentid>
    <comment_count>23</comment_count>
    <who name="Antonio Gomes">tonikitoo</who>
    <bug_when>2010-03-05 10:16:54 -0800</bug_when>
    <thetext>(In reply to comment #19)
&gt; (From update of attachment 49987 [details])
&gt; I agree with Simon. If we end-up concluding to turn on XHTML-MP we should do it
&gt; for all QtWebKit ports, not just for Symbian. Marking the patch for r-.


anttik added some nice points on IRC.

&lt;anttik&gt; tronical: it is a feature to support legacy mobile content. it should only be enabled on mobile platforms i think
&lt;anttik&gt; tronical: we get mobile only if the site thinks based on the ua string that we are a mobile device
&lt;anttik&gt; only things it seems to do are:
&lt;anttik&gt; - change the noscript behavior if the document is XHTMLMP
* zalan has quit (Ping timeout: 260 seconds)
&lt;anttik&gt; - change marquee element behavior
&lt;anttik&gt; - add xhtmlmp to the accept header
&lt;anttik&gt; tronical: on the other hand, just treating xhtmlmp content type as xhtml (or text/html) would probably be sufficient (without actually enabling the feature)</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>196654</commentid>
    <comment_count>24</comment_count>
      <attachid>50086</attachid>
    <who name="Eric Seidel (no email)">eric</who>
    <bug_when>2010-03-05 14:15:31 -0800</bug_when>
    <thetext>Comment on attachment 50086
Patch3 - enable the mime type

Why wouldn&apos;t all ports want this?</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>197154</commentid>
    <comment_count>25</comment_count>
      <attachid>50219</attachid>
    <who name="Laszlo Gombos">laszlo.gombos</who>
    <bug_when>2010-03-08 07:24:40 -0800</bug_when>
    <thetext>Created attachment 50219
add xhtml-mp mime type processing capability to all WebKit ports

Make this change available for all WebKit ports, based on Eric&apos;s feedback.

This does _not_ add XHTML-MP mime type to the accept herader list; rather this just enables processing XHTML-MP mime type if a server happens to respond back with that mime type.</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>197157</commentid>
    <comment_count>26</comment_count>
    <who name="Antti Koivisto">koivisto</who>
    <bug_when>2010-03-08 07:38:28 -0800</bug_when>
    <thetext>This seems like the best solution to me. Perhaps we could also eliminate the XHTML-MP specific code completely to reduce confusion? The minor behavior changes there seem totally pointless (do they exist just to pass some operator test cases or what?)

Any objections to accepting application/vnd.wap.xhtml+xml as a valid xhtml mime type on all platforms?</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>197448</commentid>
    <comment_count>27</comment_count>
    <who name="David Kilzer (:ddkilzer)">ddkilzer</who>
    <bug_when>2010-03-08 21:29:51 -0800</bug_when>
    <thetext>(In reply to comment #26)
&gt; Any objections to accepting application/vnd.wap.xhtml+xml as a valid xhtml mime
&gt; type on all platforms?

Maciej is the correct person to ask.</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>199795</commentid>
    <comment_count>28</comment_count>
      <attachid>50219</attachid>
    <who name="Eric Seidel (no email)">eric</who>
    <bug_when>2010-03-15 13:25:09 -0700</bug_when>
    <thetext>Comment on attachment 50219
add xhtml-mp mime type processing capability to all WebKit ports

This seems OK.  We need a layout test however.

It should be possible to write a little http test which replies with the proper mime type.</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>200575</commentid>
    <comment_count>29</comment_count>
      <attachid>50869</attachid>
    <who name="Laszlo Gombos">laszlo.gombos</who>
    <bug_when>2010-03-16 20:32:48 -0700</bug_when>
    <thetext>Created attachment 50869
previous patch + test case

Added a test case per Eric&apos;s request.

The test case fails without the code change and should pass for all ports after the code change. It registers xhtmlmp file extension with the XHTML MP mime type, so that if we need more tests later the mechanism is there to do so.</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>201108</commentid>
    <comment_count>30</comment_count>
    <who name="Kenneth Rohde Christiansen">kenneth</who>
    <bug_when>2010-03-17 18:29:16 -0700</bug_when>
    <thetext>Marciej, what is your take on this?</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>203879</commentid>
    <comment_count>31</comment_count>
      <attachid>50869</attachid>
    <who name="Eric Seidel (no email)">eric</who>
    <bug_when>2010-03-25 00:50:28 -0700</bug_when>
    <thetext>Comment on attachment 50869
previous patch + test case

Shouldn&apos;t this be dumpAsText?

Are you sure it fails w/o this change?  I&apos;m not sure how we would test if it&apos;s being treated as xhtml.</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>204914</commentid>
    <comment_count>32</comment_count>
      <attachid>51806</attachid>
    <who name="Laszlo Gombos">laszlo.gombos</who>
    <bug_when>2010-03-26 19:11:19 -0700</bug_when>
    <thetext>Created attachment 51806
test using dumpAsText</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>204915</commentid>
    <comment_count>33</comment_count>
      <attachid>51807</attachid>
    <who name="Laszlo Gombos">laszlo.gombos</who>
    <bug_when>2010-03-26 19:12:09 -0700</bug_when>
    <thetext>Created attachment 51807
test output for the failure</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>204917</commentid>
    <comment_count>34</comment_count>
    <who name="Laszlo Gombos">laszlo.gombos</who>
    <bug_when>2010-03-26 19:21:19 -0700</bug_when>
    <thetext>Thanks Eric for the review.

If xhtml-mp mime type is not accepted, the &lt;script&gt; tag in the test case does not get executed, so the output of the test is an &quot;empty rendertree&quot; - see https://bug-34262-attachments.webkit.org/attachment.cgi?id=51807

I&apos;ve changed the test case (up for review) so that if the test passes it is an output of a &quot;dumpAsText&quot; test, but this will not change the output of the failure, which will be the same as with the previous test case.</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>205232</commentid>
    <comment_count>35</comment_count>
    <who name="Simon Hausmann">hausmann</who>
    <bug_when>2010-03-28 13:53:03 -0700</bug_when>
    <thetext>mjs: ping :-)</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>207504</commentid>
    <comment_count>36</comment_count>
      <attachid>51806</attachid>
    <who name="Eric Seidel (no email)">eric</who>
    <bug_when>2010-04-01 16:54:38 -0700</bug_when>
    <thetext>Comment on attachment 51806
test using dumpAsText

LGTM.</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>207812</commentid>
    <comment_count>37</comment_count>
      <attachid>51806</attachid>
    <who name="Laszlo Gombos">laszlo.gombos</who>
    <bug_when>2010-04-02 01:44:27 -0700</bug_when>
    <thetext>Comment on attachment 51806
test using dumpAsText

Thanks Eric.</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>207823</commentid>
    <comment_count>38</comment_count>
      <attachid>51806</attachid>
    <who name="WebKit Commit Bot">commit-queue</who>
    <bug_when>2010-04-02 02:31:15 -0700</bug_when>
    <thetext>Comment on attachment 51806
test using dumpAsText

Clearing flags on attachment: 51806

Committed r56994: &lt;http://trac.webkit.org/changeset/56994&gt;</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>207824</commentid>
    <comment_count>39</comment_count>
    <who name="WebKit Commit Bot">commit-queue</who>
    <bug_when>2010-04-02 02:31:23 -0700</bug_when>
    <thetext>All reviewed patches have been landed.  Closing bug.</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>208276</commentid>
    <comment_count>40</comment_count>
    <who name="George Staikos">staikos</who>
    <bug_when>2010-04-03 07:36:58 -0700</bug_when>
    <thetext>What I see from this appears to add &quot;support&quot; for the mimetype and pass the content through as XHTML.  This can&apos;t work.  XHTML-MP will cause XML parsing errors when run through the XHTML parser.  It might fix this one case of one site but that&apos;s just a coincidence as far as I can tell.  I think you want XHTML-MP enabled, which would then give you what you want - the mimetype also.</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>208373</commentid>
    <comment_count>41</comment_count>
    <who name="Eric Seidel (no email)">eric</who>
    <bug_when>2010-04-04 13:17:22 -0700</bug_when>
    <thetext>I&apos;m happy to roll this out if that&apos;s the consensus.</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>208960</commentid>
    <comment_count>42</comment_count>
    <who name="Simon Hausmann">hausmann</who>
    <bug_when>2010-04-06 02:49:04 -0700</bug_when>
    <thetext>Revision r56994 cherry-picked into qtwebkit-2.0 with commit c99c3268f7b2f1803cf61dcaef06cad855a42313</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>209136</commentid>
    <comment_count>43</comment_count>
    <who name="George Staikos">staikos</who>
    <bug_when>2010-04-06 11:37:30 -0700</bug_when>
    <thetext>Um Hello???  As far as I can tell this change is wrong.  No comment?</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>209264</commentid>
    <comment_count>44</comment_count>
    <who name="Laszlo Gombos">laszlo.gombos</who>
    <bug_when>2010-04-06 14:53:04 -0700</bug_when>
    <thetext>George, thanks for helping out with this one and sorry for the delay.

You&apos;re correct that proper XHTML-MP support is behind the XHTML-MP flag and if as port wants to support XHTML-MP proper it should turn on that flag.

I think you&apos;re also correct that some XHTML-MP will cause parsing error if not properly treated as XHTML-MP content. I do not have data on this but the assumption was that most XHTML-MP content is actually just XHTML content served with the XHTML-MP mime type. Before the patch XHTML-MP mime type content was simply ignored by WebKit, after the change WebKit at least makes an attempt to render it.

We considered turning on XHTML-MP for all QtWebKit platforms, but advertizing XHTML-MP support in the Accept header might trigger servers to send mobile XHTML-MP content also on desktop platforms. We didn&apos;t wanted to introduces fragmentation within QtWebKit platforms by turning on XHTML-MP support for only few selected platforms (e.g. Symbian). 

I believe this patch is a good compromise for most WebKit ports  - it allows ports handle XHMTL-MP properly by turning on XHTML-MP, and at the same time it makes an attempt to handle XHTML-MP content if XHTML-MP is not specifically enabled.

We can certainly make this change only for the QtWebKit port as it has been initially proposed (patch #3), but I think other WebKit ports could benefit from this change as well.</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>209298</commentid>
    <comment_count>45</comment_count>
    <who name="George Staikos">staikos</who>
    <bug_when>2010-04-06 15:44:05 -0700</bug_when>
    <thetext>Our experience is that is not the case.  Some sites will now start giving you XHTML-MP content that is not parseable with XHTML parsers, whereas before they would send HTML.  My guess is this will cause real-world regressions on browsers such as iPhone in particular.  It caused problems for us before.</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>209405</commentid>
    <comment_count>46</comment_count>
    <who name="Charles Wei">charles.wei</who>
    <bug_when>2010-04-06 19:30:15 -0700</bug_when>
    <thetext>This fix seems a workaround for ovi.com, which sends the document with MIME type application/vnd.wap.xhtml+xml for XHTML-MP , while the content is actually XHMTL.

If a server sends the document with MIME type as application/vnd.wap.xhtml+xml, very likely it will serve XHTML-MP specific contents.  feed it to XHTML parser without XHTML-MP enabled will lose these features,which might not be what the server side expects; and might even cause other parsing problems as George noted above. 

The best solution is to enable XHTML-MP support and fix the Ovi server.</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>209617</commentid>
    <comment_count>47</comment_count>
    <who name="Janne Koskinen">koshuin</who>
    <bug_when>2010-04-07 06:55:20 -0700</bug_when>
    <thetext>(In reply to comment #46)
&gt; The best solution is to enable XHTML-MP support and fix the Ovi server.

How widely this user-agent series60/symbian check has spread that is hard to tell. ovi.com and facebook.com come up as positive, maybe a lot of others. Enabling XHTML-MP support seems is one solution, changing default user-agent string another.

I&apos;ve done a lot of testing using either WLAN connection or HSDPA and that doesn&apos;t give a real impression how slow it can be if you are bit off the grid. Went on train trip with my own test browser and it was unsusable to browse &apos;normal&apos; pages with GSM data or even with GPRS.
Another issue with normal pages is RAM footprint that they require from the devices, plus they require a lot of scrolling to use.</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>209620</commentid>
    <comment_count>48</comment_count>
    <who name="Simon Hausmann">hausmann</who>
    <bug_when>2010-04-07 06:59:31 -0700</bug_when>
    <thetext>(In reply to comment #45)
&gt; Our experience is that is not the case.  Some sites will now start giving you
&gt; XHTML-MP content that is not parseable with XHTML parsers, whereas before they
&gt; would send HTML.  My guess is this will cause real-world regressions on
&gt; browsers such as iPhone in particular.  It caused problems for us before.

Note that this change does not affect the accept header sent to the HTTP server, so I fail to see how this patch _changes_ what the servers out there send. How can it regress?</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>217035</commentid>
    <comment_count>49</comment_count>
    <who name="Kenneth Rohde Christiansen">kenneth</who>
    <bug_when>2010-04-26 05:39:47 -0700</bug_when>
    <thetext>So what are we doing about this bug report? Any new insight?</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>217168</commentid>
    <comment_count>50</comment_count>
    <who name="George Staikos">staikos</who>
    <bug_when>2010-04-26 10:04:36 -0700</bug_when>
    <thetext>I don&apos;t have much interest in debating it further.  It&apos;s strange to send content to the wrong  parser simply due to a broken website even if it doesn&apos;t affect what&apos;s sent as the accept headers.  Seems wrong to me, but I&apos;m going to leave it at that.</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>217193</commentid>
    <comment_count>51</comment_count>
    <who name="Eric Seidel (no email)">eric</who>
    <bug_when>2010-04-26 11:08:59 -0700</bug_when>
    <thetext>I have no opinions either way.  Anyone is welcome to roll this out w/o notifying or asking me further.

webkit-patch rollout 34262</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>217208</commentid>
    <comment_count>52</comment_count>
    <who name="Laszlo Gombos">laszlo.gombos</who>
    <bug_when>2010-04-26 11:34:01 -0700</bug_when>
    <thetext>As the author of the patch, I&apos;d like to keep this patch in, unless a more compelling argument is raised against it. 

This patch is simply an error handling mechanism; a fallback mechanism dealing with some servers sending content that is not entirely compliant. 

Having this fallback does not justify fixing the server to send back proper content if that is possible.</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>218984</commentid>
    <comment_count>53</comment_count>
    <who name="Laszlo Gombos">laszlo.gombos</who>
    <bug_when>2010-04-29 06:14:58 -0700</bug_when>
    <thetext>Re-closing this bug as it does not look like this patch will be rolled out or changed.</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>225337</commentid>
    <comment_count>54</comment_count>
    <who name="Timothy Lu">timothy.lu</who>
    <bug_when>2010-05-13 09:16:43 -0700</bug_when>
    <thetext>Re-opened this error.  Further investigation needed.</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>225367</commentid>
    <comment_count>55</comment_count>
    <who name="Laszlo Gombos">laszlo.gombos</who>
    <bug_when>2010-05-13 10:27:22 -0700</bug_when>
    <thetext>I talked to Timothy and conlcuded that this was re-opened by mistake. 

Re-closing it, sorry about the noise.</thetext>
  </long_desc>
      
          <attachment
              isobsolete="1"
              ispatch="1"
              isprivate="0"
          >
            <attachid>49986</attachid>
            <date>2010-03-03 22:21:09 -0800</date>
            <delta_ts>2010-03-03 22:29:48 -0800</delta_ts>
            <desc>Patch</desc>
            <filename>34262.txt</filename>
            <type>text/plain</type>
            <size>857</size>
            <attacher name="Petri Ojala">peojala11</attacher>
            
              <data encoding="base64">ZGlmZiAtLWdpdCBhL1dlYkNvcmUvQ2hhbmdlTG9nIGIvV2ViQ29yZS9DaGFuZ2VMb2cKaW5kZXgg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</data>

          </attachment>
          <attachment
              isobsolete="1"
              ispatch="1"
              isprivate="0"
          >
            <attachid>49987</attachid>
            <date>2010-03-03 22:29:48 -0800</date>
            <delta_ts>2010-03-05 14:14:55 -0800</delta_ts>
            <desc>Patch 2</desc>
            <filename>34262.txt</filename>
            <type>text/plain</type>
            <size>863</size>
            <attacher name="Petri Ojala">peojala11</attacher>
            
              <data encoding="base64">ZGlmZiAtLWdpdCBhL1dlYkNvcmUvQ2hhbmdlTG9nIGIvV2ViQ29yZS9DaGFuZ2VMb2cKaW5kZXgg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</data>
<flag name="review"
          id="33005"
          type_id="1"
          status="-"
          setter="laszlo.gombos"
    />
          </attachment>
          <attachment
              isobsolete="1"
              ispatch="1"
              isprivate="0"
          >
            <attachid>50086</attachid>
            <date>2010-03-04 21:26:27 -0800</date>
            <delta_ts>2010-03-08 07:24:40 -0800</delta_ts>
            <desc>Patch3 - enable the mime type</desc>
            <filename>ovi-2.patch</filename>
            <type>text/plain</type>
            <size>1257</size>
            <attacher name="Laszlo Gombos">laszlo.gombos</attacher>
            
              <data encoding="base64">SW5kZXg6IFdlYkNvcmUvQ2hhbmdlTG9nCj09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09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</data>

          </attachment>
          <attachment
              isobsolete="1"
              ispatch="1"
              isprivate="0"
          >
            <attachid>50219</attachid>
            <date>2010-03-08 07:24:40 -0800</date>
            <delta_ts>2010-03-16 20:32:48 -0700</delta_ts>
            <desc>add xhtml-mp mime type processing capability to all WebKit ports</desc>
            <filename>ovi-3.patch</filename>
            <type>text/plain</type>
            <size>1229</size>
            <attacher name="Laszlo Gombos">laszlo.gombos</attacher>
            
              <data encoding="base64">SW5kZXg6IFdlYkNvcmUvQ2hhbmdlTG9nCj09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09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</data>
<flag name="review"
          id="33285"
          type_id="1"
          status="-"
          setter="eric"
    />
    <flag name="commit-queue"
          id="33898"
          type_id="3"
          status="-"
          setter="eric"
    />
          </attachment>
          <attachment
              isobsolete="1"
              ispatch="1"
              isprivate="0"
          >
            <attachid>50869</attachid>
            <date>2010-03-16 20:32:48 -0700</date>
            <delta_ts>2010-03-26 19:11:19 -0700</delta_ts>
            <desc>previous patch + test case</desc>
            <filename>34262.patch</filename>
            <type>text/plain</type>
            <size>6951</size>
            <attacher name="Laszlo Gombos">laszlo.gombos</attacher>
            
              <data encoding="base64">SW5kZXg6IFdlYkNvcmUvQ2hhbmdlTG9nCj09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09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</data>
<flag name="review"
          id="34041"
          type_id="1"
          status="-"
          setter="eric"
    />
          </attachment>
          <attachment
              isobsolete="0"
              ispatch="1"
              isprivate="0"
          >
            <attachid>51806</attachid>
            <date>2010-03-26 19:11:19 -0700</date>
            <delta_ts>2010-04-02 02:31:15 -0700</delta_ts>
            <desc>test using dumpAsText</desc>
            <filename>34262-2.patch</filename>
            <type>text/plain</type>
            <size>6886</size>
            <attacher name="Laszlo Gombos">laszlo.gombos</attacher>
            
              <data encoding="base64">SW5kZXg6IFdlYkNvcmUvQ2hhbmdlTG9nCj09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09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==
</data>

          </attachment>
          <attachment
              isobsolete="0"
              ispatch="0"
              isprivate="0"
          >
            <attachid>51807</attachid>
            <date>2010-03-26 19:12:09 -0700</date>
            <delta_ts>2010-03-26 19:12:09 -0700</delta_ts>
            <desc>test output for the failure</desc>
            <filename>xhtml-actual.txt</filename>
            <type>text/plain</type>
            <size>134</size>
            <attacher name="Laszlo Gombos">laszlo.gombos</attacher>
            
              <data encoding="base64">bGF5ZXIgYXQgKDAsMCkgc2l6ZSA4MDB4NjAwCiAgUmVuZGVyVmlldyBhdCAoMCwwKSBzaXplIDgw
MHg2MDAKbGF5ZXIgYXQgKDAsMCkgc2l6ZSA4MDB4NjAwCiAgUmVuZGVyQmxvY2sge0hUTUx9IGF0
ICgwLDApIHNpemUgODAweDYwMAo=
</data>

          </attachment>
      

    </bug>

</bugzilla>