<?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>26784</bug_id>
          
          <creation_ts>2009-06-28 12:35:02 -0700</creation_ts>
          <short_desc>Enable XSSAuditor by default</short_desc>
          <delta_ts>2009-07-10 18:22:42 -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>WebKit Misc.</component>
          <version>528+ (Nightly build)</version>
          <rep_platform>All</rep_platform>
          <op_sys>All</op_sys>
          <bug_status>RESOLVED</bug_status>
          <resolution>FIXED</resolution>
          
          
          <bug_file_loc></bug_file_loc>
          <status_whiteboard></status_whiteboard>
          <keywords></keywords>
          <priority>P2</priority>
          <bug_severity>Normal</bug_severity>
          <target_milestone>---</target_milestone>
          <dependson>26807</dependson>
          
          <everconfirmed>1</everconfirmed>
          <reporter name="Adam Barth">abarth</reporter>
          <assigned_to name="Nobody">webkit-unassigned</assigned_to>
          <cc>abarth</cc>
    
    <cc>dbates</cc>
    
    <cc>sam</cc>
          

      

      

      

          <comment_sort_order>oldest_to_newest</comment_sort_order>  
          <long_desc isprivate="0" >
    <commentid>128746</commentid>
    <comment_count>0</comment_count>
    <who name="Adam Barth">abarth</who>
    <bug_when>2009-06-28 12:35:02 -0700</bug_when>
    <thetext>We should try enabling the XSSAuditor by default in the nightly to get a sense for the false positive rate.  Sam said we should do this once we have decent test coverage, and we now have 29 tests.

Please CC me and Dan on any regressions / false positives we find.  If we get a bunch of them, we can turn off the auditor again while we think about how to reduce them.

We still have one known false negative (HTML entities), but we can work on fixing that in parallel.  Also, we should support the &quot;turn off XSS filtering&quot; header that IE8 supports, but I&apos;ll file a separate bug about that.</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>128747</commentid>
    <comment_count>1</comment_count>
      <attachid>31993</attachid>
    <who name="Adam Barth">abarth</who>
    <bug_when>2009-06-28 12:35:51 -0700</bug_when>
    <thetext>Created attachment 31993
patch</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>128769</commentid>
    <comment_count>2</comment_count>
    <who name="Daniel Bates">dbates</who>
    <bug_when>2009-06-28 19:30:46 -0700</bug_when>
    <thetext>Another known false negative is HTTP header injection.

(In reply to comment #0)
&gt; We should try enabling the XSSAuditor by default in the nightly to get a sense
&gt; for the false positive rate.  Sam said we should do this once we have decent
&gt; test coverage, and we now have 29 tests.
&gt; 
&gt; Please CC me and Dan on any regressions / false positives we find.  If we get a
&gt; bunch of them, we can turn off the auditor again while we think about how to
&gt; reduce them.
&gt; 
&gt; We still have one known false negative (HTML entities), but we can work on
&gt; fixing that in parallel.  Also, we should support the &quot;turn off XSS filtering&quot;
&gt; header that IE8 supports, but I&apos;ll file a separate bug about that.
&gt; 

</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>128779</commentid>
    <comment_count>3</comment_count>
    <who name="Adam Barth">abarth</who>
    <bug_when>2009-06-28 23:05:08 -0700</bug_when>
    <thetext>(In reply to comment #2)
&gt; Another known false negative is HTTP header injection.

True.  I suppose we could try to stop that too.  I was thinking of that as more of a non-goal for the first version.</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>128783</commentid>
    <comment_count>4</comment_count>
    <who name="Eric Seidel (no email)">eric</who>
    <bug_when>2009-06-29 00:04:25 -0700</bug_when>
    <thetext>I would think it would be annoying that this would break plexode.com and hixie&apos;s live dom viewer in the nightlies.  Not a deal breaker.  But definitely important that we give sites a way to work around this.</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>128784</commentid>
    <comment_count>5</comment_count>
    <who name="Eric Seidel (no email)">eric</who>
    <bug_when>2009-06-29 00:05:57 -0700</bug_when>
    <thetext>FYI, neither plexode.com or Hixie&apos;s dom viewer send the IE8 header at this time:

% curl -I http://www.plexode.com/cgi-bin/main.py
HTTP/1.1 200 OK
Date: Mon, 29 Jun 2009 07:03:40 GMT
Server: Apache/2.2.8 (Ubuntu) DAV/2 SVN/1.4.6 mod_fastcgi/2.4.6 PHP/5.2.4-2ubuntu5.6 with Suhosin-Patch mod_ssl/2.2.8 OpenSSL/0.9.8g
Content-Type: text/html

% curl -I http://software.hixie.ch/utilities/js/live-dom-viewer/
HTTP/1.1 200 OK
Date: Mon, 29 Jun 2009 07:04:41 GMT
Server: Apache/2.0.63 (Unix) PHP/5.2.6 mod_ssl/2.0.63 OpenSSL/0.9.7e mod_fastcgi/2.4.2 DAV/2 SVN/1.4.2
Accept-Ranges: bytes
Vary: Accept-Encoding
X-Pingback: http://tracking.damowmow.com/
Content-Language: en-GB-x-Hixie
Content-Length: 13791
Content-Type: text/html; charset=utf-8
</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>128785</commentid>
    <comment_count>6</comment_count>
    <who name="Eric Seidel (no email)">eric</who>
    <bug_when>2009-06-29 00:08:49 -0700</bug_when>
    <thetext>I emailed Aaron Whyte (plexode) and Hixie to notify them of this pending change.  BTW, I&apos;m totally in favor of this change.  Although I would prefer we added HTTP Header opt-in support before.... or at least really really soon after making this change.</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>128789</commentid>
    <comment_count>7</comment_count>
    <who name="Ian &apos;Hixie&apos; Hickson">ian</who>
    <bug_when>2009-06-29 01:25:17 -0700</bug_when>
    <thetext>I&apos;m not adding anything that&apos;s explicitly asking the browsers to follow the specs.</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>128822</commentid>
    <comment_count>8</comment_count>
    <who name="Adam Barth">abarth</who>
    <bug_when>2009-06-29 09:16:17 -0700</bug_when>
    <thetext>Eric, do you have examples of where plexode.com and hixie&apos;s live dom viewer do the wrong thing with the auditor turned on?  I played with them for a few minutes and they seemed to work fine.

(You can test the auditor by enabling the pref in the above patch or by grabbing a trunk build of Chromium and passing the --enable-xss-auditor command line flag.)</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>128846</commentid>
    <comment_count>9</comment_count>
    <who name="Adam Barth">abarth</who>
    <bug_when>2009-06-29 11:23:58 -0700</bug_when>
    <thetext>We need to fix Bug 26807 first.</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>129012</commentid>
    <comment_count>10</comment_count>
      <attachid>31993</attachid>
    <who name="Eric Seidel (no email)">eric</who>
    <bug_when>2009-06-30 02:58:13 -0700</bug_when>
    <thetext>Comment on attachment 31993
patch

No need for this to be in the review queue while blocked by a crasher.  Removing r=? for now.</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>129909</commentid>
    <comment_count>11</comment_count>
      <attachid>31993</attachid>
    <who name="Adam Barth">abarth</who>
    <bug_when>2009-07-06 01:33:27 -0700</bug_when>
    <thetext>Comment on attachment 31993
patch

Marking for review again now that this is actionable again.</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>130611</commentid>
    <comment_count>12</comment_count>
      <attachid>31993</attachid>
    <who name="Sam Weinig">sam</who>
    <bug_when>2009-07-08 21:42:42 -0700</bug_when>
    <thetext>Comment on attachment 31993
patch

By the power of Grayskull!  r=me</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>130615</commentid>
    <comment_count>13</comment_count>
      <attachid>31993</attachid>
    <who name="Adam Barth">abarth</who>
    <bug_when>2009-07-08 21:58:26 -0700</bug_when>
    <thetext>Comment on attachment 31993
patch

Clearing the review flag.  We&apos;re going to hold off landing this for a few days.</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>131039</commentid>
    <comment_count>14</comment_count>
    <who name="Adam Barth">abarth</who>
    <bug_when>2009-07-10 18:22:42 -0700</bug_when>
    <thetext>Sending        WebKit/mac/ChangeLog
Sending        WebKit/mac/WebView/WebPreferences.mm
Sending        WebKit/win/ChangeLog
Sending        WebKit/win/WebPreferences.cpp
Transmitting file data ....
Committed revision 45740.</thetext>
  </long_desc>
      
          <attachment
              isobsolete="0"
              ispatch="1"
              isprivate="0"
          >
            <attachid>31993</attachid>
            <date>2009-06-28 12:35:51 -0700</date>
            <delta_ts>2009-07-08 21:58:26 -0700</delta_ts>
            <desc>patch</desc>
            <filename>bugzilla_requires_a_filename.patch</filename>
            <type>text/plain</type>
            <size>2906</size>
            <attacher name="Adam Barth">abarth</attacher>
            
              <data encoding="base64">SW5kZXg6IFdlYktpdC9tYWMvQ2hhbmdlTG9nCj09PT09PT09PT09PT09PT09PT09PT09PT09PT09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=
</data>

          </attachment>
      

    </bug>

</bugzilla>