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

<bugzilla version="5.0.6"
          urlbase="https://bugzilla.yoctoproject.org/"
          
          maintainer="it-coreprojects-helpdesk@linuxfoundation.org"
>

    <bug>
          <bug_id>7334</bug_id>
          
          <creation_ts>2015-02-19 11:18:21 +0000</creation_ts>
          <short_desc>The &apos;machine&apos; information is not searchable in the all builds, all projects and project builds tables</short_desc>
          <delta_ts>2016-01-06 10:07:27 +0000</delta_ts>
          <reporter_accessible>1</reporter_accessible>
          <cclist_accessible>1</cclist_accessible>
          <classification_id>7</classification_id>
          <classification>Build System, Metadata &amp; Runtime</classification>
          <product>Toaster</product>
          <component>toaster</component>
          <version>unspecified</version>
          <rep_platform>x86</rep_platform>
          <op_sys>Multiple</op_sys>
          <bug_status>VERIFIED</bug_status>
          <resolution>WONTFIX</resolution>
          
          
          <bug_file_loc></bug_file_loc>
          <status_whiteboard></status_whiteboard>
          <keywords></keywords>
          <priority>Medium</priority>
          <bug_severity>normal</bug_severity>
          <target_milestone>2.1</target_milestone>
          <dependson>8738</dependson>
          
          <everconfirmed>1</everconfirmed>
          <reporter name="Belen Barros Pena">belen.barros.pena</reporter>
          <assigned_to name="David Reyna">david.reyna</assigned_to>
          <cc>alexandru.damian</cc>
    
    <cc>belen.barros.pena</cc>
    
    <cc>bluelightning</cc>
    
    <cc>bogdanx.a.voiculescu</cc>
    
    <cc>cristiana.voicu</cc>
    
    <cc>david.reyna</cc>
    
    <cc>jessica.zhang</cc>
          
          <qa_contact name="Bogdan Alexandru Voiculescu">bogdanx.a.voiculescu</qa_contact>
          <cf_os>---</cf_os>
          <cf_regression_type>---</cf_regression_type>
          
          <cf_docchange>No (bug/feature does not impact docs)</cf_docchange>

      

      

      

          <comment_sort_order>oldest_to_newest</comment_sort_order>  
          <long_desc isprivate="0" >
    <commentid>48922</commentid>
    <comment_count>0</comment_count>
    <who name="Belen Barros Pena">belen.barros.pena</who>
    <bug_when>2015-02-19 11:18:21 +0000</bug_when>
    <thetext>This was working before, I am pretty sure. But something must have changed and I can no longer search the all builds or the all projects tables by machine names.</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>48924</commentid>
    <comment_count>1</comment_count>
    <who name="Belen Barros Pena">belen.barros.pena</who>
    <bug_when>2015-02-19 11:22:46 +0000</bug_when>
    <thetext>Also, the project name is not sercah</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>48925</commentid>
    <comment_count>2</comment_count>
    <who name="Belen Barros Pena">belen.barros.pena</who>
    <bug_when>2015-02-19 11:23:06 +0000</bug_when>
    <thetext>(In reply to comment #1)
&gt; Also, the project name is not sercah

(In reply to comment #1)
&gt; Also, the project name is not sercah

searchable</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>49508</commentid>
    <comment_count>3</comment_count>
    <who name="David Reyna">david.reyna</who>
    <bug_when>2015-03-18 03:20:48 +0000</bug_when>
    <thetext>Add &quot;build__machine&quot; to &quot;bldcontrol/models.py&quot;:

    search_allowed_fields = (&quot;brtarget__target&quot;, &quot;build__project__name&quot;, &quot;build__machine&quot;)

The search on project names was already fixed earlier with #6682.</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>49794</commentid>
    <comment_count>4</comment_count>
    <who name="Belen Barros Pena">belen.barros.pena</who>
    <bug_when>2015-03-26 10:55:05 +0000</bug_when>
    <thetext>The search only happens for builds, not for failed build requests. Since Toaster does not differentiate between builds and build requests (the difference is an implementation detail of no interest to users), the search looks like broken.</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>49928</commentid>
    <comment_count>5</comment_count>
    <who name="David Reyna">david.reyna</who>
    <bug_when>2015-03-30 20:48:31 +0000</bug_when>
    <thetext>&gt; &quot;The search only happens for builds, not for failed build requests&quot;

This search is not achievable with the current Toaster design, as we have discovered with other synthetic values.

The &quot;machine&quot; value for failed builds is made available to the view by special code that looks up and inserts these into the context&apos;s objects with temporary members, but this is done _after_ the search filter is applied and is therefor not searchable. Nor can this addition be done before the search filter because that filter has the side effect of removing inserted members.

The possible solutions for the future are:

  (a) allow searches on locally inserted members - the current filter removes all added members

  (b) allow searches on object routines (in this case &quot;BuildRequest.get_machine()&quot;) - the current filter only operates with physical object members.

I have therefor moved this case back to &quot;Resolved&quot;.</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>49974</commentid>
    <comment_count>6</comment_count>
    <who name="Belen Barros Pena">belen.barros.pena</who>
    <bug_when>2015-03-31 13:46:28 +0000</bug_when>
    <thetext>(In reply to comment #5)
&gt; &gt; &quot;The search only happens for builds, not for failed build requests&quot;
&gt; 
&gt; This search is not achievable with the current Toaster design, as we have
&gt; discovered with other synthetic values.

Well, this is quite bad: from the user&apos;s perspective, it looks like the search is broken, and makes Toaster look unreliable. I&apos;d rather revert the fix, to be honest, back to the machine not being searchable. At least that way search does not return what looks like incorrect results.

&gt; 
&gt; The &quot;machine&quot; value for failed builds is made available to the view by
&gt; special code that looks up and inserts these into the context&apos;s objects with
&gt; temporary members, but this is done _after_ the search filter is applied and
&gt; is therefor not searchable. Nor can this addition be done before the search
&gt; filter because that filter has the side effect of removing inserted members.
&gt; 
&gt; The possible solutions for the future are:
&gt; 
&gt;   (a) allow searches on locally inserted members - the current filter
&gt; removes all added members
&gt; 
&gt;   (b) allow searches on object routines (in this case
&gt; &quot;BuildRequest.get_machine()&quot;) - the current filter only operates with
&gt; physical object members.
&gt; 
&gt; I have therefor moved this case back to &quot;Resolved&quot;.

This is not resolved. We can not leave this as is: reopening and moving to 1.9. This bug should go into the list of known issues published with the release notes.</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>49977</commentid>
    <comment_count>7</comment_count>
    <who name="Alexandru Damian">alexandru.damian</who>
    <bug_when>2015-03-31 14:00:20 +0000</bug_when>
    <thetext>Per chat with Belen, I am reverting the patch since incomplete functionality is worse than no functionality.</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>50223</commentid>
    <comment_count>8</comment_count>
    <who name="Belen Barros Pena">belen.barros.pena</who>
    <bug_when>2015-04-10 11:10:46 +0000</bug_when>
    <thetext>*** Bug 7498 has been marked as a duplicate of this bug. ***</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>51459</commentid>
    <comment_count>9</comment_count>
    <who name="David Reyna">david.reyna</who>
    <bug_when>2015-06-03 08:03:06 +0000</bug_when>
    <thetext>The solution will also solve 7154.</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>51667</commentid>
    <comment_count>10</comment_count>
    <who name="Alexandru Damian">alexandru.damian</who>
    <bug_when>2015-06-12 13:40:30 +0000</bug_when>
    <thetext>Alex D needs a chat with David about this bug - since the builds table contents changed showing from BuildRequests to Builds, the machine field is now searchable again.</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>56594</commentid>
    <comment_count>11</comment_count>
    <who name="Stephen K Jolley">sjolley.yp.pm</who>
    <bug_when>2015-11-18 16:44:23 +0000</bug_when>
    <thetext>This has been REOPENED</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>57426</commentid>
    <comment_count>12</comment_count>
    <who name="Belen Barros Pena">belen.barros.pena</who>
    <bug_when>2015-12-22 13:45:48 +0000</bug_when>
    <thetext>As has been mentioned, Django does not support sorting calculated or derived fields out of the box. Implementing this sorting seems then a bit too much work for not that much return. So setting a &apos;won&apos;t fix&apos;.</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>57562</commentid>
    <comment_count>13</comment_count>
    <who name="Bogdan Alexandru Voiculescu">bogdanx.a.voiculescu</who>
    <bug_when>2016-01-06 10:07:27 +0000</bug_when>
    <thetext>Verified per above comment.</thetext>
  </long_desc>
      
      

    </bug>

</bugzilla>