Re: [Apache Bloodhound] #204: Activity feed currently breaks new ticket page link

[Gary Martin <ga...@wandisco.com>]

On 20/09/12 14:55, Olemis Lang wrote:
> On 9/19/12, Gary Martin <ga...@wandisco.com> wrote:
>> On 19/09/12 19:38, Olemis Lang wrote:
>>> On 9/19/12, Apache Bloodhound <bl...@incubator.apache.org>
>>> wrote:
>>>> #204: Activity feed currently breaks new ticket page link
>>>> ------------------------+-----------------------
>>>>     Reporter:  gjm        |      Owner:  nobody
>>>>         Type:  defect     |     Status:  new
>>>>     Priority:  critical   |  Milestone:  Release 2
>>>>    Component:  dashboard  |    Version:
>>>> Resolution:             |   Keywords:
>>>> ------------------------+-----------------------
>>>>
>>>> Comment (by gjm):
>>>>
>>> [...]
>>>>    ---
>>>>
>>>> bloodhound_dashboard/bhdashboard/widgets/templates/widget_timeline.html
>>>>    (revision 1387596)
>>>>    +++
>>>>
>>>> bloodhound_dashboard/bhdashboard/widgets/templates/widget_timeline.html
>>>>    (working copy)
>>>>    @@ -50,6 +50,7 @@
>>>>         </tbody>
>>>>       </table>
>>>>       <py:otherwise>
>>>>    +    <py:if test="context.resource.id">
>>>>         <py:def function="timeline_empty()">
>>>>           No events reported for <em>${summary_of(context.resource)}</em>
>>>> in
>>>>    the
>>>>           last <em>$daysback</em> days since
>>>>    @@ -59,5 +60,6 @@
>>>>         </py:def>
>>>>         <xi:include href="widget_alert.html"
>>>>             py:with="msglabel = 'Warning'; msgbody = timeline_empty()" />
>>>>    +    </py:if>
>>>>       </py:otherwise>
>>>>     </div>
>>>>    }}}
>>>>
>>> I'd rather sugesst doing something like
>>>
>>> <em py:if="context.resource.id">${summary_of(context.resource)}</em>
>>>
>>> or
>>>
>>> <em>${summary_of(context.resource) if context.resource.id else
>>> 'Unknown resource'}</em>
>>>
>>>
>> Well, that is a bit too specific and the message will no longer make
>> sense. But either way this still feels like a fix for the symptom rather
>> than the cause. I don't think that trac's TicketSystem.format_summary
>> method should fail for such input. I may of course be asking too much.
>>
> yes . I just looked at it once again , and it should not fail that ...
>
>> A little more off topic, it is also annoying that the error message
>> appears at all. There are a lot of false positives positive situations
>> where there are no events to be displayed and the message will display.
> cases like what ?
>
>> Hopefully we can remove some of those (event collection in a non-time
>> based manner may be helpful here in particular) but we should probably
>> only suggest that there is a problem with the system configuration if
>> there actually is.
>>
> you may be right , yes . I see your point .
>

OK, I guess we could do with some action on this ticket. This is an 
important bug to fix, so I'll implement one of the quick hacks. After 
that I'll raise a ticket for fixing the underlying problem.

I have already raised https://issues.apache.org/bloodhound/ticket/214 to 
look at the warning message when no events are reported.

Cheers,
     Gary