[Libreoffice-qa] next BH session and feature testing

[Dave Barton]

Hi Sophie,
Thank you for taking the time to reply. Comments in-line.

-------- Original Message  --------
From: Sophie <gautier.sophie at gmail.com>
To: libreoffice-qa at lists.freedesktop.org
Date: Mon, 13 Jun 2016 16:22:06 +0200

> Hi Dave,
> Le 13/06/2016 15:58, Dave Barton a écrit :
>> -------- Original Message  --------
>> From: Sophie <gautier.sophie at gmail.com>
>> To: LibreOffice-QA <libreoffice-qa at lists.freedesktop.org>
>> Date: Thu, 9 Jun 2016 18:25:15 +0200
>>
>>> Hi all,
>>>
>>> I'm preparing the next BH session which will take place on the 24th of
>>> June. Are there some features that you see need more testing in the 5.2
>>> version?
>>> Thanks in advance
>>> Cheers
>>> Sophie
>>
>> Thanks for the info Sophie. I see that marketing are now preparing for a
>> launch of 5.2 on the 3rd or 4th of August.
> Yes,
>>
>> I also note that the ESC propose not to have a Beta 3 and propose to
>> move directly to RC1. Will the RC1 be available for the BH session on
>> the 24th June?
> 
> Yes, there is only a shift to a more longer RC time than beta one, no
> move on the planing see:
> https://wiki.documentfoundation.org/ReleasePlan/5.2#5.2.0_release
>>
>> I am very concerned that the release date is getting close and the
>> regression bug 99310:
>> https://bugs.documentfoundation.org/show_bug.cgi?id=99310
>> remains unassigned.
> 
> Well there always will be a bug remaining for somebody ;-) This version
> is not supposed to be used on production until some months.

Sorry, but that is the definition of beta code, where there is likely to
be undiscovered bugs. In this case it seems that 5.2 is going to be
released with at least this one KNOWN REGRESSION reported when
5.2.0.0.alpha0+ was first offered for testing, more than 3 months ago.

LO 5.2.0 will become the "Fresh" edition automatically offered to anyone
wanting to download LO. This will become the first release since OOo 2.x
and Calc the only currently available spreadsheet, that doesn't have a
functioning data validity feature.

There will be many of us out promoting the new features of this release.
It will be sad if we have also have to say "Please do NOT use this
software in a production environment".

>> On the surface this looks like an easy hack, but unfortunately other
>> commitments prevent me taking the time to dig into the code to provide a
>> patch, or find and put forward a proposal to revert the commit that
>> created this regression.
> 
> Any chance you can bibisect it? that would help a lot to have it corrected.

I would be happy to bibisect this and some other known regressions, but
as stated previously I simply don't have the time right now.

> Cheers
> Sophie

Regards
Dave