Hi,
Alan wrote (07 Mar 2014 13:51:02 GMT) :
>> * HACKING has not been updated to document that one should run the
>> test suite before releasing (fixed in c31f5a5)
> This was kind of intended, as it is far from complete and the
> refactoring it is supposed to test didn't happen yet (see:
> https://labs.riseup.net/code/issues/6799)
>> * the test suite does not pass for me (see below); does it pass
>> for you?
>>
> No, and it's kind of intended:
> https://labs.riseup.net/code/issues/6799. Do you prefer I comment that
> test now?
IMO the test suite should match the current state of the code, and not
be expected to fail, else it's basically useless. Most testing
frameworks I know have a feature to mark a given set of tests as
"TODO, expected to fail" (either called "TODO" or "SKIP", depending on
the framework).
Please either make use of this feature, or comment out the buggy test.
>> So I'm giving up for now. Alan, do you think you'll have time to take
>> care of this today? Otherwise, please mark this set of tickets as
>> postponed to whatever version you think it can be ready for.
>>
> Sorry I didn't had time before to comment. Sorry to to have precized
> all that.
> I was willing to merge this so that at least most known leaks would be
> solved for 1.0. Do you think it's still possible?
I'm not be particularly excited at the idea of merging this after the
freeze, while we have had 3 months (since the last major release) to
make it ready (minus the time when the problem was reported, but that
was a while ago anyway). IMO this change is not particularly urgent,
and does not deserve a freeze exception.
But this is not my call, so maybe you'll want to 1. fix the remaining
problems ASAP; 2. talk to the current RM.
Cheers!
--
intrigeri
| GnuPG key @ https://gaffer.ptitcanardnoir.org/intrigeri/intrigeri.asc
| OTR fingerprint @ https://gaffer.ptitcanardnoir.org/intrigeri/otr.asc
