Hi there,

On 09.03.2015 09:46, Andrew Featherstone wrote:
The last time (that I can recall) the issue tracker was discussed the plan was 
that the project would be moving to a tailored
Roundup based system. What's the state of that work?


to my knowledge the state is pretty much the same as was discussed in:

  https://pairlist2.pair.net/pipermail/scons-dev/2012-September/000223.html
  https://pairlist2.pair.net/pipermail/scons-dev/2014-April/001323.html
  http://two.pairlist.net/pipermail/scons-dev/2014-May/001369.html

and

  https://pairlist2.pair.net/pipermail/scons-dev/2014-July/001511.html

So there are scripts to migrate from Tigris->Roundup, we'd just need an instance. There is still a minor problem, in that the "creation/last activity" time of all bugs gets set to the "import" time and can't be tweaked. So one can't properly sort old bugs depending on their "age" anymore...which I really regard to be a loss.

What I took from all the discussions so far, was that we definitely don't want to deal with Tigris anymore. So any step in a different direction should be a win.
Our latest plan was a migration to Roundup, because:

  a.) it's Python, and we like to support everyone who's holding the Python 
flag high (like ourselves),
b.) it supports a lot more "fields" for classifying issues than, for example, the Github tracker,... so the overall information loss would be minimal when converting the whole archive.

Here I assume, that we don't want to simply start with new bugs in a different tracker, because we (I) have to be able to search in the old archives or reference them. Tigris would still be an issue then, and we don't want that, right? There should be only *one* tracker...and I'd really like to see a "Demo" Roundup instance in action, with all our current issues imported. So it would be really cool if:

- Anatoly would express that he's still interested in vamping Roundup to support OpenID, and tell us whether he still needs help with the documentation to get this going.
- Some native speaker(s) would volunteer to help him out then.
- Someone would volunteer to setup a Roundup demo instance, so we could play around with it a bit...and get a feel for whether it's worth a try.


The team at zeroMQ tried to enshrine their development method here 
http://rfc.zeromq.org/spec:22 which makes it really easy to
understand the criteria for things being how they are and/or justify changes. 
Stating that SCons uses this process with
modifications would make sense to me.

We have our similar statement(s) at http://www.scons.org/dev.php , http://www.scons.org/guidelines.php , and not to forget http://www.scons.org/wiki/DeveloperGuide/Introduction , feel free to suggest changes or additions.

Best regards,

Dirk


_______________________________________________
Scons-dev mailing list
Scons-dev@scons.org
https://pairlist2.pair.net/mailman/listinfo/scons-dev

Reply via email to