[postgis-devel] astyle
Mark Cave-Ayland
mark.cave-ayland at siriusit.co.uk
Wed Jan 27 08:31:32 PST 2010
Paragon Corporation wrote:
> My reasoning is as follows
>
> 1) The change Paul requested was small and since 2.0 will not be released
> for another year and with our new policy this change would not make it in
> for another year.
> He would have dormant code sitting in 1.5.1 or whatever which I find
> terribly annoying and he couldn't put it in 1.5.0 because code is code
> regardless of if it changes API.
There is a well-established convention for releasing software. It goes
like this:
- Alpha = this is very new and likely to break
- Beta = this is feature complete and can be run in a test environment
(API is fixed)
- RC = this now has documentation and is packaged correctly; only
minor fixes to the documentation & packaging allowed
We have already violated RC with Paul's changes, which is generally not
a good precedent to set for people who are already testing 1.5 in beta
and production environments. So regardless of the fact it was a small
change, we still need to be considerate of this.
> 2) The re-formatting as Mateusz says is a cosmetic thing. A cosmetic thing
> that changes MANY FILES and makes code look different even if it is not.
> That's worse than changing code before a release.
>
> When you change many files randomly you have a relatively high probability
> of something going wrong and that wrongness is unpredictable. That is why I
> say it requires more testing for that.
Have you seen the resulting diff from the patch I posted? It looks
extremely sane and IMO improves readability. And the changes aren't
random - they are based on a fixed set of rules. I have no problem with
a brief set of commits correcting the formatting, however we should
avoid patches that change both spacing and code changes as much as possible.
> Besides this is something that doesn't change the API so can be done any
> time. I fail to see the sense of urgency in this.
Because it sanitises the codebase from the release point while not
changing the logic in any way.
ATB,
Mark.
--
Mark Cave-Ayland - Senior Technical Architect
PostgreSQL - PostGIS
Sirius Corporation plc - control through freedom
http://www.siriusit.co.uk
t: +44 870 608 0063
Sirius Labs: http://www.siriusit.co.uk/labs
More information about the postgis-devel
mailing list