[gdal-dev] GH repo - write access

classic Classic list List threaded Threaded
8 messages Options
Reply | Threaded
Open this post in threaded view
|

[gdal-dev] GH repo - write access

Martin Landa
Hi,

is there any policy of assigning write access to GH repository for
those of developers who had such permission for old SVN repo?

I would like to fix [1], do I need to express my interest for write
access explicitly somehow?

Thanks, Martin

[1] https://github.com/OSGeo/gdal/issues/365

--
Martin Landa
http://geo.fsv.cvut.cz/gwiki/Landa
http://gismentors.cz/mentors/landa
_______________________________________________
gdal-dev mailing list
[hidden email]
https://lists.osgeo.org/mailman/listinfo/gdal-dev
Reply | Threaded
Open this post in threaded view
|

Re: GH repo - write access

Even Rouault-2
On dimanche 8 avril 2018 22:36:27 CEST Martin Landa wrote:
> Hi,
>
> is there any policy of assigning write access to GH repository for
> those of developers who had such permission for old SVN repo?

Hi Martin,

Yep, just ask as your did.
I've just sent you an invite.
Please also add yourself in the new section of COMMITTERS where we list folks
with git push rights

Even

--
Spatialys - Geospatial professional services
http://www.spatialys.com
_______________________________________________
gdal-dev mailing list
[hidden email]
https://lists.osgeo.org/mailman/listinfo/gdal-dev
Reply | Threaded
Open this post in threaded view
|

Re: GH repo - write access

Martin Landa
In reply to this post by Martin Landa
Hi,

2018-04-08 22:38 GMT+02:00 Dan Little <[hidden email]>:
> With GitHub, you usually do a personal fork and then submit a pull request.
> Not quite as easy as direct commits to the main repository but standard
> practice.

really? I thought that PR is mechanism for those who haven't write
permission to the repo. After reviewing PR, it can be applied by
someone who has write access.

Question: So _core_ GDAL developers (with write access) should also
create PR? Than wait for feedback and any many cases apply PR on their
own?

Thanks for clarification, Ma

--
Martin Landa
http://geo.fsv.cvut.cz/gwiki/Landa
http://gismentors.cz/mentors/landa
_______________________________________________
gdal-dev mailing list
[hidden email]
https://lists.osgeo.org/mailman/listinfo/gdal-dev
Reply | Threaded
Open this post in threaded view
|

Re: GH repo - write access

Even Rouault-2
> Question: So _core_ GDAL developers (with write access) should also
> create PR? Than wait for feedback and any many cases apply PR on their
> own?

It is up to you, and your confidence in your change. There can be 3
strategies:
* directly push to origin (https://github.com/OSGeo gdal)
* do a pull request if Iyou want other people to have a chance to review and/
or have CI tests running
* or push to your own github account so that CI tests are still run on it, and
then merge if everything's OK
As far as I'm concerned, I do a mix of the 3 depending on the change, my mood
of the moment, etc...

If you directly push to origin, make sure to monitor
https://travis-ci.org/OSGeo/gdal/builds/
and
https://ci.appveyor.com/project/OSGeo/gdal/history
and be ready to take appropriate action if something breaks. There are
unfortunately some flakes sometimes, or timeouts, so a failure is not
necessarily yours, and you'll have to look at logs.

Even

--
Spatialys - Geospatial professional services
http://www.spatialys.com
_______________________________________________
gdal-dev mailing list
[hidden email]
https://lists.osgeo.org/mailman/listinfo/gdal-dev
Reply | Threaded
Open this post in threaded view
|

Re: GH repo - write access

Martin Landa
In reply to this post by Even Rouault-2
Hi Even,

2018-04-08 22:40 GMT+02:00 Even Rouault <[hidden email]>:
> Yep, just ask as your did.
> I've just sent you an invite.

thanks!

> Please also add yourself in the new section of COMMITTERS where we list folks
> with git push rights

Do you mean [1]? My name seems to be already there.

Ma

[1] https://github.com/OSGeo/gdal/blob/master/gdal/COMMITTERS

--
Martin Landa
http://geo.fsv.cvut.cz/gwiki/Landa
http://gismentors.cz/mentors/landa
_______________________________________________
gdal-dev mailing list
[hidden email]
https://lists.osgeo.org/mailman/listinfo/gdal-dev
Reply | Threaded
Open this post in threaded view
|

Re: GH repo - write access

Even Rouault-2
> Do you mean [1]? My name seems to be already there.
>

Yes, but we've added a new git related section at the top of the file. Your
name is in the old SVN section.

> Ma
>
> [1] https://github.com/OSGeo/gdal/blob/master/gdal/COMMITTERS


--
Spatialys - Geospatial professional services
http://www.spatialys.com
_______________________________________________
gdal-dev mailing list
[hidden email]
https://lists.osgeo.org/mailman/listinfo/gdal-dev
Reply | Threaded
Open this post in threaded view
|

Re: GH repo - write access

Martin Landa
In reply to this post by Even Rouault-2
Hi,

2018-04-08 23:01 GMT+02:00 Even Rouault <[hidden email]>:

> It is up to you, and your confidence in your change. There can be 3
> strategies:
> * directly push to origin (https://github.com/OSGeo gdal)
> * do a pull request if Iyou want other people to have a chance to review and/
> or have CI tests running
> * or push to your own github account so that CI tests are still run on it, and
> then merge if everything's OK
> As far as I'm concerned, I do a mix of the 3 depending on the change, my mood
> of the moment, etc...

thanks for clarification! I will choose direct push for minor changes
VFK driver probably. For rest I will use PRs. Ma

--
Martin Landa
http://geo.fsv.cvut.cz/gwiki/Landa
http://gismentors.cz/mentors/landa
_______________________________________________
gdal-dev mailing list
[hidden email]
https://lists.osgeo.org/mailman/listinfo/gdal-dev
Reply | Threaded
Open this post in threaded view
|

Re: GH repo - write access

Martin Landa
In reply to this post by Martin Landa
Hi,

2018-04-08 22:59 GMT+02:00 Dan Little <[hidden email]>:
> A good practice that I've seen many projects use is that one is not allowed
> to merge their own PRs without a review. So even a core-committer would go
> though the same steps regardless of "write access." That's particularly
> helpful when a project starts integrating CI as apart of the
> commit-acceptance practices.

OK, thanks for clarification. Ma

--
Martin Landa
http://geo.fsv.cvut.cz/gwiki/Landa
http://gismentors.cz/mentors/landa
_______________________________________________
gdal-dev mailing list
[hidden email]
https://lists.osgeo.org/mailman/listinfo/gdal-dev