Rietveld Code Review Tool
Help | Bug tracker | Discussion group | Source code | Sign in
(113)

Issue 46120044: CG: basic cleanup

Can't Edit
Can't Publish+Mail
Start Review
Created:
10 years, 4 months ago by janek
Modified:
10 years, 3 months ago
CC:
lilypond-devel_gnu.org
Visibility:
Public.

Description

CG: rename 'commit access' to 'push access' 'commit access' is svn terminology. We're using git, and in git everyone can commit but not everyone can *push*. CG merge duplicated sections about conflicts and branches * merge 'resolving conflicts' with 'merging conflicts' * merge 'Organization of remote branches' with 'Other branches' CG: remove obsolete git info ...most of these istructions were valid for git version ~1.5; nowadays working with git is much simpler (also on Windows).

Patch Set 1 #

Total comments: 14
Unified diffs Side-by-side diffs Delta from patch set Stats (+46 lines, -637 lines) Patch
M Documentation/contributor/administration.itexi View 1 chunk +1 line, -1 line 2 comments Download
M Documentation/contributor/source-code.itexi View 18 chunks +45 lines, -636 lines 12 comments Download

Messages

Total messages: 9
Keith
looks fine https://codereview.appspot.com/46120044/diff/1/Documentation/contributor/source-code.itexi File Documentation/contributor/source-code.itexi (left): https://codereview.appspot.com/46120044/diff/1/Documentation/contributor/source-code.itexi#oldcode729 Documentation/contributor/source-code.itexi:729: @end example I have often looked in ...
10 years, 4 months ago (2014-01-01 05:44:44 UTC) #1
Graham Percival
LGTM https://codereview.appspot.com/46120044/diff/1/Documentation/contributor/source-code.itexi File Documentation/contributor/source-code.itexi (left): https://codereview.appspot.com/46120044/diff/1/Documentation/contributor/source-code.itexi#oldcode729 Documentation/contributor/source-code.itexi:729: @end example On 2014/01/01 05:44:44, Keith wrote: > ...
10 years, 4 months ago (2014-01-01 06:08:09 UTC) #2
Trevor Daniels
Fine, apart from throwing away the Windows section. https://codereview.appspot.com/46120044/diff/1/Documentation/contributor/source-code.itexi File Documentation/contributor/source-code.itexi (left): https://codereview.appspot.com/46120044/diff/1/Documentation/contributor/source-code.itexi#oldcode2120 Documentation/contributor/source-code.itexi:2120: @section ...
10 years, 4 months ago (2014-01-01 10:31:48 UTC) #3
Keith
https://codereview.appspot.com/46120044/diff/1/Documentation/contributor/source-code.itexi File Documentation/contributor/source-code.itexi (left): https://codereview.appspot.com/46120044/diff/1/Documentation/contributor/source-code.itexi#oldcode729 Documentation/contributor/source-code.itexi:729: @end example On 2014/01/01 06:08:09, Graham Percival wrote: > ...
10 years, 4 months ago (2014-01-01 20:45:45 UTC) #4
benko.pal
2014/1/1 <k-ohara5a5a@oco.net>: > > Janek, when people start with `git clone` do they also get ...
10 years, 4 months ago (2014-01-01 21:16:14 UTC) #5
dak
Benkő Pál <benko.pal@gmail.com> writes: > 2014/1/1 <k-ohara5a5a@oco.net>: >> >> Janek, when people start with `git ...
10 years, 4 months ago (2014-01-01 21:35:03 UTC) #6
Keith
Good, except we want to keep the information that Git works usefully on Windows. https://codereview.appspot.com/46120044/diff/1/Documentation/contributor/source-code.itexi ...
10 years, 4 months ago (2014-01-01 22:34:11 UTC) #7
pkx166h
On 01/01/14 21:35, David Kastrup wrote: > Benkő Pál <benko.pal@gmail.com> writes: > >> 2014/1/1 <k-ohara5a5a@oco.net>: ...
10 years, 3 months ago (2014-01-02 11:49:10 UTC) #8
dak
10 years, 3 months ago (2014-01-05 20:06:00 UTC) #9
https://codereview.appspot.com/46120044/diff/1/Documentation/contributor/admi...
File Documentation/contributor/administration.itexi (right):

https://codereview.appspot.com/46120044/diff/1/Documentation/contributor/admi...
Documentation/contributor/administration.itexi:219: instructions in the CG at
@ref{Push access}.  Do not set
It's not just about pushing.  The Savannah terminology on the general Git pages
and the project specific pages is "project member access" or "member access" in
contrast to "anonymous access".

https://codereview.appspot.com/46120044/diff/1/Documentation/contributor/admi...
Documentation/contributor/administration.itexi:1574: @subsubheading Push git
access
This would seem to warrant changing as well.

https://codereview.appspot.com/46120044/diff/1/Documentation/contributor/sour...
File Documentation/contributor/source-code.itexi (right):

https://codereview.appspot.com/46120044/diff/1/Documentation/contributor/sour...
Documentation/contributor/source-code.itexi:123: command-line version of Git 1.7
or higher.}
On 2014/01/01 06:08:09, Graham Percival wrote:
> just curious, what changed in git 1.5 vs. 1.7 that's important to these
> instructions?

Possibly the default behavior for
git push
without arguments.  Also possible git clone behavior.  At any rate, LilyDev
seems to be 1.7, so we probably are not all that informed about 1.5 any more.

https://codereview.appspot.com/46120044/diff/1/Documentation/contributor/sour...
Documentation/contributor/source-code.itexi:654: The branches are kept for
archival reasons.
That's not actually correct.  It's more like "These branches are for making
stable releases.  They are only to be changed by their respective maintainers."

https://codereview.appspot.com/46120044/diff/1/Documentation/contributor/sour...
Documentation/contributor/source-code.itexi:1554: @node Push access
Member access.

https://codereview.appspot.com/46120044/diff/1/Documentation/contributor/sour...
Documentation/contributor/source-code.itexi:1745: git config push.default
matching
That's actually the worst possible setting in my experience since it tends to
push stuff you did not intend to push.  The Git documentation explicitly says
this is unsuitable for shared repositories:

           ·   matching - push all branches having the same name in both ends.
               This is for those who prepare all the branches into a
               publishable shape and then push them out with a single command.
               It is not appropriate for pushing into a repository shared by
               multiple users, since locally stalled branches will attempt a
               non-fast forward push if other users updated the branch.

               This is currently the default, but Git 2.0 will change the
               default to simple.

I think we should rather recommend "simple" here.  It's always possible to give
explicit instructions, so it is not like the user will be blocked.

"upstream" has the disadvantage that it will push feature branches straight to
master.  So "simple" is best.
Sign in to reply to this message.

Powered by Google App Engine
RSS Feeds Recent Issues | This issue
This is Rietveld f62528b