git cl push -> git cl land

git cl push -> git cl land

BUG=

Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=268835

[pgervais, 2014-04-28 23:39:36]

'git cl land' seems to be clearer than 'git cl push' what do you think?

[Dirk Pranke, 2014-04-28 23:42:34]

On 2014/04/28 23:39:36, pgervais wrote:
> 'git cl land' seems to be clearer than 'git cl push' what do you think?

"push" is the accepted git term. Diverging seems like a bad idea unless the
semantics are importantly different?

[pgervais, 2014-04-28 23:58:38]

On 2014/04/28 23:42:34, Dirk Pranke wrote:
> On 2014/04/28 23:39:36, pgervais wrote:
> > 'git cl land' seems to be clearer than 'git cl push' what do you think?
> 
> "push" is the accepted git term. Diverging seems like a bad idea unless the
> semantics are importantly different?

Bear in mind this is just an idea. 

It was the result of two long conversations between with agable@, iannucci@ and
I, triggered by me being confused by the behavior of git cl push. The surprising
behavior has been fixed now, but we (iannucci@ and me) came to the conclusion
than the interface between git and the chromium repo is rather different from a
git-to-git interface. In particular, git cl push does many different things,
which are not performed by a vanilla git push (squashing commits, rebasing,
uploading one commit at a time, etc.) Making it clear when we're talking to the
chromium repo could clarify the UI and avoid users being surprised by some
behaviors. Changing this command name makes it clear that we're not talking to a
vanilla git repo, and that special behavior is to be expected.

[szager1, 2014-04-29 00:39:41]

I'm fine with adding 'git cl land', and generally with the idea of
distinguishing this operation from a normal 'push'.  But there are already
projects using 'git cl push'.  We shouldn't break them.

[pgervais, 2014-04-29 00:44:03]

On 2014/04/29 00:39:41, szager1 wrote:
> I'm fine with adding 'git cl land', and generally with the idea of
> distinguishing this operation from a normal 'push'.  But there are already
> projects using 'git cl push'.  We shouldn't break them.

That's true, I'll add git cl land as a synonym instead of replacing it.

[pgervais, 2014-04-29 17:17:36]

On 2014/04/29 00:44:03, pgervais wrote:
> On 2014/04/29 00:39:41, szager1 wrote:
> > I'm fine with adding 'git cl land', and generally with the idea of
> > distinguishing this operation from a normal 'push'.  But there are already
> > projects using 'git cl push'.  We shouldn't break them.
> 
> That's true, I'll add git cl land as a synonym instead of replacing it.

Done. PTAL

[szager1, 2014-04-29 18:04:10]

This definitely requires a PSA to affected parties before landing.

[szager1, 2014-04-29 18:08:15]

Also, the deprecated message is a bit misleading.  What it's doing is more like:

"git cl push has been renamed to git cl land.  Currently they are treated as
synonyms, but git cl push will stop working at some point."

[pgervais, 2014-04-29 23:53:45]

Message fixed.

[szager1, 2014-04-30 06:58:53]

Sorry, I wasn't clear: the log message displayed at the terminal should give the
full explanation, not the code comment.

[pgervais, 2014-04-30 22:44:29]

Fixed the remaining issues, and added an annoying message after July 1st.

[szager1, 2014-05-01 21:11:11]

https://codereview.chromium.org/258983004/diff/70001/git_cl.py
File git_cl.py (right):

https://codereview.chromium.org/258983004/diff/70001/git_cl.py#newcode2022
git_cl.py:2022: """Temporary alias for 'land'.
One-line comment (move trailing """ up).

https://codereview.chromium.org/258983004/diff/70001/git_cl.py#newcode2033
git_cl.py:2033: 'Pausing 10 sec to help you remember :-)')
Better, I think, to hang at the terminal:

sys.stdout.write('Press Return to continue...')
sys.stdin.readline()

[pgervais, 2014-05-06 21:52:27]

On 2014/05/01 21:11:11, szager1 wrote:
> https://codereview.chromium.org/258983004/diff/70001/git_cl.py
> File git_cl.py (right):
> 
> https://codereview.chromium.org/258983004/diff/70001/git_cl.py#newcode2022
> git_cl.py:2022: """Temporary alias for 'land'.
> One-line comment (move trailing """ up).
> 
> https://codereview.chromium.org/258983004/diff/70001/git_cl.py#newcode2033
> git_cl.py:2033: 'Pausing 10 sec to help you remember :-)')
> Better, I think, to hang at the terminal:
> 
> sys.stdout.write('Press Return to continue...')
> sys.stdin.readline()

I'd rather avoid that, because it would break any automated system running 'git
cl push'. I'm not aware of any, but staying on the safe side is always good...

Beside, is this CL ready for the PSA?

[szager1, 2014-05-06 21:59:36]

lgtm

[pgervais, 2014-05-07 17:01:32]

The CQ bit was checked by pgervais@chromium.org

[commit-bot: I haz the power, 2014-05-07 17:02:16]

CQ is trying da patch. Follow status at
 https://chromium-status.appspot.com/cq/pgervais@chromium.org/258983004/70001

[commit-bot: I haz the power, 2014-05-07 17:04:04]

Change committed as 268835

[M-A Ruel, 2014-05-08 16:36:46]

Optional nit: you should use colors instead of ===. Maybe next time. :)

[iannucci, 2014-05-08 17:44:55]

On 2014/05/08 16:36:46, M-A Ruel wrote:
> Optional nit: you should use colors instead of ===. Maybe next time. :)

* in addition to... some people don't see colors so well :)

[Nico, 2014-07-16 17:53:52]

https://codereview.chromium.org/258983004/diff/70001/git_cl.py
File git_cl.py (right):

https://codereview.chromium.org/258983004/diff/70001/git_cl.py#newcode2028
git_cl.py:2028: "working after 2014/07/01.\n"
Why stop supporting the old thing, and why print this at all?