Re: What's cooking in git.git (Dec 2017, #05; Wed, 27)

[Kaartic Sivaraam]

On Wednesday 03 January 2018 02:56 PM, Daniel Knittl-Frank wrote:

On Wed, Dec 27, 2017 at 10:34 PM, Junio C Hamano  wrote:

* dk/describe-all-output-fix (2017-12-27) 1 commit
  - describe: prepend "tags/" when describing tags with embedded name

  An old regression in "git describe --all $annotated_tag^0" has been
  fixed.

  Will merge to 'next'.


Shouldn't this be merged to 'maint' since it is a bugfix (for a long
standing bug)? Or am I misinterpreting the meaning of the 'maint'
branch?



IIUC, the pipe line here is something like,

  [PATCH]

 |

 (after the patch gets some consensus
  when it's not trivial)

 |
 V

   Merge to 'pu'


 |

 (after waiting for some time to see if someone
  shouts about a build failing or complaining
   a regression about the PATCH in 'pu')

 |
 V

   Merge to 'next'

 |

 (after waiting for some time to see if someone
  shouts about a build failing or complaining
   a regression about the PATCH in 'next')

 |
/ \
   /   \
  / \
 /   \
| |

  (if it's a bugfix for  (if it's a new feature or
   an already releasedan enhancement)
   version of Git)

| |
V V
  Merge to 'maint'Merge to 'master'



Of course 'maint' and 'master' are not diverged completely. They are 
'synced' at times.


Disclaimer: I won't say I'm 100% correct with the pipeline. This is just 
what I've understood in observing the mailing list, the "What's cooking" 
emails and the history of 'git' in the short time that I've been here. 
So, there are possibilities that I've said something incorrectly. I 
guess the "Documentation/howto/maintain-git.txt" document covers it more 
comprehensively especially "The Policy" section describes the branches 
more clearly.


--
Kaartic

Quote: "Be creative. Be adventurous. Be original. And above all else, be 
young." - Wonder Woman

Re: What's cooking in git.git (Dec 2017, #05; Wed, 27)

[Daniel Knittl-Frank]

On Wed, Dec 27, 2017 at 10:34 PM, Junio C Hamano  wrote:
> * dk/describe-all-output-fix (2017-12-27) 1 commit
>  - describe: prepend "tags/" when describing tags with embedded name
>
>  An old regression in "git describe --all $annotated_tag^0" has been
>  fixed.
>
>  Will merge to 'next'.

Shouldn't this be merged to 'maint' since it is a bugfix (for a long
standing bug)? Or am I misinterpreting the meaning of the 'maint'
branch?

Daniel

--
typed with http://neo-layout.org

Re: What's cooking in git.git (Dec 2017, #05; Wed, 27)

[Jonathan Nieder]

Stefan Beller wrote:
> On Thu, Dec 28, 2017 at 11:02 AM, Junio C Hamano  wrote:
>> Elijah Newren  writes:

>>> surprised by the branch name, though.  Was 'ew/' a typo,
>>
>> Blush X-<.  Yes it is a typo.
>
> Note on that series:
> I have reviewed the first three patches (which could form an independent 
> series)
> that it would warrant a Reviewed-By: Stefan Beller 
>
> While I reviewed the earlier versions of the later patches, I would
> prefer if there is another reviewer for these as it seems like a bigger
> contribution at a core functionality.
>
> I cc'd some people who were active in some form of rename detection
> work earlier; could you review this series, please?

I'm missing context about which series you mean (though I think I can
guess).  Do you mind resending this request-for-review in a reply to
the patch thread?

That way, the request would be in context where my mail reader can
bring up the thread and it's easy for others to see that the request
happened, all in one place.

Thanks,
Jonathan

Re: What's cooking in git.git (Dec 2017, #05; Wed, 27)

[Stefan Beller]

On Thu, Dec 28, 2017 at 11:02 AM, Junio C Hamano  wrote:
> Elijah Newren  writes:
>
>> surprised by the branch name, though.  Was 'ew/' a typo,
>
> Blush X-<.  Yes it is a typo.

Note on that series:
I have reviewed the first three patches (which could form an independent series)
that it would warrant a Reviewed-By: Stefan Beller 

While I reviewed the earlier versions of the later patches, I would
prefer if there is another reviewer for these as it seems like a bigger
contribution at a core functionality.

I cc'd some people who were active in some form of rename detection
work earlier; could you review this series, please?

Thanks,
Stefan

Re: What's cooking in git.git (Dec 2017, #05; Wed, 27)

[Junio C Hamano]

Elijah Newren  writes:

> surprised by the branch name, though.  Was 'ew/' a typo,

Blush X-<.  Yes it is a typo.

Re: What's cooking in git.git (Dec 2017, #05; Wed, 27)

[Elijah Newren]

On Wed, Dec 27, 2017 at 1:34 PM, Junio C Hamano  wrote:

> * ew/empty-merge-with-dirty-index (2017-12-22) 1 commit
>  - Merge branch 'ew/empty-merge-with-dirty-index-maint' into 
> ew/empty-merge-with-dirty-index
>  (this branch uses ew/empty-merge-with-dirty-index-maint.)

> * ew/empty-merge-with-dirty-index-maint (2017-12-22) 3 commits
>  - merge-recursive: avoid incorporating uncommitted changes in a merge
>  - move index_has_changes() from builtin/am.c to merge.c for reuse
>  - t6044: recursive can silently incorporate dirty changes in a merge
>  (this branch is used by ew/empty-merge-with-dirty-index.)
>
>  "git merge -s recursive" did not correctly abort when the index is
>  dirty, if the merged tree happened to be the same as the current
>  HEAD, which has been fixed.

As promised, I looked through both to check for mis-merges or problems
in applying to maint.  The changes all look good to me.  I was
surprised by the branch name, though.  Was 'ew/' a typo, or does that
part of the branch name mean something other than what I always
assumed?