[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: Elpa packages and Makefiles
From: |
Ted Zlatanov |
Subject: |
Re: Elpa packages and Makefiles |
Date: |
Thu, 09 Oct 2014 11:40:32 -0400 |
User-agent: |
Gnus/5.130012 (Ma Gnus v0.12) Emacs/25.0.50 (darwin) |
On Thu, 09 Oct 2014 10:40:09 -0400 Stefan Monnier <monnier@iro.umontreal.ca>
wrote:
SM> The reason I prefer doing that is because I think I know what "git
SM> merge" does, whereas I have difficulty understanding what "git
SM> read-tree" or "git subtree" does.
>> `git subtree' is nice when you want subsequent resyncs of the same tree.
>> It can maintain a bijective commit ID mapping between the source repo
>> and the destination repo.
SM> Subtree merges work fine as well for subsequent resyncs of the same
SM> tree, and they don't need to maintain a bijective mapping between the
SM> two (because this mapping is the identity mapping).
A merge is not the same thing as a bijective `git subtree' mapping which
can map paths for each commit. In the ELPA case at least, you're better
off your way. I was just trying to explain what that tool does because
it's been useful to me in the past.
Ted