git/list[1] front-page[2] threads[3] people[4] search[5] about
 

Re: Strangeness with git-add and nested repositories

From
Stefan Beller <sbeller@google.com>
Date
Apr 28, 2016, 16:35 UTC
Message-ID
<CAGZ79ka9v0v4gwtbiojR5o89JWLDZYkDM2CmhySeXXKYaiLQig@mail.gmail.com>
In-Reply-To
<CAH6n4TfvybZyAYLkzyjfaP=2ZkirpacTgVQApF3bZT-j8=_Qng@mail.gmail.com>
On Wed, Apr 27, 2016 at 11:10 PM, Andrew J <andj2223@gmail.com> wrote:
Show 16 quoted lines
> Hi Stefan,
>
> On Wed, Apr 27, 2016 at 9:08 AM, Stefan Beller <sbeller@google.com> wrote:
>> I think (pure speculation), that it the error is in the context
>> (repository) switching logic.
>> What happens if you alter the order, i.e. give testfile first and then
>> the files in the nested
>> repos?
>
> Interestingly, reversing the order produces the same result (the
> testfile is added, the nested files are not).
>
> I've also noticed that running something like 'git status testfile
> nestedfiles' results in the nested files being omitted from the git
> status output; I'd expect them to be printed by git-status as
> untracked files. So it appears the problem is not isolated to git-add.

Looking at the code, git-add uses parse_pathspec, which is used my most commands[https://github.com/git/git/blob/master/builtin/add.c#L364] and I think that function needs to learn to cover paths for different repositories. Maybe we can pass in a flag, which allows parse_pathspec to get files from the different repos and then it's up to each command how to deal with that list of files.

Show 21 quoted lines
>
> To give some context, my use case is that I have a parent project that
> links to numerous chromium libraries, thus my parent project needs
> access to many of chromium's headers at build time. I wanted to make
> these headers available to other developers without them needing to
> check out all of chromium.
> So I add all the chromium headers to the parent project with something like:
> find deps/chromium/src -name "*.h" | xargs git add --
> I was weirded out to find that many of the header files weren't being
> added, as I've already described.
>
> I ultimately worked around this by doing:
> find chromium/src -name "*.h" | xargs -n 1 git add
> Since each file gets added separately, this is quite slow. So it'd be
> nice if this little bug was fixed someday :)
>
> As you probably know, Chromium is comprised of many hundreds of nested
> repos, so that aided in manifesting this issue.
>
> Thanks,
> Andrew
Previous: Andrew JNext: Junio C Hamano
Message 4 of 10 in “Strangeness with git-add and nested repositories”
  1. Andrew JApr 27, 2016
  2. Stefan BellerApr 27, 2016
  3. Andrew JApr 28, 2016
  4. Stefan BellerApr 28, 2016
  5. Junio C HamanoApr 28, 2016
  6. Junio C HamanoApr 28, 2016
  7. Andrew JApr 29, 2016
  8. Stefan BellerApr 28, 2016
  9. Junio C HamanoApr 29, 2016
  10. Junio C HamanoApr 29, 2016

Read the whole thread, see it on lore, or plain text.

$ cat FOOTERMessages come from the public archive at lore.kernel.org/git, fetched every hour. The front page is chosen and written each morning by an AI editor and can be wrong; the threads themselves are the record. About and API. For agents: an MCP server at https://gitlist.dev/mcp, and any thread, story or person page as Markdown by adding .md to its URL (or sending Accept: text/markdown). Details in /llms.txt.