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

Re: Git project and GSoC 2026

From
Lucas Seiki Oshiro <lucasseikioshiro@gmail.com>
Date
Jan 29, 2026, 20:41 UTC
Message-ID
<21D9FA76-F382-483E-817F-C3947C939D16@gmail.com>
In-Reply-To
<CAP8UFD29LtG2dRRB4f6mZAHNGqDmDxUV4ULYw3w3OYg15ZBBYg@mail.gmail.com>
> 6) Improve `git repo info` so it can show more information than now.
From my side, I have these features in my local backlog:
- remove the dependency on `the_repository`
- use the category as key
- add the path-related values (copied from git-rev-parse "Options for
  Files"):
  - git-dir
  - common-dir
  - toplevel
  - superproject-working-tree
- add more values currently obtained through
`git rev-parse --git-path`:
  - grafts file
  - index file
  - objects directory
  - hooks directory
  - git-prefix
  - other paths that are adjusted by update_common_dir()

I already started to add those path-related values [1], but I think that the major problem is deciding whether we should use relative or absolute paths.

I also think that we have room for other information that we retrieve through commands other than git-rev-parse.

> 7) Improve `git repo structure` so it can show more stats than now.

I don't know Justin's future plans for this command, but the idea was to bring some functionality from git-sizer [2] to Git.

> I would be willing to mentor any of them, but I don't have much
> knowledge on `git repo`, so I think it makes more sense for me to
> avoid 6) and 7).

If you want, I can share with you some information about git-repo-info.

I really appreciate initiatives like git-repo-structure and git-history that bring features from other tools that make Git easier to use. This week, I was talked independently with two friends about how git-blame can be misleading sometimes since it only shows the last change in a line. One of them really likes `git log -S` for "blaming" strings and thinks that it's a too powerful feature that is hidden inside git-log. The other one showed me Cregit [3], a tool for blaming based on tokens instead of lines. A "string blame" or a "token blame" could be a nice GSoC project (but maybe for future editions).

[1] https://github.com/lucasoshiro/git/compare/master...repo-info-path/ [2] https://github.com/github/git-sizer [3] https://github.com/cregit/cregit

Previous: Christian CouderNext: Christian Couder
Message 8 of 19 in “Git project and GSoC 2026”
  1. Christian CouderJan 16, 2026
  2. Lucas Seiki OshiroJan 19, 2026
  3. Christian CouderJan 19, 2026
  4. Kaartic SivaraamJan 22, 2026
  5. Christian CouderJan 22, 2026
  6. Kaartic SivaraamJan 28, 2026
  7. Christian CouderJan 29, 2026
  8. Lucas Seiki OshiroJan 29, 2026
  9. Christian CouderFeb 3, 2026
  10. Christian CouderFeb 3, 2026
  11. Christian CouderFeb 3, 2026
  12. Karthik NayakJan 30, 2026
  13. Christian CouderFeb 3, 2026
  14. Chandra PratapJan 22, 2026
  15. Christian CouderFeb 3, 2026
  16. Chandra PratapFeb 3, 2026
  17. Christian CouderFeb 3, 2026
  18. Kaartic SivaraamFeb 22, 2026
  19. Christian CouderJan 21, 2026

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.