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

Re: Strange annotated tag issue

From
RDRobert Dailey <rcdailey.lists@gmail.com>
Date
Mar 25, 2019, 13:50 UTC
Message-ID
<CAHd499BTACjf91Ohi34ozFQE_NOn-LVf-35t7h4CTtDFoMCpWw@mail.gmail.com>
In-Reply-To
<20190321192928.GA19427@sigill.intra.peff.net>
On Thu, Mar 21, 2019 at 2:29 PM Jeff King <peff@peff.net> wrote:
Show 33 quoted lines
> Tags can point to any object, including another tag. It looks like
> somebody made an annotated tag of an annotated tag (probably by
> mistake, given that they have the same tag-name).
>
> Try this:
>
>   git init
>   git commit -m commit --allow-empty
>   git tag -m inner mytag HEAD
>   git tag -f -m outer mytag mytag
>
>   git show mytag
>
> which produces similar output. You can walk the chain yourself with "git
> at-file tag 4.2.0.1900". That will have a "type" and "object" field
> which presumably point to the second commit.
>
> My guess is that somebody was trying to amend the tag commit message,
> but used the tag name to create the second one, rather than the original
> commit. I.e,. any of these would have worked for the second command to
> replace the old tag:
>
>   git tag -f -m 'new message' mytag HEAD
>
>   git tag -f -m 'new message' 2fcfd00ef84572fb88852be55315914f37e91e11
>
>   git tag -f -m 'new message' mytag mytag^{commit}
>
> If the original tag isn't signed, you could rewrite it at this point
> using one of the above commands, coupled with GIT_COMMITTER_* to munge
> the author and date.  But note that fetch doesn't update modified tags
> by default, so it may just cause confusion if you have a lot of people
> using the repository.

Thanks for explaining. This is very helpful. Am I naive to think that this should be an error? I haven't seen a valid _pragmatic_ use for tags pointing to tags. In 100% of cases (including this one), it is done out of error. As per your example, users try to "correct" an annotated tag pointing at a wrong tag or commit. What they expect is the tag to point to the other tag's commit, but that's not what they get.

From a high-level, pragmatic perspective, doesn't it make more sense to change the git behavior so that annotated tags may only point to commit objects? And in the `git tag -f -m outer mytag mytag` case in your example, this would automatically perform `mytag^{}` to ensure that the behavior the user expects is the behavior they get?

Previous: Jeff KingNext: Jeff King
Message 5 of 48 in “Strange annotated tag issue”
  1. Robert DaileyMar 21, 2019
  2. Bryan TurnerMar 21, 2019
  3. Jeff KingMar 21, 2019
  4. Jeff KingMar 21, 2019
  5. Robert DaileyMar 25, 2019
  6. Jeff KingMar 25, 2019
  7. Robert DaileyMar 25, 2019
  8. Bryan TurnerMar 25, 2019
  9. Jeff KingMar 25, 2019
  10. Ævar Arnfjörð BjarmasonMar 25, 2019
  11. Jeff KingMar 25, 2019
  12. 0/3 tag: prevent recursive tagsDenton Liu, Mar 26, 2019
  13. 1/3 tag: prevent recursive tagsDenton Liu, Mar 26, 2019
  14. Denton LiuMar 26, 2019
  15. Ævar Arnfjörð BjarmasonMar 26, 2019
  16. Elijah NewrenMar 27, 2019
  17. Ævar Arnfjörð BjarmasonMar 27, 2019
  18. Robert DaileyMar 28, 2019
  19. 2/3 t7004: ensure recursive tag behavior is workingDenton Liu, Mar 26, 2019
  20. Ævar Arnfjörð BjarmasonMar 26, 2019
  21. 3/3 git-tag.txt: document --allow-recursive-tag optionDenton Liu, Mar 26, 2019
  22. Ævar Arnfjörð BjarmasonMar 26, 2019
  23. Jeff KingMar 26, 2019
  24. 0/2 tag: prevent nested tagsDenton Liu, Apr 2, 2019
  25. 1/2 tag: fix formattingDenton Liu, Apr 2, 2019
  26. 2/2 tag: prevent nested tagsDenton Liu, Apr 2, 2019
  27. 2/2 tag: prevent nested tagsDenton Liu, Apr 2, 2019
  28. Junio C HamanoApr 3, 2019
  29. Junio C HamanoApr 3, 2019
  30. Robert DaileyApr 3, 2019
  31. Junio C HamanoApr 4, 2019
  32. Robert DaileyApr 4, 2019
  33. Junio C HamanoApr 4, 2019
  34. Robert DaileyApr 5, 2019
  35. Johannes SixtApr 3, 2019
  36. Denton LiuApr 3, 2019
  37. Jeff KingApr 4, 2019
  38. Junio C HamanoApr 4, 2019
  39. Jeff KingApr 4, 2019
  40. Junio C HamanoApr 4, 2019
  41. Jeff KingApr 4, 2019
  42. Eckhard MaaßApr 11, 2019
  43. Junio C HamanoApr 12, 2019
  44. Elijah NewrenApr 5, 2019
  45. Junio C HamanoApr 5, 2019
  46. 0/2 tag: advise on recursive taggingDenton Liu, Apr 4, 2019
  47. 1/2 tag: fix formattingDenton Liu, Apr 4, 2019
  48. 2/2 tag: advise on nested tagsDenton Liu, Apr 4, 2019

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.