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

Re: [PATCH v3 5/5] fast-import: add '--signed-tags=<mode>' option

From
Christian Couder <christian.couder@gmail.com>
Date
Oct 24, 2025, 09:27 UTC
Message-ID
<CAP8UFD01-JDZisaqMUEGd7-WJ29r0eLcXuV3RjqeWNtoJ3-QmA@mail.gmail.com>
In-Reply-To
<CABPp-BGQ=3Tuik-PCerkaK4R0b1roSVLXLKs2-+E11vDrH6WaQ@mail.gmail.com>
On Fri, Oct 24, 2025 at 4:04 AM Elijah Newren <newren@gmail.com> wrote:
>
> On Mon, Oct 13, 2025 at 4:49 AM Christian Couder
> <christian.couder@gmail.com> wrote:
Show 16 quoted lines
> > +--signed-tags=(verbatim|warn-verbatim|warn-strip|strip|abort)::
> > +       Specify how to handle signed tags.  Behaves in the same way
> > +       as the same option in linkgit:git-fast-export[1], except that
> > +       default is 'verbatim' (instead of 'abort').
>
> Sorry for not catching this earlier with the --signed-commits series
> (was otherwise occupied), but this worries me.  If we ship with this
> as the default, people will come to depend upon it, and I think it's a
> bad long term default. Long term, we'd want to check if the
> signatures are valid and keep if so and do something else if not (e.g.
> re-sign or abort or strip).  Maybe verbatim is better than abort out
> of the options you've implemented so far, but I think setting the
> default now to verbatim means people start depending on it and we
> cannot change it later.  Could we change to 'abort', for both this and
> --signed-commits, before the 2.52 release, and then re-discuss once
> you have the other options implemented?

"verbatim" was already the default long before this patch series. Any tag signature was copied as-is, as part of the tag message. So it's possible that users have relied on this for a long time.

For the --signed-commits series, "verbatim" was also the default before the series. Even if importing commit signatures has been implemented more recently and even if this is marked as experimental, it's the default in Git 2.51. So regular users could already rely on it.

The --signed-commits series has been merged to 'master' and this series has recently been merged to 'next'. They aren't part of a release, but at this point I think we should send separate patches to change the default if we want to do that.

As I plan to work soon on the new modes that would check signatures and do something based on that, and as you say that it would likely be better if such a new mode becomes the default, I am reluctant to change the default mode right now, only to have to change it again hopefully in a few weeks or months. If you want to do it, then feel free to send patches changing the default though.

> This all looks good to me, other than the default as noted above.
Thanks for your review.
Previous: Elijah NewrenNext: Junio C Hamano
Message 49 of 52 in “fast-import: start controlling how tag signatures are handled”
  1. 0/5 fast-import: start controlling how tag signatures are handledChristian Couder, Oct 7, 2025
  2. 1/5 doc: git-tag: stop focussing on GPG signed tagsChristian Couder, Oct 7, 2025
  3. Patrick SteinhardtOct 8, 2025
  4. Christian CouderOct 8, 2025
  5. Patrick SteinhardtOct 8, 2025
  6. 2/5 lib-gpg: allow tests with the GPGSM prereq firstChristian Couder, Oct 7, 2025
  7. Patrick SteinhardtOct 8, 2025
  8. Christian CouderOct 8, 2025
  9. Collin FunkOct 9, 2025
  10. Todd ZullingerOct 9, 2025
  11. Christian CouderOct 9, 2025
  12. Junio C HamanoOct 9, 2025
  13. Christian CouderOct 9, 2025
  14. 3/5 t9350: properly count annotated tagsChristian Couder, Oct 7, 2025
  15. Patrick SteinhardtOct 8, 2025
  16. Christian CouderOct 8, 2025
  17. 4/5 fast-export: handle all kinds of tag signaturesChristian Couder, Oct 7, 2025
  18. Patrick SteinhardtOct 8, 2025
  19. Christian CouderOct 8, 2025
  20. Christian CouderOct 9, 2025
  21. 5/5 fast-import: add '--signed-tags=<mode>' optionChristian Couder, Oct 7, 2025
  22. Patrick SteinhardtOct 8, 2025
  23. Christian CouderOct 8, 2025
  24. Patrick SteinhardtOct 8, 2025
  25. 0/5 fast-import: start controlling how tag signatures are handledChristian Couder, Oct 9, 2025
  26. 1/5 doc: git-tag: stop focusing on GPG signed tagsChristian Couder, Oct 9, 2025
  27. Junio C HamanoOct 10, 2025
  28. Christian CouderOct 10, 2025
  29. 2/5 lib-gpg: allow tests with GPGSM or GPGSSH prereq firstChristian Couder, Oct 9, 2025
  30. Patrick SteinhardtOct 10, 2025
  31. Todd ZullingerOct 10, 2025
  32. Junio C HamanoOct 10, 2025
  33. Todd ZullingerOct 11, 2025
  34. Junio C HamanoOct 12, 2025
  35. 3/5 t9350: properly count annotated tagsChristian Couder, Oct 9, 2025
  36. 4/5 fast-export: handle all kinds of tag signaturesChristian Couder, Oct 9, 2025
  37. 5/5 fast-import: add '--signed-tags=<mode>' optionChristian Couder, Oct 9, 2025
  38. Junio C HamanoOct 9, 2025
  39. 0/5 fast-import: start controlling how tag signatures are handledChristian Couder, Oct 13, 2025
  40. 1/5 doc: git-tag: stop focusing on GPG signed tagsChristian Couder, Oct 13, 2025
  41. Elijah NewrenOct 24, 2025
  42. 2/5 lib-gpg: allow tests with GPGSM or GPGSSH prereq firstChristian Couder, Oct 13, 2025
  43. 3/5 t9350: properly count annotated tagsChristian Couder, Oct 13, 2025
  44. Elijah NewrenOct 24, 2025
  45. 4/5 fast-export: handle all kinds of tag signaturesChristian Couder, Oct 13, 2025
  46. Elijah NewrenOct 24, 2025
  47. 5/5 fast-import: add '--signed-tags=<mode>' optionChristian Couder, Oct 13, 2025
  48. Elijah NewrenOct 24, 2025
  49. Christian CouderOct 24, 2025
  50. Junio C HamanoOct 24, 2025
  51. Christian CouderOct 13, 2025
  52. Elijah NewrenOct 24, 2025

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.