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

Re: git-am applies commit message diffs

From
Jeff King <peff@peff.net>
Date
Feb 6, 2026, 10:08 UTC
Message-ID
<20260206100837.GA2778409@coredump.intra.peff.net>
In-Reply-To
<lhujywqtd76.fsf@oldenburg.str.redhat.com>
On Fri, Feb 06, 2026 at 10:48:29AM +0100, Florian Weimer wrote:
Show 19 quoted lines
> > On Fri, Feb 06, 2026 at 09:59:31AM +0100, Florian Weimer wrote:
> >
> >> Isn't the format-patch output already unambiguous because the sequence
> >> of diffs is preceeded by the non-diff statistics section, and only then
> >> the commit message follows?  It's just not possible to process this
> >> correctly in one pass because only at the end of the input, you know
> >> that you have just seen the to-be-applied diffs.
> >
> > That diffstat is optional, and not parsed by the receiving format-patch
> > at all. Keep in mind that in the world for which it was originally
> > designed, people were not necessarily using Git to generate their
> > emails. They could be patches emailed by random folks using "diff"
> > themselves.
> 
> Is the git am format that flexible in practice?  I often have trouble
> applying patches with git am that were created with git format-patch
> and have to resort to plain old patch instead.  As a user, I definitely
> get the impression that it's not a type of tool that gets a patch
> out of an email message, no matter what the cost.

I'm sure there are corner cases it doesn't handle, but it will take input like this:

git am <<\EOF
From: Jeff King <peff@peff.net>
Date: Fri Feb 6 03:42:12 2026 -0500
Subject: my cool patch
this fixes some stuff

diff -Nru old/file new/file --- old/file 2026-02-06 04:58:56.148348259 -0500 +++ new/file 2026-02-06 04:58:59.432360938 -0500

@@ -1 +1 @@
-base
+changed
EOF


and happily produce the commit you'd expect. I generated the diff there
with GNU diff, and typed the rest. Likewise for this version with
attachments, which I generated with mutt:

git am <<\EOF
Date: Fri, 6 Feb 2026 05:03:31 -0500
From: Jeff King <peff@peff.net>
To: Jeff King <peff@peff.net>
Subject: my cool patch
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="tveeCB9LAhLXuhMJ"
Content-Disposition: inline


--tveeCB9LAhLXuhMJ
Content-Type: text/plain; charset=utf-8
Content-Disposition: inline

see the attached patch, which does blah blah blah

--tveeCB9LAhLXuhMJ
Content-Type: text/plain; charset=utf-8
Content-Disposition: attachment; filename=patch

diff -Nru old/file new/file
--- old/file	2026-02-06 04:58:56.148348259 -0500
+++ new/file	2026-02-06 04:58:59.432360938 -0500
@@ -1 +1 @@
-base
+changed

--tveeCB9LAhLXuhMJ--
EOF


I expect that Linus saw a lot of this kind of stuff in the early days.
I'd guess it's pretty rare now, but I won't be surprised if there are
some die-hards generating kernel patches with who-knows-what. ;)

-Peff
Previous: Florian WeimerNext: Kristoffer Haugsbakk
Message 42 of 65 in “git-am applies commit message diffs”
  1. Matthias BeyerFeb 6, 2026
  2. Jacob KellerFeb 6, 2026
  3. Matthias BeyerFeb 6, 2026
  4. Jeff KingFeb 6, 2026
  5. 0/3 commit-msg.sample: reject messages that would confuse "git am"Phillip Wood, Feb 7, 2026
  6. 1/3 templates: add .gitattributes entry for sample hooksPhillip Wood, Feb 7, 2026
  7. 2/3 templates: detect commit messages containing diffsPhillip Wood, Feb 7, 2026
  8. 3/3 templates: detect messages that contain a separator linePhillip Wood, Feb 7, 2026
  9. Junio C HamanoFeb 7, 2026
  10. Kristoffer HaugsbakkFeb 7, 2026
  11. Junio C HamanoFeb 9, 2026
  12. Jeff KingFeb 9, 2026
  13. Phillip WoodFeb 9, 2026
  14. Jeff KingFeb 10, 2026
  15. Jeff KingFeb 9, 2026
  16. Phillip WoodFeb 9, 2026
  17. Matthias BeyerFeb 9, 2026
  18. Jeff KingFeb 10, 2026
  19. Patrick SteinhardtFeb 9, 2026
  20. Jacob KellerFeb 10, 2026
  21. Patrick SteinhardtFeb 10, 2026
  22. Junio C HamanoFeb 10, 2026
  23. Jacob KellerFeb 11, 2026
  24. Jacob KellerFeb 11, 2026
  25. Jeff KingFeb 11, 2026
  26. Kristoffer HaugsbakkFeb 11, 2026
  27. Junio C HamanoFeb 11, 2026
  28. Jeff KingFeb 10, 2026
  29. 0/2 commit-msg.sample: reject messages that would confuse "git am"Phillip Wood, Feb 13, 2026
  30. 1/2 templates: add .gitattributes entry for sample hooksPhillip Wood, Feb 13, 2026
  31. 2/2 templates: detect commit messages containing diffsPhillip Wood, Feb 13, 2026
  32. Kristoffer HaugsbakkFeb 13, 2026
  33. Junio C HamanoFeb 13, 2026
  34. Phillip WoodFeb 14, 2026
  35. Junio C HamanoFeb 13, 2026
  36. Phillip WoodFeb 14, 2026
  37. Junio C HamanoFeb 14, 2026
  38. Junio C HamanoFeb 13, 2026
  39. Florian WeimerFeb 6, 2026
  40. Jeff KingFeb 6, 2026
  41. Florian WeimerFeb 6, 2026
  42. Jeff KingFeb 6, 2026
  43. Kristoffer HaugsbakkFeb 6, 2026
  44. Jakob HaufeFeb 6, 2026
  45. Kristoffer HaugsbakkFeb 7, 2026
  46. Kristoffer HaugsbakkFeb 7, 2026
  47. doc: add caveat about roundtripping format-patchkristofferhaugsbakk@fastmail.com, Feb 8, 2026
  48. Junio C HamanoFeb 8, 2026
  49. Kristoffer HaugsbakkFeb 8, 2026
  50. Phillip WoodFeb 9, 2026
  51. Kristoffer HaugsbakkFeb 9, 2026
  52. Phillip WoodFeb 10, 2026
  53. Kristoffer HaugsbakkFeb 10, 2026
  54. doc: add caveat about roundtripping format-patchkristofferhaugsbakk@fastmail.com, Feb 9, 2026
  55. Junio C HamanoFeb 9, 2026
  56. Kristoffer HaugsbakkFeb 9, 2026
  57. Phillip WoodFeb 10, 2026
  58. Kristoffer HaugsbakkFeb 10, 2026
  59. doc: add caveat about round-tripping format-patchkristofferhaugsbakk@fastmail.com, Feb 12, 2026
  60. Junio C HamanoFeb 12, 2026
  61. Phillip WoodFeb 13, 2026
  62. Kristoffer HaugsbakkFeb 13, 2026
  63. Junio C HamanoFeb 13, 2026
  64. Christoph Anton MittererFeb 10, 2026
  65. Kristoffer HaugsbakkFeb 10, 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.