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

Re: [PATCH 0/3] Reject non-ff pulls by default

From
John Keeping <john@keeping.me.uk>
Date
Sep 9, 2013, 21:10 UTC
Message-ID
<20130909211052.GK2582@serenity.lan>
In-Reply-To
<20130909204415.GC14536@sigill.intra.peff.net>
On Mon, Sep 09, 2013 at 04:44:16PM -0400, Jeff King wrote:
Show 29 quoted lines
> On Mon, Sep 09, 2013 at 09:24:35PM +0100, John Keeping wrote:
> 
> > > I think that would address the concern I raised, because it does not
> > > create a roadblock to new users accomplishing their task. They can
> > > ignore the warning, or choose "merge" as the default to shut up the
> > > warning (and it is easy to choose that if you are confused, because it
> > > is what git is doing by default alongside the warning).
> > 
> > I think we need to make sure that we give instructions for how to go
> > back if the default hasn't done what you wanted.  Something like this:
> > 
> >     Your pull did not fast-forward, so Git has merged '$upstream' into
> >     your branch, which may not be correct for your project.  If you
> >     would rather rebase your changes, run
> > 
> >         git rebase
> > 
> >     See "pull.mode" in git-config(1) to suppress this message in the
> >     future.
> 
> Yes, that's a good point. I don't know if just "git rebase" is the right
> advice, though; it would depend on whether we were actually pulling from
> the upstream or not.
> 
> I wonder if we have sufficient information at the time of the warning to
> print out the actual "git rebase" invocation that would rebase as if
> they had run "pull --rebase". I think we may have to do a little
> refactoring around the base selection from the reflog (IIRC, git-pull
> does not even calculate it at all if you are not using --rebase).
We can probably do something like:
    opts=
    if git merge-base --is-ancestor "$orig_head" "$merge_head"
    then
        opts=$merge_head
    else
        opts="$orig_head --onto $merge_head"
    fi

so that "git rebase $opts" is the right thing. Most users then get the simple "git rebase $merge_head" variant.

Show 11 quoted lines
> It is also depending on "git rebase" throwing away the merge commit we
> just created. Which I think should happen always if you have not
> configured anything (though perhaps we will eventually support a pull
> mode that does "rebase -p", you would not see this warning with that
> option anyway). But another option would be to simply tell them:
> 
>   git reset --keep HEAD^
>   git pull --rebase [X...]
> 
> where "[X...]" is the arguments they gave to rebase in the first place.
> That looks a little less friendly, though.

Yeah, I think we should keep it simple if possible. In my experience people are relatively happy to run a single "make things right" command but less so if there's a sequence of steps to be performed.

Previous: Jeff KingNext: Richard Hansen
Message 38 of 84 in “Reject non-ff pulls by default”
  1. 0/3 Reject non-ff pulls by defaultFelipe Contreras, Aug 31, 2013
  2. 1/3 merge: simplify ff-only optionFelipe Contreras, Aug 31, 2013
  3. 2/3 t: replace pulls with mergesFelipe Contreras, Aug 31, 2013
  4. 3/3 pull: reject non-ff pulls by defaultFelipe Contreras, Aug 31, 2013
  5. Junio C HamanoSep 3, 2013
  6. Felipe ContrerasSep 3, 2013
  7. Junio C HamanoSep 3, 2013
  8. Felipe ContrerasSep 3, 2013
  9. John KeepingSep 4, 2013
  10. Jeff KingSep 4, 2013
  11. John KeepingSep 4, 2013
  12. Felipe ContrerasSep 8, 2013
  13. Jeff KingSep 8, 2013
  14. Felipe ContrerasSep 8, 2013
  15. Jeff KingSep 8, 2013
  16. Felipe ContrerasSep 8, 2013
  17. Jeff KingSep 8, 2013
  18. Felipe ContrerasSep 8, 2013
  19. Jeff KingSep 8, 2013
  20. Felipe ContrerasSep 8, 2013
  21. Jeff KingSep 8, 2013
  22. Felipe ContrerasSep 8, 2013
  23. Jeff KingSep 9, 2013
  24. Felipe ContrerasSep 9, 2013
  25. John KeepingSep 8, 2013
  26. Jeff KingSep 9, 2013
  27. brian m. carlsonSep 8, 2013
  28. Felipe ContrerasSep 8, 2013
  29. brian m. carlsonSep 9, 2013
  30. Felipe ContrerasSep 9, 2013
  31. Felipe ContrerasSep 9, 2013
  32. brian m. carlsonSep 9, 2013
  33. Matthieu MoySep 9, 2013
  34. Junio C HamanoSep 9, 2013
  35. Jeff KingSep 9, 2013
  36. John KeepingSep 9, 2013
  37. Jeff KingSep 9, 2013
  38. John KeepingSep 9, 2013
  39. Richard HansenSep 9, 2013
  40. Matthieu MoySep 9, 2013
  41. Jeff KingSep 9, 2013
  42. Philip OakleySep 9, 2013
  43. Felipe ContrerasSep 9, 2013
  44. John KeepingSep 10, 2013
  45. Matthieu MoySep 9, 2013
  46. Junio C HamanoSep 10, 2013
  47. Felipe ContrerasSep 9, 2013
  48. Matthieu MoySep 10, 2013
  49. Felipe ContrerasSep 11, 2013
  50. Matthieu MoySep 11, 2013
  51. Felipe ContrerasSep 13, 2013
  52. Junio C HamanoSep 4, 2013
  53. Junio C HamanoSep 4, 2013
  54. Philip OakleySep 4, 2013
  55. Junio C HamanoSep 4, 2013
  56. John KeepingSep 5, 2013
  57. Junio C HamanoSep 5, 2013
  58. John KeepingSep 5, 2013
  59. Jonathan NiederSep 6, 2013
  60. Junio C HamanoSep 6, 2013
  61. John KeepingSep 7, 2013
  62. Felipe ContrerasSep 8, 2013
  63. Felipe ContrerasSep 8, 2013
  64. Philip OakleySep 8, 2013
  65. Felipe ContrerasSep 8, 2013
  66. Philip OakleySep 8, 2013
  67. Felipe ContrerasSep 8, 2013
  68. Philip OakleySep 8, 2013
  69. Philip OakleySep 8, 2013
  70. John SzakmeisterSep 5, 2013
  71. John KeepingSep 5, 2013
  72. John SzakmeisterSep 5, 2013
  73. Richard HansenSep 5, 2013
  74. Philip OakleySep 5, 2013
  75. Junio C HamanoSep 5, 2013
  76. Junio C HamanoSep 5, 2013
  77. Felipe ContrerasSep 8, 2013
  78. Richard HansenSep 8, 2013
  79. Junio C HamanoSep 8, 2013
  80. Richard HansenSep 8, 2013
  81. Philip OakleySep 8, 2013
  82. Felipe ContrerasSep 8, 2013
  83. Ramkumar RamachandraSep 8, 2013
  84. Greg TroxelSep 5, 2013

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.