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

Re: [PATCH v3 13/13] mergetool: fix running in subdir when rerere enabled

From
RHRichard Hansen <hansenr@google.com>
Date
Jan 9, 2017, 23:50 UTC
Message-ID
<f0bf49ae-7c3b-4118-bc6a-2aee8f58fc03@google.com>
In-Reply-To
<xmqq37gru819.fsf@gitster.mtv.corp.google.com>
On 2017-01-09 18:29, Junio C Hamano wrote:
Show 24 quoted lines
> Junio C Hamano <gitster@pobox.com> writes:
>
>> Junio C Hamano <gitster@pobox.com> writes:
>>
>>> I wonder if it makes more sense to always move to toplevel upfront
>>> and consistently use path from the toplevel, perhaps like the patch
>>
>> s/the patch/the attached patch/ I meant.
>>
>>> does.  The first hunk is what you wrote but only inside MERGE_RR
>>> block, and the second hunk deals with converting end-user supplied
>>> paths that are relative to the original relative to the top-level.
>>>
>>> The tweaking of $orderfile you have in the first hunk may have to be
>>> tightened mimicking the way how "eval ... --sq ... ; shift" is used
>>> in the second hunk to avoid confusion in case orderfile specified by
>>> the end user happens to be the same as a valid revname
>>> (e.g. "master").
>>
>> And here is a squash-able patch to illustrate what I mean.
>
> By the way, I didn't think this through, but how is the orderfile
> that comes from the configuration file handled when it is not an
> absolute path?

Good question; it does whatever 'git diff' does, and that case isn't documented. It's also unclear if the globs are like the .gitignore patterns.

I would expect it to act like $GIT_DIR/info/exclude, but I haven't checked.
> I think it is _wrong_ to take it as relative to
> where the user started the program.
Agreed.
> The -O<file> parameter from the
> command line, when <file> is not absolute, should be taken as
> relative to where the user _thinks_ s/he is,
Agreed.
> but when it comes from
> the diff.orderfile configuration and it is not absolute, it should
> be taken as relative to the top of the working tree.
Agreed.
Show 5 quoted lines
> As we always
> cd_to_top with the suggested SQUASH, it means that the orderfile
> that came from the configuration does not have to be touched, while
> the orderfile given via -O<file> on the command line needs
> prefixing.

Yes. By unconditionally running cd_to_top we should get the behavior we want even if 'git diff' uses the current working directory rather than the top-level directory.

I'll poke at 'git diff' to see what it does.
-Richard
Previous: Junio C HamanoNext: Richard Hansen
Message 35 of 76 in “fix mergetool+rerere+subdir regression”
  1. 0/4 fix mergetool+rerere+subdir regressionRichard Hansen, Jan 4, 2017
  2. 2/4 t7610: make tests more independent and debuggableRichard Hansen, Jan 4, 2017
  3. Stefan BellerJan 4, 2017
  4. Richard HansenJan 5, 2017
  5. Richard HansenJan 5, 2017
  6. Simon RuderichJan 5, 2017
  7. Richard HansenJan 5, 2017
  8. 3/4 t7610: add test case for rerere+mergetool+subdir bugRichard Hansen, Jan 4, 2017
  9. 1/4 t7610: update branch names to match test numberRichard Hansen, Jan 4, 2017
  10. 4/4 mergetool: fix running in subdir when rerere enabledRichard Hansen, Jan 4, 2017
  11. 0/4 fix mergetool+rerere+subdir regressionRichard Hansen, Jan 6, 2017
  12. 4/4 mergetool: fix running in subdir when rerere enabledRichard Hansen, Jan 6, 2017
  13. Johannes SixtJan 6, 2017
  14. Richard HansenJan 7, 2017
  15. 3/4 t7610: add test case for rerere+mergetool+subdir bugRichard Hansen, Jan 6, 2017
  16. 1/4 t7610: update branch names to match test numberRichard Hansen, Jan 6, 2017
  17. 2/4 t7610: make tests more independent and debuggableRichard Hansen, Jan 6, 2017
  18. Stefan BellerJan 6, 2017
  19. Richard HansenJan 7, 2017
  20. 00/13 fix mergetool+rerere+subdir regressionRichard Hansen, Jan 9, 2017
  21. 01/13 .mailmap: Use my personal email address as my canonicalRichard Hansen, Jan 9, 2017
  22. 05/13 t7610: don't rely on state from previous testRichard Hansen, Jan 9, 2017
  23. 06/13 t7610: run 'git reset --hard' after each test to clean upRichard Hansen, Jan 9, 2017
  24. 03/13 t7610: Move setup code to the 'setup' test case.Richard Hansen, Jan 9, 2017
  25. 04/13 t7610: use test_when_finished for cleanup tasksRichard Hansen, Jan 9, 2017
  26. 07/13 t7610: delete some now-unnecessary 'git reset --hard' linesRichard Hansen, Jan 9, 2017
  27. 08/13 t7610: always work on a test-specific branchRichard Hansen, Jan 9, 2017
  28. 13/13 mergetool: fix running in subdir when rerere enabledRichard Hansen, Jan 9, 2017
  29. Junio C HamanoJan 9, 2017
  30. Junio C HamanoJan 9, 2017
  31. Johannes SixtJan 9, 2017
  32. Richard HansenJan 9, 2017
  33. Junio C HamanoJan 9, 2017
  34. Junio C HamanoJan 9, 2017
  35. Richard HansenJan 9, 2017
  36. 09/13 t7610: don't assume the checked-out commitRichard Hansen, Jan 9, 2017
  37. 12/13 mergetool: take the "-O" out of $orderfileRichard Hansen, Jan 9, 2017
  38. 11/13 t7610: add test case for rerere+mergetool+subdir bugRichard Hansen, Jan 9, 2017
  39. 10/13 t7610: spell 'git reset --hard' consistentlyRichard Hansen, Jan 9, 2017
  40. 02/13 t7610: update branch names to match test numberRichard Hansen, Jan 9, 2017
  41. Stefan BellerJan 9, 2017
  42. 00/14 fix mergetool+rerere+subdir regressionRichard Hansen, Jan 9, 2017
  43. 01/14 .mailmap: Use my personal email address as my canonicalRichard Hansen, Jan 9, 2017
  44. 02/14 rev-parse doc: use "--" in the --prefix exampleRichard Hansen, Jan 9, 2017
  45. 04/14 t7610: Move setup code to the 'setup' test case.Richard Hansen, Jan 9, 2017
  46. 05/14 t7610: use test_when_finished for cleanup tasksRichard Hansen, Jan 9, 2017
  47. 08/14 t7610: delete some now-unnecessary 'git reset --hard' linesRichard Hansen, Jan 9, 2017
  48. 10/14 t7610: don't assume the checked-out commitRichard Hansen, Jan 9, 2017
  49. 13/14 mergetool: take the "-O" out of $orderfileRichard Hansen, Jan 9, 2017
  50. 11/14 t7610: spell 'git reset --hard' consistentlyRichard Hansen, Jan 9, 2017
  51. 14/14 mergetool: fix running in subdir when rerere enabledRichard Hansen, Jan 9, 2017
  52. Johannes SixtJan 10, 2017
  53. Richard HansenJan 10, 2017
  54. Junio C HamanoJan 10, 2017
  55. Johannes SixtJan 10, 2017
  56. 12/14 t7610: add test case for rerere+mergetool+subdir bugRichard Hansen, Jan 9, 2017
  57. 09/14 t7610: always work on a test-specific branchRichard Hansen, Jan 9, 2017
  58. 07/14 t7610: run 'git reset --hard' after each test to clean upRichard Hansen, Jan 9, 2017
  59. 06/14 t7610: don't rely on state from previous testRichard Hansen, Jan 9, 2017
  60. 03/14 t7610: update branch names to match test numberRichard Hansen, Jan 9, 2017
  61. 00/14 fix mergetool+rerere+subdir regressionRichard Hansen, Jan 10, 2017
  62. 01/14 .mailmap: Use my personal email address as my canonicalRichard Hansen, Jan 10, 2017
  63. 09/14 t7610: always work on a test-specific branchRichard Hansen, Jan 10, 2017
  64. 11/14 t7610: spell 'git reset --hard' consistentlyRichard Hansen, Jan 10, 2017
  65. 10/14 t7610: don't assume the checked-out commitRichard Hansen, Jan 10, 2017
  66. 12/14 t7610: add test case for rerere+mergetool+subdir bugRichard Hansen, Jan 10, 2017
  67. 13/14 mergetool: take the "-O" out of $orderfileRichard Hansen, Jan 10, 2017
  68. 14/14 mergetool: fix running in subdir when rerere enabledRichard Hansen, Jan 10, 2017
  69. 08/14 t7610: delete some now-unnecessary 'git reset --hard' linesRichard Hansen, Jan 10, 2017
  70. 07/14 t7610: run 'git reset --hard' after each test to clean upRichard Hansen, Jan 10, 2017
  71. 06/14 t7610: don't rely on state from previous testRichard Hansen, Jan 10, 2017
  72. 05/14 t7610: use test_when_finished for cleanup tasksRichard Hansen, Jan 10, 2017
  73. 04/14 t7610: Move setup code to the 'setup' test case.Richard Hansen, Jan 10, 2017
  74. 03/14 t7610: update branch names to match test numberRichard Hansen, Jan 10, 2017
  75. 02/14 rev-parse doc: pass "--" to rev-parse in the --prefix exampleRichard Hansen, Jan 10, 2017
  76. David AguilarJan 10, 2017

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.