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

Re: More precise tag following

From
EWEric Wong <normalperson@yhbt.net>
Date
Jan 29, 2007, 23:00 UTC
Message-ID
<20070129230050.GA15492@localdomain>
In-Reply-To
<204011cb0701271136m655815f6o1501de2bf699b362@mail.gmail.com>
Chris Lee <clee@kde.org> wrote:
Show 5 quoted lines
> Also, please note, the 3GB packed repo is only about 2/3 of the full
> KDE repo - I cut off the import at revision 409202, because that was
> when the KDE svn admins decided to move a bunch of modules from
> /trunk/ to /trunk/KDE/ and it screws up everything. So a *full* KDE
> history import would definitely be more than 4GB, packed.

Hmm.. This movement from /trunk to /trunk/KDE could be a good case for the (still-improving) --follow-parent feature in git-svn.

I've resigned to the fact that git-svn (and git) is not ideal for tracking the entire KDE repository. I don't think most (sane) developers check out the entire repository root when working on KDE, and cloning a 3GB pack just isn't realistic these days.

But if splitting the monster repository into separate repositories is an option for KDE, and I can make --follow-parent do what it's supposed to do very well: then it could be a nice way to make things more manageable for developers that will eventually switch to.

For tracking the trunk of kde-common, I'm running with my work-in-progress version of git-svn:

git-svn init -i kde-common svn://anonsvn.kde.org/home/kde/trunk/KDE/kde-common git-svn fetch -i kde-common --follow-parent

It seems to be following kde-common into pre-409202 revisions (down to r11472) pretty well. I'll upload the result to git.bogomips.org when I'm done.

You can get my latest git-svn here: git://git.bogomips.org/git-svn
Note: I've not used this to dcommit for serious work, so I can't tell
if it's really useful.  But fetching seems fine.

The current --follow-parent implementation does not handle some cases very well. I would like to get it to correctly track parents analogous to the way gitk was merged into git, so that subprojects merged into a bigger project are currently ignored.

Junio: there are some huge changes; so please don't merge it into master
yet, I don't feel it's ready yet, but it should be sometime in the next
week.  I would like to see this in 1.5.0, but only if it's not horribly
broken (especially w.r.t committing).
-- 
Eric Wong
Previous: Chris LeeNext: Eric Wong
Message 81 of 92 in “More precise tag following”
  1. Junio C HamanoJan 26, 2007
  2. Junio C HamanoJan 26, 2007
  3. Shawn O. PearceJan 27, 2007
  4. Junio C HamanoJan 27, 2007
  5. Jeff KingJan 27, 2007
  6. Nicolas PitreJan 27, 2007
  7. Simon 'corecode' SchubertJan 27, 2007
  8. Johannes SchindelinJan 27, 2007
  9. Simon 'corecode' SchubertJan 27, 2007
  10. Jakub NarebskiJan 27, 2007
  11. Linus TorvaldsJan 27, 2007
  12. Johannes SchindelinJan 27, 2007
  13. Simon 'corecode' SchubertJan 27, 2007
  14. Johannes SchindelinJan 27, 2007
  15. Simon 'corecode' SchubertJan 27, 2007
  16. Nicolas PitreJan 27, 2007
  17. Linus TorvaldsJan 27, 2007
  18. Linus TorvaldsJan 27, 2007
  19. Junio C HamanoJan 27, 2007
  20. Linus TorvaldsJan 27, 2007
  21. Junio C HamanoJan 28, 2007
  22. git-blame --porcelain: quote filename in c-style when needed.Junio C Hamano, Jan 28, 2007
  23. git-blame --incremental: don't use pagerRené Scharfe, Jan 28, 2007
  24. Junio C HamanoJan 28, 2007
  25. Junio C HamanoJan 28, 2007
  26. René ScharfeJan 29, 2007
  27. git blame --progressJunio C Hamano, Jan 29, 2007
  28. Simon 'corecode' SchubertJan 29, 2007
  29. Alex RiesenJan 29, 2007
  30. Matthias LederhoferJan 29, 2007
  31. Junio C HamanoJan 29, 2007
  32. René ScharfeJan 29, 2007
  33. Linus TorvaldsJan 29, 2007
  34. Junio C HamanoJan 30, 2007
  35. Linus TorvaldsJan 28, 2007
  36. Junio C HamanoJan 28, 2007
  37. Linus TorvaldsJan 28, 2007
  38. Junio C HamanoJan 28, 2007
  39. document 'blame --incremental'Junio C Hamano, Jan 28, 2007
  40. Junio C HamanoJan 28, 2007
  41. Jeff KingJan 28, 2007
  42. Junio C HamanoJan 30, 2007
  43. Shawn O. PearceJan 30, 2007
  44. Linus TorvaldsJan 30, 2007
  45. Junio C HamanoJan 28, 2007
  46. Shawn O. PearceJan 29, 2007
  47. Junio C HamanoJan 29, 2007
  48. Shawn O. PearceJan 29, 2007
  49. Linus TorvaldsJan 29, 2007
  50. Simon 'corecode' SchubertJan 29, 2007
  51. Theodore TsoJan 29, 2007
  52. Linus TorvaldsJan 29, 2007
  53. Jakub NarebskiJan 29, 2007
  54. Shawn O. PearceJan 29, 2007
  55. Jakub NarebskiJan 29, 2007
  56. Shawn O. PearceFeb 9, 2007
  57. David KågedalJan 31, 2007
  58. David KågedalJan 31, 2007
  59. Peter EriksenJan 31, 2007
  60. David KågedalJan 31, 2007
  61. Peter EriksenJan 31, 2007
  62. Jakub NarebskiJan 31, 2007
  63. David KågedalJan 31, 2007
  64. Simon 'corecode' SchubertJan 27, 2007
  65. Johannes SchindelinJan 27, 2007
  66. Simon 'corecode' SchubertJan 27, 2007
  67. Johannes SchindelinJan 27, 2007
  68. Jakub NarebskiJan 27, 2007
  69. Linus TorvaldsJan 27, 2007
  70. Linus TorvaldsJan 27, 2007
  71. Jakub NarebskiJan 27, 2007
  72. Linus TorvaldsJan 27, 2007
  73. Chris LeeJan 27, 2007
  74. Theodore TsoJan 28, 2007
  75. Linus TorvaldsJan 28, 2007
  76. David LangJan 28, 2007
  77. Nicolas PitreJan 29, 2007
  78. Linus TorvaldsJan 29, 2007
  79. Nicolas PitreJan 29, 2007
  80. Chris LeeJan 29, 2007
  81. Eric WongJan 29, 2007
  82. Eric WongJan 30, 2007
  83. Eric WongJan 30, 2007
  84. Eric WongJan 30, 2007
  85. Jakub NarebskiJan 27, 2007
  86. Jeff KingJan 27, 2007
  87. Linus TorvaldsJan 27, 2007
  88. Jeff KingJan 27, 2007
  89. Theodore TsoJan 28, 2007
  90. Randal L. SchwartzJan 28, 2007
  91. Jeff KingJan 28, 2007
  92. Shawn O. PearceJan 28, 2007

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.