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

Re: [PATCH 10/19] fast-import: introduce 'done' command

From
Jonathan Nieder <jrnieder@gmail.com>
Date
Jun 8, 2011, 20:03 UTC
Message-ID
<20110608200345.GJ27715@elie>
In-Reply-To
<1307558930-16074-11-git-send-email-srabbelier@gmail.com>
Sverre Rabbelier wrote:
Show 6 quoted lines
> Add a 'done' command that causes fast-import to stop reading from the
> stream and exit.
>
> If the new --done command line flag was passed on the command line
> (or a "feature done" declaration included at the start of the stream),
> make the 'done' command mandatory.
Hoorah!  Thanks for reviving this.
Show 6 quoted lines
> Another possible application of the 'done' command would to be allow a
> fast-import stream that is only a small part of a larger encapsulating
> stream to be easily parsed, leaving the file offset after the "done\n"
> so the other application can pick up from there.  This patch does not
> teach fast-import to do that --- fast-import still uses buffered input
> (stdio).
Possible approaches, two ugly and one sane:
 A) use setvbuf() to make input line-buffered.  Unfortunately, as
    POSIX tells us:
	Applications should note that many implementations only provide
	line buffering on input from terminal devices.
    So making it actually work involves pseudo-tty craziness.  Not a
    good idea.
 B) leave some room for buffering after the "done" command and make
    fast-import use setvbuf() to control the size of the stdio buffer.
    This is disgusting but it probably works.
 C) make sure the upstream of the pipe does not write anything after
    "done\n" until the "done\n" command has been read.  In other
    words:
	transport-helper>	done
	remote-helper>   	yep, I'm done
	transport-helper>	ah, now that you mention that, here are
	                 	a few more things you could do...

As long as there is nothing after the "done\n" in the pipe buffer when fast-import reads "done\n", it won't read too far.

This is all made complicated to debug by stdio's habit of closing streams at exit time, flushing streams at close time, and repositioning the file offset of seekable input (but not pipes) at flush time.

Hope that helps.
Previous: Sverre RabbelierNext: Sverre Rabbelier
Message 33 of 54 in “remote-helper improvements”
  1. 00/19 remote-helper improvementsSverre Rabbelier, Jun 8, 2011
  2. 01/19 transport-helper: fix minor leak in push_refs_with_exportSverre Rabbelier, Jun 8, 2011
  3. Jeff KingJun 8, 2011
  4. Sverre RabbelierJun 8, 2011
  5. 02/19 t5800: factor out some ref testsSverre Rabbelier, Jun 8, 2011
  6. 03/19 t5800: document some non-functional parts of remote helpersSverre Rabbelier, Jun 8, 2011
  7. Jonathan NiederJun 8, 2011
  8. Jonathan NiederJun 8, 2011
  9. Sverre RabbelierJun 8, 2011
  10. Sverre RabbelierJun 8, 2011
  11. Sverre RabbelierJun 9, 2011
  12. Jeff KingJun 10, 2011
  13. 04/19 teach remote-testgit to import non-HEAD refsSverre Rabbelier, Jun 8, 2011
  14. Jonathan NiederJun 8, 2011
  15. Sverre RabbelierJun 8, 2011
  16. Jeff KingJun 8, 2011
  17. Junio C HamanoJun 8, 2011
  18. Sverre RabbelierJun 9, 2011
  19. 05/19 transport-helper: don't feed bogus refs to export pushSverre Rabbelier, Jun 8, 2011
  20. 06/19 git_remote_helpers: push all refs during a non-local exportSverre Rabbelier, Jun 8, 2011
  21. Jonathan NiederJun 8, 2011
  22. Jeff KingJun 8, 2011
  23. Sverre RabbelierJun 8, 2011
  24. Jonathan NiederJun 9, 2011
  25. Sverre RabbelierJun 9, 2011
  26. Jonathan NiederJun 9, 2011
  27. Sverre RabbelierJun 9, 2011
  28. Jeff KingJun 10, 2011
  29. 07/19 remote-curl: accept empty line as terminatorSverre Rabbelier, Jun 8, 2011
  30. 08/19 git-remote-testgit: only push for non-local repositoriesSverre Rabbelier, Jun 8, 2011
  31. 09/19 git-remote-testgit: fix error handlingSverre Rabbelier, Jun 8, 2011
  32. 10/19 fast-import: introduce 'done' commandSverre Rabbelier, Jun 8, 2011
  33. Jonathan NiederJun 8, 2011
  34. Sverre RabbelierJun 8, 2011
  35. 11/19 fast-export: support done featureSverre Rabbelier, Jun 8, 2011
  36. 12/19 transport-helper: factor out push_update_refs_statusSverre Rabbelier, Jun 8, 2011
  37. 13/19 transport-helper: check status code of finish_commandSverre Rabbelier, Jun 8, 2011
  38. 14/19 transport-helper: use the new done feature where possibleSverre Rabbelier, Jun 8, 2011
  39. 15/19 transport-helper: update ref status after push with exportSverre Rabbelier, Jun 8, 2011
  40. Jonathan NiederJun 9, 2011
  41. Sverre RabbelierJun 9, 2011
  42. 16/19 transport-helper: change import semanticsSverre Rabbelier, Jun 8, 2011
  43. Jonathan NiederJun 8, 2011
  44. Sverre RabbelierJun 8, 2011
  45. 17/19 transport-helper: export is no longer always the last commandSverre Rabbelier, Jun 8, 2011
  46. Junio C HamanoJun 9, 2011
  47. Sverre RabbelierJun 9, 2011
  48. Jonathan NiederJun 9, 2011
  49. Sverre RabbelierJun 9, 2011
  50. Junio C HamanoJun 13, 2011
  51. 18/19 transport-helper: Use capname for gitdir capability tooSverre Rabbelier, Jun 8, 2011
  52. Jonathan NiederJun 8, 2011
  53. Sverre RabbelierJun 8, 2011
  54. 19/19 transport-helper: implement marks location as capabilitySverre Rabbelier, Jun 8, 2011

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.