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

Re: [PATCH v2 15/20] transport-helper: use the new done feature where possible

From
Jonathan Nieder <jrnieder@gmail.com>
Date
Jun 20, 2011, 11:45 UTC
Message-ID
<20110620114556.GA29818@elie>
In-Reply-To
<1308496725-22329-16-git-send-email-srabbelier@gmail.com>
Hi,
Sverre Rabbelier wrote:
> In other words, use fast-export --use-done-feature to add a 'done'
> command at the end of streams passed to remote helpers' "import"
> commands, and teach the remote helpers implementing "export" to use
> the 'done' command in turn when producing their streams.
Sounds like a reasonable thing to do.  One puzzle:
[...]
Show 21 quoted lines
> +++ b/transport-helper.c
> @@ -375,8 +375,9 @@ static int get_exporter(struct transport *transport,
>  	/* we need to duplicate helper->in because we want to use it after
>  	 * fastexport is done with it. */
>  	fastexport->out = dup(helper->in);
> -	fastexport->argv = xcalloc(4 + revlist_args->nr, sizeof(*fastexport->argv));
> +	fastexport->argv = xcalloc(5 + revlist_args->nr, sizeof(*fastexport->argv));
>  	fastexport->argv[argc++] = "fast-export";
> +	fastexport->argv[argc++] = "--use-done-feature";
>  	if (export_marks)
>  		fastexport->argv[argc++] = export_marks;
>  	if (import_marks)
> @@ -412,11 +413,8 @@ static int fetch_with_import(struct transport *transport,
>  		sendline(data, &buf);
>  		strbuf_reset(&buf);
>  	}
> -	if (disconnect_helper(transport))
> -		die("Error while disconnecting helper");
>  	if (finish_command(&fastimport))
>  		die("Error while running fast-import");
> -

What is this change about? Is the plan to allow other commands after a fetch_with_import? Sounds reasonable; I think it should be advertised in the log message, though.

When does the disconnect_helper call happen (to avoid leaks)? Ah, in release_helper; phew.

The disconnect_helper call writes the blank line that terminates the list of "import %s" commands to start the import, so there would need to be a

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