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

Re: [PATCH] sha1_name: don't resolve refs when core.warnambiguousrefs is false

From
Jeff King <peff@peff.net>
Date
Jan 7, 2014, 22:08 UTC
Message-ID
<20140107220856.GA10074@sigill.intra.peff.net>
In-Reply-To
<xmqqd2k3cz42.fsf@gitster.dls.corp.google.com>
On Tue, Jan 07, 2014 at 12:31:57PM -0800, Junio C Hamano wrote:
Show 10 quoted lines
> >   c. Just leave it at Brodie's patch with nothing else on top.
> >
> > My thinking in favor of (b) was basically "does anybody actually care
> > about ambiguous refs in this situation anyway?". If they do, then I
> > think (c) is my preferred choice.
> 
> OK.  I agree with that line of thinking.  Let's take it one step at
> a time, i.e. do c. and also use warn_on_object_refname_ambiguity in
> "rev-list --stdin" first and leave the simplification (i.e. b.) for
> later.

Here's a series to do that. The first three are just cleanups I noticed while looking at the problem.

While I was writing the commit messages, though, I had a thought. Maybe we could simply do the check faster for the common case that most refs do not look like object names? Right now we blindly call dwim_ref for each get_sha1 call, which is the expensive part. If we instead just loaded all of the refnames from the dwim_ref location (basically heads, tags and the top-level of "refs/"), we could build an index of all of the entries matching the 40-hex pattern. In 99% of cases, this would be zero entries, and the check would collapse to a simple integer comparison (and even if we did have one, it would be a simple binary search in memory).

Our index is more racy than actually checking the filesystem, but I don't think it matters here.

Anyway, here is the series I came up with, in the meantime. I can take a quick peek at just making it faster, too.

  [1/4]: cat-file: refactor error handling of batch_objects
  [2/4]: cat-file: fix a minor memory leak in batch_objects
  [3/4]: cat-file: restore ambiguity warning flag in batch_objects
  [4/4]: revision: turn off object/refname ambiguity check for --stdin
-Peff
Previous: Junio C HamanoNext: Jeff King
Message 9 of 36 in “sha1_name: don't resolve refs when core.warnambiguousrefs is false”
  1. sha1_name: don't resolve refs when core.warnambiguousrefs is falseBrodie Rao, Jan 7, 2014
  2. Brodie RaoJan 7, 2014
  3. Jeff KingJan 7, 2014
  4. Junio C HamanoJan 7, 2014
  5. Jeff KingJan 7, 2014
  6. Junio C HamanoJan 7, 2014
  7. Jeff KingJan 7, 2014
  8. Junio C HamanoJan 7, 2014
  9. Jeff KingJan 7, 2014
  10. 1/4 cat-file: refactor error handling of batch_objectsJeff King, Jan 7, 2014
  11. 2/4 cat-file: fix a minor memory leak in batch_objectsJeff King, Jan 7, 2014
  12. 3/4 cat-file: restore ambiguity warning flag in batch_objectsJeff King, Jan 7, 2014
  13. 4/4 revision: turn off object/refname ambiguity check for --stdinJeff King, Jan 7, 2014
  14. speeding up 40-hex ambiguity checkJeff King, Jan 7, 2014
  15. 1/5 cat-file: refactor error handling of batch_objectsJeff King, Jan 7, 2014
  16. 2/5 cat-file: fix a minor memory leak in batch_objectsJeff King, Jan 7, 2014
  17. 3/5 refs: teach for_each_ref a flag to avoid recursionJeff King, Jan 7, 2014
  18. 3/5 refs: teach for_each_ref a flag to avoid recursionJeff King, Jan 8, 2014
  19. Jeff KingJan 8, 2014
  20. Michael HaggertyJan 8, 2014
  21. Jeff KingJan 9, 2014
  22. Michael HaggertyJan 10, 2014
  23. Jeff KingJan 10, 2014
  24. Junio C HamanoJan 9, 2014
  25. Jeff KingJan 9, 2014
  26. 4/5 get_sha1: speed up ambiguous 40-hex testJeff King, Jan 7, 2014
  27. Michael HaggertyJan 8, 2014
  28. Junio C HamanoJan 9, 2014
  29. Jeff KingJan 10, 2014
  30. Jeff KingJan 14, 2014
  31. Michael HaggertyJan 14, 2014
  32. 5/5 get_sha1: drop object/refname ambiguity flagJeff King, Jan 8, 2014
  33. Michael HaggertyJan 8, 2014
  34. Duy NguyenJan 7, 2014
  35. Junio C HamanoJan 7, 2014
  36. sha1_name: don't resolve refs when core.warnambiguousrefs is falseBrodie Rao, Jan 7, 2014

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.