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

Re: [PATCH/RFC 01/12] commit-tree: allow indirect tree references

From
Mark Lodato <lodatom@gmail.com>
Date
Mar 16, 2010, 00:44 UTC
Message-ID
<ca433831003151744j80c5421n9cd1f310687dace3@mail.gmail.com>
In-Reply-To
<7v634z7uke.fsf@alter.siamese.dyndns.org>
On Sat, Mar 13, 2010 at 5:58 PM, Junio C Hamano <gitster@pobox.com> wrote:
Show 13 quoted lines
> Mark Lodato <lodatom@gmail.com> writes:
>
>> 'git commit-tree' was the last remaining built-in to accept a <tree> but
>> not a <tree-ish> (an indirect tree reference through a commit or tag
>> object.)
>
> Historically we deliberately chose not to DWIM things down to tree for
> this plumbing command to catch mistakes by Porcelain scripts
> (e.g. arguments given in wrong order---the order of commit-tree's
> arguments are, eh, unintuitive).
>
> I do not think that matters much anymore---I do not think this patch
> is unsafe.

I thought about that, but upon investigation, the _only_ command that doesn't DWIM with respect to object types (<tree> or <commit>) is commit-tree. So, I thought it might make sense to treat this like all the other commands. On the other hand, it may be a good idea to keep this special for safety. If so, then I would just amend the documentation to note that it does not automatically dereference commit or tag objects.

Previous: Junio C HamanoNext: Mark Lodato
Message 4 of 20 in “docs: use metavariables consistently”
  1. 00/12 docs: use metavariables consistentlyMark Lodato, Mar 13, 2010
  2. 01/12 commit-tree: allow indirect tree referencesMark Lodato, Mar 13, 2010
  3. Junio C HamanoMar 13, 2010
  4. Mark LodatoMar 16, 2010
  5. 02/12 grep docs: grep accepts a <tree-ish>, not a <tree>Mark Lodato, Mar 13, 2010
  6. Junio C HamanoMar 13, 2010
  7. Mark LodatoMar 16, 2010
  8. 03/12 fsck docs: remove outdated and useless diagnosticMark Lodato, Mar 13, 2010
  9. 04/12 docs: use <sha1> to mean unabbreviated IDMark Lodato, Mar 13, 2010
  10. 05/12 docs: differentiate between <tag> and <tagname>Mark Lodato, Mar 13, 2010
  11. Junio C HamanoMar 13, 2010
  12. 06/12 docs: clarify <object>, <commit>, <tree-ish>, etcMark Lodato, Mar 13, 2010
  13. 07/12 docs: use <tree> instead of <tree-ish>Mark Lodato, Mar 13, 2010
  14. 08/12 http-fetch docs: use <commit-id> consistentlyMark Lodato, Mar 13, 2010
  15. 09/12 docs: use <commit> instead of <commit-ish>Mark Lodato, Mar 13, 2010
  16. Junio C HamanoMar 13, 2010
  17. Mark LodatoMar 16, 2010
  18. 10/12 diff: use brackets for optional args in usageMark Lodato, Mar 13, 2010
  19. 11/12 docs: use ... instead of * for multiplicityMark Lodato, Mar 13, 2010
  20. 12/12 diff docs: remove <rev>{0,2} notationMark Lodato, Mar 13, 2010

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.