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

Re: [PATCH 09/10] Introduce get_octopus_merge_bases() in commit.c

From
Johannes Schindelin <johannes.schindelin@gmx.de>
Date
Jun 6, 2008, 12:28 UTC
Message-ID
<alpine.DEB.1.00.0806061316350.1783@racer>
In-Reply-To
<7vmylzw1l4.fsf@gitster.siamese.dyndns.org>
Hi,
On Thu, 5 Jun 2008, Junio C Hamano wrote:
Show 10 quoted lines
> diff --git a/commit.c b/commit.c
> index 94d5b3d..816eed1 100644
> --- a/commit.c
> +++ b/commit.c
> @@ -531,26 +531,33 @@ static struct commit *interesting(struct commit_list *list)
>  	return NULL;
>  }
>  
> -static struct commit_list *merge_bases(struct commit *one, struct commit *two)
> +static struct commit_list *merge_bases_many(struct commit *one, int n, struct commit **twos)

Clever, but for performance reasons I think it might be better to use a commit list here. This can be a local commit_list in the helper that calls merge_bases_many() for the two-head case.

Show 26 quoted lines
>  {
>  	struct commit_list *list = NULL;
>  	struct commit_list *result = NULL;
> +	int i;
>  
> -	if (one == two)
> -		/* We do not mark this even with RESULT so we do not
> -		 * have to clean it up.
> -		 */
> -		return commit_list_insert(one, &result);
> +	for (i = 0; i < n; i++) {
> +		if (one == twos[i])
> +			/* We do not mark this even with RESULT so we do not
> +			 * have to clean it up.
> +			 */
> +			return commit_list_insert(one, &result);
> +	}
>  
>  	if (parse_commit(one))
>  		return NULL;
> -	if (parse_commit(two))
> -		return NULL;
> +	for (i = 0; i < n; i++) {
> +		if (parse_commit(twos[i]))
> +			return NULL;
> +	}
Distracting curly brackets.
Show 8 quoted lines
>  	one->object.flags |= PARENT1;
> -	two->object.flags |= PARENT2;
>  	insert_by_date(one, &list);
> -	insert_by_date(two, &list);
> +	for (i = 0; i < n; i++) {
> +		twos[i]->object.flags |= PARENT2;
> +		insert_by_date(twos[i], &list);
> +	}

Ah, now that is clever: I thought we would get into a lot of problems because we would need a bit for every initial commit. But what you are basically doing is reusing PARENT2 for all the merge bases that have been found for the heads up to, but not including, the current one.

Maybe this should be documented.
But this is not yet enough, right?  You still have to have a loop
merge_bases <- calculate merge bases from the first two heads
foreach head starting from the 3rd one
	merge_bases <- calculate merge_bases_many(this_head, merge_bases)
No?

And these merge_bases are commit_lists, therefore, as mentioned above, the signature should not take an array, but the commit_lists.

Ciao, Dscho

Previous: Junio C HamanoNext: Johannes Schindelin
Message 14 of 45 in “Build in merge”
  1. 00/10 Build in mergeMiklos Vajna, Jun 5, 2008
  2. 01/10 Move split_cmdline() to alias.cMiklos Vajna, Jun 5, 2008
  3. 02/10 Move commit_list_count() to commit.cMiklos Vajna, Jun 5, 2008
  4. 03/10 Move builtin-remote's skip_prefix() to git-compat-util.hMiklos Vajna, Jun 5, 2008
  5. 04/10 Add new test to ensure git-merge handles pull.twohead and pull.octopusMiklos Vajna, Jun 5, 2008
  6. 05/10 parseopt: add a new PARSE_OPT_ARGV0_IS_AN_OPTION optionMiklos Vajna, Jun 5, 2008
  7. 06/10 Move read_cache_unmerged() to read-cache.cMiklos Vajna, Jun 5, 2008
  8. 07/10 git-fmt-merge-msg: make it useable from other builtinsMiklos Vajna, Jun 5, 2008
  9. 08/10 Introduce commit_list_append() in commit.cMiklos Vajna, Jun 5, 2008
  10. 09/10 Introduce get_octopus_merge_bases() in commit.cMiklos Vajna, Jun 5, 2008
  11. 10/10 Build in mergeMiklos Vajna, Jun 5, 2008
  12. Junio C HamanoJun 6, 2008
  13. Junio C HamanoJun 6, 2008
  14. Johannes SchindelinJun 6, 2008
  15. Johannes SchindelinJun 6, 2008
  16. Junio C HamanoJun 6, 2008
  17. Introduce get_octopus_merge_bases() in commit.cMiklos Vajna, Jun 7, 2008
  18. Johannes SchindelinJun 9, 2008
  19. Introduce get_octopus_merge_bases() in commit.cMiklos Vajna, Jun 9, 2008
  20. Junio C HamanoJun 9, 2008
  21. Johannes SchindelinJun 9, 2008
  22. Junio C HamanoJun 9, 2008
  23. Miklos VajnaJun 9, 2008
  24. Junio C HamanoJun 9, 2008
  25. Miklos VajnaJun 9, 2008
  26. Johannes SchindelinJun 9, 2008
  27. Junio C HamanoJun 9, 2008
  28. Miklos VajnaJun 10, 2008
  29. Johannes SchindelinJun 10, 2008
  30. Johannes SchindelinJun 6, 2008
  31. Miklos VajnaJun 7, 2008
  32. Junio C HamanoJun 5, 2008
  33. Miklos VajnaJun 6, 2008
  34. Junio C HamanoJun 7, 2008
  35. Miklos VajnaJun 7, 2008
  36. Junio C HamanoJun 5, 2008
  37. Miklos VajnaJun 7, 2008
  38. Andreas EricssonJun 9, 2008
  39. git-fmt-merge-msg: make it usable from other builtinsMiklos Vajna, Jun 9, 2008
  40. Junio C HamanoJun 5, 2008
  41. Move read_cache_unmerged() to read-cache.cMiklos Vajna, Jun 7, 2008
  42. Junio C HamanoJun 5, 2008
  43. Add new test to ensure git-merge handles pull.twohead and pull.octopusMiklos Vajna, Jun 7, 2008
  44. Junio C HamanoJun 5, 2008
  45. Move parse-options's skip_prefix() to git-compat-util.hMiklos Vajna, Jun 6, 2008

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.