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

[PATCH 2/2] revision: do not include sibling history in --ancestry-path output

From
Junio C Hamano <gitster@pobox.com>
Date
Aug 26, 2011, 01:00 UTC
Message-ID
<7v7h61gf5i.fsf_-_@alter.siamese.dyndns.org>
In-Reply-To
<438ea0b254ccafb3fc9f3431f8f86007cc03132b.1314290439.git.brad.king@kitware.com>

If the commit specified as the bottom of the commit range has a direct parent that has another child commit that contributed to the resulting history, "rev-list --ancestry-path" was confused and listed that side history as well.

             D---E
            /     \
        ---X---A---B---C

In this history, "rev-list --ancestry-path A..C" should list among what the corresponding command without --ancestry-path option would produce, namely, D, E, B and C, but limiting the result to those that are descendant of A (i.e. B and C). Due to the command line parser subtlety corrected by the previous commit, it also listed those that are descendant of X as well.

Signed-off-by: Junio C Hamano <gitster@pobox.com>
---
 * And this should fix the breakage you demonstrated.
 revision.c |   16 ++++++++++------
 1 files changed, 10 insertions(+), 6 deletions(-)
diff --git a/revision.c b/revision.c
index 3e87c86..48a2db4 100644
--- a/revision.c
+++ b/revision.c
@@ -724,12 +724,16 @@ static void limit_to_ancestry(struct commit_list *bottom, struct commit_list *li
  * to filter the result of "A..B" further to the ones that can actually
  * reach A.
  */
-static struct commit_list *collect_bottom_commits(struct commit_list *list)
+static struct commit_list *collect_bottom_commits(struct rev_info *revs)
 {
-	struct commit_list *elem, *bottom = NULL;
-	for (elem = list; elem; elem = elem->next)
-		if (elem->item->object.flags & UNINTERESTING)
-			commit_list_insert(elem->item, &bottom);
+	struct commit_list *bottom = NULL;
+	int i;
+	for (i = 0; i < revs->cmdline.nr; i++) {
+		struct rev_cmdline_entry *elem = &revs->cmdline.rev[i];
+		if ((elem->flags & UNINTERESTING) &&
+		    elem->item->type == OBJ_COMMIT)
+			commit_list_insert((struct commit *)elem->item, &bottom);
+	}
 	return bottom;
 }
 
@@ -743,7 +747,7 @@ static int limit_list(struct rev_info *revs)
 	struct commit_list *bottom = NULL;
 
 	if (revs->ancestry_path) {
-		bottom = collect_bottom_commits(list);
+		bottom = collect_bottom_commits(revs);
 		if (!bottom)
 			die("--ancestry-path given but there are no bottom commits");
 	}
-- 
1.7.6.1.385.gb7fcd0
Previous: Junio C HamanoNext: Brad King
Message 15 of 25 in “submodule: Demonstrate known breakage during recursive merge”
  1. submodule: Demonstrate known breakage during recursive mergeBrad King, Aug 24, 2011
  2. Heiko VoigtAug 24, 2011
  3. Junio C HamanoAug 24, 2011
  4. Heiko VoigtAug 24, 2011
  5. Brad KingAug 24, 2011
  6. Heiko VoigtAug 24, 2011
  7. Brad KingAug 24, 2011
  8. Heiko VoigtAug 24, 2011
  9. rev-list: Demonstrate breakage with --ancestry-path --allBrad King, Aug 25, 2011
  10. Junio C HamanoAug 25, 2011
  11. Junio C HamanoAug 25, 2011
  12. 1/2 revision: keep track of the end-user input from the command lineJunio C Hamano, Aug 26, 2011
  13. Sverre RabbelierAug 26, 2011
  14. Junio C HamanoAug 26, 2011
  15. 2/2 revision: do not include sibling history in --ancestry-path outputJunio C Hamano, Aug 26, 2011
  16. Brad KingAug 26, 2011
  17. Junio C HamanoAug 24, 2011
  18. allow multiple calls to submodule merge search for the same pathHeiko Voigt, Aug 25, 2011
  19. Junio C HamanoAug 25, 2011
  20. Junio C HamanoAug 25, 2011
  21. Junio C HamanoAug 25, 2011
  22. submodule: Demonstrate known breakage during recursive mergeBrad King, Aug 25, 2011
  23. submodule: Search for merges only at end of recursive mergeBrad King, Aug 26, 2011
  24. Junio C HamanoAug 26, 2011
  25. submodule: Search for merges only at end of recursive mergeBrad King, Aug 26, 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.