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

[PATCH v4 1/6] pull: drop confusing prefix parameter of die_on_unclean_work_tree()

From
Johannes Schindelin <johannes.schindelin@gmx.de>
Date
Oct 7, 2016, 16:08 UTC
Message-ID
<72a2d6318a16230e6bd6a1d13c52ddc42be1cbe7.1475856491.git.johannes.schindelin@gmx.de>
In-Reply-To
<cover.1475856491.git.johannes.schindelin@gmx.de>

In cmd_pull(), when verifying that there are no changes preventing a rebasing pull, we diligently pass the prefix parameter to the die_on_unclean_work_tree() function which in turn diligently passes it to the has_unstaged_changes() and has_uncommitted_changes() functions.

The casual reader might now be curious (as this developer was) whether that means that calling `git pull --rebase` in a subdirectory will ignore unstaged changes in other parts of the working directory. And be puzzled that `git pull --rebase` (correctly) complains about those changes outside of the current directory.

The puzzle is easily resolved: while we take pains to pass around the prefix and even pass it to init_revisions(), the fact that no paths are passed to init_revisions() ensures that the prefix is simply ignored.

That, combined with the fact that we will *always* want a *full* working directory check before running a rebasing pull, is reason enough to simply do away with the actual prefix parameter and to pass NULL instead, as if we were running this from the top-level working directory anyway.

Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
---
 builtin/pull.c | 16 ++++++++--------
 1 file changed, 8 insertions(+), 8 deletions(-)
diff --git a/builtin/pull.c b/builtin/pull.c
index 398aae1..d4bd635 100644
--- a/builtin/pull.c
+++ b/builtin/pull.c
@@ -328,12 +328,12 @@ static int git_pull_config(const char *var, const char *value, void *cb)
 /**
  * Returns 1 if there are unstaged changes, 0 otherwise.
  */
-static int has_unstaged_changes(const char *prefix)
+static int has_unstaged_changes(void)
 {
 	struct rev_info rev_info;
 	int result;
 
-	init_revisions(&rev_info, prefix);
+	init_revisions(&rev_info, NULL);
 	DIFF_OPT_SET(&rev_info.diffopt, IGNORE_SUBMODULES);
 	DIFF_OPT_SET(&rev_info.diffopt, QUICK);
 	diff_setup_done(&rev_info.diffopt);
@@ -344,7 +344,7 @@ static int has_unstaged_changes(const char *prefix)
 /**
  * Returns 1 if there are uncommitted changes, 0 otherwise.
  */
-static int has_uncommitted_changes(const char *prefix)
+static int has_uncommitted_changes(void)
 {
 	struct rev_info rev_info;
 	int result;
@@ -352,7 +352,7 @@ static int has_uncommitted_changes(const char *prefix)
 	if (is_cache_unborn())
 		return 0;
 
-	init_revisions(&rev_info, prefix);
+	init_revisions(&rev_info, NULL);
 	DIFF_OPT_SET(&rev_info.diffopt, IGNORE_SUBMODULES);
 	DIFF_OPT_SET(&rev_info.diffopt, QUICK);
 	add_head_to_pending(&rev_info);
@@ -365,7 +365,7 @@ static int has_uncommitted_changes(const char *prefix)
  * If the work tree has unstaged or uncommitted changes, dies with the
  * appropriate message.
  */
-static void die_on_unclean_work_tree(const char *prefix)
+static void die_on_unclean_work_tree(void)
 {
 	struct lock_file *lock_file = xcalloc(1, sizeof(*lock_file));
 	int do_die = 0;
@@ -375,12 +375,12 @@ static void die_on_unclean_work_tree(const char *prefix)
 	update_index_if_able(&the_index, lock_file);
 	rollback_lock_file(lock_file);
 
-	if (has_unstaged_changes(prefix)) {
+	if (has_unstaged_changes()) {
 		error(_("Cannot pull with rebase: You have unstaged changes."));
 		do_die = 1;
 	}
 
-	if (has_uncommitted_changes(prefix)) {
+	if (has_uncommitted_changes()) {
 		if (do_die)
 			error(_("Additionally, your index contains uncommitted changes."));
 		else
@@ -875,7 +875,7 @@ int cmd_pull(int argc, const char **argv, const char *prefix)
 			die(_("Updating an unborn branch with changes added to the index."));
 
 		if (!autostash)
-			die_on_unclean_work_tree(prefix);
+			die_on_unclean_work_tree();
 
 		if (get_rebase_fork_point(rebase_fork_point, repo, *refspecs))
 			hashclr(rebase_fork_point);
-- 
2.10.0.windows.1.325.ge6089c1
Previous: Johannes SchindelinNext: Johannes Schindelin
Message 45 of 52 in “Pull out require_clean_work_tree() functionality from builtin/pull.c”
  1. 0/6 Pull out require_clean_work_tree() functionality from builtin/pull.cJohannes Schindelin, Aug 25, 2016
  2. 1/6 pull: drop confusing prefix parameter of die_on_unclean_work_tree()Johannes Schindelin, Aug 25, 2016
  3. Junio C HamanoAug 29, 2016
  4. 3/6 Make the require_clean_work_tree() function truly reusableJohannes Schindelin, Aug 25, 2016
  5. 2/6 pull: make code more similar to the shell script againJohannes Schindelin, Aug 25, 2016
  6. Junio C HamanoAug 29, 2016
  7. Johannes SchindelinAug 30, 2016
  8. Johannes SchindelinSep 9, 2016
  9. 4/6 require_clean_work_tree: ensure that the index was readJohannes Schindelin, Aug 25, 2016
  10. Junio C HamanoAug 29, 2016
  11. Junio C HamanoAug 30, 2016
  12. Johannes SchindelinAug 30, 2016
  13. Johannes SchindelinAug 30, 2016
  14. Junio C HamanoAug 30, 2016
  15. Johannes SchindelinAug 30, 2016
  16. 5/6 Export also the has_un{staged,committed}_changed() functionsJohannes Schindelin, Aug 25, 2016
  17. Junio C HamanoAug 29, 2016
  18. 6/6 wt-status: teach has_{unstaged,uncommitted}_changes() about submodulesJohannes Schindelin, Aug 25, 2016
  19. 0/5 Pull out require_clean_work_tree() functionality from builtin/pull.cJohannes Schindelin, Sep 11, 2016
  20. 1/5 pull: drop confusing prefix parameter of die_on_unclean_work_tree()Johannes Schindelin, Sep 11, 2016
  21. 2/5 pull: make code more similar to the shell script againJohannes Schindelin, Sep 11, 2016
  22. Junio C HamanoSep 12, 2016
  23. 3/5 Make the require_clean_work_tree() function truly reusableJohannes Schindelin, Sep 11, 2016
  24. Junio C HamanoSep 12, 2016
  25. 4/5 Export also the has_un{staged,committed}_changed() functionsJohannes Schindelin, Sep 11, 2016
  26. 5/5 wt-status: teach has_{unstaged,uncommitted}_changes() about submodulesJohannes Schindelin, Sep 11, 2016
  27. Junio C HamanoSep 12, 2016
  28. 0/6 Pull out require_clean_work_tree() functionality from builtin/pull.cJohannes Schindelin, Oct 4, 2016
  29. 1/6 pull: drop confusing prefix parameter of die_on_unclean_work_tree()Johannes Schindelin, Oct 4, 2016
  30. 2/6 pull: make code more similar to the shell script againJohannes Schindelin, Oct 4, 2016
  31. 3/6 Make the require_clean_work_tree() function reusableJohannes Schindelin, Oct 4, 2016
  32. Junio C HamanoOct 4, 2016
  33. Johannes SchindelinOct 5, 2016
  34. Junio C HamanoOct 5, 2016
  35. 4/6 Export also the has_un{staged,committed}_changed() functionsJohannes Schindelin, Oct 4, 2016
  36. Junio C HamanoOct 4, 2016
  37. Jakub NarębskiOct 5, 2016
  38. 5/6 wt-status: teach has_{unstaged,uncommitted}_changes() about submodulesJohannes Schindelin, Oct 4, 2016
  39. 6/6 wt-status: begin error messages with lower-caseJohannes Schindelin, Oct 4, 2016
  40. Jakub NarębskiOct 5, 2016
  41. Johannes SchindelinOct 6, 2016
  42. Junio C HamanoOct 4, 2016
  43. Johannes SchindelinOct 5, 2016
  44. 0/6 Pull out require_clean_work_tree() functionality from builtin/pull.cJohannes Schindelin, Oct 7, 2016
  45. 1/6 pull: drop confusing prefix parameter of die_on_unclean_work_tree()Johannes Schindelin, Oct 7, 2016
  46. 2/6 pull: make code more similar to the shell script againJohannes Schindelin, Oct 7, 2016
  47. 3/6 wt-status: make the require_clean_work_tree() function reusableJohannes Schindelin, Oct 7, 2016
  48. 4/6 wt-status: export also the has_un{staged,committed}_changes() functionsJohannes Schindelin, Oct 7, 2016
  49. 5/6 wt-status: teach has_{unstaged,uncommitted}_changes() about submodulesJohannes Schindelin, Oct 7, 2016
  50. 6/6 wt-status: begin error messages with lower-caseJohannes Schindelin, Oct 7, 2016
  51. Junio C HamanoOct 7, 2016
  52. Jakub NarębskiOct 7, 2016

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.