Re: Bogus error: Untracked working tree file '....' would be overwritten by merge. Aborting

[Nguyen Thai Ngoc Duy <pclouds@gmail.com>]

On Tue, Nov 16, 2010 at 11:26:19AM +0100, demerphq wrote:
> The error in the subject line is generated if one has a git repo
> checked out to a commit that adds a new file and one does something
> like:
> 
> git reset HEAD^
> 
> and then a merge operation that involves going forward onto or past HEAD.
> 
> Why is this error generated when the file is *exactly* the same as the
> file that would overwrite it?
> 
> Obviously it makes sense to throw this error when data would be lost,
> but when they are identical what is the point?

Something like this may help. Completely untested but could be a
starting point. See [1] for a more generic case, where users just want
git it ignore updating worktree and go ahead.
-- 
Duy

[1] http://thread.gmane.org/gmane.comp.version-control.git/160568/focus=160725

-- 8< --
diff --git a/unpack-trees.c b/unpack-trees.c
index 803445a..f9451fc 100644
--- a/unpack-trees.c
+++ b/unpack-trees.c
@@ -980,6 +980,12 @@ static int same(struct cache_entry *a, struct cache_entry *b)
 }
 
 
+static int identical_content(struct cache_entry *ce, struct stat *st)
+{
+	unsigned char sha1[20];
+	return !index_path(sha1, ce->name, st, 0) && !hashcmp(sha1, ce->sha1);
+}
+
 /*
  * When a CE gets turned into an unmerged entry, we
  * want it to be up-to-date
@@ -1006,6 +1012,10 @@ static int verify_uptodate_1(struct cache_entry *ce,
 		 */
 		if (S_ISGITLINK(ce->ce_mode))
 			return 0;
+
+		if (identical_content(ce, &st))
+			return 0;
+
 		errno = 0;
 	}
 	if (errno == ENOENT)
@@ -1195,6 +1205,8 @@ static int verify_absent_1(struct cache_entry *ce,
 				return 0;
 		}
 
+		if (identical_content(ce, &st))
+			return 0;
 		return o->gently ? -1 :
 			add_rejected_path(o, error_type, ce->name);
 	}
-- 8< --