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

[PATCH] add: check return value of launch_editor

From
Jeff King <peff@peff.net>
Date
May 13, 2015, 01:21 UTC
Message-ID
<20150513012158.GA3066@peff.net>
In-Reply-To
<CAA8EjDS9HePk6ZFa0kOy7GDa9_c_TsJ7uv7t+N9RL-3z=ZrQFQ@mail.gmail.com>
On Tue, May 12, 2015 at 08:31:26PM -0400, Russ Cox wrote:
> The root cause seems to be that builtin/add.c's edit_patch does not
> check the result of  the launch_editor call. It probably should.
Yes, definitely. Patch is below.
> The call to launch_editor in builtin/config.c should probably also be
> checked, although there it's not as big a deal, provided the editor
> did not modify the file; reapplying the same file should be a no-op.
> This is different from git add -e.

This one is trickier. Certainly we could propagate the return value and say "hey, editing did not work". But we otherwise have no operation to abort. This is because "git config --edit" does not actually take our normal dot-lock, but rather edits the file in place.

IMHO this is questionable, as proper locking seems like half of the purpose of "config --edit" (the other half being that you do not have to type the location of the config file yourself). But I wonder if switching it would make people unhappy, as it would probably break editor rules to do syntax highlighting (e.g., if they are looking for "*/.gitconfig" in the path).

-- >8 --
Subject: add: check return value of launch_editor

When running "add -e", if launching the editor fails, we do not notice and continue as if the output is what the user asked for. The likely case is that the editor did not touch the contents at all, and we end up adding everything.

Reported-by: Russ Cox <rsc@golang.org>
Signed-off-by: Jeff King <peff@peff.net>
---
 builtin/add.c       | 3 ++-
 t/t3702-add-edit.sh | 7 +++++++
 2 files changed, 9 insertions(+), 1 deletion(-)
diff --git a/builtin/add.c b/builtin/add.c
index 3390933..4bd98b7 100644
--- a/builtin/add.c
+++ b/builtin/add.c
@@ -208,7 +208,8 @@ static int edit_patch(int argc, const char **argv, const char *prefix)
 	if (run_diff_files(&rev, 0))
 		die(_("Could not write patch"));
 
-	launch_editor(file, NULL, NULL);
+	if (launch_editor(file, NULL, NULL))
+		die(_("editing patch failed"));
 
 	if (stat(file, &st))
 		die_errno(_("Could not stat '%s'"), file);
diff --git a/t/t3702-add-edit.sh b/t/t3702-add-edit.sh
index 4ee47cc..3cb74ca 100755
--- a/t/t3702-add-edit.sh
+++ b/t/t3702-add-edit.sh
@@ -118,4 +118,11 @@ test_expect_success 'add -e' '
 
 '
 
+test_expect_success 'add -e notices editor failure' '
+	git reset --hard &&
+	echo change >>file &&
+	test_must_fail env GIT_EDITOR=false git add -e &&
+	test_expect_code 1 git diff --exit-code
+'
+
 test_done
-- 
2.4.0.192.g5f8138b
Previous: Russ CoxNext: Johannes Schindelin
Message 2 of 3 in “'EDITOR=false git add -e' adds to index”
  1. Russ CoxMay 13, 2015
  2. add: check return value of launch_editorJeff King, May 13, 2015
  3. Johannes SchindelinMay 13, 2015

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.