From c033a2f62dda774de8cccc83b7f1918377995d6c Mon Sep 17 00:00:00 2001 From: Prathamesh Chavan Date: Tue, 8 May 2018 17:29:49 -0700 Subject: [PATCH] submodule foreach: correct '$path' in nested submodules from a subdirectory When running 'git submodule foreach --recursive' from a subdirectory of your repository, nested submodules get a bogus value for $path: For a submodule 'sub' that contains a nested submodule 'nested', running 'git -C dir submodule foreach echo $path' from the root of the superproject would report path='../nested' for the nested submodule. The first part '../' is derived from the logic computing the relative path from $pwd to the root of the superproject. The second part is the submodule path inside the submodule. This value is of little use and is hard to document. Also, in git-submodule.txt, $path is documented to be the "name of the submodule directory relative to the superproject", but "the superproject" is ambiguous. To resolve both these issues, we could: (a) Change "the superproject" to "its immediate superproject", so $path would be "nested" instead of "../nested". (b) Change "the superproject" to "the superproject the original command was run from", so $path would be "sub/nested" instead of "../nested". (c) Change "the superproject" to "the directory the original command was run from", so $path would be "../sub/nested" instead of "../nested". The behavior for (c) was attempted to be introduced in 091a6eb0fe (submodule: drop the top-level requirement, 2013-06-16) with the intent for $path to be relative from $pwd to the submodule worktree, but that did not work for nested submodules, as the intermittent submodules were not included in the path. If we were to fix the meaning of the $path using (a), we would break any existing submodule user that runs foreach from non-root of the superproject as the non-nested submodule '../sub' would change its path to 'sub'. If we were to fix the meaning of $path using (b), then we would break any user that uses nested submodules (even from the root directory) as the 'nested' would become 'sub/nested'. If we were to fix the meaning of $path using (c), then we would break the same users as in (b) as 'nested' would become 'sub/nested' from the root directory of the superproject. All groups can be found in the wild. The author has no data if one group outweighs the other by large margin, and offending each one seems equally bad at first. However in the authors imagination it is better to go with (a) as running from a sub directory sounds like it is carried out by a human rather than by some automation task. With a human on the keyboard the feedback loop is short and the changed behavior can be adapted to quickly unlike some automation that can break silently. Discussed-with: Ramsay Jones Signed-off-by: Prathamesh Chavan Signed-off-by: Stefan Beller Signed-off-by: Junio C Hamano Signed-off-by: Stefan Beller Signed-off-by: Junio C Hamano --- git-submodule.sh | 1 - t/t7407-submodule-foreach.sh | 36 ++++++++++++++++++++++++++++++++++-- 2 files changed, 34 insertions(+), 3 deletions(-) diff --git a/git-submodule.sh b/git-submodule.sh index 24914963ca..331d71c908 100755 --- a/git-submodule.sh +++ b/git-submodule.sh @@ -345,7 +345,6 @@ cmd_foreach() prefix="$prefix$sm_path/" sanitize_submodule_env cd "$sm_path" && - sm_path=$(git submodule--helper relative-path "$sm_path" "$wt_prefix") && # we make $path available to scripts ... path=$sm_path && if test $# -eq 1 diff --git a/t/t7407-submodule-foreach.sh b/t/t7407-submodule-foreach.sh index 6ba5daf42e..5144cc6926 100755 --- a/t/t7407-submodule-foreach.sh +++ b/t/t7407-submodule-foreach.sh @@ -82,9 +82,9 @@ test_expect_success 'test basic "submodule foreach" usage' ' cat >expect <expect <../../actual + ) && + test_i18ncmp expect actual +' cat > expect <