Teach stash to parse -m/--message like commit does
PATCH v2 1/1 Teach stash to parse -m/--message like commit does
git stash push -m foo uses "foo" as the message for the stash. But git stash push -m"foo" does not parse successfully. Similarly git stash push --message="My stash message" also fails. The stash documentation doesn't suggest this syntax should work, but gitcli does and my fingers have learned this pattern long ago for commit.
Teach git stash and git store to parse -mFoo and --message=Foo the same as git commit would do. Even though it's an internal function, add similar support to create_stash() for consistency.
---
Added tests for 'stash push' and 'stash store'. Added a note that create_stash is included but unnecessary.
git-stash.sh | 18 +++++++++++ t/t3903-stash.sh | 93 ++++++++++++++++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 111 insertions(+)
git-stash.sh +18 -0
@@ -76,6 +76,12 @@ create_stash () { shift stash_msg=${1?"BUG: create_stash () -m requires an argument"} ;; + -m*) + stash_msg=${1#-m} + ;; + --message=*) + stash_msg=${1#--message=} + ;; -u|--include-untracked) shift untracked=${1?"BUG: create_stash () -u requires an argument"}
@@ -193,6 +199,12 @@ store_stash () { shift stash_msg="$1" ;; + -m*) + stash_msg=${1#-m} + ;; + --message=*) + stash_msg=${1#--message=} + ;; -q|--quiet) quiet=t ;;
@@ -251,6 +263,12 @@ push_stash () { test -z ${1+x} && usage stash_msg=$1 ;; + -m*) + stash_msg=${1#-m} + ;; + --message=*) + stash_msg=${1#--message=} + ;; --help) show_help ;;
t/t3903-stash.sh +93 -0
99 lines not on this page · raw message
Signature · 1 line
2.15.0.471.g17a719cfe.dirty
Quoted 9 lines · Phil Hord
git stash push -m foouses "foo" as the message for the stash. Butgit stash push -m"foo"does not parse successfully. Similarlygit stash push --message="My stash message"also fails. The stash documentation doesn't suggest this syntax should work, but gitcli does and my fingers have learned this pattern long ago forcommit.Teach
git stashandgit storeto parse -mFoo and --message=Foo the same asgit commitwould do. Even though it's an internal function, add similar support tocreate_stash()for consistency.
I definitely approve of the goal. The implementation looks pretty straightforward given the current parsing scheme.
Many of our other scripts lean on "rev-parse --parseopt" to handle options. E.g.:
OPTIONS="\
git foo [options]
--
m,message= stash message
"
foo() {
for i in "$@"; do echo " pre: $i"; done
eval "$(echo -n "$OPTIONS" | git rev-parse --parseopt -- "$@")"
for i in "$@"; do echo "post: $i"; done
}foo -mmsg
foo -m msg
foo --message=msg
foo --message msg
should convert each of those into "-m msg". It also handles unique partial options like "--mess", though IMHO that is not that big a deal.
Would it be possible to convert stash to use --parseopt? I'm fine if the answer is "no", or even "yes, but it's tricky so let's do this in the meantime". But I think that's the endgame we should be shooting for (or, of course, doing the whole thing in C, which I think somebody else is working on).
-Peff
Quoted 9 lines · Phil Hord
git stash push -m foouses "foo" as the message for the stash. Butgit stash push -m"foo"does not parse successfully. Similarlygit stash push --message="My stash message"also fails. The stash documentation doesn't suggest this syntax should work, but gitcli does and my fingers have learned this pattern long ago forcommit.Teach
git stashandgit storeto parse -mFoo and --message=Foo the same asgit commitwould do. Even though it's an internal function, add similar support tocreate_stash()for consistency.
I sense some typo around "git store", but I am not exactly sure what the right spelling should be. "git stash -m..", "git stash save -m..", "git stash push -m.." and "git stash store -m..", if you want a full enumeration, but stepping back a bit, mentioning "git stash" ought to be sufficient. A need to spell all of them whose handling of -m you fixed would imply there may be some others whose handling of -m is still broken, which is not a good place for us to end up with.
Quoted 1 line
Reviewd-by Junio C Hamano <[email protected]>
Heh, I didn't review this version and certainly not its test.
I didn't see anything questionable there after a quick read, though.
Thanks.