On Thu, 21 Aug 2025 at 15:51, Patrick Steinhardt <ps@xxxxxx> wrote: > > On Thu, Aug 21, 2025 at 02:22:46PM +0530, Meet Soni wrote: > > The new `git refs exists` subcommand must have identical behavior to its > > predecessor, `git show-ref --exists`. To avoid duplicating the entire > > test suite, refactor the existing tests into a shareable helper script. > > > > Extract the tests for the `--exists` flag from `t1403-show-ref.sh` into > > a new `show-ref-exists-tests.sh` helper. The command under test is > > parameterized using the `$git_show_ref_exists` variable. > > > > Source new helper to both `t1403-show-ref.sh` and the new test file, > > s/new/this/, otherwise this reads grammatically wrong to me. > > > test_done > > diff --git a/t/t1462-refs-exists.sh b/t/t1462-refs-exists.sh > > new file mode 100755 > > index 0000000000..c00d76cc9e > > --- /dev/null > > +++ b/t/t1462-refs-exists.sh > > @@ -0,0 +1,22 @@ > > +#!/bin/sh > > + > > +test_description='refs exists' > > +GIT_TEST_DEFAULT_INITIAL_BRANCH_NAME=main > > +export GIT_TEST_DEFAULT_INITIAL_BRANCH_NAME > > + > > +. ./test-lib.sh > > + > > +git_show_ref_exists='git refs exists' > > + > > +test_expect_success setup ' > > + test_commit --annotate A && > > + git checkout -b side && > > + test_commit --annotate B && > > + git checkout main && > > + test_commit C && > > + git branch B A^0 > > +' > > It's a bit weird that this setup needs to be replicated. I guess it > comes from the fact that t1403 also has a bunch of other tests for > git-show-ref(1) that are irrelevant to existence checks. > > How about we instead split out the existence-checks in t1403 into a > separate test suite first and then pull out the whole logic from it in a > subsequent commit so that we can also share the setup? > Thanks for the review. You're right, splitting the test commit into two is cleaner. I'll make that change and send out a v2 shortly. > Patrick Thanks, Meet.
