git.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: "Karl Hasselström" <kha@treskal.com>
To: Catalin Marinas <catalin.marinas@gmail.com>
Cc: git@vger.kernel.org, Junio C Hamano <gitster@pobox.com>
Subject: [StGit PATCH] Fix problems in t1000-branch-create
Date: Sun, 13 Apr 2008 15:43:53 +0200	[thread overview]
Message-ID: <20080413134104.15175.84334.stgit@yoghurt> (raw)
In-Reply-To: <20080413060223.GA27382@diana.vm.bytemark.co.uk>

There were a number of problems with t1000-branch-create:

  * It assumed that refs were not packed. I added calls to show-ref to
    spot any packed refs.

  * It reused the same name for several consecutive branch creation
    tests. This causes a domino effect if one of the subtests fail. I
    made the subtests use different names.

  * It tried to make sure that we couldn't create a branch if there
    was a broken ref by the same name. Unfortunately that test fails
    -- we don't create the branch, but we point HEAD to it. This
    causes many subsequent operations to fail. I marked this test as a
    known failure, and added some reset code so that subsequent
    subtests don't die because of this.

Signed-off-by: Karl Hasselström <kha@treskal.com>

---

Thanks to Junio for suggesting that the test should be repaired rather
than deleted. It finds a real bug! (A rather harmless one, but still.)

 t/t1000-branch-create.sh |   56 ++++++++++++++++++++++++++++++----------------
 1 files changed, 36 insertions(+), 20 deletions(-)


diff --git a/t/t1000-branch-create.sh b/t/t1000-branch-create.sh
index d6cf34a..5a097a4 100755
--- a/t/t1000-branch-create.sh
+++ b/t/t1000-branch-create.sh
@@ -21,65 +21,81 @@ test_expect_success \
     'Create a spurious patches/ entry' '
     stg branch master &&
     stg init &&
-    find .git -name foo | xargs rm -rf &&
-    mkdir -p .git/patches && touch .git/patches/foo
+    mkdir -p .git/patches && touch .git/patches/foo1
 '
 
 test_expect_success \
     'Try to create an stgit branch with a spurious patches/ entry' '
-    ! stg branch -c foo
+    ! stg branch -c foo1
 '
 
 test_expect_success \
     'Check that no part of the branch was created' '
-    test "`find .git -name foo | tee /dev/stderr`" = ".git/patches/foo" &&
-    ( grep foo .git/HEAD; test $? = 1 )
+    test "$(find .git -name foo1 | tee /dev/stderr)" = ".git/patches/foo1" &&
+    test "$(git show-ref | grep foo1 | wc -l)" = 0 &&
+    test "$(git symbolic-ref HEAD)" = "refs/heads/master"
 '
 
 test_expect_success \
     'Create a git branch' '
-    find .git -name foo | xargs rm -rf &&
-    cp .git/refs/heads/master .git/refs/heads/foo
+    git update-ref refs/heads/foo2 refs/heads/master
 '
 
 test_expect_success \
     'Try to create an stgit branch with an existing git branch by that name' '
-    ! stg branch -c foo
+    ! stg branch -c foo2
 '
 
 test_expect_success \
     'Check that no part of the branch was created' '
-    test "`find .git -name foo | tee /dev/stderr`" = ".git/refs/heads/foo" &&
-    ( grep foo .git/HEAD; test $? = 1 )
+    test "$(find .git -name foo2 | tee /dev/stderr \
+        | grep -v ^\\.git/refs/heads/foo2$ \
+        | grep -v ^\\.git/logs/refs/heads/foo2$ | wc -l)" = 0 &&
+    test "$(git show-ref | grep foo2 | wc -l)" = 1 &&
+    test "$(git symbolic-ref HEAD)" = "refs/heads/master"
 '
 
 test_expect_success \
     'Create an invalid refs/heads/ entry' '
-    find .git -name foo | xargs rm -rf &&
-    touch .git/refs/heads/foo &&
-    ! stg branch -c foo
+    touch .git/refs/heads/foo3 &&
+    ! stg branch -c foo3
 '
 
+test_expect_failure \
+    'Check that no part of the branch was created' '
+    test "$(find .git -name foo3 | tee /dev/stderr \
+        | grep -v ^\\.git/refs/heads/foo3$ | wc -l)" = 0 &&
+    test "$(git show-ref | grep foo3 | wc -l)" = 0 &&
+    test "$(git symbolic-ref HEAD)" = "refs/heads/master"
+'
+
+# Workaround for the test failure to make the rest of the subtests
+# succeed. (HEAD was erroneously overwritten with the bad foo3 ref, so
+# we need to reset it.)
+git symbolic-ref HEAD refs/heads/master
+
 test_expect_success \
     'Setup two commits including removal of generated files' '
     git init &&
-    touch a.c a.o &&
-    git add a.c a.o &&
+    touch file1 file2 &&
+    git add file1 file2 &&
     git commit -m 1 &&
-    git rm a.c a.o &&
+    git rm file1 file2 &&
     git commit -m 2 &&
-    touch a.o
+    touch file2
 '
 
 test_expect_success \
     'Create branch down the stack, behind the conflict caused by the generated file' '
-    ! stg branch --create bar master^
+    ! stg branch --create foo4 master^
 '
 
 test_expect_success \
     'Check the branch was not created' '
-    test ! -r .git/refs/heads/bar &&
-    test ! -r a.c
+    test ! -e file1 &&
+    test "$(find .git -name foo4 | tee /dev/stderr | wc -l)" = 0 &&
+    test "$(git show-ref | grep foo4 | wc -l)" = 0 &&
+    test "$(git symbolic-ref HEAD)" = "refs/heads/master"
 '
 
 test_done

      reply	other threads:[~2008-04-13 13:45 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2008-04-12 12:44 [StGit PATCH] Remove broken branch creation subtest Karl Hasselström
2008-04-12 18:06 ` Junio C Hamano
2008-04-13  6:02   ` Karl Hasselström
2008-04-13 13:43     ` Karl Hasselström [this message]

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20080413134104.15175.84334.stgit@yoghurt \
    --to=kha@treskal.com \
    --cc=catalin.marinas@gmail.com \
    --cc=git@vger.kernel.org \
    --cc=gitster@pobox.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).