Re: [PATCH] t4202 (log): add test for --follow over a merge

2013-04-22 Thread Jonathan Nieder
Hi,

Ramkumar Ramachandra wrote:

> The --follow feature can be used to follow the history of a file over
> a merge commit, and is useful even when the file hasn't been
> copied/renamed.  Add a test to show off this feature.

I can't claim to have followed the discussion, but I don't understand
the above.  If the file's name hasn't changed, the most one can hope
for is that --follow doesn't hurt, no?

[...]
> + cat >expect <<-\EOF &&
> + df26551 add a line to the beginning of ichi
> + 882d8d9 add a line to the end of ichi
> + 2fbe8c0 third
> + f7dab8e second
> + 3a2fdcb initial
> + EOF
> + test_cmp expect actual

t/README explains:

| As with any programming projects, existing programs are the best
| source of the information.  However, do _not_ emulate
| t-basic.sh when writing your tests.  The test is special in
| that it tries to validate the very core of GIT.
[...]
|  other tests that simply rely on basic parts of the core
| GIT working properly should not have that level of intimate
| knowledge of the core GIT internals.  If all the test scripts
| hardcoded the object IDs like t-basic.sh does, that defeats
| the purpose of t-basic.sh, which is to isolate that level of
| validation in one place.

Hard-coding object names makes it painful to tweak a given test and
tests that come before it, lest a timestamp or the phase of moon
change and cause the object names to drift.   Don't do it, except
in t.

Hope that helps,
Jonathan
--
To unsubscribe from this list: send the line "unsubscribe git" in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html


[PATCH] t4202 (log): add test for --follow over a merge

2013-04-22 Thread Ramkumar Ramachandra
The --follow feature can be used to follow the history of a file over
a merge commit, and is useful even when the file hasn't been
copied/renamed.  Add a test to show off this feature.

Signed-off-by: Ramkumar Ramachandra 
---
 I haven't updated the documentation, because I can't claim to
 understand --follow fully without reading the code (doing that now).
 In the meantime, I'm contributing a patch I wrote out originally as
 an example for Thomas Rast to prove a point.

 Where should this go?  What is t4206-log-follow-harder-copies (it
 just has one test which isn't even that special)?  Should we move all
 the --follow tests out of this file and put it in a separate file?

 t/t4202-log.sh | 29 +
 1 file changed, 29 insertions(+)

diff --git a/t/t4202-log.sh b/t/t4202-log.sh
index 523c1be..5b041fd 100755
--- a/t/t4202-log.sh
+++ b/t/t4202-log.sh
@@ -168,6 +168,35 @@ test_expect_success 'git log --follow' '
 
 '
 
+test_expect_success '--follow over merge' '
+   git checkout -b featurebranch
+   echo foodle >>ichi &&
+   git add ichi &&
+   test_tick &&
+   git commit -m "add a line to the end of ichi" &&
+   echo moodle >unrelated &&
+   git add unrelated &&
+   test_tick &&
+   git commit -m quux &&
+   git checkout master &&
+   mv ichi ichi.bak &&
+   echo gooble >ichi &&
+   cat ichi.bak >>ichi &&
+   git add ichi &&
+   test_tick &&
+   git commit -m "add a line to the beginning of ichi" &&
+   git merge featurebranch &&
+   git log --follow --oneline ichi >actual &&
+   cat >expect <<-\EOF &&
+   df26551 add a line to the beginning of ichi
+   882d8d9 add a line to the end of ichi
+   2fbe8c0 third
+   f7dab8e second
+   3a2fdcb initial
+   EOF
+   test_cmp expect actual
+'
+
 test_expect_failure 'log pathspec in tree read into prefix' '
git checkout --orphan subtree &&
git rm -rf . &&
-- 
1.8.2.1.546.gea3475a

--
To unsubscribe from this list: send the line "unsubscribe git" in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html