[Svnmerge] Re: Tests for --record-only

[Daniel Rall]

On Wed, 29 Mar 2006, Giovanni Bajo wrote:

> Daniel,
> 
> I just realized I forget in the review about the tests. You should really add a
> test in the testsuite which verifies the behaviour of --record-only, so that it
> doesn't get accidentally broken in the future. Can you find a little time for
> that?
> 
> Thanks again for your patch!

Giovanni, your wish is my command.  :-P

I committed a new test for 'merge --record-only' along with the new
functionality in r19051.  For reference, here it is:

--- svnmerge_test.py	(revision 19050)
+++ svnmerge_test.py	(revision 19051)
@@ -521,6 +521,15 @@
         self.svnmerge("merge -F -vv -r8-9", match=r"svn log.*-r8:9")
         self.svnmerge("avail -vv -r2", nonmatch=r"svn log")
 
+    def testMergeRecordOnly(self):
+        """Check that flagging revisions as manually merged works."""
+        self.svnmerge("init")
+        self.svnmerge("avail -vv -r9", match=r"svn log.*-r9:9")
+        self.svnmerge("merge --record-only -F -vv -r9",
+                      nonmatch=r"svn merge -r 8:9")
+        out = self.svnmerge("avail -r9")
+        self.assertEqual(out.strip(), "")
+
     def testBidirectionalMerges(self):
         """Check that reflected revisions are recognized properly for bidirectional merges."""

-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
Url : /pipermail/svnmerge/attachments/20060328/99be288c/attachment.pgp