reopen

diff --git a/doc/bugs/initial_get_of_unlocked_file_fails_to_populate_pointer.mdwn b/doc/bugs/initial_get_of_unlocked_file_fails_to_populate_pointer.mdwn
index d2eea4c95da7d355d97311bff98abec518f6f409..885205068d3a16650b58612f44443abb06c17ca0 100644 (file)
--- a/doc/bugs/initial_get_of_unlocked_file_fails_to_populate_pointer.mdwn
+++ b/doc/bugs/initial_get_of_unlocked_file_fails_to_populate_pointer.mdwn
@@ -14,4 +14,32 @@ empty list. So something to do with database write caching.
 
 Somehow, not having init call `scanAnnexedFiles` makes this bug go away.
 
-> [[fixed|done]] --[[Joey]]
+> fixed --[[Joey]]
+
+----
+
+I have reopened this bug, it seems the previous fix was not right.
+See [[!commit b3c4579c7907147a496bdf2c73b42238d8b239d6]] for that
+fix, which had doubts at the time in the commit message. --[[Joey]] 
+
+Here is a test case:
+
+       git init foo
+       cd foo
+       git annex init
+       echo 1 > foo
+       echo 2 > bar
+       echo 3 > baz
+       git annex add
+       git annex unlock
+       git commit -m add
+       cd ..
+       git clone foo bar
+       cd bar
+       git-annex get
+       cat *
+
+This outputs 1 for foo, followed by annex pointer files for files bar and baz.
+
+The previous fix attempt did make foo get populated, before that none
+of the files were populated.