Revert "reorder another test"

This reverts commit 3e63f00f63.
This commit is contained in:
Joey Hess 2021-04-23 01:01:29 -04:00
parent bb4f2af602
commit da0a696c96
No known key found for this signature in database
GPG key ID: DB12DB0FF05F8F38
2 changed files with 4 additions and 20 deletions

View file

@ -247,6 +247,10 @@ test runannex mkr mkk =
, check "storeKey when already present" $ \r k ->
whenwritable r $ runBool (store r k)
, check ("present " ++ show True) $ \r k -> present r k True
, check "retrieveKeyFile" $ \r k -> do
lockContentForRemoval k noop removeAnnex
get r k
, check "fsck downloaded object" fsck
, check "retrieveKeyFile resume from 0" $ \r k -> do
tmp <- fromRawFilePath <$> prepTmp k
liftIO $ writeFile tmp ""
@ -270,10 +274,6 @@ test runannex mkr mkk =
lockContentForRemoval k noop removeAnnex
get r k
, check "fsck downloaded object" fsck
, check "retrieveKeyFile" $ \r k -> do
lockContentForRemoval k noop removeAnnex
get r k
, check "fsck downloaded object" fsck
, check "removeKey when present" $ \r k ->
whenwritable r $ runBool (remove r k)
, check ("present " ++ show False) $ \r k ->

View file

@ -1,16 +0,0 @@
[[!comment format=mdwn
username="joey"
subject="""comment 4"""
date="2021-04-22T13:40:33Z"
content="""
Problem is confirmed to not be from the 33% test, the 0% test failed when
run before it, also in the removeAnnex part. And removeAnnex didn't change
lately, so it seems probably an earlier test sets up the failure.
I've moved the 0% test to be the first test that retrieves the file, let's
see if it succeeds that way, and if so, it must be that retrieveKeyFile is
leaking a file handle, despite using withBinaryFile which is supposed to
close handles automatically.
ghc version used for this windows build is 8.8.4 btw.
"""]]