fix build failure by avoiding refutable pattern match

Annex/TransferrerPool.hs

@@ -53,7 +53,7 @@ withTransferrer a = do
 	withTransferrer' False nocheck program nonBatchCommandMaker pool a
 
 withTransferrer'
-	:: (MonadIO m, MonadFail m, MonadMask m)
+	:: (MonadIO m, MonadMask m)
 	=> Bool
 	-- ^ When minimizeprocesses is True, only one Transferrer is left
 	-- running in the pool at a time. So if this needed to start a
@@ -67,8 +67,11 @@ withTransferrer'
 	-> m a
 withTransferrer' minimizeprocesses mkcheck program batchmaker pool a = do
 	(mi, leftinpool) <- liftIO $ atomically (popTransferrerPool pool)
-	i@(TransferrerPoolItem (Just t) check) <- liftIO $ case mi of
+	(i@(TransferrerPoolItem _ check), t) <- liftIO $ case mi of
-		Nothing -> mkTransferrerPoolItem mkcheck =<< mkTransferrer program batchmaker
+		Nothing -> do
+			t <- mkTransferrer program batchmaker
+			i <- mkTransferrerPoolItem mkcheck t
+			return (i, t)
 		Just i -> checkTransferrerPoolItem program batchmaker i
 	a t `finally` returntopool leftinpool check t i
   where
@@ -85,10 +88,10 @@ withTransferrer' minimizeprocesses mkcheck program batchmaker pool a = do
 
 {- Check if a Transferrer from the pool is still ok to be used.
  - If not, stop it and start a new one. -}
-checkTransferrerPoolItem :: FilePath -> BatchCommandMaker -> TransferrerPoolItem -> IO TransferrerPoolItem
+checkTransferrerPoolItem :: FilePath -> BatchCommandMaker -> TransferrerPoolItem -> IO (TransferrerPoolItem, Transferrer)
 checkTransferrerPoolItem program batchmaker i = case i of
 	TransferrerPoolItem (Just t) check -> ifM check
-		( return i
+		( return (i, t)
 		, do
 			shutdownTransferrer t
 			new check
@@ -97,7 +100,7 @@ checkTransferrerPoolItem program batchmaker i = case i of
   where
 	new check = do
 		t <- mkTransferrer program batchmaker
-		return $ TransferrerPoolItem (Just t) check
+		return (TransferrerPoolItem (Just t) check, t)
 
 {- Requests that a Transferrer perform a Transfer, and waits for it to
  - finish.

doc/bugs/Build_failing_because_MonadFail_is_not_in_scope.mdwn

@@ -27,3 +27,5 @@ index 973f75629..0de145461 100644
 
 [[!meta author=jwodder]]
 [[!tag projects/datalad]]
+
+> [[fixed|done]] --[[Joey]]

doc/bugs/Build_failing_because_MonadFail_is_not_in_scope/comment_1_3ebe2e53fd8e8ce673267ac2715d964c._comment

@@ -0,0 +1,12 @@
+[[!comment format=mdwn
+ username="joey"
+ subject="""comment 1"""
+ date="2020-12-09T16:40:24Z"
+ content="""
+Indeed, and thank you for the patch.
+
+However, MonadFail often suggests a code smell, here it's the `Just ... <-`
+pattern match which could potentially fail. In fact, it's not possible for
+it to, so I have instead made that clear in a way that the type checker
+can understand.
+"""]]