annex.hardlink extended to also try to use hard links when copying from the repository to a remote.

Also, it used to only check that one of the repos was not in direct mode;
now when either repo is direct mode, annex.hardlink won't have an effect.

4 files changed, 31 insertions, 18 deletions

diff --git a/Remote/Git.hs b/Remote/Git.hs
index 0fecf1ed1..4187a5178 100644
--- a/Remote/Git.hs
+++ b/Remote/Git.hs
@@ -368,9 +368,7 @@ copyFromRemote' r key file dest meterupdate
 	| not $ Git.repoIsUrl (repo r) = guardUsable (repo r) (return False) $ do
 		params <- Ssh.rsyncParams r Download
 		u <- getUUID
-#ifndef mingw32_HOST_OS
-		hardlink <- annexHardLink <$> Annex.getGitConfig
-#endif
+		hardlink <- wantHardLink
 		-- run copy from perspective of remote
 		onLocal r $ do
 			ensureInitialized
@@ -378,19 +376,9 @@ copyFromRemote' r key file dest meterupdate
 			case v of
 				Nothing -> return False
 				Just (object, checksuccess) -> do
-					let copier = rsyncOrCopyFile params object dest
-#ifndef mingw32_HOST_OS
-					let linker = createLink object dest >> return True
-					go <- ifM (pure hardlink <&&> not <$> isDirect)
-						( return $ \m -> liftIO (catchBoolIO linker)
-							<||> copier m
-						, return copier
-						)
-#else
-					let go = copier
-#endif
+					copier <- mkCopier hardlink params object dest
 					runTransfer (Transfer Download u key)
-						file noRetry noObserver go
+						file noRetry noObserver copier
 						<&&> checksuccess
 	| Git.repoIsSsh (repo r) = feedprogressback $ \feeder -> do
 		direct <- isDirect
@@ -506,6 +494,7 @@ copyToRemote' r key file p
 		checksuccessio <- Annex.withCurrentState checksuccess
 		params <- Ssh.rsyncParams r Upload
 		u <- getUUID
+		hardlink <- wantHardLink
 		-- run copy from perspective of remote
 		onLocal r $ ifM (Annex.Content.inAnnex key)
 			( return True
@@ -514,7 +503,7 @@ copyToRemote' r key file p
 				runTransfer (Transfer Download u key) file noRetry noObserver $ const $
 					Annex.Content.saveState True `after`
 						Annex.Content.getViaTmpChecked (liftIO checksuccessio) key
-							(\d -> rsyncOrCopyFile params object d p)
+							(\dest -> mkCopier hardlink params object dest >>= \a -> a p)
 			)
 
 fsckOnRemote :: Git.Repo -> [CommandParam] -> Annex (IO Bool)
@@ -622,3 +611,23 @@ commitOnCleanup r a = go `after` a
 				withQuietOutput createProcessSuccess $
 					proc shellcmd $
 						toCommand shellparams
+
+wantHardLink :: Annex Bool
+wantHardLink = (annexHardLink <$> Annex.getGitConfig) <&&> (not <$> isDirect)
+
+-- If either the remote or local repository wants to use hard links,
+-- the copier will do so, falling back to copying.
+mkCopier :: Bool -> [CommandParam] -> FilePath -> FilePath -> Annex (MeterUpdate -> Annex Bool)
+mkCopier remotewanthardlink rsyncparams object dest = do
+	let copier = rsyncOrCopyFile rsyncparams object dest
+#ifndef mingw32_HOST_OS
+	localwanthardlink <- wantHardLink
+	let linker = createLink object dest >> return True
+	ifM (pure (remotewanthardlink || localwanthardlink) <&&> not <$> isDirect)
+		( return $ \m -> liftIO (catchBoolIO linker)
+			<||> copier m
+		, return copier
+		)
+#else
+	return copier
+#endif
diff --git a/debian/changelog b/debian/changelog
index eb2fed8ff..92b6b16ab 100644
--- a/debian/changelog
+++ b/debian/changelog
@@ -20,6 +20,8 @@ git-annex (5.20150825) UNRELEASED; urgency=medium
   * sync: Add --no-commit, --no-pull, --no-push options to turn off parts of
     the sync process, as well as supporting --commit, --pull, --push, and
     --no-content options to specify the (current) default behavior.
+  * annex.hardlink extended to also try to use hard links when copying from
+    the repository to a remote.
 
  -- Joey Hess <id@joeyh.name>  Tue, 01 Sep 2015 14:46:18 -0700
 
diff --git a/doc/git-annex.mdwn b/doc/git-annex.mdwn
index a916b74a4..794950d76 100644
--- a/doc/git-annex.mdwn
+++ b/doc/git-annex.mdwn
@@ -875,8 +875,8 @@ Here are all the supported configuration settings.
 
 * `annex.hardlink`
 
-  Set this to `true` to make file contents be hard linked into the
-  repository when possible, instead of a more expensive copy.
+  Set this to `true` to make file contents be hard linked between the
+  repository and its remotes when possible, instead of a more expensive copy.
 
   Use with caution -- This can invalidate numcopies counting, since
   with hard links, fewer copies of a file can exist. So, it is a good
diff --git a/doc/todo/annex.hardlink_should_also_affect_copy_to_origin.mdwn b/doc/todo/annex.hardlink_should_also_affect_copy_to_origin.mdwn
index bee099312..d79c13f92 100644
--- a/doc/todo/annex.hardlink_should_also_affect_copy_to_origin.mdwn
+++ b/doc/todo/annex.hardlink_should_also_affect_copy_to_origin.mdwn
@@ -3,3 +3,5 @@ repository when eg `git annex copy --from origin`. This should also
 be done when copying files --to origin (or other remotes on the same
 filesystem). This way, a quick shared clone can be used to add/modify
 files, and cheaply send the changes back to the parent repo. --[[Joey]]
+
+> [[done]] --[[Joey]]