Sped up git-annex add in direct mode and v6 by using git hash-object --batch.

Speeds up hashSymlink and hashPointerFile.

Annex.hs

@@ -42,6 +42,7 @@ import qualified Git
 import qualified Git.Config
 import Annex.Fixup
 import Git.CatFile
+import Git.HashObject
 import Git.CheckAttr
 import Git.CheckIgnore
 import qualified Git.Hook
@@ -106,6 +107,7 @@ data AnnexState = AnnexState
 	, branchstate :: BranchState
 	, repoqueue :: Maybe Git.Queue.Queue
 	, catfilehandles :: M.Map FilePath CatFileHandle
+	, hashobjecthandle :: Maybe HashObjectHandle
 	, checkattrhandle :: Maybe CheckAttrHandle
 	, checkignorehandle :: Maybe (Maybe CheckIgnoreHandle)
 	, forcebackend :: Maybe String
@@ -151,6 +153,7 @@ newState c r = AnnexState
 	, branchstate = startBranchState
 	, repoqueue = Nothing
 	, catfilehandles = M.empty
+	, hashobjecthandle = Nothing
 	, checkattrhandle = Nothing
 	, checkignorehandle = Nothing
 	, forcebackend = Nothing

Annex/Concurrent.hs

@@ -11,6 +11,7 @@ import Annex.Common
 import Annex
 import Annex.CatFile
 import Annex.CheckAttr
+import Annex.HashObject
 import Annex.CheckIgnore
 import qualified Annex.Queue
 
@@ -64,4 +65,5 @@ mergeState st = do
 	closehandles = do
 		catFileStop
 		checkAttrStop
+		hashObjectStop
 		checkIgnoreStop

Annex/HashObject.hs

@@ -12,38 +12,31 @@ module Annex.HashObject (
 	hashObjectStop,
 ) where
 
-import qualified Data.ByteString.Lazy as L
-import qualified Data.Map as M
-import System.PosixCompat.Types
-
 import Annex.Common
-import qualified Git
 import qualified Git.HashObject
 import qualified Annex
 import Git.Types
-import Git.FilePath
-import qualified Git.Ref
-import Annex.Link
 
 hashObjectHandle :: Annex Git.HashObject.HashObjectHandle
 hashObjectHandle = maybe startup return =<< Annex.getState Annex.hashobjecthandle
   where
 	startup = do
-		inRepo $ Git.hashObjectStart
+		h <- inRepo $ Git.HashObject.hashObjectStart
 		Annex.changeState $ \s -> s { Annex.hashobjecthandle = Just h }
 		return h
 
 hashObjectStop :: Annex ()
-hashObjectStop = maybe noop stop =<< Annex.hashobjecthandle
+hashObjectStop = maybe noop stop =<< Annex.getState Annex.hashobjecthandle
   where
 	stop h = do
-		liftIO $ Git.hashObjectStop h
+		liftIO $ Git.HashObject.hashObjectStop h
 		Annex.changeState $ \s -> s { Annex.hashobjecthandle = Nothing }
+		return ()
 
 hashFile :: FilePath -> Annex Sha
 hashFile f = do
 	h <- hashObjectHandle
-	Git.HashObject.hashFile h f
+	liftIO $ Git.HashObject.hashFile h f
 
 {- Note that the content will be written to a temp file.
  - So it may be faster to use Git.HashObject.hashObject for large
@@ -51,4 +44,4 @@ hashFile f = do
 hashBlob :: String -> Annex Sha
 hashBlob content = do
 	h <- hashObjectHandle
-	Git.HashObject.hashFile h content
+	liftIO $ Git.HashObject.hashBlob h content

Annex/Link.hs

@@ -18,11 +18,11 @@ module Annex.Link where
 
 import Annex.Common
 import qualified Annex
-import qualified Git.HashObject
 import qualified Git.UpdateIndex
 import qualified Annex.Queue
 import Git.Types
 import Git.FilePath
+import Annex.HashObject
 
 import qualified Data.ByteString.Lazy as L
 import Data.Int
@@ -105,12 +105,7 @@ addAnnexLink linktarget file = do
 
 {- Injects a symlink target into git, returning its Sha. -}
 hashSymlink :: LinkTarget -> Annex Sha
-hashSymlink linktarget = inRepo $ Git.HashObject.hashObject BlobObject $ 
-	toInternalGitPath linktarget
-
-hashSymlink' :: Git.HashObject.HashObjectHandle -> LinkTarget -> Annex Sha
-hashSymlink' h linktarget = liftIO $ Git.HashObject.hashBlob h $
-	toInternalGitPath linktarget
+hashSymlink linktarget = hashBlob (toInternalGitPath linktarget)
 
 {- Stages a symlink to an annexed object, using a Sha of its target. -}
 stageSymlink :: FilePath -> Sha -> Annex ()
@@ -120,8 +115,7 @@ stageSymlink file sha =
 
 {- Injects a pointer file content into git, returning its Sha. -}
 hashPointerFile :: Key -> Annex Sha
-hashPointerFile key = inRepo $ Git.HashObject.hashObject BlobObject $
-	formatPointer key
+hashPointerFile key = hashBlob (formatPointer key)
 
 {- Stages a pointer file, using a Sha of its content -}
 stagePointerFile :: FilePath -> Sha -> Annex ()

Annex/View.hs

@@ -19,7 +19,7 @@ import qualified Git.LsFiles
 import qualified Git.Ref
 import Git.UpdateIndex
 import Git.Sha
-import Git.HashObject
+import Annex.HashObject
 import Git.Types
 import Git.FilePath
 import Annex.WorkTree
@@ -340,38 +340,36 @@ applyView' mkviewedfile getfilemetadata view = do
 	(l, clean) <- inRepo $ Git.LsFiles.inRepo [top]
 	liftIO . nukeFile =<< fromRepo gitAnnexViewIndex
 	uh <- withViewIndex $ inRepo Git.UpdateIndex.startUpdateIndex
-	hasher <- inRepo hashObjectStart
 	forM_ l $ \f -> do
 		relf <- getTopFilePath <$> inRepo (toTopFilePath f)
-		go uh hasher relf =<< lookupFile f
+		go uh relf =<< lookupFile f
 	liftIO $ do
-		hashObjectStop hasher
 		void $ stopUpdateIndex uh
 		void clean
 	genViewBranch view
   where
 	genviewedfiles = viewedFiles view mkviewedfile -- enables memoization
-	go uh hasher f (Just k) = do
+	go uh f (Just k) = do
 		metadata <- getCurrentMetaData k
 		let metadata' = getfilemetadata f `unionMetaData` metadata
 		forM_ (genviewedfiles f metadata') $ \fv -> do
 			f' <- fromRepo $ fromTopFilePath $ asTopFilePath fv
-			stagesymlink uh hasher f' =<< calcRepo (gitAnnexLink f' k)
-	go uh hasher f Nothing
+			stagesymlink uh f' =<< calcRepo (gitAnnexLink f' k)
+	go uh f Nothing
 		| "." `isPrefixOf` f = do
 			s <- liftIO $ getSymbolicLinkStatus f
 			if isSymbolicLink s
-				then stagesymlink uh hasher f =<< liftIO (readSymbolicLink f)
+				then stagesymlink uh f =<< liftIO (readSymbolicLink f)
 				else do
-					sha <- liftIO $ Git.HashObject.hashFile hasher f
+					sha <- hashFile f
 					let blobtype = if isExecutable (fileMode s)
 						then ExecutableBlob
 						else FileBlob
 					liftIO . Git.UpdateIndex.streamUpdateIndex' uh
 						=<< inRepo (Git.UpdateIndex.stageFile sha blobtype f)
 		| otherwise = noop
-	stagesymlink uh hasher f linktarget = do
-		sha <- hashSymlink' hasher linktarget
+	stagesymlink uh f linktarget = do
+		sha <- hashSymlink linktarget
 		liftIO . Git.UpdateIndex.streamUpdateIndex' uh
 			=<< inRepo (Git.UpdateIndex.stageSymlink f sha)
 

debian/changelog

@@ -13,6 +13,8 @@ git-annex (6.20160230) UNRELEASED; urgency=medium
     lost in last release.
   * Always try to thaw content, even when annex.crippledfilesystem is set.
   * Correct git-annex info to include unlocked files in v6 repository.
+  * Sped up git-annex add in direct mode and v6 by using
+    git hash-object --batch.
 
  -- Joey Hess <id@joeyh.name>  Mon, 29 Feb 2016 13:00:30 -0400