better worm keys

2 files changed, 26 insertions, 6 deletions

diff --git a/Backend/Worm.hs b/Backend/Worm.hs
index ba79428ef..89fe4bf57 100644
--- a/Backend/Worm.hs
+++ b/Backend/Worm.hs
@@ -3,16 +3,36 @@
 
 module Backend.Worm (backend) where
 
+import Control.Monad.State
 import qualified Backend.File
 import BackendTypes
 import Utility
 import System.FilePath
+import System.Posix.Files
+import Data.Digest.Pure.SHA -- slow, but we only checksum filenames
+import qualified Data.ByteString.Lazy.Char8 as B
 
 backend = Backend.File.backend {
 	name = "WORM",
 	getKey = keyValue
 }
 
--- direct mapping from basename of filename to key
+-- A SHA1 of the basename of the filename, plus the file size and
+-- modification time, is used as the unique part of the key. That
+-- allows multiple files with the same names to have different keys,
+-- while also allowing a file to be moved around while retaining the
+-- same key.
+--
+-- The basename of the filename is also included in the key, so it's clear
+-- what the original filename was when a user sees the value.
 keyValue :: FilePath -> Annex (Maybe Key)
-keyValue file = return $ Just $ Key ((name backend), (takeFileName file))
+keyValue file = do
+	stat <- liftIO $ getFileStatus file
+	return $ Just $ Key ((name backend), key stat)
+	where
+ 		key stat = (checksum $ uniqueid stat) ++ sep ++ base
+		checksum s = show $ sha1 $ B.pack s
+		uniqueid stat = (show $ fileSize stat) ++ sep ++
+			(show $ modificationTime stat)
+		base = takeFileName file
+		sep = ":"
diff --git a/git-annex.mdwn b/git-annex.mdwn
index fba9648db..2079b5b46 100644
--- a/git-annex.mdwn
+++ b/git-annex.mdwn
@@ -94,10 +94,10 @@ Multiple pluggable backends are supported, and more than one can be used
 to store different files' contents in a given repository.
 
 * `WORM` ("Write Once, Read Many") This backend stores the file's content
-  in `.git/annex/`, and assumes that any file with the same basename
-  has the same content. So with this backend, files can be moved around,
-  but should never be added to or changed. This is the default, and
-  the least expensive backend.
+  in `.git/annex/`, and assumes that any file with the same basename,
+  size, and modification time has the same content. So with this backend,
+  files can be moved around, but should never be added to or changed.
+  This is the default, and the least expensive backend.
 * `sha1sum` -- This backend stores the file's content in
   `.git/annex/`, with a name based on its sha1 checksum. This backend allows
   modifications of files to be tracked. Its need to generate checksums