git-annex/Limit.hs
Joey Hess 067aabdd48
wip RawFilePath 2x git-annex find speedup
Finally builds (oh the agoncy of making it build), but still very
unmergable, only Command.Find is included and lots of stuff is badly
hacked to make it compile.

Benchmarking vs master, this git-annex find is significantly faster!
Specifically:

	num files	old	new	speedup
	48500		4.77	3.73	28%
	12500		1.36	1.02	66%
	20		0.075	0.074	0% (so startup time is unchanged)

That's without really finishing the optimization. Things still to do:

* Eliminate all the fromRawFilePath, toRawFilePath, encodeBS,
  decodeBS conversions.
* Use versions of IO actions like getFileStatus that take a RawFilePath.
* Eliminate some Data.ByteString.Lazy.toStrict, which is a slow copy.
* Use ByteString for parsing git config to speed up startup.

It's likely several of those will speed up git-annex find further.
And other commands will certianly benefit even more.
2019-11-26 16:01:58 -04:00

377 lines
13 KiB
Haskell

{- user-specified limits on files to act on
-
- Copyright 2011-2019 Joey Hess <id@joeyh.name>
-
- Licensed under the GNU AGPL version 3 or higher.
-}
module Limit where
import Annex.Common
import qualified Annex
import qualified Utility.Matcher
import qualified Remote
import Annex.Content
import Annex.WorkTree
import Annex.Action
import Annex.UUID
import Annex.Magic
import Annex.Link
import Logs.Trust
import Annex.NumCopies
import Types.Key
import Types.TrustLevel
import Types.Group
import Types.FileMatcher
import Types.MetaData
import Annex.MetaData
import Logs.MetaData
import Logs.Group
import Logs.Unused
import Logs.Location
import Git.Types (RefDate(..))
import Utility.Glob
import Utility.HumanTime
import Utility.DataUnits
import Data.Time.Clock.POSIX
import qualified Data.Set as S
import qualified Data.Map as M
{- Some limits can look at the current status of files on
- disk, or in the annex. This allows controlling which happens. -}
data LimitBy = LimitDiskFiles | LimitAnnexFiles
{- Checks if there are user-specified limits. -}
limited :: Annex Bool
limited = (not . Utility.Matcher.isEmpty) <$> getMatcher'
{- Gets a matcher for the user-specified limits. The matcher is cached for
- speed; once it's obtained the user-specified limits can't change. -}
getMatcher :: Annex (MatchInfo -> Annex Bool)
getMatcher = Utility.Matcher.matchM <$> getMatcher'
getMatcher' :: Annex (Utility.Matcher.Matcher (MatchInfo -> Annex Bool))
getMatcher' = go =<< Annex.getState Annex.limit
where
go (CompleteMatcher matcher) = return matcher
go (BuildingMatcher l) = do
let matcher = Utility.Matcher.generate (reverse l)
Annex.changeState $ \s ->
s { Annex.limit = CompleteMatcher matcher }
return matcher
{- Adds something to the limit list, which is built up reversed. -}
add :: Utility.Matcher.Token (MatchInfo -> Annex Bool) -> Annex ()
add l = Annex.changeState $ \s -> s { Annex.limit = prepend $ Annex.limit s }
where
prepend (BuildingMatcher ls) = BuildingMatcher $ l:ls
prepend _ = error "internal"
{- Adds a new syntax token. -}
addSyntaxToken :: String -> Annex ()
addSyntaxToken = either error add . Utility.Matcher.syntaxToken
{- Adds a new limit. -}
addLimit :: Either String (MatchFiles Annex) -> Annex ()
addLimit = either giveup (\l -> add $ Utility.Matcher.Operation $ l S.empty)
{- Add a limit to skip files that do not match the glob. -}
addInclude :: String -> Annex ()
addInclude = addLimit . limitInclude
limitInclude :: MkLimit Annex
limitInclude glob = Right $ const $ matchGlobFile glob
{- Add a limit to skip files that match the glob. -}
addExclude :: String -> Annex ()
addExclude = addLimit . limitExclude
limitExclude :: MkLimit Annex
limitExclude glob = Right $ const $ not <$$> matchGlobFile glob
matchGlobFile :: String -> MatchInfo -> Annex Bool
matchGlobFile glob = go
where
cglob = compileGlob glob CaseSensative -- memoized
go (MatchingFile fi) = pure $ matchGlob cglob (matchFile fi)
go (MatchingInfo p) = matchGlob cglob <$> getInfo (providedFilePath p)
go (MatchingKey _ (AssociatedFile Nothing)) = pure False
go (MatchingKey _ (AssociatedFile (Just af))) = pure $ matchGlob cglob (fromRawFilePath af)
addMimeType :: String -> Annex ()
addMimeType = addMagicLimit "mimetype" getMagicMimeType providedMimeType
addMimeEncoding :: String -> Annex ()
addMimeEncoding = addMagicLimit "mimeencoding" getMagicMimeEncoding providedMimeEncoding
addMagicLimit :: String -> (Magic -> FilePath -> Annex (Maybe String)) -> (ProvidedInfo -> OptInfo String) -> String -> Annex ()
addMagicLimit limitname querymagic selectprovidedinfo glob = do
magic <- liftIO initMagicMime
addLimit $ matchMagic limitname querymagic' selectprovidedinfo magic glob
where
querymagic' magic f = liftIO (isPointerFile (toRawFilePath f)) >>= \case
-- Avoid getting magic of a pointer file, which would
-- wrongly be detected as text.
Just _ -> return Nothing
-- When the file is an annex symlink, get magic of the
-- object file.
Nothing -> isAnnexLink (toRawFilePath f) >>= \case
Just k -> withObjectLoc k $ querymagic magic
Nothing -> querymagic magic f
matchMagic :: String -> (Magic -> FilePath -> Annex (Maybe String)) -> (ProvidedInfo -> OptInfo String) -> Maybe Magic -> MkLimit Annex
matchMagic _limitname querymagic selectprovidedinfo (Just magic) glob = Right $ const go
where
cglob = compileGlob glob CaseSensative -- memoized
go (MatchingKey _ _) = pure False
go (MatchingFile fi) = catchBoolIO $
maybe False (matchGlob cglob)
<$> querymagic magic (currFile fi)
go (MatchingInfo p) =
matchGlob cglob <$> getInfo (selectprovidedinfo p)
matchMagic limitname _ _ Nothing _ =
Left $ "unable to load magic database; \""++limitname++"\" cannot be used"
addUnlocked :: Annex ()
addUnlocked = addLimit $ Right $ const $ matchLockStatus False
addLocked :: Annex ()
addLocked = addLimit $ Right $ const $ matchLockStatus True
matchLockStatus :: Bool -> MatchInfo -> Annex Bool
matchLockStatus _ (MatchingKey _ _) = pure False
matchLockStatus _ (MatchingInfo _) = pure False
matchLockStatus wantlocked (MatchingFile fi) = liftIO $ do
islocked <- isPointerFile (toRawFilePath (currFile fi)) >>= \case
Just _key -> return False
Nothing -> isSymbolicLink
<$> getSymbolicLinkStatus (currFile fi)
return (islocked == wantlocked)
{- Adds a limit to skip files not believed to be present
- in a specfied repository. Optionally on a prior date. -}
addIn :: String -> Annex ()
addIn s = do
u <- Remote.nameToUUID name
hereu <- getUUID
addLimit $ if u == hereu && null date
then use inhere
else use (inuuid u)
where
(name, date) = separate (== '@') s
use a = Right $ checkKey . a
inuuid u notpresent key
| null date = do
us <- Remote.keyLocations key
return $ u `elem` us && u `S.notMember` notpresent
| otherwise = do
us <- loggedLocationsHistorical (RefDate date) key
return $ u `elem` us
inhere notpresent key
| S.null notpresent = inAnnex key
| otherwise = do
u <- getUUID
if u `S.member` notpresent
then return False
else inAnnex key
{- Limit to content that is currently present on a uuid. -}
limitPresent :: Maybe UUID -> MatchFiles Annex
limitPresent u _ = checkKey $ \key -> do
hereu <- getUUID
if u == Just hereu || isNothing u
then inAnnex key
else do
us <- Remote.keyLocations key
return $ maybe False (`elem` us) u
{- Limit to content that is in a directory, anywhere in the repository tree -}
limitInDir :: FilePath -> MatchFiles Annex
limitInDir dir = const go
where
go (MatchingFile fi) = checkf $ matchFile fi
go (MatchingKey _ (AssociatedFile Nothing)) = return False
go (MatchingKey _ (AssociatedFile (Just af))) = checkf (fromRawFilePath af)
go (MatchingInfo p) = checkf =<< getInfo (providedFilePath p)
checkf = return . elem dir . splitPath . takeDirectory
{- Adds a limit to skip files not believed to have the specified number
- of copies. -}
addCopies :: String -> Annex ()
addCopies = addLimit . limitCopies
limitCopies :: MkLimit Annex
limitCopies want = case splitc ':' want of
-- Note that in case of a group having the same name as a trust
-- level, it's parsed as a trust level, not as a group.
[v, n] -> case parsetrustspec v of
Just checker -> go n $ checktrust checker
Nothing -> go n $ checkgroup (toGroup v)
[n] -> go n $ const $ return True
_ -> Left "bad value for copies"
where
go num good = case readish num of
Nothing -> Left "bad number for copies"
Just n -> Right $ \notpresent -> checkKey $
go' n good notpresent
go' n good notpresent key = do
us <- filter (`S.notMember` notpresent)
<$> (filterM good =<< Remote.keyLocations key)
return $ length us >= n
checktrust checker u = checker <$> lookupTrust u
checkgroup g u = S.member g <$> lookupGroups u
parsetrustspec s
| "+" `isSuffixOf` s = (<=) <$> readTrustLevel (beginning s)
| otherwise = (==) <$> readTrustLevel s
{- Adds a limit to match files that need more copies made. -}
addLackingCopies :: Bool -> String -> Annex ()
addLackingCopies approx = addLimit . limitLackingCopies approx
limitLackingCopies :: Bool -> MkLimit Annex
limitLackingCopies approx want = case readish want of
Just needed -> Right $ \notpresent mi -> flip checkKey mi $
go mi needed notpresent
Nothing -> Left "bad value for number of lacking copies"
where
go mi needed notpresent key = do
NumCopies numcopies <- if approx
then approxNumCopies
else case mi of
MatchingFile fi -> getGlobalFileNumCopies $ matchFile fi
MatchingKey _ _ -> approxNumCopies
MatchingInfo {} -> approxNumCopies
us <- filter (`S.notMember` notpresent)
<$> (trustExclude UnTrusted =<< Remote.keyLocations key)
return $ numcopies - length us >= needed
approxNumCopies = fromMaybe defaultNumCopies <$> getGlobalNumCopies
{- Match keys that are unused.
-
- This has a nice optimisation: When a file exists,
- its key is obviously not unused.
-}
limitUnused :: MatchFiles Annex
limitUnused _ (MatchingFile _) = return False
limitUnused _ (MatchingKey k _) = S.member k <$> unusedKeys
limitUnused _ (MatchingInfo p) = do
k <- getInfo (providedKey p)
S.member k <$> unusedKeys
{- Limit that matches any version of any file or key. -}
limitAnything :: MatchFiles Annex
limitAnything _ _ = return True
{- Limit that never matches. -}
limitNothing :: MatchFiles Annex
limitNothing _ _ = return False
{- Adds a limit to skip files not believed to be present in all
- repositories in the specified group. -}
addInAllGroup :: String -> Annex ()
addInAllGroup groupname = addLimit $ limitInAllGroup groupMap groupname
limitInAllGroup :: Annex GroupMap -> MkLimit Annex
limitInAllGroup getgroupmap groupname = Right $ \notpresent mi -> do
m <- getgroupmap
let want = fromMaybe S.empty $ M.lookup (toGroup groupname) $ uuidsByGroup m
if S.null want
then return True
-- optimisation: Check if a wanted uuid is notpresent.
else if not (S.null (S.intersection want notpresent))
then return False
else checkKey (check want) mi
where
check want key = do
present <- S.fromList <$> Remote.keyLocations key
return $ S.null $ want `S.difference` present
{- Adds a limit to skip files not using a specified key-value backend. -}
addInBackend :: String -> Annex ()
addInBackend = addLimit . limitInBackend
limitInBackend :: MkLimit Annex
limitInBackend name = Right $ const $ checkKey check
where
check key = pure $ fromKey keyVariety key == variety
variety = parseKeyVariety (encodeBS name)
{- Adds a limit to skip files not using a secure hash. -}
addSecureHash :: Annex ()
addSecureHash = addLimit $ Right limitSecureHash
limitSecureHash :: MatchFiles Annex
limitSecureHash _ = checkKey $ pure . cryptographicallySecure . fromKey keyVariety
{- Adds a limit to skip files that are too large or too small -}
addLargerThan :: String -> Annex ()
addLargerThan = addLimit . limitSize LimitAnnexFiles (>)
addSmallerThan :: String -> Annex ()
addSmallerThan = addLimit . limitSize LimitAnnexFiles (<)
limitSize :: LimitBy -> (Maybe Integer -> Maybe Integer -> Bool) -> MkLimit Annex
limitSize lb vs s = case readSize dataUnits s of
Nothing -> Left "bad size"
Just sz -> Right $ go sz
where
go sz _ (MatchingFile fi) = case lb of
LimitAnnexFiles -> lookupFileKey fi >>= \case
Just key -> checkkey sz key
Nothing -> return False
LimitDiskFiles -> do
filesize <- liftIO $ catchMaybeIO $ getFileSize (currFile fi)
return $ filesize `vs` Just sz
go sz _ (MatchingKey key _) = checkkey sz key
go sz _ (MatchingInfo p) =
getInfo (providedFileSize p)
>>= \sz' -> return (Just sz' `vs` Just sz)
checkkey sz key = return $ fromKey keySize key `vs` Just sz
addMetaData :: String -> Annex ()
addMetaData = addLimit . limitMetaData
limitMetaData :: MkLimit Annex
limitMetaData s = case parseMetaDataMatcher s of
Left e -> Left e
Right (f, matching) -> Right $ const $ checkKey (check f matching)
where
check f matching k = not . S.null
. S.filter matching
. metaDataValues f <$> getCurrentMetaData k
addTimeLimit :: Duration -> Annex ()
addTimeLimit duration = do
start <- liftIO getPOSIXTime
let cutoff = start + durationToPOSIXTime duration
addLimit $ Right $ const $ const $ do
now <- liftIO getPOSIXTime
if now > cutoff
then do
warning $ "Time limit (" ++ fromDuration duration ++ ") reached!"
shutdown True
liftIO $ exitWith $ ExitFailure 101
else return True
addAccessedWithin :: Duration -> Annex ()
addAccessedWithin duration = do
now <- liftIO getPOSIXTime
addLimit $ Right $ const $ checkKey $ check now
where
check now k = inAnnexCheck k $ \f ->
liftIO $ catchDefaultIO False $ do
s <- getFileStatus f
let accessed = realToFrac (accessTime s)
let delta = now - accessed
return $ delta <= secs
secs = fromIntegral (durationSeconds duration)
lookupFileKey :: FileInfo -> Annex (Maybe Key)
lookupFileKey = lookupFile . toRawFilePath . currFile
checkKey :: (Key -> Annex Bool) -> MatchInfo -> Annex Bool
checkKey a (MatchingFile fi) = lookupFileKey fi >>= maybe (return False) a
checkKey a (MatchingKey k _) = a k
checkKey a (MatchingInfo p) = a =<< getInfo (providedKey p)