prevent dropping required content of other file using same content

When two files have the same content, and a required content expression
matches one but not the other, dropping the latter file will fail as it
would also remove the content of the required file.

This will slow down drop (w/o --auto), dropunused, mirror, and move, by one
keys db lookup per file. But I did include an optimisation to avoid a
double db lookup in the drop --auto / sync --content case. I suspect that
dropunused could also use PreferredContentChecked True, but haven't
entirely thought it through and it's rarely used with enough files for the
optimisation to matter.

Sponsored-by: Dartmouth College's Datalad project
This commit is contained in:
Joey Hess 2021-05-25 10:57:06 -04:00
parent 7029ef1c3d
commit cedc28a783
No known key found for this signature in database
GPG key ID: DB12DB0FF05F8F38
9 changed files with 105 additions and 60 deletions

View file

@ -21,8 +21,6 @@ import Annex.Content
import Annex.Wanted
import Annex.Notification
import qualified Data.Set as S
cmd :: Command
cmd = withGlobalOptions [jobsOption, jsonOptions, annexedMatchingOptions] $
command "drop" SectionCommon
@ -88,31 +86,32 @@ start' o from key afile ai si =
checkDropAuto (autoMode o) from afile key $ \numcopies mincopies ->
stopUnless wantdrop $
case from of
Nothing -> startLocal afile ai si numcopies mincopies key []
Just remote -> startRemote afile ai si numcopies mincopies key remote
Nothing -> startLocal pcc afile ai si numcopies mincopies key []
Just remote -> startRemote pcc afile ai si numcopies mincopies key remote
where
wantdrop
| autoMode o = wantDrop False (Remote.uuid <$> from) (Just key) afile Nothing
| otherwise = return True
pcc = PreferredContentChecked (autoMode o)
startKeys :: DropOptions -> Maybe Remote -> (SeekInput, Key, ActionItem) -> CommandStart
startKeys o from (si, key, ai) = start' o from key (AssociatedFile Nothing) ai si
startLocal :: AssociatedFile -> ActionItem -> SeekInput -> NumCopies -> MinCopies -> Key -> [VerifiedCopy] -> CommandStart
startLocal afile ai si numcopies mincopies key preverified =
startLocal :: PreferredContentChecked -> AssociatedFile -> ActionItem -> SeekInput -> NumCopies -> MinCopies -> Key -> [VerifiedCopy] -> CommandStart
startLocal pcc afile ai si numcopies mincopies key preverified =
starting "drop" (OnlyActionOn key ai) si $
performLocal key afile numcopies mincopies preverified
performLocal pcc key afile numcopies mincopies preverified
startRemote :: AssociatedFile -> ActionItem -> SeekInput -> NumCopies -> MinCopies -> Key -> Remote -> CommandStart
startRemote afile ai si numcopies mincopies key remote =
startRemote :: PreferredContentChecked -> AssociatedFile -> ActionItem -> SeekInput -> NumCopies -> MinCopies -> Key -> Remote -> CommandStart
startRemote pcc afile ai si numcopies mincopies key remote =
starting ("drop " ++ Remote.name remote) (OnlyActionOn key ai) si $
performRemote key afile numcopies mincopies remote
performRemote pcc key afile numcopies mincopies remote
performLocal :: Key -> AssociatedFile -> NumCopies -> MinCopies -> [VerifiedCopy] -> CommandPerform
performLocal key afile numcopies mincopies preverified = lockContentForRemoval key fallback $ \contentlock -> do
performLocal :: PreferredContentChecked -> Key -> AssociatedFile -> NumCopies -> MinCopies -> [VerifiedCopy] -> CommandPerform
performLocal pcc key afile numcopies mincopies preverified = lockContentForRemoval key fallback $ \contentlock -> do
u <- getUUID
(tocheck, verified) <- verifiableCopies key [u]
doDrop u (Just contentlock) key afile numcopies mincopies [] (preverified ++ verified) tocheck
doDrop pcc u (Just contentlock) key afile numcopies mincopies [] (preverified ++ verified) tocheck
( \proof -> do
fastDebug "Command.Drop" $ unwords
[ "Dropping from here"
@ -134,12 +133,12 @@ performLocal key afile numcopies mincopies preverified = lockContentForRemoval k
-- to be done except for cleaning up.
fallback = next $ cleanupLocal key
performRemote :: Key -> AssociatedFile -> NumCopies -> MinCopies -> Remote -> CommandPerform
performRemote key afile numcopies mincopies remote = do
performRemote :: PreferredContentChecked -> Key -> AssociatedFile -> NumCopies -> MinCopies -> Remote -> CommandPerform
performRemote pcc key afile numcopies mincopies remote = do
-- Filter the uuid it's being dropped from out of the lists of
-- places assumed to have the key, and places to check.
(tocheck, verified) <- verifiableCopies key [uuid]
doDrop uuid Nothing key afile numcopies mincopies [uuid] verified tocheck
doDrop pcc uuid Nothing key afile numcopies mincopies [uuid] verified tocheck
( \proof -> do
fastDebug "Command.Drop" $ unwords
[ "Dropping from remote"
@ -169,12 +168,11 @@ cleanupRemote key remote ok = do
- verify that enough copies of a key exist to allow it to be
- safely removed (with no data loss).
-
- Also checks if it's required content, and refuses to drop if so.
-
- --force overrides and always allows dropping.
-}
doDrop
:: UUID
:: PreferredContentChecked
-> UUID
-> Maybe ContentRemovalLock
-> Key
-> AssociatedFile
@ -185,10 +183,10 @@ doDrop
-> [UnVerifiedCopy]
-> (Maybe SafeDropProof -> CommandPerform, CommandPerform)
-> CommandPerform
doDrop dropfrom contentlock key afile numcopies mincopies skip preverified check (dropaction, nodropaction) =
doDrop pcc dropfrom contentlock key afile numcopies mincopies skip preverified check (dropaction, nodropaction) =
ifM (Annex.getState Annex.force)
( dropaction Nothing
, ifM (checkRequiredContent dropfrom key afile)
, ifM (checkRequiredContent pcc dropfrom key afile)
( verifyEnoughCopiesToDrop nolocmsg key
contentlock numcopies mincopies
skip preverified check
@ -203,18 +201,27 @@ doDrop dropfrom contentlock key afile numcopies mincopies skip preverified check
showLongNote "(Use --force to override this check, or adjust numcopies.)"
a
checkRequiredContent :: UUID -> Key -> AssociatedFile -> Annex Bool
checkRequiredContent u k afile =
ifM (isRequiredContent (Just u) S.empty (Just k) afile False)
( requiredContent
, return True
)
{- Checking preferred content also checks required content, so when
- auto mode causes preferred content to be checked, it's redundant
- for checkRequiredContent to separately check required content, and
- providing this avoids that extra work. -}
newtype PreferredContentChecked = PreferredContentChecked Bool
requiredContent :: Annex Bool
requiredContent = do
showLongNote "That file is required content, it cannot be dropped!"
showLongNote "(Use --force to override this check, or adjust required content configuration.)"
return False
checkRequiredContent :: PreferredContentChecked -> UUID -> Key -> AssociatedFile -> Annex Bool
checkRequiredContent (PreferredContentChecked True) _ _ _ = return True
checkRequiredContent (PreferredContentChecked False) u k afile =
checkDrop isRequiredContent False (Just u) (Just k) afile Nothing >>= \case
Nothing -> return True
Just afile' -> do
if afile == afile'
then showLongNote "That file is required content. It cannot be dropped!"
else showLongNote $ "That file has the same content as another file"
++ case afile' of
AssociatedFile (Just f) -> " (" ++ fromRawFilePath f ++ "),"
AssociatedFile Nothing -> ""
++ " which is required content. It cannot be dropped!"
showLongNote "(Use --force to override this check, or adjust required content configuration.)"
return False
{- In auto mode, only runs the action if there are enough
- copies on other semitrusted repositories. -}

View file

@ -49,7 +49,7 @@ perform :: Maybe Remote -> NumCopies -> MinCopies -> Key -> CommandPerform
perform from numcopies mincopies key = case from of
Just r -> do
showAction $ "from " ++ Remote.name r
Command.Drop.performRemote key (AssociatedFile Nothing) numcopies mincopies r
Command.Drop.performRemote pcc key (AssociatedFile Nothing) numcopies mincopies r
Nothing -> ifM (inAnnex key)
( droplocal
, ifM (objectFileExists key)
@ -63,7 +63,8 @@ perform from numcopies mincopies key = case from of
)
)
where
droplocal = Command.Drop.performLocal key (AssociatedFile Nothing) numcopies mincopies []
droplocal = Command.Drop.performLocal pcc key (AssociatedFile Nothing) numcopies mincopies []
pcc = Command.Drop.PreferredContentChecked False
performOther :: (Key -> Git.Repo -> RawFilePath) -> Key -> CommandPerform
performOther filespec key = do

View file

@ -69,7 +69,7 @@ startKey o afile (si, key, ai) = case fromToOptions o of
( Command.Move.toStart Command.Move.RemoveNever afile key ai si =<< getParsed r
, do
(numcopies, mincopies) <- getnummincopies
Command.Drop.startRemote afile ai si numcopies mincopies key =<< getParsed r
Command.Drop.startRemote pcc afile ai si numcopies mincopies key =<< getParsed r
)
FromRemote r -> checkFailedTransferDirection ai Download $ do
haskey <- flip Remote.hasKey key =<< getParsed r
@ -82,10 +82,11 @@ startKey o afile (si, key, ai) = case fromToOptions o of
Right False -> ifM (inAnnex key)
( do
(numcopies, mincopies) <- getnummincopies
Command.Drop.startLocal afile ai si numcopies mincopies key []
Command.Drop.startLocal pcc afile ai si numcopies mincopies key []
, stop
)
where
getnummincopies = case afile of
AssociatedFile Nothing -> (,) <$> getNumCopies <*> getMinCopies
AssociatedFile (Just af) -> getFileNumMinCopies af
pcc = Command.Drop.PreferredContentChecked False

View file

@ -293,7 +293,7 @@ toHereStart removewhen afile key ai si =
- repository reduces the number of copies, and should fail if
- that would violate numcopies settings.
-
- On the other hand, when the destiation repository does not already
- On the other hand, when the destination repository does not already
- have a copy of a file, it can be dropped without making numcopies
- worse, so the move is allowed even if numcopies is not met.
-
@ -311,7 +311,7 @@ toHereStart removewhen afile key ai si =
-}
willDropMakeItWorse :: UUID -> UUID -> DestStartedWithCopy -> Key -> AssociatedFile -> Annex DropCheck
willDropMakeItWorse srcuuid destuuid (DestStartedWithCopy deststartedwithcopy) key afile =
ifM (Command.Drop.checkRequiredContent srcuuid key afile)
ifM (Command.Drop.checkRequiredContent (Command.Drop.PreferredContentChecked False) srcuuid key afile)
( if deststartedwithcopy
then unlessforced DropCheckNumCopies
else ifM checktrustlevel