git-annex/Logs/Transitions.hs

111 lines
3.6 KiB
Haskell
Raw Normal View History

2013-08-28 17:19:02 +00:00
{- git-annex transitions log
-
- This is used to record transitions that have been performed on the
- git-annex branch, and when the transition was first started.
-
- We can quickly detect when the local branch has already had an transition
- done that is listed in the remote branch by checking that the local
- branch contains the same transition, with the same or newer start time.
-
2021-12-28 16:15:51 +00:00
- Copyright 2013-2021 Joey Hess <id@joeyh.name>
2013-08-28 17:19:02 +00:00
-
- Licensed under the GNU AGPL version 3 or higher.
2013-08-28 17:19:02 +00:00
-}
{-# LANGUAGE OverloadedStrings #-}
2013-08-28 17:19:02 +00:00
module Logs.Transitions where
import Annex.Common
import Annex.VectorClock
import Logs.Line
2021-12-28 16:15:51 +00:00
import qualified Git
import Git.Types (fromRef)
import Annex.CatFile
2013-08-28 17:19:02 +00:00
import qualified Data.Set as S
import Data.Either
import Data.ByteString (ByteString)
import Data.ByteString.Builder
import qualified Data.ByteString.Lazy as L
import qualified Data.Attoparsec.ByteString.Lazy as A
import qualified Data.Attoparsec.ByteString.Char8 as A8
transitionsLog :: RawFilePath
2013-08-28 17:19:02 +00:00
transitionsLog = "transitions.log"
data Transition
= ForgetGitHistory
| ForgetDeadRemotes
deriving (Show, Ord, Eq, Read)
data TransitionLine = TransitionLine
{ transitionStarted :: VectorClock
-- New transitions that we don't know about yet are preserved.
, transition :: Either ByteString Transition
} deriving (Ord, Eq)
2013-08-28 17:19:02 +00:00
type Transitions = S.Set TransitionLine
describeTransition :: Transition -> String
describeTransition ForgetGitHistory = "forget git history"
describeTransition ForgetDeadRemotes = "forget dead remotes"
noTransitions :: Transitions
noTransitions = S.empty
deal better with clock skew situations, using vector clocks * Deal with clock skew, both forwards and backwards, when logging information to the git-annex branch. * GIT_ANNEX_VECTOR_CLOCK can now be set to a fixed value (eg 1) rather than needing to be advanced each time a new change is made. * Misuse of GIT_ANNEX_VECTOR_CLOCK will no longer confuse git-annex. When changing a file in the git-annex branch, the vector clock to use is now determined by first looking at the current time (or GIT_ANNEX_VECTOR_CLOCK when set), and comparing it to the newest vector clock already in use in that file. If a newer time stamp was already in use, advance it forward by a second instead. When the clock is set to a time in the past, this avoids logging with an old timestamp, which would risk that log line later being ignored in favor of "newer" line that is really not newer. When a log entry has been made with a clock that was set far ahead in the future, this avoids newer information being logged with an older timestamp and so being ignored in favor of that future-timestamped information. Once all clocks get fixed, this will result in the vector clocks being incremented, until finally enough time has passed that time gets back ahead of the vector clock value, and then it will return to usual operation. (This latter situation is not ideal, but it seems the best that can be done. The issue with it is, since all writers will be incrementing the last vector clock they saw, there's no way to tell when one writer made a write significantly later in time than another, so the earlier write might arbitrarily be picked when merging. This problem is why git-annex uses timestamps in the first place, rather than pure vector clocks.) Advancing forward by 1 second is somewhat arbitrary. setDead advances a timestamp by just 1 picosecond, and the vector clock could too. But then it would interfere with setDead, which wants to be overrulled by any change. So it could use 2 picoseconds or something, but that seems weird. It could just as well advance it forward by a minute or whatever, but then it would be harder for real time to catch up with the vector clock when forward clock slew had happened. A complication is that many log files contain several different peices of information, and it may be best to only use vector clocks for the same peice of information. For example, a key's location log file contains InfoPresent/InfoMissing for each UUID, and it only looks at the vector clocks for the UUID that is being changed, and not other UUIDs. Although exactly where the dividing line is can be hard to determine. Consider metadata logs, where a field "tag" can have multiple values set at different times. Should it advance forward past the last tag? Probably. What about when a different field is set, should it look at the clocks of other fields? Perhaps not, but currently it does, and this does not seems like it will cause any problems. Another one I'm not entirely sure about is the export log, which is keyed by (fromuuid, touuid). So if multiple repos are exporting to the same remote, different vector clocks can be used for that remote. It looks like that's probably ok, because it does not try to determine what order things occurred when there was an export conflict. Sponsored-by: Jochen Bartl on Patreon
2021-08-03 20:45:20 +00:00
addTransition :: CandidateVectorClock -> Transition -> Transitions -> Transitions
addTransition c t s = S.insert (TransitionLine c' (Right t)) s
where
oldcs = map transitionStarted $
filter (\l -> transition l == Right t) (S.toList s)
c' = advanceVectorClock c oldcs
2013-08-28 17:19:02 +00:00
buildTransitions :: Transitions -> Builder
buildTransitions = mconcat . map genline . S.elems
where
genline tl = buildt (transition tl) <> charUtf8 ' '
<> buildVectorClock (transitionStarted tl) <> charUtf8 '\n'
buildt (Left b) = byteString b
buildt (Right t) = byteString (encodeBS (show t))
parseTransitions :: L.ByteString -> Transitions
parseTransitions = fromMaybe S.empty . A.maybeResult . A.parse
(S.fromList <$> parseLogLines transitionLineParser)
parseTransitionsStrictly :: String -> L.ByteString -> Transitions
parseTransitionsStrictly source b =
let ts = parseTransitions b
in if S.null $ S.filter (isLeft . transition) ts
then ts
else giveup $ "unknown transitions listed in " ++ source ++ "; upgrade git-annex!"
transitionLineParser :: A.Parser TransitionLine
transitionLineParser = do
t <- (parsetransition <$> A.takeByteString)
_ <- A8.char ' '
c <- vectorClockParser
return $ TransitionLine c t
2013-08-28 17:19:02 +00:00
where
parsetransition b = case readish (decodeBS b) of
Just t -> Right t
Nothing -> Left b
2013-08-28 17:19:02 +00:00
combineTransitions :: [Transitions] -> Transitions
combineTransitions = S.unions
{- Unknown types of transitions are omitted. -}
knownTransitionList :: Transitions -> [Transition]
knownTransitionList = nub . rights . map transition . S.elems
2013-08-28 17:19:02 +00:00
{- Typically ran with Annex.Branch.change, but we can't import Annex.Branch
- here since it depends on this module. -}
recordTransitions :: (RawFilePath -> (L.ByteString -> Builder) -> Annex ()) -> Transitions -> Annex ()
2013-09-26 03:19:01 +00:00
recordTransitions changer t = changer transitionsLog $
buildTransitions . S.union t . parseTransitionsStrictly "local"
2021-12-28 16:15:51 +00:00
getRefTransitions :: Git.Ref -> Annex Transitions
getRefTransitions ref =
parseTransitionsStrictly (fromRef ref)
2021-12-28 16:15:51 +00:00
<$> catFile ref transitionsLog