git-annex/Key.hs

199 lines
5.8 KiB
Haskell
Raw Normal View History

2017-02-24 17:42:30 +00:00
{- git-annex Keys
-
convert key2file and file2key to use builder and attoparsec The new parser is significantly stricter than the old one: The old file2key allowed the fields to come in any order, but the new one requires the fixed order that git-annex has always used. Hopefully this will not cause any breakage. And the old file2key allowed eg SHA1-m1-m2-m3-m4-m5-m6--xxxx while the new does not allow duplication of fields. This could potentially improve security, because allowing lots of extra junk like that in a key could potentially be used in a SHA1 collision attack, although the current attacks need binary data and not this kind of structured numeric data. Speed improved of course, and fairly substantially, in microbenchmarks: benchmarking old/key2file time 2.264 μs (2.257 μs .. 2.273 μs) 1.000 R² (1.000 R² .. 1.000 R²) mean 2.265 μs (2.260 μs .. 2.275 μs) std dev 21.17 ns (13.06 ns .. 39.26 ns) benchmarking new/key2file' time 1.744 μs (1.741 μs .. 1.747 μs) 1.000 R² (1.000 R² .. 1.000 R²) mean 1.745 μs (1.742 μs .. 1.751 μs) std dev 13.55 ns (9.099 ns .. 21.89 ns) benchmarking old/file2key time 6.114 μs (6.102 μs .. 6.129 μs) 1.000 R² (1.000 R² .. 1.000 R²) mean 6.118 μs (6.106 μs .. 6.143 μs) std dev 55.00 ns (30.08 ns .. 100.2 ns) benchmarking new/file2key' time 1.791 μs (1.782 μs .. 1.801 μs) 1.000 R² (0.999 R² .. 1.000 R²) mean 1.792 μs (1.785 μs .. 1.804 μs) std dev 32.46 ns (20.59 ns .. 50.82 ns) variance introduced by outliers: 19% (moderately inflated)
2019-01-11 20:33:42 +00:00
- Copyright 2011-2019 Joey Hess <id@joeyh.name>
2017-02-24 17:42:30 +00:00
-
- Licensed under the GNU AGPL version 3 or higher.
2017-02-24 17:42:30 +00:00
-}
{-# OPTIONS_GHC -fno-warn-orphans #-}
module Key (
Key(..),
AssociatedFile(..),
2017-02-24 17:42:30 +00:00
stubKey,
buildKey,
keyParser,
serializeKey,
2019-01-14 17:17:47 +00:00
serializeKey',
deserializeKey,
deserializeKey',
2017-02-24 17:42:30 +00:00
nonChunkKey,
chunkKeyOffset,
isChunkKey,
isKeyPrefix,
convert key2file and file2key to use builder and attoparsec The new parser is significantly stricter than the old one: The old file2key allowed the fields to come in any order, but the new one requires the fixed order that git-annex has always used. Hopefully this will not cause any breakage. And the old file2key allowed eg SHA1-m1-m2-m3-m4-m5-m6--xxxx while the new does not allow duplication of fields. This could potentially improve security, because allowing lots of extra junk like that in a key could potentially be used in a SHA1 collision attack, although the current attacks need binary data and not this kind of structured numeric data. Speed improved of course, and fairly substantially, in microbenchmarks: benchmarking old/key2file time 2.264 μs (2.257 μs .. 2.273 μs) 1.000 R² (1.000 R² .. 1.000 R²) mean 2.265 μs (2.260 μs .. 2.275 μs) std dev 21.17 ns (13.06 ns .. 39.26 ns) benchmarking new/key2file' time 1.744 μs (1.741 μs .. 1.747 μs) 1.000 R² (1.000 R² .. 1.000 R²) mean 1.745 μs (1.742 μs .. 1.751 μs) std dev 13.55 ns (9.099 ns .. 21.89 ns) benchmarking old/file2key time 6.114 μs (6.102 μs .. 6.129 μs) 1.000 R² (1.000 R² .. 1.000 R²) mean 6.118 μs (6.106 μs .. 6.143 μs) std dev 55.00 ns (30.08 ns .. 100.2 ns) benchmarking new/file2key' time 1.791 μs (1.782 μs .. 1.801 μs) 1.000 R² (0.999 R² .. 1.000 R²) mean 1.792 μs (1.785 μs .. 1.804 μs) std dev 32.46 ns (20.59 ns .. 50.82 ns) variance introduced by outliers: 19% (moderately inflated)
2019-01-11 20:33:42 +00:00
splitKeyNameExtension,
2017-02-24 17:42:30 +00:00
prop_isomorphic_key_encode
2017-02-24 17:42:30 +00:00
) where
import qualified Data.Text as T
convert key2file and file2key to use builder and attoparsec The new parser is significantly stricter than the old one: The old file2key allowed the fields to come in any order, but the new one requires the fixed order that git-annex has always used. Hopefully this will not cause any breakage. And the old file2key allowed eg SHA1-m1-m2-m3-m4-m5-m6--xxxx while the new does not allow duplication of fields. This could potentially improve security, because allowing lots of extra junk like that in a key could potentially be used in a SHA1 collision attack, although the current attacks need binary data and not this kind of structured numeric data. Speed improved of course, and fairly substantially, in microbenchmarks: benchmarking old/key2file time 2.264 μs (2.257 μs .. 2.273 μs) 1.000 R² (1.000 R² .. 1.000 R²) mean 2.265 μs (2.260 μs .. 2.275 μs) std dev 21.17 ns (13.06 ns .. 39.26 ns) benchmarking new/key2file' time 1.744 μs (1.741 μs .. 1.747 μs) 1.000 R² (1.000 R² .. 1.000 R²) mean 1.745 μs (1.742 μs .. 1.751 μs) std dev 13.55 ns (9.099 ns .. 21.89 ns) benchmarking old/file2key time 6.114 μs (6.102 μs .. 6.129 μs) 1.000 R² (1.000 R² .. 1.000 R²) mean 6.118 μs (6.106 μs .. 6.143 μs) std dev 55.00 ns (30.08 ns .. 100.2 ns) benchmarking new/file2key' time 1.791 μs (1.782 μs .. 1.801 μs) 1.000 R² (0.999 R² .. 1.000 R²) mean 1.792 μs (1.785 μs .. 1.804 μs) std dev 32.46 ns (20.59 ns .. 50.82 ns) variance introduced by outliers: 19% (moderately inflated)
2019-01-11 20:33:42 +00:00
import qualified Data.ByteString as S
import qualified Data.ByteString.Char8 as S8
import qualified Data.ByteString.Lazy as L
import Data.ByteString.Builder
import Data.ByteString.Builder.Extra
import qualified Data.Attoparsec.ByteString as A
import qualified Data.Attoparsec.ByteString.Char8 as A8
import Foreign.C.Types
2017-02-24 17:42:30 +00:00
import Common
import Types.Key
import Utility.QuickCheck
import Utility.Bloom
import Utility.Aeson
2017-02-24 17:42:30 +00:00
import qualified Utility.SimpleProtocol as Proto
stubKey :: Key
stubKey = Key
convert key2file and file2key to use builder and attoparsec The new parser is significantly stricter than the old one: The old file2key allowed the fields to come in any order, but the new one requires the fixed order that git-annex has always used. Hopefully this will not cause any breakage. And the old file2key allowed eg SHA1-m1-m2-m3-m4-m5-m6--xxxx while the new does not allow duplication of fields. This could potentially improve security, because allowing lots of extra junk like that in a key could potentially be used in a SHA1 collision attack, although the current attacks need binary data and not this kind of structured numeric data. Speed improved of course, and fairly substantially, in microbenchmarks: benchmarking old/key2file time 2.264 μs (2.257 μs .. 2.273 μs) 1.000 R² (1.000 R² .. 1.000 R²) mean 2.265 μs (2.260 μs .. 2.275 μs) std dev 21.17 ns (13.06 ns .. 39.26 ns) benchmarking new/key2file' time 1.744 μs (1.741 μs .. 1.747 μs) 1.000 R² (1.000 R² .. 1.000 R²) mean 1.745 μs (1.742 μs .. 1.751 μs) std dev 13.55 ns (9.099 ns .. 21.89 ns) benchmarking old/file2key time 6.114 μs (6.102 μs .. 6.129 μs) 1.000 R² (1.000 R² .. 1.000 R²) mean 6.118 μs (6.106 μs .. 6.143 μs) std dev 55.00 ns (30.08 ns .. 100.2 ns) benchmarking new/file2key' time 1.791 μs (1.782 μs .. 1.801 μs) 1.000 R² (0.999 R² .. 1.000 R²) mean 1.792 μs (1.785 μs .. 1.804 μs) std dev 32.46 ns (20.59 ns .. 50.82 ns) variance introduced by outliers: 19% (moderately inflated)
2019-01-11 20:33:42 +00:00
{ keyName = mempty
, keyVariety = OtherKey mempty
2017-02-24 17:42:30 +00:00
, keySize = Nothing
, keyMtime = Nothing
, keyChunkSize = Nothing
, keyChunkNum = Nothing
}
-- Gets the parent of a chunk key.
nonChunkKey :: Key -> Key
nonChunkKey k = k
{ keyChunkSize = Nothing
, keyChunkNum = Nothing
}
2017-02-24 17:42:30 +00:00
-- Where a chunk key is offset within its parent.
chunkKeyOffset :: Key -> Maybe Integer
chunkKeyOffset k = (*)
<$> keyChunkSize k
<*> (pred <$> keyChunkNum k)
isChunkKey :: Key -> Bool
isChunkKey k = isJust (keyChunkSize k) && isJust (keyChunkNum k)
-- Checks if a string looks like at least the start of a key.
isKeyPrefix :: String -> Bool
isKeyPrefix s = [fieldSep, fieldSep] `isInfixOf` s
fieldSep :: Char
fieldSep = '-'
{- Builds a ByteString from a Key.
-
- The name field is always shown last, separated by doubled fieldSeps,
- and is the only field allowed to contain the fieldSep.
-}
buildKey :: Key -> Builder
buildKey k = byteString (formatKeyVariety (keyVariety k))
convert key2file and file2key to use builder and attoparsec The new parser is significantly stricter than the old one: The old file2key allowed the fields to come in any order, but the new one requires the fixed order that git-annex has always used. Hopefully this will not cause any breakage. And the old file2key allowed eg SHA1-m1-m2-m3-m4-m5-m6--xxxx while the new does not allow duplication of fields. This could potentially improve security, because allowing lots of extra junk like that in a key could potentially be used in a SHA1 collision attack, although the current attacks need binary data and not this kind of structured numeric data. Speed improved of course, and fairly substantially, in microbenchmarks: benchmarking old/key2file time 2.264 μs (2.257 μs .. 2.273 μs) 1.000 R² (1.000 R² .. 1.000 R²) mean 2.265 μs (2.260 μs .. 2.275 μs) std dev 21.17 ns (13.06 ns .. 39.26 ns) benchmarking new/key2file' time 1.744 μs (1.741 μs .. 1.747 μs) 1.000 R² (1.000 R² .. 1.000 R²) mean 1.745 μs (1.742 μs .. 1.751 μs) std dev 13.55 ns (9.099 ns .. 21.89 ns) benchmarking old/file2key time 6.114 μs (6.102 μs .. 6.129 μs) 1.000 R² (1.000 R² .. 1.000 R²) mean 6.118 μs (6.106 μs .. 6.143 μs) std dev 55.00 ns (30.08 ns .. 100.2 ns) benchmarking new/file2key' time 1.791 μs (1.782 μs .. 1.801 μs) 1.000 R² (0.999 R² .. 1.000 R²) mean 1.792 μs (1.785 μs .. 1.804 μs) std dev 32.46 ns (20.59 ns .. 50.82 ns) variance introduced by outliers: 19% (moderately inflated)
2019-01-11 20:33:42 +00:00
<> 's' ?: (integerDec <$> keySize k)
<> 'm' ?: (integerDec . (\(CTime t) -> fromIntegral t) <$> keyMtime k)
<> 'S' ?: (integerDec <$> keyChunkSize k)
<> 'C' ?: (integerDec <$> keyChunkNum k)
<> sepbefore (sepbefore (byteString (keyName k)))
where
sepbefore s = char7 fieldSep <> s
c ?: (Just b) = sepbefore (char7 c <> b)
_ ?: Nothing = mempty
serializeKey :: Key -> String
serializeKey = decodeBL' . serializeKey'
convert key2file and file2key to use builder and attoparsec The new parser is significantly stricter than the old one: The old file2key allowed the fields to come in any order, but the new one requires the fixed order that git-annex has always used. Hopefully this will not cause any breakage. And the old file2key allowed eg SHA1-m1-m2-m3-m4-m5-m6--xxxx while the new does not allow duplication of fields. This could potentially improve security, because allowing lots of extra junk like that in a key could potentially be used in a SHA1 collision attack, although the current attacks need binary data and not this kind of structured numeric data. Speed improved of course, and fairly substantially, in microbenchmarks: benchmarking old/key2file time 2.264 μs (2.257 μs .. 2.273 μs) 1.000 R² (1.000 R² .. 1.000 R²) mean 2.265 μs (2.260 μs .. 2.275 μs) std dev 21.17 ns (13.06 ns .. 39.26 ns) benchmarking new/key2file' time 1.744 μs (1.741 μs .. 1.747 μs) 1.000 R² (1.000 R² .. 1.000 R²) mean 1.745 μs (1.742 μs .. 1.751 μs) std dev 13.55 ns (9.099 ns .. 21.89 ns) benchmarking old/file2key time 6.114 μs (6.102 μs .. 6.129 μs) 1.000 R² (1.000 R² .. 1.000 R²) mean 6.118 μs (6.106 μs .. 6.143 μs) std dev 55.00 ns (30.08 ns .. 100.2 ns) benchmarking new/file2key' time 1.791 μs (1.782 μs .. 1.801 μs) 1.000 R² (0.999 R² .. 1.000 R²) mean 1.792 μs (1.785 μs .. 1.804 μs) std dev 32.46 ns (20.59 ns .. 50.82 ns) variance introduced by outliers: 19% (moderately inflated)
2019-01-11 20:33:42 +00:00
serializeKey' :: Key -> L.ByteString
serializeKey' = toLazyByteStringWith (safeStrategy 128 smallChunkSize) L.empty . buildKey
convert key2file and file2key to use builder and attoparsec The new parser is significantly stricter than the old one: The old file2key allowed the fields to come in any order, but the new one requires the fixed order that git-annex has always used. Hopefully this will not cause any breakage. And the old file2key allowed eg SHA1-m1-m2-m3-m4-m5-m6--xxxx while the new does not allow duplication of fields. This could potentially improve security, because allowing lots of extra junk like that in a key could potentially be used in a SHA1 collision attack, although the current attacks need binary data and not this kind of structured numeric data. Speed improved of course, and fairly substantially, in microbenchmarks: benchmarking old/key2file time 2.264 μs (2.257 μs .. 2.273 μs) 1.000 R² (1.000 R² .. 1.000 R²) mean 2.265 μs (2.260 μs .. 2.275 μs) std dev 21.17 ns (13.06 ns .. 39.26 ns) benchmarking new/key2file' time 1.744 μs (1.741 μs .. 1.747 μs) 1.000 R² (1.000 R² .. 1.000 R²) mean 1.745 μs (1.742 μs .. 1.751 μs) std dev 13.55 ns (9.099 ns .. 21.89 ns) benchmarking old/file2key time 6.114 μs (6.102 μs .. 6.129 μs) 1.000 R² (1.000 R² .. 1.000 R²) mean 6.118 μs (6.106 μs .. 6.143 μs) std dev 55.00 ns (30.08 ns .. 100.2 ns) benchmarking new/file2key' time 1.791 μs (1.782 μs .. 1.801 μs) 1.000 R² (0.999 R² .. 1.000 R²) mean 1.792 μs (1.785 μs .. 1.804 μs) std dev 32.46 ns (20.59 ns .. 50.82 ns) variance introduced by outliers: 19% (moderately inflated)
2019-01-11 20:33:42 +00:00
{- This is a strict parser for security reasons; a key
- can contain only 4 fields, which all consist only of numbers.
- Any key containing other fields, or non-numeric data will fail
- to parse.
-
- If a key contained non-numeric fields, they could be used to
- embed data used in a SHA1 collision attack, which would be a
- problem since the keys are committed to git.
-}
keyParser :: A.Parser Key
keyParser = do
convert key2file and file2key to use builder and attoparsec The new parser is significantly stricter than the old one: The old file2key allowed the fields to come in any order, but the new one requires the fixed order that git-annex has always used. Hopefully this will not cause any breakage. And the old file2key allowed eg SHA1-m1-m2-m3-m4-m5-m6--xxxx while the new does not allow duplication of fields. This could potentially improve security, because allowing lots of extra junk like that in a key could potentially be used in a SHA1 collision attack, although the current attacks need binary data and not this kind of structured numeric data. Speed improved of course, and fairly substantially, in microbenchmarks: benchmarking old/key2file time 2.264 μs (2.257 μs .. 2.273 μs) 1.000 R² (1.000 R² .. 1.000 R²) mean 2.265 μs (2.260 μs .. 2.275 μs) std dev 21.17 ns (13.06 ns .. 39.26 ns) benchmarking new/key2file' time 1.744 μs (1.741 μs .. 1.747 μs) 1.000 R² (1.000 R² .. 1.000 R²) mean 1.745 μs (1.742 μs .. 1.751 μs) std dev 13.55 ns (9.099 ns .. 21.89 ns) benchmarking old/file2key time 6.114 μs (6.102 μs .. 6.129 μs) 1.000 R² (1.000 R² .. 1.000 R²) mean 6.118 μs (6.106 μs .. 6.143 μs) std dev 55.00 ns (30.08 ns .. 100.2 ns) benchmarking new/file2key' time 1.791 μs (1.782 μs .. 1.801 μs) 1.000 R² (0.999 R² .. 1.000 R²) mean 1.792 μs (1.785 μs .. 1.804 μs) std dev 32.46 ns (20.59 ns .. 50.82 ns) variance introduced by outliers: 19% (moderately inflated)
2019-01-11 20:33:42 +00:00
-- key variety cannot be empty
v <- (parseKeyVariety <$> A8.takeWhile1 (/= fieldSep))
s <- parsesize
m <- parsemtime
cs <- parsechunksize
cn <- parsechunknum
_ <- A8.char fieldSep
_ <- A8.char fieldSep
n <- A.takeByteString
if validKeyName v n
then return $ Key
{ keyName = n
, keyVariety = v
, keySize = s
, keyMtime = m
, keyChunkSize = cs
, keyChunkNum = cn
}
else fail "invalid keyName"
2017-02-24 17:42:30 +00:00
where
convert key2file and file2key to use builder and attoparsec The new parser is significantly stricter than the old one: The old file2key allowed the fields to come in any order, but the new one requires the fixed order that git-annex has always used. Hopefully this will not cause any breakage. And the old file2key allowed eg SHA1-m1-m2-m3-m4-m5-m6--xxxx while the new does not allow duplication of fields. This could potentially improve security, because allowing lots of extra junk like that in a key could potentially be used in a SHA1 collision attack, although the current attacks need binary data and not this kind of structured numeric data. Speed improved of course, and fairly substantially, in microbenchmarks: benchmarking old/key2file time 2.264 μs (2.257 μs .. 2.273 μs) 1.000 R² (1.000 R² .. 1.000 R²) mean 2.265 μs (2.260 μs .. 2.275 μs) std dev 21.17 ns (13.06 ns .. 39.26 ns) benchmarking new/key2file' time 1.744 μs (1.741 μs .. 1.747 μs) 1.000 R² (1.000 R² .. 1.000 R²) mean 1.745 μs (1.742 μs .. 1.751 μs) std dev 13.55 ns (9.099 ns .. 21.89 ns) benchmarking old/file2key time 6.114 μs (6.102 μs .. 6.129 μs) 1.000 R² (1.000 R² .. 1.000 R²) mean 6.118 μs (6.106 μs .. 6.143 μs) std dev 55.00 ns (30.08 ns .. 100.2 ns) benchmarking new/file2key' time 1.791 μs (1.782 μs .. 1.801 μs) 1.000 R² (0.999 R² .. 1.000 R²) mean 1.792 μs (1.785 μs .. 1.804 μs) std dev 32.46 ns (20.59 ns .. 50.82 ns) variance introduced by outliers: 19% (moderately inflated)
2019-01-11 20:33:42 +00:00
parseopt p = (Just <$> (A8.char fieldSep *> p)) <|> pure Nothing
parsesize = parseopt $ A8.char 's' *> A8.decimal
parsemtime = parseopt $ CTime <$> (A8.char 'm' *> A8.decimal)
parsechunksize = parseopt $ A8.char 'S' *> A8.decimal
parsechunknum = parseopt $ A8.char 'C' *> A8.decimal
2017-02-24 17:42:30 +00:00
deserializeKey :: String -> Maybe Key
deserializeKey = deserializeKey' . encodeBS'
convert key2file and file2key to use builder and attoparsec The new parser is significantly stricter than the old one: The old file2key allowed the fields to come in any order, but the new one requires the fixed order that git-annex has always used. Hopefully this will not cause any breakage. And the old file2key allowed eg SHA1-m1-m2-m3-m4-m5-m6--xxxx while the new does not allow duplication of fields. This could potentially improve security, because allowing lots of extra junk like that in a key could potentially be used in a SHA1 collision attack, although the current attacks need binary data and not this kind of structured numeric data. Speed improved of course, and fairly substantially, in microbenchmarks: benchmarking old/key2file time 2.264 μs (2.257 μs .. 2.273 μs) 1.000 R² (1.000 R² .. 1.000 R²) mean 2.265 μs (2.260 μs .. 2.275 μs) std dev 21.17 ns (13.06 ns .. 39.26 ns) benchmarking new/key2file' time 1.744 μs (1.741 μs .. 1.747 μs) 1.000 R² (1.000 R² .. 1.000 R²) mean 1.745 μs (1.742 μs .. 1.751 μs) std dev 13.55 ns (9.099 ns .. 21.89 ns) benchmarking old/file2key time 6.114 μs (6.102 μs .. 6.129 μs) 1.000 R² (1.000 R² .. 1.000 R²) mean 6.118 μs (6.106 μs .. 6.143 μs) std dev 55.00 ns (30.08 ns .. 100.2 ns) benchmarking new/file2key' time 1.791 μs (1.782 μs .. 1.801 μs) 1.000 R² (0.999 R² .. 1.000 R²) mean 1.792 μs (1.785 μs .. 1.804 μs) std dev 32.46 ns (20.59 ns .. 50.82 ns) variance introduced by outliers: 19% (moderately inflated)
2019-01-11 20:33:42 +00:00
deserializeKey' :: S.ByteString -> Maybe Key
deserializeKey' = eitherToMaybe . A.parseOnly keyParser
convert key2file and file2key to use builder and attoparsec The new parser is significantly stricter than the old one: The old file2key allowed the fields to come in any order, but the new one requires the fixed order that git-annex has always used. Hopefully this will not cause any breakage. And the old file2key allowed eg SHA1-m1-m2-m3-m4-m5-m6--xxxx while the new does not allow duplication of fields. This could potentially improve security, because allowing lots of extra junk like that in a key could potentially be used in a SHA1 collision attack, although the current attacks need binary data and not this kind of structured numeric data. Speed improved of course, and fairly substantially, in microbenchmarks: benchmarking old/key2file time 2.264 μs (2.257 μs .. 2.273 μs) 1.000 R² (1.000 R² .. 1.000 R²) mean 2.265 μs (2.260 μs .. 2.275 μs) std dev 21.17 ns (13.06 ns .. 39.26 ns) benchmarking new/key2file' time 1.744 μs (1.741 μs .. 1.747 μs) 1.000 R² (1.000 R² .. 1.000 R²) mean 1.745 μs (1.742 μs .. 1.751 μs) std dev 13.55 ns (9.099 ns .. 21.89 ns) benchmarking old/file2key time 6.114 μs (6.102 μs .. 6.129 μs) 1.000 R² (1.000 R² .. 1.000 R²) mean 6.118 μs (6.106 μs .. 6.143 μs) std dev 55.00 ns (30.08 ns .. 100.2 ns) benchmarking new/file2key' time 1.791 μs (1.782 μs .. 1.801 μs) 1.000 R² (0.999 R² .. 1.000 R²) mean 1.792 μs (1.785 μs .. 1.804 μs) std dev 32.46 ns (20.59 ns .. 50.82 ns) variance introduced by outliers: 19% (moderately inflated)
2019-01-11 20:33:42 +00:00
{- This splits any extension out of the keyName, returning the
- keyName minus extension, and the extension (including leading dot).
-}
splitKeyNameExtension :: Key -> (S.ByteString, S.ByteString)
splitKeyNameExtension = splitKeyNameExtension' . keyName
splitKeyNameExtension' :: S.ByteString -> (S.ByteString, S.ByteString)
splitKeyNameExtension' keyname = S8.span (/= '.') keyname
{- Limits the length of the extension in the keyName to mitigate against
- SHA1 collision attacks.
2017-02-24 17:42:30 +00:00
-
- In such an attack, the extension of the key could be made to contain
- the collision generation data, with the result that a signed git commit
- including such keys would not be secure.
2017-02-24 17:42:30 +00:00
-
- The maximum extension length ever generated for such a key was 8
- characters, but they may be unicode which could use up to 4 bytes each,
- so 32 bytes. 64 bytes is used here to give a little future wiggle-room.
- The SHA1 common-prefix attack needs 128 bytes of data.
2017-02-24 17:42:30 +00:00
-}
convert key2file and file2key to use builder and attoparsec The new parser is significantly stricter than the old one: The old file2key allowed the fields to come in any order, but the new one requires the fixed order that git-annex has always used. Hopefully this will not cause any breakage. And the old file2key allowed eg SHA1-m1-m2-m3-m4-m5-m6--xxxx while the new does not allow duplication of fields. This could potentially improve security, because allowing lots of extra junk like that in a key could potentially be used in a SHA1 collision attack, although the current attacks need binary data and not this kind of structured numeric data. Speed improved of course, and fairly substantially, in microbenchmarks: benchmarking old/key2file time 2.264 μs (2.257 μs .. 2.273 μs) 1.000 R² (1.000 R² .. 1.000 R²) mean 2.265 μs (2.260 μs .. 2.275 μs) std dev 21.17 ns (13.06 ns .. 39.26 ns) benchmarking new/key2file' time 1.744 μs (1.741 μs .. 1.747 μs) 1.000 R² (1.000 R² .. 1.000 R²) mean 1.745 μs (1.742 μs .. 1.751 μs) std dev 13.55 ns (9.099 ns .. 21.89 ns) benchmarking old/file2key time 6.114 μs (6.102 μs .. 6.129 μs) 1.000 R² (1.000 R² .. 1.000 R²) mean 6.118 μs (6.106 μs .. 6.143 μs) std dev 55.00 ns (30.08 ns .. 100.2 ns) benchmarking new/file2key' time 1.791 μs (1.782 μs .. 1.801 μs) 1.000 R² (0.999 R² .. 1.000 R²) mean 1.792 μs (1.785 μs .. 1.804 μs) std dev 32.46 ns (20.59 ns .. 50.82 ns) variance introduced by outliers: 19% (moderately inflated)
2019-01-11 20:33:42 +00:00
validKeyName :: KeyVariety -> S.ByteString -> Bool
validKeyName kv name
| hasExt kv =
let ext = snd $ splitKeyNameExtension' name
in S.length ext <= 64
2017-02-24 17:42:30 +00:00
| otherwise = True
instance Arbitrary Key where
arbitrary = Key
convert key2file and file2key to use builder and attoparsec The new parser is significantly stricter than the old one: The old file2key allowed the fields to come in any order, but the new one requires the fixed order that git-annex has always used. Hopefully this will not cause any breakage. And the old file2key allowed eg SHA1-m1-m2-m3-m4-m5-m6--xxxx while the new does not allow duplication of fields. This could potentially improve security, because allowing lots of extra junk like that in a key could potentially be used in a SHA1 collision attack, although the current attacks need binary data and not this kind of structured numeric data. Speed improved of course, and fairly substantially, in microbenchmarks: benchmarking old/key2file time 2.264 μs (2.257 μs .. 2.273 μs) 1.000 R² (1.000 R² .. 1.000 R²) mean 2.265 μs (2.260 μs .. 2.275 μs) std dev 21.17 ns (13.06 ns .. 39.26 ns) benchmarking new/key2file' time 1.744 μs (1.741 μs .. 1.747 μs) 1.000 R² (1.000 R² .. 1.000 R²) mean 1.745 μs (1.742 μs .. 1.751 μs) std dev 13.55 ns (9.099 ns .. 21.89 ns) benchmarking old/file2key time 6.114 μs (6.102 μs .. 6.129 μs) 1.000 R² (1.000 R² .. 1.000 R²) mean 6.118 μs (6.106 μs .. 6.143 μs) std dev 55.00 ns (30.08 ns .. 100.2 ns) benchmarking new/file2key' time 1.791 μs (1.782 μs .. 1.801 μs) 1.000 R² (0.999 R² .. 1.000 R²) mean 1.792 μs (1.785 μs .. 1.804 μs) std dev 32.46 ns (20.59 ns .. 50.82 ns) variance introduced by outliers: 19% (moderately inflated)
2019-01-11 20:33:42 +00:00
<$> (encodeBS <$> (listOf1 $ elements $ ['A'..'Z'] ++ ['a'..'z'] ++ ['0'..'9'] ++ "-_\r\n \t"))
<*> (parseKeyVariety . encodeBS <$> (listOf1 $ elements ['A'..'Z'])) -- BACKEND
2017-02-24 17:42:30 +00:00
<*> ((abs <$>) <$> arbitrary) -- size cannot be negative
<*> ((abs . fromInteger <$>) <$> arbitrary) -- mtime cannot be negative
2017-02-24 17:42:30 +00:00
<*> ((abs <$>) <$> arbitrary) -- chunksize cannot be negative
<*> ((succ . abs <$>) <$> arbitrary) -- chunknum cannot be 0 or negative
instance Hashable Key where
2019-01-14 17:17:47 +00:00
hashIO32 = hashIO32 . serializeKey'
hashIO64 = hashIO64 . serializeKey'
2017-02-24 17:42:30 +00:00
instance ToJSON' Key where
toJSON' = toJSON' . serializeKey
2017-02-24 17:42:30 +00:00
instance FromJSON Key where
parseJSON (String t) = maybe mempty pure $ deserializeKey $ T.unpack t
2017-02-24 17:42:30 +00:00
parseJSON _ = mempty
instance Proto.Serializable Key where
serialize = serializeKey
deserialize = deserializeKey
2017-02-24 17:42:30 +00:00
prop_isomorphic_key_encode :: Key -> Bool
prop_isomorphic_key_encode k = Just k == (deserializeKey . serializeKey) k