cndict 0.4.7 → 0.4.8
raw patch · 2 files changed
+6/−2 lines, 2 filesPVP ok
version bump matches the API change (PVP)
API changes (from Hackage documentation)
Files
- cndict.cabal +1/−1
- src/Data/Chinese/CCDict.hs +5/−1
cndict.cabal view
@@ -2,7 +2,7 @@ -- documentation, see http://haskell.org/cabal/users-guide/ name: cndict-version: 0.4.7+version: 0.4.8 synopsis: Chinese/Mandarin <-> English dictionary, Chinese lexer. -- description: license: PublicDomain
src/Data/Chinese/CCDict.hs view
@@ -291,7 +291,11 @@ -- Enhanced tokenizer, mixed non-determistic and greedy algorithm tokenizer' :: CCDict -> Text -> [Token]-tokenizer' trie inp = collapseNonDet (tokenizerNondet trie inp)+tokenizer' trie inp = compress $ collapseNonDet (tokenizerNondet trie inp)+ where+ compress [] = []+ compress (UnknownWord a:UnknownWord b:xs) = compress (UnknownWord (a `T.append` b):xs)+ compress (x:xs) = x:compress xs tokenizerNondet :: CCDict -> Text -> [NonDet] tokenizerNondet trie inp = map _compactNonDet $ go inp