cndict 0.7.4 → 0.7.5

raw patch · 4 files changed

+19/−4 lines, 4 files

Files

cndict.cabal +1/−1
data/dict.sorted too large to diff
src/Data/Chinese/CCDict.hs +6/−2
src/Data/Chinese/Segmentation.hs +12/−1

cndict.cabal view

@@ -2,7 +2,7 @@ -- documentation, see http://haskell.org/cabal/users-guide/  name:                cndict-version:             0.7.4+version:             0.7.5 synopsis:            Chinese/Mandarin <-> English dictionary, Chinese lexer. -- description: license:             PublicDomain

data/dict.sorted view

file too large to diff

src/Data/Chinese/CCDict.hs view

@@ -119,6 +119,12 @@     where       (lower, upper) = bounds ccDict +allVariants :: [Variant]+allVariants = worker 0+  where+    worker nth | nth > snd (bounds ccDict) = []+    worker nth = parseVariant (ccDictNth nth ccDict) : worker (nth+1)+ scrapeEntry :: CCDict -> Int -> Text -> Maybe Entry scrapeEntry dict nth key =     case variants of@@ -233,5 +239,3 @@ -- /first/second/third/ -> [first, second, third] splitDefinition :: Text -> [Text] splitDefinition = filter (not . T.null) . T.splitOn "/" . T.dropAround isSpace--

src/Data/Chinese/Segmentation.hs view

@@ -76,7 +76,10 @@       ,["这", "位子"]       ,["十", "分钟"]       ,["一", "点钟"]-      ,["合上", "书"]]+      ,["合上", "书"]+      ,["第二", "天性"]+      ,["都", "会"]+      ,["上", "都"]]  greedyGroups :: [[[Token]]] -> [[[Token]]] greedyGroups = map worker@@ -216,6 +219,14 @@         , ("我合上书准备离开", "我 合上 书 准备 离开")         , ("他的话","他 的 话")         , ("你用什么方法学习","你 用 什么 方法 学习")+        , ("AAA","AAA")+        , ("BBB","BBB")++        , ("习惯是第二天性", "习惯 是 第二 天性")+        , ("一切都会好的", "一切 都 会 好 的")+        , ("上帝什么都会", "上帝 什么 都 会")+        -- , ("他比我高一个头", "他 比 我 高 一 个 头") -- What's the right way to tokenize here?+        , ("每张桌子上都有菜单", "每 张 桌子 上 都 有 菜单")         ]