countable-inflections 0.1.0 → 0.2.0
raw patch · 5 files changed
+81/−78 lines, 5 filesdep +pcre-utilsdep +regex-pcre-builtindep −pcre-lightPVP ok
version bump matches the API change (PVP)
Dependencies added: pcre-utils, regex-pcre-builtin
Dependencies removed: pcre-light
API changes (from Hackage documentation)
Files
- README.md +14/−7
- Text/Countable.hs +17/−25
- Text/Countable/Data.hs +44/−44
- countable-inflections.cabal +3/−2
- test/Text/CountableSpec.hs +3/−0
README.md view
@@ -26,16 +26,18 @@ These can also be given custom inflection matchers -```+```haskell λ: :t singularizeWith [Inflection] -> Text -> Text ``` There are 3 different types of transformations: -Match (takes a PCRE regex and a replacement string)+### Match -```+Takes a PCRE regex and a replacement string.++```haskell λ: :t makeMatchMapping [(RegexPattern, RegexReplace)] -> [Inflection] @@ -44,9 +46,11 @@ "octopi" ``` -Irregular (from singular to plural with no greater pattern)+### Irregular -```+From singular to plural with no greater pattern.++```haskell λ: :t makeIrregularMapping [(Singular, Plural)] -> [Inflection] @@ -55,8 +59,11 @@ "zombies" ``` -Uncountable (doesn't have a mapping, word stays the same) so it has the type:-```+### Uncountable++Doesn't have a mapping, word stays the same) so it has the type:++```haskell [Text] -> [Inflection] ```
Text/Countable.hs view
@@ -22,12 +22,13 @@ ) where -import Data.Maybe (catMaybes, fromMaybe)-import Data.Monoid ((<>))+import Data.Maybe (catMaybes, fromMaybe, isNothing) import Data.Text as T import Data.Text.Encoding (decodeUtf8, encodeUtf8) import Text.Countable.Data-import Text.Regex.PCRE.Light+import Text.Regex.PCRE.ByteString+import Text.Regex.PCRE.ByteString.Utils (substitute')+import System.IO.Unsafe type RegexPattern = T.Text type RegexReplace = T.Text@@ -38,6 +39,8 @@ = Simple (Singular, Plural) | Match (Maybe Regex, RegexReplace) +data MatchType = HasMatch | HasNoMatch+ -- | pluralize a word given a default mapping pluralize :: T.Text -> T.Text pluralize = pluralizeWith mapping@@ -108,34 +111,23 @@ runSub (Just reg, rep) t = matchWithReplace (reg, rep) t matchWithReplace :: (Regex, RegexReplace) -> T.Text -> Maybe T.Text-matchWithReplace (reg, rep) t = do- matched : grouping <- regexMatch t reg- return $ (t `without` matched) <> groupReplace grouping rep+matchWithReplace (reg, rep) t = case regexMatch t reg of+ HasNoMatch -> Nothing+ HasMatch -> toMaybe $ substitute' reg (encodeUtf8 t) (encodeUtf8 rep) where- without t1 t2 = if t2 == "" then t1 else T.replace t2 "" t1--groupReplace :: [T.Text] -> T.Text -> T.Text-groupReplace g rep =- case g of- [] -> rep- (g':_) -> g' <> additions- where- rep' = tailMaybe $ T.split ((==) '\1') rep- additions = T.concat $ fromMaybe [] rep'+ toMaybe = either (const Nothing) (Just . decodeUtf8) -regexMatch :: T.Text -> Regex -> Maybe [T.Text]-regexMatch t r = do- bs <- match r (encodeUtf8 t) []- return $ fmap decodeUtf8 bs+regexMatch :: T.Text -> Regex -> MatchType+regexMatch t r = case match of+ Left _ -> HasNoMatch+ Right m -> if isNothing m then HasNoMatch else HasMatch+ where match = unsafePerformIO $ execute r (encodeUtf8 t) regexPattern :: T.Text -> Maybe Regex-regexPattern pat = toMaybe $ compileM (encodeUtf8 pat) [caseless]+regexPattern pat = toMaybe reg where toMaybe = either (const Nothing) Just+ reg = unsafePerformIO $ compile compCaseless execBlank (encodeUtf8 pat) headMaybe :: [a] -> Maybe a headMaybe [] = Nothing headMaybe (x:_) = Just x--tailMaybe :: [a] -> Maybe [a]-tailMaybe [] = Nothing-tailMaybe (_:xs) = Just xs
Text/Countable/Data.hs view
@@ -20,25 +20,25 @@ defaultPlurals' = [ ("$", "s") , ("s$", "s")- , ("^(ax|test)is$", "\1es")- , ("(octop|vir)us$", "\1i")- , ("(octop|vir)i$", "\1i")- , ("(alias|status)$", "\1es")- , ("(bu)s$", "\1ses")- , ("(buffal|tomat)o$", "\1oes")- , ("([ti])um$", "\1a")- , ("([ti])a$", "\1a")+ , ("^(ax|test)is$", "\\1es")+ , ("(octop|vir)us$", "\\1i")+ , ("(octop|vir)i$", "\\1i")+ , ("(alias|status)$", "\\1es")+ , ("(bu)s$", "\\1ses")+ , ("(buffal|tomat)o$", "\\1oes")+ , ("([ti])um$", "\\1a")+ , ("([ti])a$", "\\1a") , ("sis$", "ses")- , ("(?:([^f])fe|([lr])f)$", "\1\2ves")- , ("(hive)$", "\1s")- , ("([^aeiouy]|qu)y$", "\1ies")- , ("(x|ch|ss|sh)$", "\1es")- , ("(matr|vert|ind)(?:ix|ex)$", "\1ices")- , ("^(m|l)ouse$", "\1ice")- , ("^(m|l)ice$", "\1ice")- , ("^(ox)$", "\1en")- , ("^(oxen)$", "\1")- , ("(quiz)$", "\1zes")+ , ("(?:([^f])fe|([lr])f)$", "\\1\2ves")+ , ("(hive)$", "\\1s")+ , ("([^aeiouy]|qu)y$", "\\1ies")+ , ("(x|ch|ss|sh)$", "\\1es")+ , ("(matr|vert|ind)(?:ix|ex)$", "\\1ices")+ , ("^(m|l)ouse$", "\\1ice")+ , ("^(m|l)ice$", "\\1ice")+ , ("^(ox)$", "\\1en")+ , ("^(oxen)$", "\\1")+ , ("(quiz)$", "\\1zes") ] -- |These default inflections stolen from the Ruby inflection library - see@@ -46,32 +46,32 @@ defaultSingulars' :: [(Text, Text)] defaultSingulars' = [ ("s$", "")- , ("(ss)$", "\1")- , ("(n)ews$", "\1ews")- , ("([ti])a$", "\1um")- , ("((a)naly|(b)a|(d)iagno|(p)arenthe|(p)rogno|(s)ynop|(t)he)(sis|ses)$", "\1sis")- , ("(^analy)(sis|ses)$", "\1sis")- , ("([^f])ves$", "\1fe")- , ("(hive)s$", "\1")- , ("(tive)s$", "\1")- , ("([lr])ves$", "\1f")- , ("([^aeiouy]|qu)ies$", "\1y")- , ("(s)eries$", "\1eries")- , ("(m)ovies$", "\1ovie")- , ("(x|ch|ss|sh)es$", "\1")- , ("^(m|l)ice$", "\1ouse")- , ("(bus)(es)?$", "\1")- , ("(o)es$", "\1")- , ("(shoe)s$", "\1")- , ("(cris|test)(is|es)$", "\1is")- , ("^(a)x[ie]s$", "\1xis")- , ("(octop|vir)(us|i)$", "\1us")- , ("(alias|status)(es)?$", "\1")- , ("^(ox)en", "\1")- , ("(vert|ind)ices$", "\1ex")- , ("(matr)ices$", "\1ix")- , ("(quiz)zes$", "\1")- , ("(database)s$", "\1")+ , ("(ss)$", "\\1")+ , ("(n)ews$", "\\1ews")+ , ("([ti])a$", "\\1um")+ , ("((a)naly|(b)a|(d)iagno|(p)arenthe|(p)rogno|(s)ynop|(t)he)(sis|ses)$", "\\1sis")+ , ("(^analy)(sis|ses)$", "\\1sis")+ , ("([^f])ves$", "\\1fe")+ , ("(hive)s$", "\\1")+ , ("(tive)s$", "\\1")+ , ("([lr])ves$", "\\1f")+ , ("([^aeiouy]|qu)ies$", "\\1y")+ , ("(s)eries$", "\\1eries")+ , ("(m)ovies$", "\\1ovie")+ , ("(x|ch|ss|sh)es$", "\\1")+ , ("^(m|l)ice$", "\\1ouse")+ , ("(bus)(es)?$", "\\1")+ , ("(o)es$", "\\1")+ , ("(shoe)s$", "\\1")+ , ("(cris|test)(is|es)$", "\\1is")+ , ("^(a)x[ie]s$", "\\1xis")+ , ("(octop|vir)(us|i)$", "\\1us")+ , ("(alias|status)(es)?$", "\\1")+ , ("^(ox)en", "\\1")+ , ("(vert|ind)ices$", "\\1ex")+ , ("(matr)ices$", "\\1ix")+ , ("(quiz)zes$", "\\1")+ , ("(database)s$", "\\1") ] -- |These default irregular inflections stolen from the Ruby inflection library - see
countable-inflections.cabal view
@@ -1,5 +1,5 @@ name: countable-inflections-version: 0.1.0+version: 0.2.0 synopsis: Countable Text Inflections description: Provides methods for singularizing and pluralizing text.@@ -35,7 +35,8 @@ base >= 4.6 && < 4.10 , exceptions >= 0.6 && < 0.9 , text >= 0.2 && < 1.3- , pcre-light+ , regex-pcre-builtin+ , pcre-utils , bytestring default-language: Haskell2010
test/Text/CountableSpec.hs view
@@ -36,6 +36,9 @@ singularize "mice" `shouldBe` "mouse" singularize "oxen" `shouldBe` "ox" singularize "branches" `shouldBe` "branch"+ singularize "transactions" `shouldBe` "transaction"+ singularize "sessions" `shouldBe` "session"+ singularize "users" `shouldBe` "user" irregularCases :: Spec