Merge pull request #183 from purescript-contrib/string-parser-mv

Move parsers from `Parsing.String` to `Parsing.String.Basic`

Author: jamesdbrock

CHANGELOG.md

@@ -29,6 +29,14 @@ Breaking changes:
 - Replace the `regex` parser. (#170 by @jamesdbrock)
 - Reorganize Combinators for #154 (#182 by @jamesdbrock)
 - Add the `index` field to `Position`. (#171 by @jamesdbrock)
+- Move the parsers
+  * `whiteSpace`
+  * `skipSpaces`
+  * `oneOf`
+  * `oneOfCodePoints`
+  * `noneOf`
+  * `noneOfCodePoints`
+  from `Parsing.String` to `Parsing.String.Basic`. (#183 by @jamesdbrock)
 
 New features:
 

bench/Json/Parsing.purs

@@ -12,7 +12,8 @@ import Data.String.Regex.Flags (noFlags)
 import Data.Tuple (Tuple(..))
 import Parsing (ParserT, fail)
 import Parsing.Combinators (between, choice, sepBy, try)
-import Parsing.String (regex, skipSpaces, string)
+import Parsing.String (regex, string)
+import Parsing.String.Basic (skipSpaces)
 import Partial.Unsafe (unsafeCrashWith)
 
 json :: forall m. Monad m => ParserT String m Json

src/Parsing/Indent.purs

@@ -69,7 +69,8 @@ import Data.Maybe (Maybe(..))
 import Parsing (ParserT, fail, position)
 import Parsing.Combinators (option, optionMaybe)
 import Parsing.Pos (Position(..), initialPos)
-import Parsing.String (oneOf, string)
+import Parsing.String (string)
+import Parsing.String.Basic (oneOf)
 
 -- | Indentation sensitive parser type. Usually @ m @ will
 -- | be @ Identity @ as with any @ ParserT @

src/Parsing/Language.purs

@@ -13,8 +13,8 @@ import Prelude
 
 import Control.Alt ((<|>))
 import Parsing (ParserT)
-import Parsing.String (char, oneOf)
-import Parsing.String.Basic (alphaNum, letter)
+import Parsing.String (char)
+import Parsing.String.Basic (alphaNum, letter, oneOf)
 import Parsing.Token (GenLanguageDef(..), LanguageDef, TokenParser, makeTokenParser, unGenLanguageDef)
 
 -----------------------------------------------------------

src/Parsing/String.purs

@@ -34,21 +34,15 @@
 -- | - carriage-return (`0x0D`)
 -- | - carriage-return-newline (`0x0D 0x0A`)
 module Parsing.String
-  ( string
-  , eof
-  , rest
+  ( char
+  , string
   , anyChar
   , anyCodePoint
   , satisfy
   , satisfyCodePoint
-  , char
   , takeN
-  , whiteSpace
-  , skipSpaces
-  , oneOf
-  , oneOfCodePoints
-  , noneOf
-  , noneOfCodePoints
+  , rest
+  , eof
   , match
   , regex
   , consumeWith
@@ -57,22 +51,20 @@ module Parsing.String
 import Prelude hiding (between)
 
 import Control.Monad.State (get)
-import Data.Array (elem, notElem)
 import Data.Array.NonEmpty as NonEmptyArray
-import Data.CodePoint.Unicode (isSpace)
 import Data.Either (Either(..))
 import Data.Enum (fromEnum, toEnum)
 import Data.Function.Uncurried (mkFn5, runFn2)
 import Data.Maybe (Maybe(..), fromJust)
-import Data.String (CodePoint, Pattern(..), codePointAt, length, null, singleton, splitAt, stripPrefix, takeWhile, uncons)
+import Data.String (CodePoint, Pattern(..), codePointAt, length, null, splitAt, stripPrefix, uncons)
 import Data.String as String
 import Data.String.CodeUnits as SCU
 import Data.String.Regex as Regex
 import Data.String.Regex.Flags (RegexFlags)
-import Data.Tuple (Tuple(..), fst)
+import Data.Tuple (Tuple(..))
 import Partial.Unsafe (unsafePartial)
 import Parsing (ParseError(..), ParseState(..), ParserT(..))
-import Parsing.Combinators ((<?>), (<~?>))
+import Parsing.Combinators ((<?>))
 import Parsing.Pos (Position(..))
 
 -- | Match “end-of-file,” the end of the input stream.
@@ -159,34 +151,6 @@ takeN n = consumeWith \input -> do
   else
     Left $ "Could not take " <> show n <> " characters"
 
--- | Match zero or more whitespace characters satisfying
--- | `Data.CodePoint.Unicode.isSpace`. Always succeeds.
-whiteSpace :: forall m. ParserT String m String
-whiteSpace = fst <$> match skipSpaces
-
--- | Skip whitespace characters and throw them away. Always succeeds.
-skipSpaces :: forall m. ParserT String m Unit
-skipSpaces = consumeWith \input -> do
-  let consumed = takeWhile isSpace input
-  let remainder = SCU.drop (SCU.length consumed) input
-  Right { value: unit, consumed, remainder }
-
--- | Match one of the BMP `Char`s in the array.
-oneOf :: forall m. Array Char -> ParserT String m Char
-oneOf ss = satisfy (flip elem ss) <~?> \_ -> "one of " <> show ss
-
--- | Match any BMP `Char` not in the array.
-noneOf :: forall m. Array Char -> ParserT String m Char
-noneOf ss = satisfy (flip notElem ss) <~?> \_ -> "none of " <> show ss
-
--- | Match one of the Unicode characters in the array.
-oneOfCodePoints :: forall m. Array CodePoint -> ParserT String m CodePoint
-oneOfCodePoints ss = satisfyCodePoint (flip elem ss) <~?> \_ -> "one of " <> show (singleton <$> ss)
-
--- | Match any Unicode character not in the array.
-noneOfCodePoints :: forall m. Array CodePoint -> ParserT String m CodePoint
-noneOfCodePoints ss = satisfyCodePoint (flip notElem ss) <~?> \_ -> "none of " <> show (singleton <$> ss)
-
 -- | Updates a `Position` by adding the columns and lines in `String`.
 updatePosString :: Position -> String -> String -> Position
 updatePosString pos before after = case uncons before of

src/Parsing/String/Basic.purs

@@ -15,22 +15,30 @@ module Parsing.String.Basic
   , alphaNum
   , intDecimal
   , number
-  , module Parsing.String
+  , whiteSpace
+  , skipSpaces
+  , oneOf
+  , oneOfCodePoints
+  , noneOf
+  , noneOfCodePoints
   ) where
 
 import Prelude
 
+import Data.Array (elem, notElem)
 import Data.CodePoint.Unicode (isAlpha, isAlphaNum, isDecDigit, isHexDigit, isLower, isOctDigit, isSpace, isUpper)
+import Data.Either (Either(..))
 import Data.Int as Data.Int
 import Data.Maybe (Maybe(..))
 import Data.Number (infinity, nan)
 import Data.Number as Data.Number
-import Data.String (CodePoint)
+import Data.String (CodePoint, singleton, takeWhile)
 import Data.String.CodePoints (codePointFromChar)
-import Data.Tuple (Tuple(..))
+import Data.String.CodeUnits as SCU
+import Data.Tuple (Tuple(..), fst)
 import Parsing (ParserT, fail)
-import Parsing.Combinators (choice, skipMany, (<?>))
-import Parsing.String (noneOf, noneOfCodePoints, oneOf, oneOfCodePoints, skipSpaces, whiteSpace)
+import Parsing.Combinators (choice, skipMany, (<?>), (<~?>))
+import Parsing.String (consumeWith, match, satisfy, satisfyCodePoint)
 import Parsing.String as Parser.String
 
 -- | Parse a digit.  Matches any char that satisfies `Data.CodePoint.Unicode.isDecDigit`.
@@ -94,8 +102,8 @@ number =
     , Parser.String.string "NaN" *> pure nan
     , do
         Tuple section _ <- Parser.String.match do
-          _ <- Parser.String.oneOf [ '+', '-', '.', '0', '1', '2', '3', '4', '5', '6', '7', '8', '9' ]
-          skipMany $ Parser.String.oneOf [ 'e', 'E', '+', '-', '.', '0', '1', '2', '3', '4', '5', '6', '7', '8', '9' ]
+          _ <- oneOf [ '+', '-', '.', '0', '1', '2', '3', '4', '5', '6', '7', '8', '9' ]
+          skipMany $ oneOf [ 'e', 'E', '+', '-', '.', '0', '1', '2', '3', '4', '5', '6', '7', '8', '9' ]
         -- https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Global_Objects/parseFloat
         case Data.Number.fromString section of
           Nothing -> fail $ "Could not parse Number " <> section
@@ -113,12 +121,40 @@ number =
 intDecimal :: forall m. ParserT String m Int
 intDecimal = do
   Tuple section _ <- Parser.String.match do
-    _ <- Parser.String.oneOf [ '+', '-', '0', '1', '2', '3', '4', '5', '6', '7', '8', '9' ]
-    skipMany $ Parser.String.oneOf [ '0', '1', '2', '3', '4', '5', '6', '7', '8', '9' ]
+    _ <- oneOf [ '+', '-', '0', '1', '2', '3', '4', '5', '6', '7', '8', '9' ]
+    skipMany $ oneOf [ '0', '1', '2', '3', '4', '5', '6', '7', '8', '9' ]
   case Data.Int.fromString section of
     Nothing -> fail $ "Could not parse Int " <> section
     Just x -> pure x
 
 -- | Helper function
 satisfyCP :: forall m. (CodePoint -> Boolean) -> ParserT String m Char
 satisfyCP p = Parser.String.satisfy (p <<< codePointFromChar)
+
+-- | Match zero or more whitespace characters satisfying
+-- | `Data.CodePoint.Unicode.isSpace`. Always succeeds.
+whiteSpace :: forall m. ParserT String m String
+whiteSpace = fst <$> match skipSpaces
+
+-- | Skip whitespace characters and throw them away. Always succeeds.
+skipSpaces :: forall m. ParserT String m Unit
+skipSpaces = consumeWith \input -> do
+  let consumed = takeWhile isSpace input
+  let remainder = SCU.drop (SCU.length consumed) input
+  Right { value: unit, consumed, remainder }
+
+-- | Match one of the BMP `Char`s in the array.
+oneOf :: forall m. Array Char -> ParserT String m Char
+oneOf ss = satisfy (flip elem ss) <~?> \_ -> "one of " <> show ss
+
+-- | Match any BMP `Char` not in the array.
+noneOf :: forall m. Array Char -> ParserT String m Char
+noneOf ss = satisfy (flip notElem ss) <~?> \_ -> "none of " <> show ss
+
+-- | Match one of the Unicode characters in the array.
+oneOfCodePoints :: forall m. Array CodePoint -> ParserT String m CodePoint
+oneOfCodePoints ss = satisfyCodePoint (flip elem ss) <~?> \_ -> "one of " <> show (singleton <$> ss)
+
+-- | Match any Unicode character not in the array.
+noneOfCodePoints :: forall m. Array CodePoint -> ParserT String m CodePoint
+noneOfCodePoints ss = satisfyCodePoint (flip notElem ss) <~?> \_ -> "none of " <> show (singleton <$> ss)

src/Parsing/Token.purs

@@ -46,9 +46,9 @@ import Data.Tuple (Tuple(..))
 import Parsing (ParseState(..), ParserT, consume, fail)
 import Parsing.Combinators (between, choice, notFollowedBy, option, sepBy, sepBy1, skipMany, skipMany1, try, tryRethrow, (<?>), (<??>))
 import Parsing.Pos (Position)
-import Parsing.String (char, noneOf, oneOf, satisfy, satisfyCodePoint, string)
+import Parsing.String (char, satisfy, satisfyCodePoint, string)
+import Parsing.String.Basic (alphaNum, digit, hexDigit, letter, noneOf, octDigit, oneOf, space, upper)
 import Parsing.String.Basic as Basic
-import Parsing.String.Basic (digit, hexDigit, octDigit, upper, space, letter, alphaNum)
 
 -- | A parser which returns the first token in the stream.
 token :: forall m a. (a -> Position) -> ParserT (List a) m a

test/Main.purs

@@ -30,8 +30,8 @@ import Parsing.Combinators (between, chainl, chainl1, chainr, chainr1, choice, e
 import Parsing.Expr (Assoc(..), Operator(..), buildExprParser)
 import Parsing.Language (haskellDef, haskellStyle, javaStyle)
 import Parsing.Pos (Position(..), initialPos)
-import Parsing.String (anyChar, anyCodePoint, char, eof, regex, noneOfCodePoints, oneOfCodePoints, rest, satisfy, string, takeN, whiteSpace)
-import Parsing.String.Basic (intDecimal, number, letter)
+import Parsing.String (anyChar, anyCodePoint, char, eof, regex, rest, satisfy, string, takeN)
+import Parsing.String.Basic (intDecimal, number, letter, noneOfCodePoints, oneOfCodePoints, whiteSpace)
 import Parsing.Token (TokenParser, makeTokenParser, match, token, when)
 import Parsing.Token as Parser.Token
 import Partial.Unsafe (unsafePartial)