takeWhile

Author: jamesdbrock

CHANGELOG.md

@@ -10,6 +10,8 @@ Breaking changes:
 
 New features:
 
+- Add `Parsing.String.Basic.takeWhile` (#218 by @jamesdbrock)
+
 Other improvements:
 
 ## [v10.1.0](https://github.com/purescript-contrib/purescript-parsing/releases/tag/v10.1.0) - 2022-11-10

src/Parsing.purs

@@ -430,8 +430,7 @@ failWithPosition message pos = throwError (ParseError message pos)
 -- |
 -- | lmap (parseErrorHuman input 30) $ runParser input do
 -- |   inContext ("Megacity list: " <> _) do
--- |     cityname <- inContext ("city name: " <> _) do
--- |       fst <$> match (skipMany letter)
+-- |     cityname <- inContext ("city name: " <> _) (takeWhile isLetter)
 -- |     skipSpaces
 -- |     population <- inContext ("population: " <> _) intDecimal
 -- |     pure $ Tuple cityname population

src/Parsing/String/Basic.purs

@@ -2,7 +2,9 @@
 -- |
 -- | #### unicode dependency
 -- |
--- | Some of the parsers in this module depend on the __unicode__ package.
+-- | Some of the parsers in this module depend on the
+-- | [__unicode__](https://pursuit.purescript.org/packages/purescript-unicode)
+-- | package.
 -- | The __unicode__ package is large; about half a megabyte unminified.
 -- | If code which depends on __parsing__ is “tree-shaken”
 -- | “dead-code-eliminated,” then
@@ -24,6 +26,7 @@ module Parsing.String.Basic
   , alphaNum
   , intDecimal
   , number
+  , takeWhile
   , whiteSpace
   , skipSpaces
   , oneOf
@@ -41,9 +44,11 @@ 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, singleton, takeWhile)
+import Data.String (CodePoint, singleton)
+import Data.String as String
 import Data.String.CodePoints (codePointFromChar)
 import Data.String.CodeUnits as SCU
+import Data.String.CodeUnits as String.CodeUnits
 import Data.Tuple (fst)
 import Parsing (ParserT, fail)
 import Parsing.Combinators (choice, tryRethrow, (<?>), (<|>), (<~?>))
@@ -161,7 +166,7 @@ whiteSpace = fst <$> match skipSpaces
 -- | Always succeeds. Will only consume when some characters are skipped.
 skipSpaces :: forall m. ParserT String m Unit
 skipSpaces = consumeWith \input -> do
-  let consumed = takeWhile isSpace input
+  let consumed = String.takeWhile isSpace input
   let remainder = SCU.drop (SCU.length consumed) input
   Right { value: unit, consumed, remainder }
 
@@ -180,3 +185,32 @@ oneOfCodePoints ss = satisfyCodePoint (flip elem ss) <~?> \_ -> "one of " <> sho
 -- | 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)
+
+-- | Take the longest `String` for which the characters satisfy the
+-- | predicate.
+-- |
+-- | See [__`Data.CodePoint.Unicode`__](https://pursuit.purescript.org/packages/purescript-unicode/docs/Data.CodePoint.Unicode)
+-- | for useful predicates.
+-- |
+-- | Example:
+-- |
+-- | ```
+-- | runParser "Tackling the Awkward Squad" do
+-- |   takeWhile Data.CodePoint.Unicode.isLetter
+-- | ```
+-- | ---
+-- | ```
+-- | Right "Tackling"
+-- | ```
+takeWhile :: forall m. (CodePoint -> Boolean) -> ParserT String m String
+takeWhile predicate =
+  consumeWith \s ->
+    let
+      value = String.takeWhile predicate s
+    in
+      Right
+        { consumed: value
+        , remainder: SCU.drop (SCU.length value) s
+        , value
+        }
+

test/Main.purs

@@ -13,6 +13,7 @@ import Control.Monad.State (State, lift, modify, runState)
 import Data.Array (some, toUnfoldable)
 import Data.Array as Array
 import Data.Bifunctor (lmap, rmap)
+import Data.CodePoint.Unicode as CodePoint.Unicode
 import Data.Either (Either(..), either, fromLeft, hush)
 import Data.Foldable (oneOf)
 import Data.List (List(..), fromFoldable, (:))
@@ -41,7 +42,7 @@ import Parsing.Combinators.Array as Combinators.Array
 import Parsing.Expr (Assoc(..), Operator(..), buildExprParser)
 import Parsing.Language (haskellDef, haskellStyle, javaStyle)
 import Parsing.String (anyChar, anyCodePoint, anyTill, char, eof, match, parseErrorHuman, regex, rest, satisfy, string, takeN)
-import Parsing.String.Basic (intDecimal, letter, noneOfCodePoints, number, oneOfCodePoints, skipSpaces, whiteSpace)
+import Parsing.String.Basic (intDecimal, letter, noneOfCodePoints, number, oneOfCodePoints, skipSpaces, takeWhile, whiteSpace)
 import Parsing.String.Basic as String.Basic
 import Parsing.String.Replace (breakCap, replace, replaceT, splitCap, splitCapT)
 import Parsing.Token (TokenParser, makeTokenParser, token, when)
@@ -712,8 +713,7 @@ main = do
     assertEqual' "region 1"
       { actual: runParser input do
           inContext ("Megacity list: " <> _) do
-            cityname <- inContext ("city name: " <> _) do
-              fst <$> match (Combinators.skipMany letter)
+            cityname <- inContext ("city name: " <> _) (takeWhile CodePoint.Unicode.isLetter)
             skipSpaces
             population <- inContext ("population: " <> _) intDecimal
             pure $ Tuple cityname population
@@ -725,6 +725,12 @@ main = do
     , expected: Left $ ParseError "Expected 'c'" (Position { index: 1, column: 2, line: 1 })
     }
 
+  assertEqual' "takeWhile 1"
+    { actual: runParser "Tackling the Awkward" do
+        takeWhile CodePoint.Unicode.isLetter <* string " the Awkward"
+    , expected: Right "Tackling"
+    }
+
   log "\nTESTS number\n"
 
   -- assert' "Number.fromString" $ Just infinity == Data.Number.fromString "Infinity"