Non-empty combinators return NonEmptyList (#102)

* Non-empty combinators return NonEmptyList

* Updated change log

* Separate 0+ and 1+ combinator implementations

* Removed `toList` from imports

Author: robertdp

CHANGELOG.md

@@ -5,6 +5,7 @@ Notable changes to this project are documented in this file. The format is based
 ## [Unreleased]
 
 Breaking changes (😱!!!):
+- Non-empty combinators return `NonEmptyList` ([#102](https://github.com/purescript-contrib/purescript-parsing/pull/102))
 
 New features:
 

src/Text/Parsing/Parser/Combinators.purs

@@ -29,7 +29,8 @@ import Control.Monad.State (StateT(..), runStateT)
 import Control.Plus (empty, (<|>))
 import Data.Either (Either(..))
 import Data.Foldable (class Foldable, foldl)
-import Data.List (List(..), (:), many, some, singleton)
+import Data.List (List(..), (:), many)
+import Data.List.NonEmpty (NonEmptyList, cons', singleton)
 import Data.Maybe (Maybe(..))
 import Data.Newtype (unwrap)
 import Data.Tuple (Tuple(..))
@@ -99,30 +100,40 @@ lookAhead p = (ParserT <<< ExceptT <<< StateT) \s -> do
 -- | digit `sepBy` string ","
 -- | ```
 sepBy :: forall m s a sep. Monad m => ParserT s m a -> ParserT s m sep -> ParserT s m (List a)
-sepBy p sep = sepBy1 p sep <|> pure Nil
+sepBy p sep =
+  (do a <- p
+      as <- many $ sep *> p
+      pure (a : as)) <|> pure Nil
 
 -- | Parse phrases delimited by a separator, requiring at least one match.
-sepBy1 :: forall m s a sep. Monad m => ParserT s m a -> ParserT s m sep -> ParserT s m (List a)
+sepBy1 :: forall m s a sep. Monad m => ParserT s m a -> ParserT s m sep -> ParserT s m (NonEmptyList a)
 sepBy1 p sep = do
   a <- p
   as <- many $ sep *> p
-  pure (a : as)
+  pure (cons' a as)
 
 -- | Parse phrases delimited and optionally terminated by a separator.
 sepEndBy :: forall m s a sep. Monad m => ParserT s m a -> ParserT s m sep -> ParserT s m (List a)
-sepEndBy p sep = sepEndBy1 p sep <|> pure Nil
+sepEndBy p sep =
+  (do a <- p
+      as <- many $ sep *> p
+      optional sep
+      pure (a : as)) <|> pure Nil
 
 -- | Parse phrases delimited and optionally terminated by a separator, requiring at least one match.
-sepEndBy1 :: forall m s a sep. Monad m => ParserT s m a -> ParserT s m sep -> ParserT s m (List a)
+sepEndBy1 :: forall m s a sep. Monad m => ParserT s m a -> ParserT s m sep -> ParserT s m (NonEmptyList a)
 sepEndBy1 p sep = do
   a <- p
-  (do _ <- sep
-      as <- sepEndBy p sep
-      pure (a : as)) <|> pure (singleton a)
+  (do as <- many $ sep *> p
+      optional sep
+      pure (cons' a as)) <|> pure (singleton a)
 
 -- | Parse phrases delimited and terminated by a separator, requiring at least one match.
-endBy1 :: forall m s a sep. Monad m => ParserT s m a -> ParserT s m sep -> ParserT s m (List a)
-endBy1 p sep = some $ p <* sep
+endBy1 :: forall m s a sep. Monad m => ParserT s m a -> ParserT s m sep -> ParserT s m (NonEmptyList a)
+endBy1 p sep = do
+  a <- p <* sep
+  as <- many $ p <* sep
+  pure (cons' a as)
 
 -- | Parse phrases delimited and terminated by a separator.
 endBy :: forall m s a sep. Monad m => ParserT s m a -> ParserT s m sep -> ParserT s m (List a)
@@ -193,8 +204,9 @@ manyTill p end = scan
               pure (x:xs)
 
 -- | Parse several phrases until the specified terminator matches, requiring at least one match.
-many1Till :: forall s a m e. Monad m => ParserT s m a -> ParserT s m e -> ParserT s m (List a)
+many1Till :: forall s a m e. Monad m => ParserT s m a -> ParserT s m e -> ParserT s m (NonEmptyList a)
 many1Till p end = do
   x <- p
   xs <- manyTill p end
-  pure (x:xs)
+  pure (cons' x xs)
+

src/Text/Parsing/Parser/Token.purs

@@ -36,6 +36,7 @@ import Data.Identity (Identity)
 import Data.Int (toNumber)
 import Data.List (List(..))
 import Data.List as List
+import Data.List.NonEmpty (NonEmptyList)
 import Data.Maybe (Maybe(..), maybe)
 import Data.String (null, toLower)
 import Data.String.CodeUnits as SCU
@@ -258,15 +259,15 @@ type GenTokenParser s m
         semiSep          :: forall a . ParserT s m a -> ParserT s m (List a),
         -- | Lexeme parser `semiSep1 p` parses *one* or more occurrences of `p`
         -- | separated by `semi`. Returns a list of values pureed by `p`.
-        semiSep1         :: forall a . ParserT s m a -> ParserT s m (List a),
+        semiSep1         :: forall a . ParserT s m a -> ParserT s m (NonEmptyList a),
         -- | Lexeme parser `commaSep p` parses *zero* or more occurrences of
         -- | `p` separated by `comma`. Returns a list of values pureed
         -- | by `p`.
         commaSep         :: forall a . ParserT s m a -> ParserT s m (List a),
         -- | Lexeme parser `commaSep1 p` parses *one* or more occurrences of
         -- | `p` separated by `comma`. Returns a list of values pureed
         -- | by `p`.
-        commaSep1        :: forall a . ParserT s m a -> ParserT s m (List a)
+        commaSep1        :: forall a . ParserT s m a -> ParserT s m (NonEmptyList a)
     }
 
 -----------------------------------------------------------
@@ -369,10 +370,10 @@ makeTokenParser (LanguageDef languageDef)
     semiSep :: forall a . ParserT String m a -> ParserT String m (List a)
     semiSep p = sepBy p semi
 
-    commaSep1 :: forall a . ParserT String m a -> ParserT String m (List a)
+    commaSep1 :: forall a . ParserT String m a -> ParserT String m (NonEmptyList a)
     commaSep1 p = sepBy1 p comma
 
-    semiSep1 :: forall a . ParserT String m a -> ParserT String m (List a)
+    semiSep1 :: forall a . ParserT String m a -> ParserT String m (NonEmptyList a)
     semiSep1 p = sepBy1 p semi
 
     -----------------------------------------------------------

test/Main.purs

@@ -7,6 +7,7 @@ import Control.Lazy (fix)
 import Data.Array (some)
 import Data.Either (Either(..))
 import Data.List (List(..), fromFoldable, many)
+import Data.List.NonEmpty (cons, cons')
 import Data.Maybe (Maybe(..))
 import Data.String.CodeUnits (fromCharArray, singleton)
 import Data.Tuple (Tuple(..))
@@ -357,10 +358,10 @@ tokenParserSemiSepTest = do
 tokenParserSemiSep1Test :: TestM
 tokenParserSemiSep1Test = do
     -- parse semi sep1
-    parseTest "foo; foo" (fromFoldable ["foo", "foo"]) $ testTokenParser.semiSep1 $ string "foo"
+    parseTest "foo; foo" (cons "foo" (cons' "foo" Nil)) $ testTokenParser.semiSep1 $ string "foo"
 
     -- parse semi sep1 with newline
-    parseTest "foo; \nfoo" (fromFoldable ["foo", "foo"]) $ testTokenParser.semiSep1 $ string "foo"
+    parseTest "foo; \nfoo" (cons "foo" (cons' "foo" Nil)) $ testTokenParser.semiSep1 $ string "foo"
 
     -- no parse on empty string
     parseErrorTestPosition (testTokenParser.semiSep1 $ string "foo") "" $ mkPos 1
@@ -379,10 +380,10 @@ tokenParserCommaSepTest = do
 tokenParserCommaSep1Test :: TestM
 tokenParserCommaSep1Test = do
     -- parse comma sep1
-    parseTest "foo, foo" (fromFoldable ["foo", "foo"]) $ testTokenParser.commaSep1 $ string "foo"
+    parseTest "foo, foo" (cons "foo" (cons' "foo" Nil)) $ testTokenParser.commaSep1 $ string "foo"
 
     -- parse comma sep1 with newline
-    parseTest "foo, \nfoo" (fromFoldable ["foo", "foo"]) $ testTokenParser.commaSep1 $ string "foo"
+    parseTest "foo, \nfoo" (cons "foo" (cons' "foo" Nil)) $ testTokenParser.commaSep1 $ string "foo"
 
     -- no parse on empty string
     parseErrorTestPosition (testTokenParser.commaSep1 $ string "foo") "" $ mkPos 1
@@ -438,8 +439,8 @@ main = do
   parseTest "(ab)" (Just "b") $ parens do
     _ <- string "a"
     optionMaybe $ string "b"
-  parseTest "a,a,a" (Cons "a" (Cons "a" (Cons "a" Nil))) $ string "a" `sepBy1` string ","
-  parseTest "a,a,a," (Cons "a" (Cons "a" (Cons "a" Nil))) $ do
+  parseTest "a,a,a" (cons "a" (cons "a" (cons' "a" Nil))) $ string "a" `sepBy1` string ","
+  parseTest "a,a,a," (cons "a" (cons "a" (cons' "a" Nil))) $ do
     as <- string "a" `endBy1` string ","
     eof
     pure as