fix: adjusted descriptions to match parsley:5.0.0

j-mie6 · Jan 15, 2024 · 26e1e66 · 26e1e66
1 parent 8bfe439
commit 26e1e66
Show file tree

Hide file tree

Showing 3 changed files with 30 additions and 27 deletions.
diff --git a/src/Text/Gigaparsec/Token/Descriptions.hs b/src/Text/Gigaparsec/Token/Descriptions.hs
@@ -96,21 +96,25 @@ plainNumeric = NumericDesc { literalBreakChar = NoBreakChar
                                                                       , chars = ['e', 'E']
                                                                       , base = 10
                                                                       , expSign = PlusOptional
+                                                                      , expLeadingZerosAllowd = True
                                                                       }
                            , hexadecimalExponentDesc = ExponentsSupported { compulsory = True
                                                                           , chars = ['p', 'P']
                                                                           , base = 2
                                                                           , expSign = PlusOptional
+                                                                          , expLeadingZerosAllowd = True
                                                                           }
                            , octalExponentDesc = ExponentsSupported { compulsory = True
                                                                     , chars = ['e', 'E', 'p', 'P']
                                                                     , base = 2
                                                                     , expSign = PlusOptional
+                                                                    , expLeadingZerosAllowd = True
                                                                     }
                            , binaryExponentDesc = ExponentsSupported { compulsory = True
                                                                      , chars = ['e', 'E', 'p', 'P']
                                                                      , base = 2
                                                                      , expSign = PlusOptional
+                                                                     , expLeadingZerosAllowd = True
                                                                      }
                            }
 
@@ -120,6 +124,7 @@ data ExponentDesc = NoExponents
                                        , chars :: !(Set Char)
                                        , base :: !Int
                                        , expSign :: !PlusSignPresence
+                                       , expLeadingZerosAllowd :: !Bool
                                        }
 
 type BreakCharDesc :: *
@@ -134,24 +139,23 @@ data PlusSignPresence = PlusRequired | PlusOptional | PlusIllegal
 type TextDesc :: *
 data TextDesc = TextDesc { escapeSequences :: {-# UNPACK #-} !EscapeDesc
                          , characterLiteralEnd :: !Char
-                         , stringEnds :: !(Set String)
-                         , multiStringEnds :: !(Set String)
+                         , stringEnds :: !(Set (String, String))
+                         , multiStringEnds :: !(Set (String, String))
                          , graphicCharacter :: !CharPredicate
                          }
 
 plainText :: TextDesc
 plainText = TextDesc { escapeSequences = plainEscape
                      , characterLiteralEnd = '\''
-                     , stringEnds = ["\""]
+                     , stringEnds = [("\"", "\"")]
                      , multiStringEnds = []
                      , graphicCharacter = Just (>= ' ')
                      }
 
 type EscapeDesc :: *
 data EscapeDesc = EscapeDesc { escBegin :: !Char
                              , literals :: !(Set Char)
-                             , singleMap :: !(Map Char Char)
-                             , multiMap :: !(Map String Char)
+                             , mapping :: !(Map String Char)
                              , decimalEscape :: !NumericEscape
                              , hexadecimalEscape :: !NumericEscape
                              , octalEscape :: !NumericEscape
@@ -163,8 +167,7 @@ data EscapeDesc = EscapeDesc { escBegin :: !Char
 plainEscape :: EscapeDesc
 plainEscape = EscapeDesc { escBegin = '\\'
                          , literals = ['\\']
-                         , singleMap = []
-                         , multiMap = []
+                         , mapping = []
                          , decimalEscape = NumericIllegal
                          , hexadecimalEscape = NumericIllegal
                          , octalEscape = NumericIllegal
@@ -186,21 +189,21 @@ type NumberOfDigits :: *
 data NumberOfDigits = Unbounded | Exactly !(NonEmpty Word) | AtMost !Word
 
 type SpaceDesc :: *
-data SpaceDesc = SpaceDesc { commentStart :: !String
-                           , commentEnd :: !String
-                           , commentLine :: !String
-                           , commentLineAllowsEOF :: !Bool
-                           , nestedComments :: !Bool
+data SpaceDesc = SpaceDesc { lineCommentStart :: !String
+                           , lineCommentAllowsEOF :: !Bool
+                           , multiLineCommentStart :: !String
+                           , multiLineCommentEnd :: !String
+                           , multiLineNestedComments :: !Bool
                            , space :: !CharPredicate
                            , whitespaceIsContextDependent :: !Bool
                            }
 
 plainSpace :: SpaceDesc
-plainSpace = SpaceDesc { commentStart = ""
-                       , commentEnd = ""
-                       , commentLine = ""
-                       , commentLineAllowsEOF = True
-                       , nestedComments = False
+plainSpace = SpaceDesc { lineCommentStart = ""
+                       , lineCommentAllowsEOF = True
+                       , multiLineCommentStart = ""
+                       , multiLineCommentEnd = ""
+                       , multiLineNestedComments = False
                        , space = Just isSpace
                        , whitespaceIsContextDependent = False
                        }

diff --git a/src/Text/Gigaparsec/Token/Lexer.hs b/src/Text/Gigaparsec/Token/Lexer.hs
@@ -153,33 +153,33 @@ We have the following invariances to be checked up front:
 commentParser :: Desc.SpaceDesc -> Parsec ()
 commentParser Desc.SpaceDesc{..} =
   require (multiEnabled || singleEnabled) "skipComments" noComments $
-    require (not (multiEnabled && isPrefixOf commentStart commentLine)) "skipComments" noOverlap $
+    require (not (multiEnabled && isPrefixOf multiLineCommentStart lineCommentStart)) "skipComments" noOverlap $
       hide (multiLine <|> singleLine)
   where
     -- can't make these strict until guard is gone
-    openComment = atomic (string commentStart)
-    closeComment = atomic (string commentEnd)
+    openComment = atomic (string multiLineCommentStart)
+    closeComment = atomic (string multiLineCommentEnd)
     multiLine = guard multiEnabled *> openComment *> wellNested 1
     wellNested :: Int -> Parsec ()
     wellNested 0 = unit
     wellNested n = closeComment *> wellNested (n - 1)
-               <|> guard nestedComments *> openComment *> wellNested (n + 1)
+               <|> guard multiLineNestedComments *> openComment *> wellNested (n + 1)
                <|> item *> wellNested n
     singleLine = guard singleEnabled
-              *> atomic (string commentLine)
+              *> atomic (string lineCommentStart)
               *> skipManyTill item endOfLineComment
 
     endOfLineComment
-      | commentLineAllowsEOF = void endOfLine <|> eof
+      | lineCommentAllowsEOF = void endOfLine <|> eof
       | otherwise            = void endOfLine
 
-    multiEnabled = not (null commentStart || null commentEnd)
-    singleEnabled = not (null commentLine)
+    multiEnabled = not (null multiLineCommentStart || null multiLineCommentEnd)
+    singleEnabled = not (null lineCommentStart)
     noComments = "one of single- or multi-line comments must be enabled"
     noOverlap = "single-line comments must not overlap with multi-line comments"
 
 supportsComments :: Desc.SpaceDesc -> Bool
-supportsComments Desc.SpaceDesc{..} = not (null commentLine && null commentStart)
+supportsComments Desc.SpaceDesc{..} = not (null lineCommentStart && null multiLineCommentStart)
 
 type UnsupportedOperation :: *
 newtype UnsupportedOperation = UnsupportedOperation String deriving stock Eq

diff --git a/src/Text/Gigaparsec/Token/Numeric.hs b/src/Text/Gigaparsec/Token/Numeric.hs
@@ -1,6 +1,6 @@
 {-# LANGUAGE Safe #-}
 {-# LANGUAGE DataKinds, KindSignatures, ConstraintKinds, MultiParamTypeClasses, AllowAmbiguousTypes, FlexibleInstances, FlexibleContexts, UndecidableInstances, ApplicativeDo #-}
--- TODO: refine
+-- TODO: refine, move to Internal
 module Text.Gigaparsec.Token.Numeric (module Text.Gigaparsec.Token.Numeric) where
 
 import Text.Gigaparsec (Parsec, mapMaybeS, unit, void, atomic, (<|>), ($>))