From 01685ab88228fb602cb0e408d93560e76e1371a1 Mon Sep 17 00:00:00 2001
From: Josh Rahm <joshuarahm@gmail.com>
Date: Sat, 17 Dec 2022 22:38:48 -0700
Subject: WIP: Basic parser implemented.

The parser is completely untested and probably broken, but it's probably
pretty close becasue it does typecheck. This is a Work-in-progress.
---
 src/Language/Fiddle/Tokenizer.hs | 31 ++++++++++++++++++++-----------
 1 file changed, 20 insertions(+), 11 deletions(-)

(limited to 'src/Language/Fiddle/Tokenizer.hs')

diff --git a/src/Language/Fiddle/Tokenizer.hs b/src/Language/Fiddle/Tokenizer.hs
index d3239fd..ec41042 100644
--- a/src/Language/Fiddle/Tokenizer.hs
+++ b/src/Language/Fiddle/Tokenizer.hs
@@ -1,29 +1,34 @@
 {-# LANGUAGE DeriveFunctor #-}
+{-# LANGUAGE OverloadedStrings #-}
+
 module Language.Fiddle.Tokenizer where
 
 import Data.Char (isDigit)
 import Data.Text (Text)
+import qualified Data.Text
 import Language.Fiddle.Types
 import Text.Parsec
 import qualified Text.Parsec
 
 data T
   = KWAssertPos
-  | Ident !String
+  | TokIdent !Text
   | KWAt
-  | KWBittype
+  | KWBits
   | KWEnum
-  | CommentTok !String
-  | DocCommentTok !String
+  | TokComment !Text
+  | TokDocComment !Text
   | KWLocation
   | KWObject
   | KWObjtype
   | KWOption
   | KWPackage
   | KWReg
+  | KWReserved
   | KWRo
   | KWWo
-  | LitNum !String
+  | KWRw
+  | TokLitNum !Text
   | TokColon
   | TokComma
   | TokEq
@@ -48,10 +53,11 @@ parseToken = spaces *> tok parseToken' <* spaces
 
       Token t . SourceSpan p1 <$> getPosition
 
+    parseAlNumTok :: Text -> T
     parseAlNumTok str =
       case str of
         "at" -> KWAt
-        "bittype" -> KWBittype
+        "bits" -> KWBits
         "enum" -> KWEnum
         "location" -> KWLocation
         "object" -> KWObject
@@ -61,19 +67,22 @@ parseToken = spaces *> tok parseToken' <* spaces
         "reg" -> KWReg
         "ro" -> KWRo
         "wo" -> KWWo
-        (h : _) | isDigit h -> LitNum str
-        ident -> Ident ident
+        "rw" -> KWRw
+        "reserved" -> KWReserved
+        "assert_pos" -> KWAssertPos
+        (Data.Text.head -> h) | isDigit h -> TokLitNum str
+        ident -> TokIdent ident
 
     parseComment =
       try
         ( do
             string "//"
-            CommentTok <$> manyTill anyChar (char '\n')
+            TokComment . Data.Text.pack <$> manyTill anyChar (char '\n')
         )
         <|> try
           ( do
               string "/**"
-              DocCommentTok <$> manyTill anyChar (try $ string "*/")
+              TokDocComment . Data.Text.pack <$> manyTill anyChar (try $ string "*/")
           )
 
     parseSymbol =
@@ -93,7 +102,7 @@ parseToken = spaces *> tok parseToken' <* spaces
         a $> b = a >> return b
 
     parseToken' =
-      fmap parseAlNumTok (many1 (alphaNum <|> char '_'))
+      fmap (parseAlNumTok . Data.Text.pack) (many1 (alphaNum <|> char '_'))
         <|> parseComment
         <|> parseSymbol
 
-- 
cgit