From be944226687b89acf4216250a9a2113e41509ae4 Mon Sep 17 00:00:00 2001
From: rhiannon morris <rhi@rhiannon.website>
Date: Thu, 16 Mar 2023 18:34:49 +0100
Subject: [PATCH] move name lexing stuff to Quox.Name

---
 lib/Quox/Name.idr         | 33 +++++++++++++++++++++++++++++++++
 lib/Quox/Parser/Lexer.idr | 34 ++--------------------------------
 2 files changed, 35 insertions(+), 32 deletions(-)

diff --git a/lib/Quox/Name.idr b/lib/Quox/Name.idr
index 20d16d8..757f612 100644
--- a/lib/Quox/Name.idr
+++ b/lib/Quox/Name.idr
@@ -3,6 +3,8 @@ module Quox.Name
 import public Data.SnocList
 import Data.List
 import Derive.Prelude
+import Quox.CharExtra
+import Text.Lexer
 
 %hide TT.Name
 
@@ -95,3 +97,34 @@ fromListP (x ::: xs) = go [<] x xs where
 export %inline
 fromList : List1 String -> Name
 fromList = fromPName . fromListP
+
+
+export
+syntaxChars : List Char
+syntaxChars = ['(', ')', '[', ']', '{', '}', '"', '\'', ',', '.', ';']
+
+export
+isSymStart, isSymCont : Char -> Bool
+isSymStart c = not (c `elem` syntaxChars) && isSymChar c
+isSymCont  c = c == '\'' || isSymStart c
+
+export
+idStart, idCont, idEnd, idContEnd : Lexer
+idStart   = pred isIdStart
+idCont    = pred isIdCont
+idEnd     = pred $ \c => c `elem` unpack "?!#"
+idContEnd = idCont <|> idEnd
+
+export
+symStart, symCont : Lexer
+symStart  = pred isSymStart
+symCont   = pred isSymCont
+
+export
+baseName : Lexer
+baseName = idStart <+> many idCont <+> many idEnd
+       <|> symStart <+> many symCont
+
+export
+name : Lexer
+name = baseName <+> many (is '.' <+> baseName)
diff --git a/lib/Quox/Parser/Lexer.idr b/lib/Quox/Parser/Lexer.idr
index 1353b3b..873a549 100644
--- a/lib/Quox/Parser/Lexer.idr
+++ b/lib/Quox/Parser/Lexer.idr
@@ -59,39 +59,9 @@ match t f = Tokenizer.match t (Just . f)
 %hide Tokenizer.match
 
 
-export %inline
-syntaxChars : List Char
-syntaxChars = ['(', ')', '[', ']', '{', '}', '"', '\'', ',', '.', ';']
-
-private
-isSymStart, isSymCont : Char -> Bool
-isSymStart c = not (c `elem` syntaxChars) && isSymChar c
-isSymCont  c = c == '\'' || isSymStart c
-
-private
-idStart, idCont, idEnd, idContEnd : Lexer
-idStart   = pred isIdStart
-idCont    = pred isIdCont
-idEnd     = pred $ \c => c `elem` unpack "?!#"
-idContEnd = idCont <|> idEnd
-
-private
-symStart, symCont : Lexer
-symStart  = pred isSymStart
-symCont   = pred isSymCont
-
-private
-baseNameL : Lexer
-baseNameL = idStart <+> many idCont <+> many idEnd
-        <|> symStart <+> many symCont
-
-private
-nameL : Lexer
-nameL = baseNameL <+> many (is '.' <+> baseNameL)
-
 private
 name : Tokenizer TokenW
-name = match nameL $ Name . fromListP . split (== '.') . normalizeNfc
+name = match name $ Name . fromListP . split (== '.') . normalizeNfc
 
 ||| [todo] escapes other than `\"` and (accidentally) `\\`
 export
@@ -113,7 +83,7 @@ nat = match (some (range '0' '9')) (Nat . cast)
 
 private
 tag : Tokenizer TokenW
-tag = match (is '\'' <+> nameL)     (Tag . drop 1)
+tag = match (is '\'' <+> name)      (Tag . drop 1)
   <|> match (is '\'' <+> stringLit) (Tag . fromStringLit . drop 1)