Modifier letter in middle of identifier is ok

Refactoring only. Cleanup some loose ends from #10196.

Initially the idea was to only allow modifier letters at the end of
identifiers. Since we later decided to allow modifier letters also in
the middle of identifiers (because not doing so would not fix the
regression completely), the names `suffix` and `okIdSuffixChar` don't
seem appropriate anymore.

Remove TODO. Move test from should_fail to should_compile.

compiler/basicTypes/Lexeme.hs

@@ -28,7 +28,6 @@ module Lexeme (
   ) where
 
 import FastString
-import Util ((<||>))
 
 import Data.Char
 import qualified Data.Set as Set
@@ -183,8 +182,7 @@ okConSymOcc str = all okSymChar str &&
 -- but not worrying about case or clashing with reserved words?
 okIdOcc :: String -> Bool
 okIdOcc str
-  -- TODO. #10196. Only allow modifier letters in the suffix of an identifier.
-  = let hashes = dropWhile (okIdChar <||> okIdSuffixChar) str in
+  = let hashes = dropWhile okIdChar str in
     all (== '#') hashes   -- -XMagicHash allows a suffix of hashes
                           -- of course, `all` says "True" to an empty list
 
@@ -194,19 +192,13 @@ okIdChar :: Char -> Bool
 okIdChar c = case generalCategory c of
   UppercaseLetter -> True
   LowercaseLetter -> True
-  OtherLetter     -> True
   TitlecaseLetter -> True
+  ModifierLetter  -> True -- See #10196
+  OtherLetter     -> True
   DecimalNumber   -> True
   OtherNumber     -> True
   _               -> c == '\'' || c == '_'
 
--- | Is this character acceptable in the suffix of an identifier.
--- See alexGetByte in Lexer.x
-okIdSuffixChar :: Char -> Bool
-okIdSuffixChar c = case generalCategory c of
-  ModifierLetter  -> True  -- See #10196
-  _               -> False
-
 -- | Is this character acceptable in a symbol (after the first char)?
 -- See alexGetByte in Lexer.x
 okSymChar :: Char -> Bool

compiler/parser/Lexer.x

@@ -155,9 +155,8 @@ $binit     = 0-1
 $octit     = 0-7
 $hexit     = [$decdigit A-F a-f]
 
-$suffix    = \x07 -- Trick Alex into handling Unicode. See alexGetByte.
--- TODO #10196. Only allow modifier letters in the suffix of an identifier.
-$idchar    = [$small $large $digit $suffix \']
+$modifier  = \x07 -- Trick Alex into handling Unicode. See alexGetByte.
+$idchar    = [$small $large $digit $modifier \']
 
 $pragmachar = [$small $large $digit]
 
@@ -1875,7 +1874,7 @@ alexGetByte (AI loc s)
         symbol          = '\x04'
         space           = '\x05'
         other_graphic   = '\x06'
-        suffix          = '\x07'
+        modifier        = '\x07'
 
         adj_c
           | c <= '\x06' = non_graphic
@@ -1892,7 +1891,7 @@ alexGetByte (AI loc s)
                   UppercaseLetter       -> upper
                   LowercaseLetter       -> lower
                   TitlecaseLetter       -> upper
-                  ModifierLetter        -> suffix -- see #10196
+                  ModifierLetter        -> modifier -- see #10196
                   OtherLetter           -> lower -- see #1103
                   NonSpacingMark        -> other_graphic
                   SpacingCombiningMark  -> other_graphic

testsuite/tests/parser/should_compile/T10196.hs

@@ -11,3 +11,8 @@ f =
       xᵪ = xᵢ
       xᵣ = xᵪ
   in xᵣ
+
+-- Modifier letters are also allowed in the middle of an identifier.
+-- This should not be lexed as 2 separate identifiers.
+xᵦx :: Int
+xᵦx = 1

testsuite/tests/parser/should_fail/all.T

@@ -91,6 +91,5 @@ test('T8506', normal, compile_fail, [''])
 test('T9225', normal, compile_fail, [''])
 test('T10196Fail1', normal, compile_fail, [''])
 test('T10196Fail2', normal, compile_fail, [''])
-test('T10196Fail3', expect_broken(10196), compile_fail, [''])
 test('T10498a', normal, compile_fail, [''])
 test('T10498b', normal, compile_fail, [''])