ref(moduleName): break into multiple smaller functions

upperId has been changed to only use isUpper because an non alphabetical
character would be false anyway
This commit is contained in:
Primrose 2025-09-24 20:47:55 +08:00
parent 368e5bc9a0
commit 7ceb9b0277
Signed by: primrose
GPG key ID: 4E887A4CA9714ADA
2 changed files with 23 additions and 26 deletions

View file

@ -56,7 +56,7 @@ data Token
| MathsBracketOpen
| MathsBracketClose
| NumericEntity Int
| Module
| Module Text
| QuoteOpen
| QuoteClose
| Space
@ -180,31 +180,31 @@ anchors =
[ between anchorHash anchorHash (Anchor <$> anyUntil anchorHash)
]
moduleName :: Parser Text
moduleName = intercalate "." . fmap Text.pack <$> upperId `sepBy1` char '.'
upperId :: Parser String
upperId = (:) <$> satisfy isUpper <*> many1 identifierChar
identifierChar :: Parser Char
identifierChar = satisfy (\c -> isAlphaNum c || c == '_')
-- "Module.Name"
-- "Module.Name#anchor"
-- "Module.Name\#anchor" -- this has been deprecated for 9 years, thanks Ben
modules :: Lexer
modules = do
startPos <- startPosition $ char '"'
(modPos, modName) <- located modId
anch <- option [] do
anchPos <- startPosition (string "#" <|> string' "\\#")
txt <- Text.pack <$> many (satisfy (\c -> c /= '"' && not (isSpace c)))
pure [(anchPos, Anchor txt)]
void $ char '"'
pure $ [(startPos, Module), (modPos, Token modName)] <> anch
modules = between (char '"') (char '"') inner
where
modId = intercalate "." <$> (fmap Text.pack <$> (conId `sepBy1` (char '.')))
inner = do
module_ <- located $ Module <$> moduleName
mAnchor <- optionMaybe (located $ anchorHash *> (Anchor <$> anchorText))
pure $ case mAnchor of
Just anchor -> [module_, anchor]
Nothing -> [module_]
conId :: Parser String
conId =
(:)
<$> satisfy (\c -> isAlpha c && isUpper c)
<*> many1 conChar
conChar :: Parser Char
conChar = satisfy (\c -> isAlphaNum c || c == '_')
anchorText :: Parser Text
anchorText = Text.pack <$> many (satisfy (\c -> c /= '"' && not (isSpace c)))
linkRaw :: Lexer
linkRaw =