Safe Haskell | Safe-Inferred |
---|---|
Language | Haskell2010 |
If you are only interested in full parses of some parser p
, enforce
full parses on the Parser
level using the
eof
combinator as in
. This ensures error reporting.parseImproved
(p <*
eof
) input
Report bugs to gitlab
or p.rednaz@googlemail.com
, please.
Synopsis
- parseImproved :: (ErrorsPretty s, Ord s) => Parser s a -> [s] -> [(a, [s])]
- parseAndTraceImproved :: (ErrorsPretty s, Ord s) => Config -> Parser s a -> [s] -> [(a, [s])]
- parseWithConfig :: Ord s => Config -> Parser s a -> [s] -> Either (ParseErrorBundle [s]) (NonEmpty (a, [s]))
- parseAndTrace :: (Show s, Ord s) => Config -> Parser s a -> [s] -> [(a, [s])]
- parse :: (Show s, Ord s) => Parser s a -> [s] -> [(a, [s])]
- type Parser s r = [s] -> ([(r, [s])], DifferenceList (ParseError [s]))
- anySymbol :: Parser s s
- satisfy :: (s -> Bool) -> Parser s s
- empty :: Parser s a
- failp :: Parser s a
- succeed :: a -> Parser s a
- pure :: a -> Parser s a
- fail :: String -> Parser s a
- expected :: [s] -> Parser s a
- (<|>) :: Parser s a -> Parser s a -> Parser s a
- (<<|>) :: Parser s a -> Parser s a -> Parser s a
- (<*>) :: Parser s (b -> a) -> Parser s b -> Parser s a
- (<$>) :: (a -> b) -> Parser s a -> Parser s b
- (>>=) :: Parser s a -> (a -> Parser s b) -> Parser s b
- look :: Parser s [s]
Running parsers
parseImproved :: (ErrorsPretty s, Ord s) => Parser s a -> [s] -> [(a, [s])] Source #
Runs a parser on a given string printing error messages to standard
error (stderr) like parse
but makes error messages bearable for
Parser Char
. parseImproved
is always preferable to parse
.
Notice that, when using parseImproved
, you might need to add Ord
and
ErrorsPretty
constraints to your own functions and ensure your own
data types are deriving (
.Ord
, Show
)
The ErrorsPretty
constraint is automatically fulfilled by Show
instances. But if you see the following GHC error, you usually need to
add a (
constraint to your function and ErrorsPretty
s)import
ParseLib.Error (
.ErrorsPretty
)
Overlapping instances for ErrorsPretty s arising from a use of ‘parseImproved’
ErrorsPretty
is not defined in this package but in
uu-tc-error-error
. We
did this so you can switch back and forth between this library and
uu-tc
without the need to
remove ErrorsPretty
constraints from your code. Just permanently keep
uu-tc-error-error
in your .cabal
file. It does not conflict with
uu-tc
because there are
no module name collisions.
parseAndTraceImproved :: (ErrorsPretty s, Ord s) => Config -> Parser s a -> [s] -> [(a, [s])] Source #
Runs a parser on a given string printing error messages to standard
error (stderr) like parseAndTrace
but makes error messages bearable
for Parser Char
. parseAndTraceImproved
is always preferable to
parse
.
The ErrorsPretty
constraint is automatically fulfilled by Show
instances. But if you see the following GHC error, you usually need to
add a (
constraint to your function and ErrorsPretty
s)import
ParseLib.Error (
.ErrorsPretty
)
Overlapping instances for ErrorsPretty s arising from a use of ‘parseImproved’
ErrorsPretty
is not defined in this package but in
uu-tc-error-error
. We
did this so you can switch back and forth between this library and
uu-tc
without the need to
remove ErrorsPretty
constraints from your code. Just permanently keep
uu-tc-error-error
in your .cabal
file. It does not conflict with
uu-tc
because there are
no module name collisions.
parseWithConfig :: Ord s => Config -> Parser s a -> [s] -> Either (ParseErrorBundle [s]) (NonEmpty (a, [s])) Source #
Runs a parser on a given string. Pretty print the error information
with errorBundlePrettyImproved
.
parseAndTrace :: (Show s, Ord s) => Config -> Parser s a -> [s] -> [(a, [s])] Source #
Runs a parser on a given string printing error messages to standard
error (stderr) like parseAndTraceImproved
but with much worse error
messages for Parser Char
. parseAndTraceImproved
is always preferable
to parseAndTrace
.
parse :: (Show s, Ord s) => Parser s a -> [s] -> [(a, [s])] Source #
Runs a parser on a given string printing error messages to standard
error (stderr) like parseImproved
but with much worse error messages
for Parser Char
. parseImproved
is always preferable to parse
.
Notice that, when using parse
, you might need to add Ord
and Show
constraints to your own functions and ensure your own data types are
deriving (
.Ord
, Show
)
The type of parsers
type Parser s r = [s] -> ([(r, [s])], DifferenceList (ParseError [s])) Source #
An input string is mapped to a list of successful parses.
For each succesful parse, we return the result of type r
,
and the remaining input string. The input must be a list of
symbols.
Elementary parsers
satisfy :: (s -> Bool) -> Parser s s Source #
Takes a predicate and returns a parser that parses a single symbol satisfying that predicate.
pure :: a -> Parser s a Source #
Same as succeed
; provided for compatiblity with the applicative
interface.
Parser combinators
(<|>) :: Parser s a -> Parser s a -> Parser s a infixr 3 Source #
Choice between two parsers with the same result type.
(<<|>) :: Parser s a -> Parser s a -> Parser s a infixr 3 Source #
Biased choice. If the left hand side parser succeeds, the right hand side is not considered. Use with care!
(>>=) :: Parser s a -> (a -> Parser s b) -> Parser s b infixl 1 Source #
Monadic bind. Do not use this combinator unless absolutely
required. Most sequencing can be done with <*>
.