megaparsec/tests/Util.hs
mrkkrp a3254f5371 Move language extensions to files
This turns out that it's easier for readers to lookup the list of
extensions in current file than go to Cabal file.
2016-04-10 19:26:54 +07:00

269 lines
9.5 KiB
Haskell
Raw Blame History

This file contains ambiguous Unicode characters

This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.

--
-- QuickCheck tests for Megaparsec, utility functions for parser testing.
--
-- Copyright © 20152016 Megaparsec contributors
--
-- Redistribution and use in source and binary forms, with or without
-- modification, are permitted provided that the following conditions are
-- met:
--
-- * Redistributions of source code must retain the above copyright notice,
-- this list of conditions and the following disclaimer.
--
-- * Redistributions in binary form must reproduce the above copyright
-- notice, this list of conditions and the following disclaimer in the
-- documentation and/or other materials provided with the distribution.
--
-- THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS “AS IS” AND ANY
-- EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
-- WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
-- DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDERS BE LIABLE FOR ANY
-- DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
-- DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
-- OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
-- HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT,
-- STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN
-- ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
-- POSSIBILITY OF SUCH DAMAGE.
{-# LANGUAGE CPP #-}
{-# LANGUAGE FlexibleContexts #-}
{-# LANGUAGE RankNTypes #-}
module Util
( checkParser
, checkParser'
, checkCase
, simpleParse
, checkChar
, checkString
, updatePosString
, (/=\)
, (!=!)
, abcRow
, posErr
, uneCh
, uneStr
, uneSpec
, uneEof
, exCh
, exStr
, exSpec
, exEof
, msg
, showToken )
where
import Control.Monad.Reader
import Control.Monad.Trans.Identity
import Data.Foldable (foldl')
import Data.Maybe (maybeToList)
import qualified Control.Monad.State.Lazy as L
import qualified Control.Monad.State.Strict as S
import qualified Control.Monad.Writer.Lazy as L
import qualified Control.Monad.Writer.Strict as S
import Test.QuickCheck
import Test.HUnit (Assertion, (@?=))
import Text.Megaparsec.Error
import Text.Megaparsec.Pos
import Text.Megaparsec.Prim
import Text.Megaparsec.ShowToken
import Text.Megaparsec.String
#if !MIN_VERSION_base(4,8,0)
import Control.Applicative ((<$>), (<*))
#endif
-- | @checkParser p r s@ tries to run parser @p@ on input @s@ to parse
-- entire @s@. Result of the parsing is compared with expected result @r@,
-- it should match, otherwise the property doesn't hold and the test fails.
checkParser :: (Eq a, Show a)
=> Parser a -- ^ Parser to test
-> Either ParseError a -- ^ Expected result of parsing
-> String -- ^ Input for the parser
-> Property -- ^ Resulting property
checkParser p r s = simpleParse p s === r
-- | A variant of 'checkParser' that runs given parser code with all
-- standard instances of 'MonadParsec'. Useful when testing primitive
-- combinators.
checkParser' :: (Eq a, Show a)
=> (forall m. MonadParsec String m Char => m a) -- ^ Parser to test
-> Either ParseError a -- ^ Expected result of parsing
-> String -- ^ Input for the parser
-> Property -- ^ Resulting property
checkParser' p r s = conjoin
[ checkParser p r s
, checkParser (runIdentityT p) r s
, checkParser (runReaderT p ()) r s
, checkParser (L.evalStateT p ()) r s
, checkParser (S.evalStateT p ()) r s
, checkParser (evalWriterTL p) r s
, checkParser (evalWriterTS p) r s ]
-- | Similar to 'checkParser', but produces HUnit's 'Assertion's instead.
checkCase :: (Eq a, Show a)
=> (forall m. MonadParsec String m Char => m a) -- ^ Parser to test
-> Either ParseError a -- ^ Expected result of parsing
-> String -- ^ Input for the parser
-> Assertion -- ^ Resulting assertion
checkCase p r s = do
parse p "" s @?= r
parse (runIdentityT p) "" s @?= r
parse (runReaderT p ()) "" s @?= r
parse (L.evalStateT p ()) "" s @?= r
parse (S.evalStateT p ()) "" s @?= r
parse (evalWriterTL p) "" s @?= r
parse (evalWriterTS p) "" s @?= r
evalWriterTL :: Monad m => L.WriterT [Int] m a -> m a
evalWriterTL = liftM fst . L.runWriterT
evalWriterTS :: Monad m => S.WriterT [Int] m a -> m a
evalWriterTS = liftM fst . S.runWriterT
-- | @simpleParse p s@ runs parser @p@ on input @s@ and returns corresponding
-- result of type @Either ParseError a@, where @a@ is type of parsed
-- value. This parser tries to parser end of file too and name of input file
-- is always empty string.
simpleParse :: Parser a -> String -> Either ParseError a
simpleParse p = parse (p <* eof) ""
-- | @checkChar p test label s@ runs parser @p@ on input @s@ and checks if
-- the parser correctly parses single character that satisfies @test@. The
-- character may be labelled, in this case @label@ is used to check quality
-- of error messages.
checkChar :: Parser Char -> (Char -> Bool)
-> Maybe String -> String -> Property
checkChar p f l' s = checkParser p r s
where h = head s
l = exSpec <$> maybeToList l'
r | null s = posErr 0 s (uneEof : l)
| length s == 1 && f h = Right h
| not (f h) = posErr 0 s (uneCh h : l)
| otherwise = posErr 1 s [uneCh (s !! 1), exEof]
-- | @checkString p a test label s@ runs parser @p@ on input @s@ and checks if
-- the result is equal to @a@ and also quality of error messages. @test@ is
-- used to compare tokens. @label@ is used as expected representation of
-- parser's result in error messages.
checkString :: Parser String -> String -> (Char -> Char -> Bool)
-> String -> String -> Property
checkString p a' test l s' = checkParser p (w a' 0 s') s'
where w [] _ [] = Right s'
w [] i (s:_) = posErr i s' [uneCh s, exEof]
w _ 0 [] = posErr 0 s' [uneEof, exSpec l]
w _ i [] = posErr 0 s' [uneStr (take i s'), exSpec l]
w (a:as) i (s:ss)
| test a s = w as i' ss
| otherwise = posErr 0 s' [uneStr (take i' s'), exSpec l]
where i' = succ i
-- | A helper function that is used to advance 'SourcePos' given a 'String'.
updatePosString
:: Int -- ^ Tab width
-> SourcePos -- ^ Initial position
-> String -- ^ 'String' — collection of tokens to process
-> SourcePos -- ^ Final position
updatePosString w = foldl' f
where f p t = snd (defaultUpdatePos w p t)
infix 4 /=\ -- preserve whitespace on automatic trim
-- | @p /=\\ x@ runs parser @p@ on empty input and compares its result
-- (which should be successful) with @x@. Succeeds when the result is equal
-- to @x@, prints counterexample on failure.
(/=\) :: (Eq a, Show a) => Parser a -> a -> Property
p /=\ x = simpleParse p "" === Right x
infix 4 !=!
-- | @n !=! m@ represents property that holds when results of running @n@
-- and @m@ parsers are identical. This is useful when checking monad laws
-- for example.
(!=!) :: (Eq a, Show a) => Parser a -> Parser a -> Property
n !=! m = simpleParse n "" === simpleParse m ""
-- | @abcRow a b c@ generates string consisting of character “a” repeated
-- @a@ times, character “b” repeated @b@ times, and finally character “c”
-- repeated @c@ times.
abcRow :: Enum a => a -> a -> a -> String
abcRow a b c = f a 'a' ++ f b 'b' ++ f c 'c'
where f x = replicate (fromEnum x)
-- | @posErr pos s ms@ is an easy way to model result of parser that
-- fails. @pos@ is how many tokens (characters) has been consumed before
-- failure. @s@ is input of the parser. @ms@ is a list, collection of
-- 'Message's. See 'uneStr', 'uneCh', 'uneSpec', 'exStr', 'exCh', and
-- 'exSpec' for easy ways to create error messages.
posErr :: Int -> String -> [Message] -> Either ParseError a
posErr pos s = Left . foldr addErrorMessage (newErrorUnknown errPos)
where errPos = updatePosString defaultTabWidth (initialPos "") (take pos s)
-- | @uneCh s@ returns message created with 'Unexpected' constructor that
-- tells the system that char @s@ is unexpected.
uneCh :: Char -> Message
uneCh s = Unexpected $ showToken s
-- | @uneStr s@ returns message created with 'Unexpected' constructor that
-- tells the system that string @s@ is unexpected.
uneStr :: String -> Message
uneStr s = Unexpected $ showToken s
-- | @uneSpec s@ returns message created with 'Unexpected' constructor that
-- tells the system that @s@ is unexpected. This is different from 'uneStr'
-- in that it doesn't use 'showToken' but rather pass its argument unaltered
-- allowing for “special” labels.
uneSpec :: String -> Message
uneSpec = Unexpected
-- | @uneEof@ represents message “unexpected end of input”.
uneEof :: Message
uneEof = Unexpected "end of input"
-- | @exCh s@ returns message created with 'Expected' constructor that tells
-- the system that character @s@ is expected.
exCh :: Char -> Message
exCh s = Expected $ showToken s
-- | @exStr s@ returns message created with 'Expected' constructor that tells
-- the system that string @s@ is expected.
exStr :: String -> Message
exStr s = Expected $ showToken s
-- | @exSpec s@ returns message created with 'Expected' constructor that tells
-- the system that @s@ is expected. This is different from 'exStr' in that
-- it doesn't use 'showToken' but rather pass its argument unaltered
-- allowing for “special” labels.
exSpec :: String -> Message
exSpec = Expected
-- | @exEof@ represents message “expecting end of input”.
exEof :: Message
exEof = Expected "end of input"
-- | @msg s@ return message created with 'Message' constructor.
msg :: String -> Message
msg = Message