Commit ed1b3db362bf9773695cf5333dc98a960820b7a3 - s-cargot

Big final bunch of refactors/documentation bits in preparation for Hackage release Getty Ritter 8 years ago

13 changed file(s) with 887 addition(s) and 650 deletion(s). Collapse all Expand all

-41

~~Data/SCargot/Basic.hs~~ less more

1		{-# LANGUAGE OverloadedStrings #-}
2
3		module Data.SCargot.Basic
4		( -- * Spec
5		-- $descr
6		basicSpec
7		) where
8
9		import Control.Applicative ((<$>))
10		import Data.Char (isAlphaNum)
11		import Text.Parsec (many1, satisfy)
12		import Data.Text (Text, pack)
13
14		import Data.SCargot.Repr.Basic (SExpr)
15		import Data.SCargot.General ( SExprSpec
16		, mkSpec
17		)
18		import Data.SCargot.Comments (withLispComments)
19
20		isAtomChar :: Char -> Bool
21		isAtomChar c = isAlphaNum c
22		\|\| c == '-' \|\| c == '*' \|\| c == '/'
23		\|\| c == '+' \|\| c == '<' \|\| c == '>'
24		\|\| c == '=' \|\| c == '!' \|\| c == '?'
25
26		-- $descr
27		-- The 'basicSpec' describes S-expressions whose atoms are simply
28		-- text strings that contain alphanumeric characters and a small
29		-- set of punctuation. It does no parsing of numbers or other data
30		-- types, and will accept tokens that typical Lisp implementations
31		-- would find nonsensical (like @77foo@).
32		--
33		-- Atoms recognized by the 'basicSpec' are any string matching the
34		-- regular expression @[A-Za-z0-9+*<>/=!?-]+@.
35
36		-- \| A 'SExprSpec' that understands atoms to be sequences of
37		-- alphanumeric characters as well as the punctuation
38		-- characters @[-*/+<>=!?]@, and does no processing of them.
39		basicSpec :: SExprSpec Text (SExpr Text)
40		basicSpec = mkSpec pToken id
41		where pToken = pack <$> many1 (satisfy isAtomChar)

+27

-26

Data/SCargot/Comments.hs less more

4	4	( -- $intro
5	5
6	6	-- * Lisp-Style Syntax
	7
7	8	-- $lisp
8	9	withLispComments
9	10	-- * Other Existing Comment Syntaxes

35	36	, string
36	37	)
37	38
38		import Data.SCargot.General ( Comment
39		, SExprSpec
40		, setComment
41		)
	39	import Data.SCargot.Parse ( Comment
	40	, SExprParser
	41	, setComment
	42	)
42	43
43	44	-- \| Given a string, produce a comment parser that matches that
44	45	-- initial string and ignores everything until the end of the

73	74	-- \| Lisp-style line-oriented comments start with @;@ and last
74	75	-- until the end of the line. This is usually the comment
75	76	-- syntax you want.
76		withLispComments :: SExpr~~Spec t a -> SExprSpec~~ t a
	77	withLispComments :: SExprParser t a -> SExprParser t a
77	78	withLispComments = setComment (lineComment ";")
78	79
79	80	-- \| C++-like line-oriented comment start with @//@ and last
80	81	-- until the end of the line.
81		withCLikeLineComments :: SExpr~~Spec t a -> SExprSpec~~ t a
	82	withCLikeLineComments :: SExprParser t a -> SExprParser t a
82	83	withCLikeLineComments = setComment (lineComment "//")
83	84
84	85	-- \| C-like block comments start with @/@ and end with @/@.
85	86	-- They do not nest.
86		withCLikeBlockComments :: SExpr~~Spec t a -> SExprSpec~~ t a
	87	withCLikeBlockComments :: SExprParser t a -> SExprParser t a
87	88	withCLikeBlockComments = setComment (simpleBlockComment "/" "/")
88	89
89	90	-- \| C-like comments include both line- and block-comments, the
90	91	-- former starting with @//@ and the latter contained within
91	92	-- @//* ... *//@.
92		withCLikeComments :: SExpr~~Spec t a -> SExprSpec~~ t a
	93	withCLikeComments :: SExprParser t a -> SExprParser t a
93	94	withCLikeComments = setComment (lineComment "//" <\|>
94	95	simpleBlockComment "/" "/")
95	96
96	97	-- \| Haskell line-oriented comments start with @--@ and last
97	98	-- until the end of the line.
98		withHaskellLineComments :: SExpr~~Spec t a -> SExprSpec~~ t a
	99	withHaskellLineComments :: SExprParser t a -> SExprParser t a
99	100	withHaskellLineComments = setComment (lineComment "--")
100	101
101	102	-- \| Haskell block comments start with @{-@ and end with @-}@.
102	103	-- They do not nest.
103		withHaskellBlockComments :: SExpr~~Spec t a -> SExprSpec~~ t a
	104	withHaskellBlockComments :: SExprParser t a -> SExprParser t a
104	105	withHaskellBlockComments = setComment (simpleBlockComment "{-" "-}")
105	106
106	107	-- \| Haskell comments include both the line-oriented @--@ comments
107	108	-- and the block-oriented @{- ... -}@ comments
108		withHaskellComments :: SExpr~~Spec t a -> SExprSpec~~ t a
	109	withHaskellComments :: SExprParser t a -> SExprParser t a
109	110	withHaskellComments = setComment (lineComment "--" <\|>
110	111	simpleBlockComment "{-" "-}")
111	112
112	113	-- \| Many scripting and shell languages use these, which begin with
113	114	-- @#@ and last until the end of the line.
114		withOctothorpeComments :: SExpr~~Spec t a -> SExprSpec~~ t a
	115	withOctothorpeComments :: SExprParser t a -> SExprParser t a
115	116	withOctothorpeComments = setComment (lineComment "#")
116	117
117	118
118	119	{- $intro
119	120
120		By default a 'SExpr~~Spec~~' will not understand any kind of comment
	121	By default a 'SExprParser' will not understand any kind of comment
121	122	syntax. Most varieties of s-expression will, however, want some kind
122	123	of commenting capability, so the below functions will produce a new
123		'SExpr~~Spec~~' which understands various kinds of comment syntaxes.
	124	'SExprParser' which understands various kinds of comment syntaxes.
124	125
125	126	For example:
126	127
127		> mySpec :: SExprSpec Text (SExpr Text)
128		> mySpec = asWellFormed $ mkSpec (pack <$> many1 alphaNum) id
	128	> mySpec :: SExprParser Text (SExpr Text)
	129	> mySpec = asWellFormed $ mkParser (pack <$> many1 alphaNum)
129	130	>
130		> myLispySpec :: SExpr~~Spec~~ Text (SExpr Text)
	131	> myLispySpec :: SExprParser Text (SExpr Text)
131	132	> myLispySpec = withLispComments mySpec
132	133	>
133		> myCLikeSpec :: SExpr~~Spec~~ Text (SExpr Text)
	134	> myCLikeSpec :: SExprParser Text (SExpr Text)
134	135	> myCLikeSpec = withCLikeComment mySpec
135	136
136	137	We can then use these to parse s-expressions with different kinds of
137	138	comment syntaxes:
138	139
139		> > decode mySpec "(foo ; a lisp comment\n bar)\n"
140		> Left "(line 1, column 6):\nunexpected \";\"\nexpecting space or atom"
141		> > decode myLispySpec "(foo ; a lisp comment\n bar)\n"
142		> Right [WFSList [WFSAtom "foo", WFSAtom "bar"]]
143		> > decode mySpec "(foo /* a c-like\n comment */ bar)\n"
144		> Left "(line 1, column 6):\nunexpected \"/\"\nexpecting space or atom"
145		> > decode myCLikeSpec "(foo /* a c-like\n comment */ bar)\n"
146		> Right [WFSList [WFSAtom "foo", WFSAtom "bar"]]
	140	>>> decode mySpec "(foo ; a lisp comment\n bar)\n"
	141	Left "(line 1, column 6):\nunexpected \";\"\nexpecting space or atom"
	142	>>> decode myLispySpec "(foo ; a lisp comment\n bar)\n"
	143	Right [WFSList [WFSAtom "foo", WFSAtom "bar"]]
	144	>>> decode mySpec "(foo /* a c-like\n comment */ bar)\n"
	145	Left "(line 1, column 6):\nunexpected \"/\"\nexpecting space or atom"
	146	>>> decode myCLikeSpec "(foo /* a c-like\n comment */ bar)\n"
	147	Right [WFSList [WFSAtom "foo", WFSAtom "bar"]]
147	148
148	149	-}
149	150

+22

-3

Data/SCargot/Common.hs less more

4	4	, parseR6RSIdent
5	5	, parseR7RSIdent
6	6	-- * Numeric Literal Parsers
	7	, signed
	8	, prefixedNumber
	9	, signedPrefixedNumber
7	10	, binNumber
8	11	, signedBinNumber
9	12	, octNumber

14	17	, signedDozNumber
15	18	, hexNumber
16	19	, signedHexNumber
17		, signed
18	20	) where
19	21
20	22	import Data.Char

42	44
43	45	-- \| Parse an identifier according to the R6RS Scheme standard. An
44	46	-- R6RS identifier may include inline hexadecimal escape sequences
45		-- so that, for example, @foo@ is equivalent to @f\x6f;o@, and is
	47	-- so that, for example, @foo@ is equivalent to @f\\x6f;o@, and is
46	48	-- more liberal than R5RS as to which Unicode characters it may
47	49	-- accept.
48	50	parseR6RSIdent :: Parser Text

141	143	signed :: Num a => Parser a -> Parser a
142	144	signed p = ($) <$> sign <*> p
143	145
	146	-- \| Parses a number in the same way as 'prefixedNumber', with an optional
	147	-- leading @+@ or @-@.
	148	signedPrefixedNumber :: Parser Integer
	149	signedPrefixedNumber = signed prefixedNumber
	150
	151	-- \| Parses a number, determining which numeric base to use by examining
	152	-- the literal's prefix: @0x@ for a hexadecimal number, @0z@ for a
	153	-- dozenal number, @0o@ for an octal number, and @0b@ for a binary
	154	-- number (as well as the upper-case versions of the same.) If the
	155	-- base is omitted entirely, then it is treated as a decimal number.
	156	prefixedNumber :: Parser Integer
	157	prefixedNumber = (string "0x" <\|> string "0X") *> hexNumber
	158	<\|> (string "0o" <\|> string "0O") *> octNumber
	159	<\|> (string "0z" <\|> string "0Z") *> dozNumber
	160	<\|> (string "0b" <\|> string "0B") *> binNumber
	161	<\|> decNumber
	162
144	163	-- \| A parser for non-signed binary numbers
145	164	binNumber :: Parser Integer
146	165	binNumber = number 2 (char '0' <\|> char '1')
147	166
148	167	-- \| A parser for signed binary numbers, with an optional leading @+@ or @-@.
149	168	signedBinNumber :: Parser Integer
150		signedBinNumber = ~~($) <$> sign <*>~~ binNumber
	169	signedBinNumber = signed binNumber
151	170
152	171	-- \| A parser for non-signed octal numbers
153	172	octNumber :: Parser Integer

-286

~~Data/SCargot/General.hs~~ less more

1		{-# LANGUAGE ViewPatterns #-}
2		{-# LANGUAGE OverloadedStrings #-}
3
4		module Data.SCargot.General
5		( -- * SExprSpec
6		SExprSpec
7		, mkSpec
8		, convertSpec
9		, addReader
10		, setComment
11		-- * Specific SExprSpec Conversions
12		, asRich
13		, asWellFormed
14		, withQuote
15		-- * Using a SExprSpec
16		, decode
17		, decodeOne
18		, encode
19		, encodeOne
20		-- * Useful Type Aliases
21		, Reader
22		, Comment
23		, Serializer
24		) where
25
26		import Control.Applicative ((<), (>), (<*>), (<$>), pure)
27		import Control.Monad ((>=>))
28		import Data.Char (isAlpha, isDigit, isAlphaNum)
29		import Data.Map.Strict (Map)
30		import qualified Data.Map.Strict as M
31		import Data.Maybe (fromJust)
32		import Data.Monoid ((<>))
33		import Data.String (IsString)
34		import Data.Text (Text, pack, unpack)
35		import qualified Data.Text as T
36		import Text.Parsec ( (<\|>)
37		, (<?>)
38		, char
39		, eof
40		, lookAhead
41		, many1
42		, runParser
43		, skipMany
44		)
45		import Text.Parsec.Char (anyChar, space)
46		import Text.Parsec.Text (Parser)
47
48		import Data.SCargot.Repr ( SExpr(..)
49		, RichSExpr
50		, WellFormedSExpr
51		, fromRich
52		, toRich
53		, fromWellFormed
54		, toWellFormed
55		)
56
57		type ReaderMacroMap atom = Map Char (Reader atom)
58
59		-- \| A 'Reader' represents a reader macro: it takes a parser for
60		-- the S-Expression type and performs as much or as little
61		-- parsing as it would like, and then returns an S-expression.
62		type Reader atom = (Parser (SExpr atom) -> Parser (SExpr atom))
63
64		-- \| A 'Comment' represents any kind of skippable comment. This
65		-- parser __must__ be able to fail if a comment is not being
66		-- recognized, and it __must__ not consume any input.
67		type Comment = Parser ()
68
69		-- \| A 'Serializer' is any function which can serialize an Atom
70		-- to 'Text'.
71		type Serializer atom = atom -> Text
72
73		-- \| A 'SExprSpec' describes a parser and emitter for a particular
74		-- variant of S-Expressions. The @atom@ type corresponds to a
75		-- Haskell type used to represent the atoms, and the @carrier@
76		-- type corresponds to the parsed S-Expression structure. The
77		-- 'SExprSpec' type is deliberately opaque so that it must be
78		-- constructed and modified with other helper functions.
79		data SExprSpec atom carrier = SExprSpec
80		{ sesPAtom :: Parser atom
81		, sesSAtom :: Serializer atom
82		, readerMap :: ReaderMacroMap atom
83		, comment :: Maybe Comment
84		, postparse :: SExpr atom -> Either String carrier
85		, preserial :: carrier -> SExpr atom
86		}
87
88		-- \| Create a basic 'SExprSpec' when given a parser and serializer
89		-- for an atom type. A small minimal 'SExprSpec' that recognizes
90		-- any alphanumeric sequence as a valid atom looks like:
91		--
92		-- > simpleSpec :: SExprSpec Text (SExpr Text)
93		-- > simpleSpec = mkSpec (pack <$> many1 isAlphaNum) id
94		mkSpec :: Parser atom -> Serializer atom -> SExprSpec atom (SExpr atom)
95		mkSpec p s = SExprSpec
96		{ sesPAtom = p <?> "atom"
97		, sesSAtom = s
98		, readerMap = M.empty
99		, comment = Nothing
100		, postparse = return
101		, preserial = id
102		}
103
104		-- \| Modify the carrier type for a 'SExprSpec'. This is
105		-- used internally to convert between various 'SExpr' representations,
106		-- but could also be used externally to add an extra conversion layer
107		-- onto a 'SExprSpec'.
108		--
109		-- The following defines an S-expression spec that recognizes the
110		-- language of binary addition trees. It does so by first transforming
111		-- the internal S-expression representation using 'asWellFormed', and
112		-- then providing a conversion between the 'WellFormedSExpr' type and
113		-- an @Expr@ AST. Notice that the below parser uses 'String' as its
114		-- underlying atom type, instead of some token type.
115		--
116		-- > data Expr = Add Expr Expr \| Num Int deriving (Eq, Show)
117		-- >
118		-- > toExpr :: WellFormedSExpr String -> Either String Expr
119		-- > toExpr (L [A "+", l, r]) = Add <$> toExpr l <*> toExpr r
120		-- > toExpr (A c) \| all isDigit c = pure (Num (read c))
121		-- > toExpr c = Left ("Invalid expr: " ++ show c)
122		-- >
123		-- > fromExpr :: Expr -> WellFormedSExpr String
124		-- > fromExpr (Add l r) = L [A "+", fromExpr l, fromExpr r]
125		-- > fromExpr (Num n) = A (show n)
126		-- >
127		-- > mySpec :: SExprSpec String Expr
128		-- > mySpec = convertSpec toExpr fromExpr $ asWellFormed $ mkSpec parser pack
129		-- > where parser = many1 (satisfy isValidChar)
130		-- > isValidChar c = isDigit c \|\| c == '+'
131		convertSpec :: (b -> Either String c) -> (c -> b)
132		-> SExprSpec a b -> SExprSpec a c
133		convertSpec f g spec = spec
134		{ postparse = postparse spec >=> f
135		, preserial = preserial spec . g
136		}
137
138		-- \| Convert the final output representation from the 'SExpr' type
139		-- to the 'RichSExpr' type.
140		asRich :: SExprSpec a (SExpr b) -> SExprSpec a (RichSExpr b)
141		asRich = convertSpec (return . toRich) fromRich
142
143		-- \| Convert the final output representation from the 'SExpr' type
144		-- to the 'WellFormedSExpr' type.
145		asWellFormed :: SExprSpec a (SExpr b) -> SExprSpec a (WellFormedSExpr b)
146		asWellFormed = convertSpec toWellFormed fromWellFormed
147
148		-- \| Add the ability to execute some particular reader macro, as
149		-- defined by its initial character and the 'Parser' which returns
150		-- the parsed S-Expression. The 'Reader' is passed a 'Parser' which
151		-- can be recursively called to parse more S-Expressions, and begins
152		-- parsing after the reader character has been removed from the
153		-- stream.
154		--
155		-- The following defines an S-expression variant that treats
156		-- @'expr@ as being sugar for @(quote expr)@. Note that this is done
157		-- already in a more general way by the 'withQuote' function, but
158		-- it is a good illustration of using reader macros in practice:
159		--
160		-- > mySpec :: SExprSpec String (SExpr Text)
161		-- > mySpec = addReader '\'' reader $ mkSpec (many1 alphaNum) pack
162		-- > where reader p = quote <$> p
163		-- > quote e = SCons (SAtom "quote") (SCons e SNil)
164		addReader :: Char -> Reader a -> SExprSpec a c -> SExprSpec a c
165		addReader c reader spec = spec
166		{ readerMap = M.insert c reader (readerMap spec) }
167
168		-- \| Add the ability to ignore some kind of comment. This gets
169		-- factored into whitespace parsing, and it's very important that
170		-- the parser supplied __be able to fail__ (as otherwise it will
171		-- cause an infinite loop), and also that it __not consume any input__
172		-- (which may require it to be wrapped in 'try'.)
173		--
174		-- The following code defines an S-expression variant that skips
175		-- C++-style comments, i.e. those which begin with @//@ and last
176		-- until the end of a line:
177		--
178		-- > t :: SExprSpec String (SExpr Text)
179		-- > t = setComment comm $ mkSpec (many1 alphaNum) pack
180		-- > where comm = try (string "//" > manyTill newline > pure ())
181
182		setComment :: Comment -> SExprSpec a c -> SExprSpec a c
183		setComment c spec = spec { comment = Just (c <?> "comment") }
184
185		-- \| Add the ability to understand a quoted S-Expression. In general,
186		-- many Lisps use @'sexpr@ as sugar for @(quote sexpr)@. This is
187		-- a convenience function which allows you to easily add quoted
188		-- expressions to a 'SExprSpec', provided that you supply which
189		-- atom you want substituted in for the symbol @quote@.
190		withQuote :: IsString t => SExprSpec t (SExpr t) -> SExprSpec t (SExpr t)
191		withQuote = addReader '\'' (fmap go)
192		where go s = SCons "quote" (SCons s SNil)
193
194		peekChar :: Parser (Maybe Char)
195		peekChar = Just <$> lookAhead anyChar <\|> pure Nothing
196
197		parseGenericSExpr ::
198		Parser atom -> ReaderMacroMap atom -> Parser () -> Parser (SExpr atom)
199		parseGenericSExpr atom reader skip = do
200		let sExpr = parseGenericSExpr atom reader skip <?> "s-expr"
201		skip
202		c <- peekChar
203		r <- case c of
204		Nothing -> fail "Unexpected end of input"
205		Just '(' -> char '(' >> skip >> parseList sExpr skip
206		Just (flip M.lookup reader -> Just r) -> anyChar >> r sExpr
207		_ -> SAtom `fmap` atom
208		skip
209		return r
210
211		parseList :: Parser (SExpr atom) -> Parser () -> Parser (SExpr atom)
212		parseList sExpr skip = do
213		i <- peekChar
214		case i of
215		Nothing -> fail "Unexpected end of input"
216		Just ')' -> char ')' >> return SNil
217		_ -> do
218		car <- sExpr
219		skip
220		c <- peekChar
221		case c of
222		Just '.' -> do
223		char '.'
224		cdr <- sExpr
225		skip
226		char ')'
227		skip
228		return (SCons car cdr)
229		Just ')' -> do
230		char ')'
231		skip
232		return (SCons car SNil)
233		_ -> do
234		cdr <- parseList sExpr skip
235		return (SCons car cdr)
236
237		-- \| Given a CommentMap, create the corresponding parser to
238		-- skip those comments (if they exist).
239		buildSkip :: Maybe (Parser ()) -> Parser ()
240		buildSkip Nothing = skipMany space
241		buildSkip (Just c) = alternate
242		where alternate = skipMany space >> ((c >> alternate) <\|> return ())
243
244		doParse :: Parser a -> Text -> Either String a
245		doParse p t = case runParser p () "" t of
246		Left err -> Left (show err)
247		Right x -> Right x
248
249		-- \| Decode a single S-expression. If any trailing input is left after
250		-- the S-expression (ignoring comments or whitespace) then this
251		-- will fail: for those cases, use 'decode', which returns a list of
252		-- all the S-expressions found at the top level.
253		decodeOne :: SExprSpec atom carrier -> Text -> Either String carrier
254		decodeOne spec = doParse (parser <* eof) >=> (postparse spec)
255		where parser = parseGenericSExpr
256		(sesPAtom spec)
257		(readerMap spec)
258		(buildSkip (comment spec))
259
260		-- \| Decode several S-expressions according to a given 'SExprSpec'. This
261		-- will return a list of every S-expression that appears at the top-level
262		-- of the document.
263		decode :: SExprSpec atom carrier -> Text -> Either String [carrier]
264		decode spec =
265		doParse (many1 parser <* eof) >=> mapM (postparse spec)
266		where parser = parseGenericSExpr
267		(sesPAtom spec)
268		(readerMap spec)
269		(buildSkip (comment spec))
270
271		-- \| Encode (without newlines) a single S-expression.
272		encodeSExpr :: SExpr atom -> (atom -> Text) -> Text
273		encodeSExpr SNil _ = "()"
274		encodeSExpr (SAtom s) t = t s
275		encodeSExpr (SCons x xs) t = go xs (encodeSExpr x t)
276		where go (SAtom s) rs = "(" <> rs <> " . " <> t s <> ")"
277		go SNil rs = "(" <> rs <> ")"
278		go (SCons x xs) rs = go xs (rs <> " " <> encodeSExpr x t)
279
280		-- \| Emit an S-Expression in a machine-readable way. This does no
281		-- pretty-printing or indentation, and produces no comments.
282		encodeOne :: SExprSpec atom carrier -> carrier -> Text
283		encodeOne spec c = encodeSExpr (preserial spec c) (sesSAtom spec)
284
285		encode :: SExprSpec atom carrier -> [carrier] -> Text
286		encode spec cs = T.concat (map (encodeOne spec) cs)

-140

~~Data/SCargot/HaskLike.hs~~ less more

1		{-# LANGUAGE OverloadedStrings #-}
2
3		module Data.SCargot.HaskLike ( -- $info
4		haskLikeSpec
5		, HaskLikeAtom(..)
6		) where
7
8		import Control.Applicative ((<$>), (<*>), (<$))
9		import Data.Maybe (catMaybes)
10		import Data.String (IsString(..))
11		import Data.Text (Text, pack)
12		import Text.Parsec
13		import Text.Parsec.Text (Parser)
14
15		import Prelude hiding (concatMap)
16
17		import Data.SCargot.Common
18		import Data.SCargot.Repr.Basic (SExpr)
19		import Data.SCargot.General (SExprSpec, mkSpec)
20
21		{- $info
22
23		This module is intended for simple, ad-hoc configuration or data formats
24		that might not need their on rich structure but might benefit from a few
25		various kinds of literals. The 'haskLikeSpec' understands identifiers as
26		defined by R5RS, as well as string, integer, and floating-point literals
27		as defined by the Haskell spec. It does _not_ natively understand other
28		data types, such as booleans, vectors, bitstrings, or the like.
29
30		-}
31
32
33		-- \| An atom type that understands Haskell-like values as well as
34		-- Scheme-like identifiers.
35		data HaskLikeAtom
36		= HSIdent Text -- ^ An identifier, parsed according to the R5RS Scheme
37		-- standard
38		\| HSString Text -- ^ A string, parsed according to the syntax for string
39		-- literals in the Haskell report
40		\| HSInt Integer -- ^ An arbitrary-sized integer value, parsed according to
41		-- the syntax for integer literals in the Haskell report
42		\| HSFloat Double -- ^ A double-precision floating-point value, parsed
43		-- according to the syntax for floats in the Haskell
44		-- report
45		deriving (Eq, Show)
46
47		instance IsString HaskLikeAtom where
48		fromString = HSIdent . fromString
49
50		pString :: Parser Text
51		pString = pack . catMaybes <$> between (char '"') (char '"') (many (val <\|> esc))
52		where val = Just <$> satisfy (\ c -> c /= '"' && c /= '\\' && c > '\026')
53		esc = do char '\\'
54		Nothing <$ (gap <\|> char '&') <\|>
55		Just <$> code
56		gap = many1 space >> char '\\'
57		code = eEsc <\|> eNum <\|> eCtrl <\|> eAscii
58		eCtrl = char '^' >> unCtrl <$> upper
59		eNum = (toEnum . fromInteger) <$>
60		(decNumber <\|> (char 'o' >> octNumber)
61		<\|> (char 'x' >> hexNumber))
62		eEsc = choice [ char a >> return b \| (a, b) <- escMap ]
63		eAscii = choice [ try (string a >> return b)
64		\| (a, b) <- asciiMap ]
65		unCtrl c = toEnum (fromEnum c - fromEnum 'A' + 1)
66
67		escMap :: [(Char, Char)]
68		escMap = zip "abfntv\\\"\'" "\a\b\f\n\r\t\v\\\"\'"
69
70		asciiMap :: [(String, Char)]
71		asciiMap = zip
72		["BS","HT","LF","VT","FF","CR","SO","SI","EM"
73		,"FS","GS","RS","US","SP","NUL","SOH","STX","ETX"
74		,"EOT","ENQ","ACK","BEL","DLE","DC1","DC2","DC3"
75		,"DC4","NAK","SYN","ETB","CAN","SUB","ESC","DEL"]
76		("\BS\HT\LF\VT\FF\CR\SO\SI\EM\FS\GS\RS\US\SP\NUL\SOH" ++
77		"\STX\ETX\EOT\ENQ\ACK\BEL\DLE\DC1\DC2\DC3\DC4\NAK" ++
78		"\SYN\ETB\CAN\SUB\ESC\DEL")
79
80		pFloat :: Parser Double
81		pFloat = do
82		n <- decNumber
83		withDot n <\|> noDot n
84		where withDot n = do
85		char '.'
86		m <- decNumber
87		e <- option 1.0 exponent
88		return ((fromIntegral n + asDec m 0) * e)
89		noDot n = do
90		e <- exponent
91		return (fromIntegral n * e)
92		exponent = do
93		oneOf "eE"
94		s <- power
95		x <- decNumber
96		return (10 ** s (fromIntegral x))
97		asDec 0 k = k
98		asDec n k =
99		asDec (n `div` 10) ((fromIntegral (n `rem` 10) + k) * 0.1)
100
101		power :: Num a => Parser (a -> a)
102		power = negate <$ char '-' <\|> id <$ char '+' <\|> return id
103
104		pInt :: Parser Integer
105		pInt = do
106		s <- power
107		n <- pZeroNum <\|> decNumber
108		return (fromIntegral (s n))
109
110		pZeroNum :: Parser Integer
111		pZeroNum = char '0' >>
112		( (oneOf "xX" >> hexNumber)
113		<\|> (oneOf "oO" >> octNumber)
114		<\|> decNumber
115		<\|> return 0
116		)
117
118		pHaskLikeAtom :: Parser HaskLikeAtom
119		pHaskLikeAtom
120		= HSFloat <$> (try pFloat <?> "float")
121		<\|> HSInt <$> (try pInt <?> "integer")
122		<\|> HSString <$> (pString <?> "string literal")
123		<\|> HSIdent <$> (parseR5RSIdent <?> "token")
124
125		sHaskLikeAtom :: HaskLikeAtom -> Text
126		sHaskLikeAtom (HSIdent t) = t
127		sHaskLikeAtom (HSString s) = pack (show s)
128		sHaskLikeAtom (HSInt i) = pack (show i)
129		sHaskLikeAtom (HSFloat f) = pack (show f)
130
131		-- \| This `SExprSpec` understands s-expressions that contain
132		-- Scheme-like tokens, as well as string literals, integer
133		-- literals, and floating-point literals. Each of these values
134		-- is parsed according to the lexical rules in the Haskell
135		-- report, so the same set of string escapes, numeric bases,
136		-- and floating-point options are available. This spec does
137		-- not parse comments and does not understand any reader
138		-- macros.
139		haskLikeSpec :: SExprSpec HaskLikeAtom (SExpr HaskLikeAtom)
140		haskLikeSpec = mkSpec pHaskLikeAtom sHaskLikeAtom

+55

-0

Data/SCargot/Language/Basic.hs less more

	1	{-# LANGUAGE OverloadedStrings #-}
	2
	3	module Data.SCargot.Language.Basic
	4	( -- * Spec
	5	-- $descr
	6	basicParser
	7	, basicPrinter
	8	) where
	9
	10	import Control.Applicative ((<$>))
	11	import Data.Char (isAlphaNum)
	12	import Text.Parsec (many1, satisfy)
	13	import Data.Text (Text, pack)
	14
	15	import Data.SCargot.Repr.Basic (SExpr)
	16	import Data.SCargot ( SExprParser
	17	, SExprPrinter
	18	, mkParser
	19	, flatPrint
	20	)
	21	import Data.SCargot.Comments (withLispComments)
	22
	23	isAtomChar :: Char -> Bool
	24	isAtomChar c = isAlphaNum c
	25	\|\| c == '-' \|\| c == '*' \|\| c == '/'
	26	\|\| c == '+' \|\| c == '<' \|\| c == '>'
	27	\|\| c == '=' \|\| c == '!' \|\| c == '?'
	28
	29	-- $descr
	30	-- The 'basicSpec' describes S-expressions whose atoms are simply
	31	-- text strings that contain alphanumeric characters and a small
	32	-- set of punctuation. It does no parsing of numbers or other data
	33	-- types, and will accept tokens that typical Lisp implementations
	34	-- would find nonsensical (like @77foo@).
	35	--
	36	-- Atoms recognized by the 'basicSpec' are any string matching the
	37	-- regular expression @[A-Za-z0-9+*<>/=!?-]+@.
	38
	39	-- \| A 'SExprParser' that understands atoms to be sequences of
	40	-- alphanumeric characters as well as the punctuation
	41	-- characters @[-*/+<>=!?]@, and does no processing of them.
	42	--
	43	-- >>> decode basicParser "(1 elephant)"
	44	-- Right [SCons (SAtom "1") (SCons (SAtom "elephant") SNil)]
	45	basicParser :: SExprParser Text (SExpr Text)
	46	basicParser = mkParser pToken
	47	where pToken = pack <$> many1 (satisfy isAtomChar)
	48
	49	-- \| A 'SExprPrinter' that prints textual atoms directly (without quoting
	50	-- or any other processing) onto a single line.
	51	--
	52	-- >>> encode basicPrinter [L [A "1", A "elephant"]]
	53	-- "(1 elephant)"
	54	basicPrinter :: SExprPrinter Text (SExpr Text)
	55	basicPrinter = flatPrint id

+156

-0

Data/SCargot/Language/HaskLike.hs less more

	1	{-# LANGUAGE OverloadedStrings #-}
	2
	3	module Data.SCargot.Language.HaskLike
	4	( -- $info
	5	HaskLikeAtom(..)
	6	, haskLikeParser
	7	, haskLikePrinter
	8	) where
	9
	10	import Control.Applicative ((<$>), (<*>), (<$))
	11	import Data.Maybe (catMaybes)
	12	import Data.String (IsString(..))
	13	import Data.Text (Text, pack)
	14	import Text.Parsec
	15	import Text.Parsec.Text (Parser)
	16
	17	import Prelude hiding (concatMap)
	18
	19	import Data.SCargot.Common
	20	import Data.SCargot.Repr.Basic (SExpr)
	21	import Data.SCargot (SExprParser, SExprPrinter, mkParser, flatPrint)
	22
	23	{- $info
	24
	25	This module is intended for simple, ad-hoc configuration or data formats
	26	that might not need their on rich structure but might benefit from a few
	27	various kinds of literals. The 'haskLikeParser' understands identifiers as
	28	defined by R5RS, as well as string, integer, and floating-point literals
	29	as defined by the Haskell spec. It does __not__ natively understand other
	30	data types, such as booleans, vectors, bitstrings.
	31
	32	-}
	33
	34
	35	-- \| An atom type that understands Haskell-like values as well as
	36	-- Scheme-like identifiers.
	37	data HaskLikeAtom
	38	= HSIdent Text -- ^ An identifier, parsed according to the R5RS Scheme
	39	-- standard
	40	\| HSString Text -- ^ A string, parsed according to the syntax for string
	41	-- literals in the Haskell report
	42	\| HSInt Integer -- ^ An arbitrary-sized integer value, parsed according to
	43	-- the syntax for integer literals in the Haskell report
	44	\| HSFloat Double -- ^ A double-precision floating-point value, parsed
	45	-- according to the syntax for floats in the Haskell
	46	-- report
	47	deriving (Eq, Show)
	48
	49	instance IsString HaskLikeAtom where
	50	fromString = HSIdent . fromString
	51
	52	pString :: Parser Text
	53	pString = pack . catMaybes <$> between (char '"') (char '"') (many (val <\|> esc))
	54	where val = Just <$> satisfy (\ c -> c /= '"' && c /= '\\' && c > '\026')
	55	esc = do char '\\'
	56	Nothing <$ (gap <\|> char '&') <\|>
	57	Just <$> code
	58	gap = many1 space >> char '\\'
	59	code = eEsc <\|> eNum <\|> eCtrl <\|> eAscii
	60	eCtrl = char '^' >> unCtrl <$> upper
	61	eNum = (toEnum . fromInteger) <$>
	62	(decNumber <\|> (char 'o' >> octNumber)
	63	<\|> (char 'x' >> hexNumber))
	64	eEsc = choice [ char a >> return b \| (a, b) <- escMap ]
	65	eAscii = choice [ try (string a >> return b)
	66	\| (a, b) <- asciiMap ]
	67	unCtrl c = toEnum (fromEnum c - fromEnum 'A' + 1)
	68
	69	escMap :: [(Char, Char)]
	70	escMap = zip "abfntv\\\"\'" "\a\b\f\n\r\t\v\\\"\'"
	71
	72	asciiMap :: [(String, Char)]
	73	asciiMap = zip
	74	["BS","HT","LF","VT","FF","CR","SO","SI","EM"
	75	,"FS","GS","RS","US","SP","NUL","SOH","STX","ETX"
	76	,"EOT","ENQ","ACK","BEL","DLE","DC1","DC2","DC3"
	77	,"DC4","NAK","SYN","ETB","CAN","SUB","ESC","DEL"]
	78	("\BS\HT\LF\VT\FF\CR\SO\SI\EM\FS\GS\RS\US\SP\NUL\SOH" ++
	79	"\STX\ETX\EOT\ENQ\ACK\BEL\DLE\DC1\DC2\DC3\DC4\NAK" ++
	80	"\SYN\ETB\CAN\SUB\ESC\DEL")
	81
	82	pFloat :: Parser Double
	83	pFloat = do
	84	n <- decNumber
	85	withDot n <\|> noDot n
	86	where withDot n = do
	87	char '.'
	88	m <- decNumber
	89	e <- option 1.0 exponent
	90	return ((fromIntegral n + asDec m 0) * e)
	91	noDot n = do
	92	e <- exponent
	93	return (fromIntegral n * e)
	94	exponent = do
	95	oneOf "eE"
	96	s <- power
	97	x <- decNumber
	98	return (10 ** s (fromIntegral x))
	99	asDec 0 k = k
	100	asDec n k =
	101	asDec (n `div` 10) ((fromIntegral (n `rem` 10) + k) * 0.1)
	102
	103	power :: Num a => Parser (a -> a)
	104	power = negate <$ char '-' <\|> id <$ char '+' <\|> return id
	105
	106	pInt :: Parser Integer
	107	pInt = do
	108	s <- power
	109	n <- pZeroNum <\|> decNumber
	110	return (fromIntegral (s n))
	111
	112	pZeroNum :: Parser Integer
	113	pZeroNum = char '0' >>
	114	( (oneOf "xX" >> hexNumber)
	115	<\|> (oneOf "oO" >> octNumber)
	116	<\|> decNumber
	117	<\|> return 0
	118	)
	119
	120	pHaskLikeAtom :: Parser HaskLikeAtom
	121	pHaskLikeAtom
	122	= HSFloat <$> (try pFloat <?> "float")
	123	<\|> HSInt <$> (try pInt <?> "integer")
	124	<\|> HSString <$> (pString <?> "string literal")
	125	<\|> HSIdent <$> (parseR5RSIdent <?> "token")
	126
	127	sHaskLikeAtom :: HaskLikeAtom -> Text
	128	sHaskLikeAtom (HSIdent t) = t
	129	sHaskLikeAtom (HSString s) = pack (show s)
	130	sHaskLikeAtom (HSInt i) = pack (show i)
	131	sHaskLikeAtom (HSFloat f) = pack (show f)
	132
	133	-- \| This `SExprParser` understands s-expressions that contain
	134	-- Scheme-like tokens, as well as string literals, integer
	135	-- literals, and floating-point literals. Each of these values
	136	-- is parsed according to the lexical rules in the Haskell
	137	-- report, so the same set of string escapes, numeric bases,
	138	-- and floating-point options are available. This spec does
	139	-- not parse comments and does not understand any reader
	140	-- macros.
	141	--
	142	-- >>> decode haskLikeParser "(0x01 \"\\x65lephant\")"
	143	-- Right [SCons (SAtom (HSInt 1)) (SCons (SAtom (HSString "elephant")) SNil)]
	144	haskLikeParser :: SExprParser HaskLikeAtom (SExpr HaskLikeAtom)
	145	haskLikeParser = mkParser pHaskLikeAtom
	146
	147	-- \| This 'SExprPrinter' emits s-expressions that contain Scheme-like
	148	-- tokens as well as string literals, integer literals, and floating-point
	149	-- literals, which will be emitted as the literals produced by Haskell's
	150	-- 'show' function. This printer will produce a flat s-expression with
	151	-- no indentation of any kind.
	152	--
	153	-- >>> encode haskLikePrinter [L [A (HSInt 1), A (HSString "elephant")]]
	154	-- "(1 \"elephant\")"
	155	haskLikePrinter :: SExprPrinter HaskLikeAtom (SExpr HaskLikeAtom)
	156	haskLikePrinter = flatPrint sHaskLikeAtom

+265

-0

Data/SCargot/Parse.hs less more

	1	{-# LANGUAGE ViewPatterns #-}
	2	{-# LANGUAGE OverloadedStrings #-}
	3
	4	module Data.SCargot.Parse
	5	( -- * Parsing
	6	decode
	7	, decodeOne
	8	-- * Parsing Control
	9	, SExprParser
	10	, Reader
	11	, Comment
	12	, mkParser
	13	, setCarrier
	14	, addReader
	15	, setComment
	16	-- * Specific SExprParser Conversions
	17	, asRich
	18	, asWellFormed
	19	, withQuote
	20	) where
	21
	22	import Control.Applicative ((<), (>), (<*>), (<$>), pure)
	23	import Control.Monad ((>=>))
	24	import Data.Char (isAlpha, isDigit, isAlphaNum)
	25	import Data.Map.Strict (Map)
	26	import qualified Data.Map.Strict as M
	27	import Data.Maybe (fromJust)
	28	import Data.Monoid ((<>))
	29	import Data.String (IsString)
	30	import Data.Text (Text, pack, unpack)
	31	import qualified Data.Text as T
	32	import Text.Parsec ( (<\|>)
	33	, (<?>)
	34	, char
	35	, eof
	36	, lookAhead
	37	, many1
	38	, runParser
	39	, skipMany
	40	)
	41	import Text.Parsec.Char (anyChar, space)
	42	import Text.Parsec.Text (Parser)
	43
	44	import Data.SCargot.Repr ( SExpr(..)
	45	, RichSExpr
	46	, WellFormedSExpr
	47	, fromRich
	48	, toRich
	49	, fromWellFormed
	50	, toWellFormed
	51	)
	52
	53	type ReaderMacroMap atom = Map Char (Reader atom)
	54
	55	-- \| A 'Reader' represents a reader macro: it takes a parser for
	56	-- the S-Expression type and performs as much or as little
	57	-- parsing as it would like, and then returns an S-expression.
	58	type Reader atom = (Parser (SExpr atom) -> Parser (SExpr atom))
	59
	60	-- \| A 'Comment' represents any kind of skippable comment. This
	61	-- parser __must__ be able to fail if a comment is not being
	62	-- recognized, and it __must__ not consume any input in case
	63	-- of failure.
	64	type Comment = Parser ()
	65
	66	-- \| A 'SExprParser' describes a parser for a particular value
	67	-- that has been serialized as an s-expression. The @atom@ parameter
	68	-- corresponds to a Haskell type used to represent the atoms,
	69	-- and the @carrier@ parameter corresponds to the parsed S-Expression
	70	-- structure.
	71	data SExprParser atom carrier = SExprParser
	72	{ sesPAtom :: Parser atom
	73	, readerMap :: ReaderMacroMap atom
	74	, comment :: Maybe Comment
	75	, postparse :: SExpr atom -> Either String carrier
	76	}
	77
	78	-- \| Create a basic 'SExprParser' when given a parser
	79	-- for an atom type.
	80	--
	81	-- >>> import Text.Parsec (alphaNum, many1)
	82	-- >>> let parser = mkParser (many1 alphaNum)
	83	-- >>> decode parser "(ele phant)"
	84	-- Right [SCons (SAtom "ele") (SCons (SAtom "phant") SNil)]
	85	mkParser :: Parser atom -> SExprParser atom (SExpr atom)
	86	mkParser parser = SExprParser
	87	{ sesPAtom = parser
	88	, readerMap = M.empty
	89	, comment = Nothing
	90	, postparse = return
	91	}
	92
	93	-- \| Modify the carrier type for a 'SExprParser'. This is
	94	-- used internally to convert between various 'SExpr' representations,
	95	-- but could also be used externally to add an extra conversion layer
	96	-- onto a 'SExprParser'.
	97	--
	98	-- >>> import Text.Parsec (alphaNum, many1)
	99	-- >>> import Data.SCargot.Repr (toRich)
	100	-- >>> let parser = setCarrier (return . toRich) (mkParser (many1 alphaNum))
	101	-- >>> decode parser "(ele phant)"
	102	-- Right [RSlist [RSAtom "ele",RSAtom "phant"]]
	103	setCarrier :: (b -> Either String c) -> SExprParser a b -> SExprParser a c
	104	setCarrier f spec = spec { postparse = postparse spec >=> f }
	105
	106	-- \| Convert the final output representation from the 'SExpr' type
	107	-- to the 'RichSExpr' type.
	108	--
	109	-- >>> import Text.Parsec (alphaNum, many1)
	110	-- >>> let parser = asRich (mkParser (many1 alphaNum))
	111	-- >>> decode parser "(ele phant)"
	112	-- Right [RSlist [RSAtom "ele",RSAtom "phant"]]
	113	asRich :: SExprParser a (SExpr b) -> SExprParser a (RichSExpr b)
	114	asRich = setCarrier (return . toRich)
	115
	116	-- \| Convert the final output representation from the 'SExpr' type
	117	-- to the 'WellFormedSExpr' type.
	118	--
	119	-- >>> import Text.Parsec (alphaNum, many1)
	120	-- >>> let parser = asWellFormed (mkParser (many1 alphaNum))
	121	-- >>> decode parser "(ele phant)"
	122	-- Right [WFSList [WFSAtom "ele",WFSAtom "phant"]]
	123	asWellFormed :: SExprParser a (SExpr b) -> SExprParser a (WellFormedSExpr b)
	124	asWellFormed = setCarrier toWellFormed
	125
	126	-- \| Add the ability to execute some particular reader macro, as
	127	-- defined by its initial character and the 'Parser' which returns
	128	-- the parsed S-Expression. The 'Reader' is passed a 'Parser' which
	129	-- can be recursively called to parse more S-Expressions, and begins
	130	-- parsing after the reader character has been removed from the
	131	-- stream.
	132	--
	133	-- >>> import Text.Parsec (alphaNum, char, many1)
	134	-- >>> let vecReader p = (char ']' > pure SNil) <\|> (SCons <$> p <> vecReader p)
	135	-- >>> let parser = addReader '[' vecReader (mkParser (many1 alphaNum))
	136	-- >>> decode parser "(an [ele phant])"
	137	-- Right [SCons (SAtom "an") (SCons (SCons (SAtom "ele") (SCons (SAtom "phant") SNil)) SNil)]
	138
	139	addReader :: Char -> Reader a -> SExprParser a c -> SExprParser a c
	140	addReader c reader spec = spec
	141	{ readerMap = M.insert c reader (readerMap spec) }
	142
	143	-- \| Add the ability to ignore some kind of comment. This gets
	144	-- factored into whitespace parsing, and it's very important that
	145	-- the parser supplied __be able to fail__ (as otherwise it will
	146	-- cause an infinite loop), and also that it __not consume any input__
	147	-- (which may require it to be wrapped in 'try'.)
	148	--
	149	-- >>> import Text.Parsec (alphaNum, anyChar, manyTill, many1, string)
	150	-- >>> let comment = string "//" > manyTill anyChar newline > pure ()
	151	-- >>> let parser = setComment comment (mkParser (many1 alphaNum))
	152	-- >>> decode parser "(ele //a comment\n phant)"
	153	-- Right [SCons (SAtom "ele") (SCons (SAtom "phant") SNil)]
	154
	155	setComment :: Comment -> SExprParser a c -> SExprParser a c
	156	setComment c spec = spec { comment = Just (c <?> "comment") }
	157
	158	-- \| Add the ability to understand a quoted S-Expression.
	159	-- Many Lisps use @'sexpr@ as sugar for @(quote sexpr)@. This
	160	-- assumes that the underlying atom type implements the "IsString"
	161	-- class, and will create the @quote@ atom using @fromString "quote"@.
	162	--
	163	-- >>> import Text.Parsec (alphaNum, many1)
	164	-- >>> let parser = withQuote (mkParser (many1 alphaNum))
	165	-- >>> decode parser "'elephant"
	166	-- Right [SCons (SAtom "quote") (SCons (SAtom "foo") SNil)]
	167	withQuote :: IsString t => SExprParser t (SExpr t) -> SExprParser t (SExpr t)
	168	withQuote = addReader '\'' (fmap go)
	169	where go s = SCons "quote" (SCons s SNil)
	170
	171	peekChar :: Parser (Maybe Char)
	172	peekChar = Just <$> lookAhead anyChar <\|> pure Nothing
	173
	174	parseGenericSExpr ::
	175	Parser atom -> ReaderMacroMap atom -> Parser () -> Parser (SExpr atom)
	176	parseGenericSExpr atom reader skip = do
	177	let sExpr = parseGenericSExpr atom reader skip <?> "s-expr"
	178	skip
	179	c <- peekChar
	180	r <- case c of
	181	Nothing -> fail "Unexpected end of input"
	182	Just '(' -> char '(' >> skip >> parseList sExpr skip
	183	Just (flip M.lookup reader -> Just r) -> anyChar >> r sExpr
	184	_ -> SAtom `fmap` atom
	185	skip
	186	return r
	187
	188	parseList :: Parser (SExpr atom) -> Parser () -> Parser (SExpr atom)
	189	parseList sExpr skip = do
	190	i <- peekChar
	191	case i of
	192	Nothing -> fail "Unexpected end of input"
	193	Just ')' -> char ')' >> return SNil
	194	_ -> do
	195	car <- sExpr
	196	skip
	197	c <- peekChar
	198	case c of
	199	Just '.' -> do
	200	char '.'
	201	cdr <- sExpr
	202	skip
	203	char ')'
	204	skip
	205	return (SCons car cdr)
	206	Just ')' -> do
	207	char ')'
	208	skip
	209	return (SCons car SNil)
	210	_ -> do
	211	cdr <- parseList sExpr skip
	212	return (SCons car cdr)
	213
	214	-- \| Given a CommentMap, create the corresponding parser to
	215	-- skip those comments (if they exist).
	216	buildSkip :: Maybe (Parser ()) -> Parser ()
	217	buildSkip Nothing = skipMany space
	218	buildSkip (Just c) = alternate
	219	where alternate = skipMany space >> ((c >> alternate) <\|> return ())
	220
	221	doParse :: Parser a -> Text -> Either String a
	222	doParse p t = case runParser p () "" t of
	223	Left err -> Left (show err)
	224	Right x -> Right x
	225
	226	-- \| Decode a single S-expression. If any trailing input is left after
	227	-- the S-expression (ignoring comments or whitespace) then this
	228	-- will fail: for those cases, use 'decode', which returns a list of
	229	-- all the S-expressions found at the top level.
	230	decodeOne :: SExprParser atom carrier -> Text -> Either String carrier
	231	decodeOne spec = doParse (parser <* eof) >=> (postparse spec)
	232	where parser = parseGenericSExpr
	233	(sesPAtom spec)
	234	(readerMap spec)
	235	(buildSkip (comment spec))
	236
	237	-- \| Decode several S-expressions according to a given 'SExprParser'. This
	238	-- will return a list of every S-expression that appears at the top-level
	239	-- of the document.
	240	decode :: SExprParser atom carrier -> Text -> Either String [carrier]
	241	decode spec =
	242	doParse (many1 parser <* eof) >=> mapM (postparse spec)
	243	where parser = parseGenericSExpr
	244	(sesPAtom spec)
	245	(readerMap spec)
	246	(buildSkip (comment spec))
	247
	248	{-
	249	-- \| Encode (without newlines) a single S-expression.
	250	encodeSExpr :: SExpr atom -> (atom -> Text) -> Text
	251	encodeSExpr SNil _ = "()"
	252	encodeSExpr (SAtom s) t = t s
	253	encodeSExpr (SCons x xs) t = go xs (encodeSExpr x t)
	254	where go (SAtom s) rs = "(" <> rs <> " . " <> t s <> ")"
	255	go SNil rs = "(" <> rs <> ")"
	256	go (SCons x xs) rs = go xs (rs <> " " <> encodeSExpr x t)
	257
	258	-- \| Emit an S-Expression in a machine-readable way. This does no
	259	-- pretty-printing or indentation, and produces no comments.
	260	encodeOne :: SExprParser atom carrier -> carrier -> Text
	261	encodeOne spec c = encodeSExpr (preserial spec c) (sesSAtom spec)
	262
	263	encode :: SExprParser atom carrier -> [carrier] -> Text
	264	encode spec cs = T.concat (map (encodeOne spec) cs)
	265	-}

-144

~~Data/SCargot/Pretty.hs~~ less more

1		{-# LANGUAGE RecordWildCards #-}
2		{-# LANGUAGE OverloadedStrings #-}
3		{-# LANGUAGE ScopedTypeVariables #-}
4
5		module Data.SCargot.Pretty
6		( -- * Pretty-Printing
7		prettyPrintSExpr
8		-- * Pretty-Printing Control
9		, LayoutOptions(..)
10		, Indent(..)
11		-- * Default Printing Strategies
12		, basicPrint
13		, flatPrint
14		) where
15
16		import Data.Monoid ((<>))
17		import Data.Text (Text)
18		import qualified Data.Text as T
19
20		import Data.SCargot.Repr
21
22		-- \| The 'Indent' type is used to determine how to indent subsequent
23		-- s-expressions in a list, after printing the head of the list.
24		data Indent
25		= Swing -- ^ A 'Swing' indent will indent subsequent exprs some fixed
26		-- amount more than the current line.
27		--
28		-- > (foo
29		-- > bar
30		-- > baz
31		-- > quux)
32		\| SwingAfter Int -- ^ A 'SwingAfter' @n@ indent will try to print the
33		-- first @n@ expressions after the head on the same
34		-- line as the head, and all after will be swung.
35		-- 'SwingAfter' @0@ is equivalent to 'Swing'.
36		--
37		-- > (foo bar
38		-- > baz
39		-- > quux)
40		\| Align -- ^ An 'Align' indent will print the first expression after
41		-- the head on the same line, and subsequent expressions will
42		-- be aligned with that one.
43		--
44		-- > (foo bar
45		-- > baz
46		-- > quux)
47		deriving (Eq, Show)
48
49		-- \| A 'LayoutOptions' value describes the strategy taken in
50		-- pretty-printing a 'SExpr'.
51		data LayoutOptions a = LayoutOptions
52		{ atomPrinter :: a -> Text
53		-- ^ How to serialize a given atom to 'Text'.
54		, swingIndent :: SExpr a -> Indent
55		-- ^ How to indent subsequent expressions, as determined by
56		-- the head of the list.
57		, indentAmount :: Int
58		-- ^ How much to indent after a swung indentation.
59		, maxWidth :: Maybe Int
60		-- ^ The maximum width (if any) If this is 'None' then
61		-- the resulting s-expression will always be printed
62		-- on a single line.
63		}
64
65		-- \| A default 'LayoutOptions' struct that will always print a 'SExpr'
66		-- as a single line.
67		flatPrint :: (a -> Text) -> LayoutOptions a
68		flatPrint printer = LayoutOptions
69		{ atomPrinter = printer
70		, swingIndent = const Swing
71		, indentAmount = 2
72		, maxWidth = Nothing
73		}
74
75		-- \| A default 'LayoutOptions' struct that will always swing subsequent
76		-- expressions onto later lines if they're too long, indenting them
77		-- by two spaces.
78		basicPrint :: (a -> Text) -> LayoutOptions a
79		basicPrint printer = LayoutOptions
80		{ atomPrinter = printer
81		, swingIndent = const Swing
82		, indentAmount = 2
83		, maxWidth = Just 80
84		}
85
86		-- Sort of like 'unlines' but without the trailing newline
87		joinLines :: [Text] -> Text
88		joinLines = T.intercalate "\n"
89
90		-- Indents a line by n spaces
91		indent :: Int -> Text -> Text
92		indent n ts = T.replicate n " " <> ts
93
94		-- Indents every line n spaces, and adds a newline to the beginning
95		-- used in swung indents
96		indentAll :: Int -> [Text] -> Text
97		indentAll n = ("\n" <>) . joinLines . map (indent n)
98
99		-- Indents every line but the first by some amount
100		-- used in aligned indents
101		indentSubsequent :: Int -> [Text] -> Text
102		indentSubsequent _ [] = ""
103		indentSubsequent _ [t] = t
104		indentSubsequent n (t:ts) = joinLines (t : go ts)
105		where go = map (indent n)
106
107		-- oh god this code is so disgusting
108		-- i'm sorry to everyone i let down by writing this
109		-- i swear i'll do better in the future i promise i have to
110		-- for my sake and for everyone's
111
112		-- \| Pretty-print a 'SExpr' according to the options in a
113		-- 'LayoutOptions' value.
114		prettyPrintSExpr :: LayoutOptions a -> SExpr a -> Text
115		prettyPrintSExpr LayoutOptions { .. } = pHead 0
116		where pHead _ SNil = "()"
117		pHead _ (SAtom a) = atomPrinter a
118		pHead ind (SCons x xs) = gather ind x xs id
119		gather _ _ (SAtom _) _ = error "no dotted pretty printing yet!"
120		gather ind h (SCons x xs) k = gather ind h xs (k . (x:))
121		gather ind h SNil k = "(" <> hd <> body <> ")"
122		where hd = indentSubsequent ind [pHead (ind+1) h]
123		lst = k []
124		flat = T.unwords (map (pHead (ind+1)) lst)
125		headWidth = T.length hd + 1
126		indented =
127		case swingIndent h of
128		SwingAfter n ->
129		let (l, ls) = splitAt n lst
130		t = T.unwords (map (pHead (ind+1)) l)
131		ts = indentAll (ind + indentAmount)
132		(map (pHead (ind + indentAmount)) ls)
133		in t <> ts
134		Swing ->
135		indentAll (ind + indentAmount)
136		(map (pHead (ind + indentAmount)) lst)
137		Align ->
138		indentSubsequent (ind + headWidth + 1)
139		(map (pHead (ind + headWidth + 1)) lst)
140		body
141		\| length lst == 0 = ""
142		\| Just maxAmt <- maxWidth
143		, T.length flat + ind > maxAmt = " " <> indented
144		\| otherwise = " " <> flat

+219

-0

Data/SCargot/Print.hs less more

	1	{-# LANGUAGE RecordWildCards #-}
	2	{-# LANGUAGE OverloadedStrings #-}
	3	{-# LANGUAGE ScopedTypeVariables #-}
	4
	5	module Data.SCargot.Print
	6	( -- * Pretty-Printing
	7	encodeOne
	8	, encode
	9	-- * Pretty-Printing Control
	10	, SExprPrinter
	11	, Indent(..)
	12	, setFromCarrier
	13	, setMaxWidth
	14	, removeMaxWidth
	15	, setIndentAmount
	16	, setIndentStrategy
	17	-- * Default Printing Strategies
	18	, basicPrint
	19	, flatPrint
	20	) where
	21
	22	import Data.Monoid ((<>))
	23	import Data.Text (Text)
	24	import qualified Data.Text as T
	25
	26	import Data.SCargot.Repr
	27
	28	-- \| The 'Indent' type is used to determine how to indent subsequent
	29	-- s-expressions in a list, after printing the head of the list.
	30	data Indent
	31	= Swing -- ^ A 'Swing' indent will indent subsequent exprs some fixed
	32	-- amount more than the current line.
	33	--
	34	-- > (foo
	35	-- > bar
	36	-- > baz
	37	-- > quux)
	38	\| SwingAfter Int -- ^ A 'SwingAfter' @n@ indent will try to print the
	39	-- first @n@ expressions after the head on the same
	40	-- line as the head, and all after will be swung.
	41	-- 'SwingAfter' @0@ is equivalent to 'Swing'.
	42	--
	43	-- > (foo bar
	44	-- > baz
	45	-- > quux)
	46	\| Align -- ^ An 'Align' indent will print the first expression after
	47	-- the head on the same line, and subsequent expressions will
	48	-- be aligned with that one.
	49	--
	50	-- > (foo bar
	51	-- > baz
	52	-- > quux)
	53	deriving (Eq, Show)
	54
	55	-- \| A 'SExprPrinter' value describes how to print a given value as an
	56	-- s-expression. The @carrier@ type parameter indicates the value
	57	-- that will be printed, and the @atom@ parameter indicates the type
	58	-- that will represent tokens in an s-expression structure.
	59	data SExprPrinter atom carrier = SExprPrinter
	60	{ atomPrinter :: atom -> Text
	61	-- ^ How to serialize a given atom to 'Text'.
	62	, fromCarrier :: carrier -> SExpr atom
	63	-- ^ How to turn a carrier type back into a 'Sexpr'.
	64	, swingIndent :: SExpr atom -> Indent
	65	-- ^ How to indent subsequent expressions, as determined by
	66	-- the head of the list.
	67	, indentAmount :: Int
	68	-- ^ How much to indent after a swung indentation.
	69	, maxWidth :: Maybe Int
	70	-- ^ The maximum width (if any) If this is 'None' then
	71	-- the resulting s-expression will always be printed
	72	-- on a single line.
	73	}
	74
	75	-- \| A default 'LayoutOptions' struct that will always print a 'SExpr'
	76	-- as a single line.
	77	flatPrint :: (atom -> Text) -> SExprPrinter atom (SExpr atom)
	78	flatPrint printer = SExprPrinter
	79	{ atomPrinter = printer
	80	, fromCarrier = id
	81	, swingIndent = const Swing
	82	, indentAmount = 2
	83	, maxWidth = Nothing
	84	}
	85
	86	-- \| A default 'LayoutOptions' struct that will always swing subsequent
	87	-- expressions onto later lines if they're too long, indenting them
	88	-- by two spaces.
	89	basicPrint :: (atom -> Text) -> SExprPrinter atom (SExpr atom)
	90	basicPrint printer = SExprPrinter
	91	{ atomPrinter = printer
	92	, fromCarrier = id
	93	, swingIndent = const Swing
	94	, indentAmount = 2
	95	, maxWidth = Just 80
	96	}
	97
	98	-- \| Modify the carrier type of a 'SExprPrinter' by describing how
	99	-- to convert the new type back to the previous type. For example,
	100	-- to pretty-print a well-formed s-expression, we can modify the
	101	-- 'SExprPrinter' value as follows:
	102	--
	103	-- >>> let printer = setFromCarrier fromWellFormed (basicPrint id)
	104	-- >>> encodeOne printer (WFSList [WFSAtom "ele", WFSAtom "phant"])
	105	-- "(ele phant)"
	106	setFromCarrier :: (c -> b) -> SExprPrinter a b -> SExprPrinter a c
	107	setFromCarrier fc pr = pr { fromCarrier = fromCarrier pr . fc }
	108
	109	-- \| Dictate a maximum width for pretty-printed s-expressions.
	110	--
	111	-- >>> let printer = setMaxWidth 8 (basicPrint id)
	112	-- >>> encodeOne printer (L [A "one", A "two", A "three"])
	113	-- "(one \n two\n three)"
	114	setMaxWidth :: Int -> SExprPrinter atom carrier -> SExprPrinter atom carrier
	115	setMaxWidth n pr = pr { maxWidth = Just n }
	116
	117	-- \| Allow the serialized s-expression to be arbitrarily wide. This
	118	-- makes all pretty-printing happen on a single line.
	119	--
	120	-- >>> let printer = removeMaxWidth (basicPrint id)
	121	-- >>> encodeOne printer (L [A "one", A "two", A "three"])
	122	-- "(one two three)"
	123	removeMaxWidth :: SExprPrinter atom carrier -> SExprPrinter atom carrier
	124	removeMaxWidth pr = pr { maxWidth = Nothing }
	125
	126	-- \| Set the number of spaces that a subsequent line will be indented
	127	-- after a swing indentation.
	128	--
	129	-- >>> let printer = setMaxWidth 12 (basicPrint id)
	130	-- >>> encodeOne printer (L [A "elephant", A "pachyderm"])
	131	-- "(elephant \n pachyderm)"
	132	-- >>> encodeOne (setIndentAmount 4) (L [A "elephant", A "pachyderm"])
	133	-- "(elephant \n pachyderm)"
	134	setIndentAmount :: Int -> SExprPrinter atom carrier -> SExprPrinter atom carrier
	135	setIndentAmount n pr = pr { indentAmount = n }
	136
	137	-- \| Dictate how to indent subsequent lines based on the leading
	138	-- subexpression in an s-expression. For details on how this works,
	139	-- consult the documentation of the 'Indent' type.
	140	--
	141	-- >>> let indent (A "def") = SwingAfter 1; indent _ = Swing
	142	-- >>> let printer = setIndentStrategy indent (setMaxWidth 8 (basicPrint id))
	143	-- >>> encodeOne printer (L [ A "def", L [ A "func", A "arg" ], A "body" ])
	144	-- "(def (func arg)\n body)"
	145	-- >>> encodeOne printer (L [ A "elephant", A "among", A "pachyderms" ])
	146	-- "(elephant \n among\n pachyderms)"
	147	setIndentStrategy :: (SExpr atom -> Indent) -> SExprPrinter atom carrier -> SExprPrinter atom carrier
	148	setIndentStrategy st pr = pr { swingIndent = st }
	149
	150	-- Sort of like 'unlines' but without the trailing newline
	151	joinLines :: [Text] -> Text
	152	joinLines = T.intercalate "\n"
	153
	154	-- Indents a line by n spaces
	155	indent :: Int -> Text -> Text
	156	indent n ts = T.replicate n " " <> ts
	157
	158	-- Indents every line n spaces, and adds a newline to the beginning
	159	-- used in swung indents
	160	indentAll :: Int -> [Text] -> Text
	161	indentAll n = ("\n" <>) . joinLines . map (indent n)
	162
	163	-- Indents every line but the first by some amount
	164	-- used in aligned indents
	165	indentSubsequent :: Int -> [Text] -> Text
	166	indentSubsequent _ [] = ""
	167	indentSubsequent _ [t] = t
	168	indentSubsequent n (t:ts) = joinLines (t : go ts)
	169	where go = map (indent n)
	170
	171	-- oh god this code is so disgusting
	172	-- i'm sorry to everyone i let down by writing this
	173	-- i swear i'll do better in the future i promise i have to
	174	-- for my sake and for everyone's
	175
	176	-- \| Pretty-print a 'SExpr' according to the options in a
	177	-- 'LayoutOptions' value.
	178	prettyPrintSExpr :: SExprPrinter a (SExpr a) -> SExpr a -> Text
	179	prettyPrintSExpr SExprPrinter { .. } = pHead 0
	180	where pHead _ SNil = "()"
	181	pHead _ (SAtom a) = atomPrinter a
	182	pHead ind (SCons x xs) = gather ind x xs id
	183	gather _ _ (SAtom _) _ = error "no dotted pretty printing yet!"
	184	gather ind h (SCons x xs) k = gather ind h xs (k . (x:))
	185	gather ind h SNil k = "(" <> hd <> body <> ")"
	186	where hd = indentSubsequent ind [pHead (ind+1) h]
	187	lst = k []
	188	flat = T.unwords (map (pHead (ind+1)) lst)
	189	headWidth = T.length hd + 1
	190	indented =
	191	case swingIndent h of
	192	SwingAfter n ->
	193	let (l, ls) = splitAt n lst
	194	t = T.unwords (map (pHead (ind+1)) l)
	195	ts = indentAll (ind + indentAmount)
	196	(map (pHead (ind + indentAmount)) ls)
	197	in t <> ts
	198	Swing ->
	199	indentAll (ind + indentAmount)
	200	(map (pHead (ind + indentAmount)) lst)
	201	Align ->
	202	indentSubsequent (ind + headWidth + 1)
	203	(map (pHead (ind + headWidth + 1)) lst)
	204	body
	205	\| length lst == 0 = ""
	206	\| Just maxAmt <- maxWidth
	207	, T.length flat + ind > maxAmt = " " <> indented
	208	\| otherwise = " " <> flat
	209
	210	-- \| Turn a single s-expression into a string according to a given
	211	-- 'SExprPrinter'.
	212	encodeOne :: SExprPrinter atom carrier -> carrier -> Text
	213	encodeOne s@(SExprPrinter { .. }) =
	214	prettyPrintSExpr (s { fromCarrier = id }) . fromCarrier
	215
	216	-- \| Turn a list of s-expressions into a single string according to
	217	-- a given 'SExprPrinter'.
	218	encode :: SExprPrinter atom carrier -> [carrier] -> Text
	219	encode spec = T.intercalate "\n\n" . map (encodeOne spec)

+42

-3

Data/SCargot/Repr.hs less more

5	5	{-# LANGUAGE TypeFamilies #-}
6	6
7	7	module Data.SCargot.Repr
8		( -- * Elementary SExpr representation
	8	( -- $reprs
	9	-- * Elementary SExpr representation
9	10	SExpr(..)
10	11	-- * Rich SExpr representation
11	12	, RichSExpr(..)

51	52	-- represent a well-formed cons list, and 'RSDotted'
52	53	-- to represent an improper list of the form
53	54	-- @(a b c . d)@. This representation is based on
54		-- the shape of the parsed S-Expression, and not on
55		-- how it was represented, so @(a . (b))@ is going to
	55	-- the structure of the parsed S-Expression, and not on
	56	-- how it was originally represented: thus, @(a . (b))@ is going to
56	57	-- be represented as @RSList[RSAtom a, RSAtom b]@
57	58	-- despite having been originally represented as a
58	59	-- dotted list.

138	139	fromWellFormed (WFSAtom a) = SAtom a
139	140	fromWellFormed (WFSList xs) =
140	141	foldr SCons SNil (map fromWellFormed xs)
	142
	143	{- $reprs
	144
	145	This module contains several different representations for
	146	s-expressions. The s-cargot library underlying uses the
	147	'SExpr' type as its representation type, which is a binary
	148	tree representation with an arbitrary type for its leaves.
	149
	150	This type is not always convenient to manipulate in Haskell
	151	code, this module defines two alternate representations
	152	which turn a sequence of nested right-branching cons pairs
	153	into Haskell lists: that is to say, they transform between
	154
	155	@
	156	SCons a (SCons b (SCons c SNil)) \<=\> RSList [a, b, c]
	157	@
	158
	159	These two types differ in how they handle non-well-formed
	160	lists, i.e. lists that end with an atom. The 'RichSExpr'
	161	format handles this with a special constructor for lists
	162	that end in an atom:
	163
	164	@
	165	SCons a (SCons b (SAtom c)) \<=\> RSDotted [a, b] c
	166	@
	167
	168	On the other hand, the 'WellFormedSExpr' type elects
	169	not to handle this case. This is unusual for Lisp source code,
	170	but is a reasonable choice for configuration or data
	171	storage formats that use s-expressions, where
	172	non-well-formed lists would be an unnecessary
	173	complication.
	174
	175	To make working with these types less verbose, there are other
	176	modules that export pattern aliases and helper functions: these
	177	can be found at "Data.SCargot.Repr.Basic",
	178	"Data.SCargot.Repr.Rich", and "Data.SCargot.Repr.WellFormed".
	179	-}

+92

-3

Data/SCargot.hs less more

1	1	module Data.SCargot
2		( module Data.SCargot.General
3		) where
	2	( -- * SCargot Basics
4	3
5		import Data.SCargot.General
	4	-- $intro
	5
	6	-- * Parsing and Printing
	7	decode
	8	, decodeOne
	9	, encode
	10	, encodeOne
	11	-- * Parser Construction
	12	-- ** Specifying a Parser
	13	, SExprParser
	14	, Reader
	15	, Comment
	16	, mkParser
	17	, setCarrier
	18	, addReader
	19	, setComment
	20	, asRich
	21	, asWellFormed
	22	, withQuote
	23	-- * Printer Construction
	24	-- * Specifying a Pretty-Printer
	25	, SExprPrinter
	26	, Indent(..)
	27	, basicPrint
	28	, flatPrint
	29	, setFromCarrier
	30	, setMaxWidth
	31	, removeMaxWidth
	32	, setIndentAmount
	33	, setIndentStrategy
	34	) where
	35
	36	import Data.SCargot.Parse
	37	import Data.SCargot.Print
	38
	39	{- $intro
	40
	41	The S-Cargot library is a library for parsing and emitting
	42	<https://en.wikipedia.org/wiki/S-expression s-expressions>, designed
	43	to be as flexible as possible. Despite some efforts at
	44	<http://people.csail.mit.edu/rivest/Sexp.txt standardization>,
	45	s-expressions are a general approach to describing a data format
	46	that can very often differ in subtle, incompatible ways: the
	47	s-expressions understood by Common Lisp are different from the
	48	s-expressions understood by Scheme, and even the different
	49	revisions of the Scheme language understand s-expressions in a
	50	slightly different way. To accomodate this, the S-Cargot library
	51	provides a toolbox for defining variations on s-expressions,
	52	complete with the ability to select various comment syntaxes, reader
	53	macros, and atom types.
	54
	55	If all you want is to read some s-expressions and don't care about
	56	the edge cases of the format, or all you want is a new configuration
	57	format, try the "Data.SCargot.Language.Basic" or "Data.SCargot.Language.HaskLike"
	58	modules, which define an s-expression language whose atoms are
	59	plain strings and Haskell literals, respectively.
	60
	61	The S-Cargot library works by specifying values which contain all
	62	the information needed to either parse or print an s-expression.
	63	The actual s-expression structure is parsed as a structure of
	64	<https://en.wikipedia.org/wiki/Cons cons cells> as represented
	65	by the 'SExpr' type, but can alternately be exposed as the
	66	isomorphic 'RichSExpr' type or the less expressive but
	67	easier-to-work-with 'WellFormedSExpr' type. Modules devoted
	68	to each representation type (in "Data.SCargot.Repr.Basic",
	69	"Data.SCargot.Repr.Rich", and "Data.SCargot.Repr.WellFormed")
	70	provide helper functions, lenses, and pattern synonyms to make
	71	creating and processing these values easier.
	72
	73	The details of how to parse a given structure are represented
	74	by building up a 'SExprParser' value, which is defined in
	75	"Data.SCargot.Parse" and re-exported here. A minimal
	76	'SExprParser' defines only how to parse the atoms of the
	77	language; helper functions can define comment syntaxes,
	78	reader macros, and transformations over the parsed structure.
	79
	80	The details of how to print a given structure are represented
	81	by building up a 'SExprPrinter' value, which is defined in
	82	"Data.SCargot.Print" and re-exported here. A minimal
	83	'SExprPrinter' defines only how to print the atoms of the
	84	language; helper functions help with the layout of the
	85	pretty-printed s-expression in terms of how to indent the
	86	surrounding expression.
	87
	88	Other helper modules define useful primitives for building up
	89	s-expression languages: the "Data.SCargot.Common" module provides
	90	parsers for common literals, while the "Data.SCargot.Comments"
	91	module provides parsers for comment syntaxes borrowed from
	92	various other languages.
	93
	94	-}

-4

s-cargot.cabal less more

1	1	name: s-cargot
2	2	version: 0.1.0.0
3	3	synopsis: A flexible, extensible s-expression library.
	4	homepage: https://github.com/aisamanra/s-cargot
4	5	description: S-Cargot is a library for working with s-expressions in
5	6	a modular and extensible way, opting for genericity and
6	7	flexibility instead of speed. Instead of understanding

18	19	build-type: Simple
19	20	cabal-version: >=1.10
20	21
	22	source-repository head
	23	type: git
	24	location: git://github.com/aisamanra/s-cargot.git
	25
21	26	library
22	27	exposed-modules: Data.SCargot,
23	28	Data.SCargot.Repr,
24	29	Data.SCargot.Repr.Basic,
25	30	Data.SCargot.Repr.Rich,
26	31	Data.SCargot.Repr.WellFormed,
27		Data.SCargot.General,
28		Data.SCargot.Pretty,
29		Data.SCargot.~~Basic~~,
	32	Data.SCargot.Parse,
	33	Data.SCargot.Print,
30	34	Data.SCargot.Comments,
31	35	Data.SCargot.Common,
32		Data.SCargot.~~HaskLike~~
	36	Data.SCargot.Language.Basic,
	37	Data.SCargot.Language.HaskLike
33	38	build-depends: base >=4.7 && <5,
34	39	parsec,
35	40	text,