summaryrefslogtreecommitdiff
path: root/src/Text/Pandoc/Readers/Textile.hs
blob: 6e5ab2791c23cb8d5000a769e7997ace44c7a19c (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
{-
Copyright (C) 2010 Paul Rivier <paul*rivier#demotera*com> | tr '*#' '.@'

This program is free software; you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by
the Free Software Foundation; either version 2 of the License, or
(at your option) any later version.

This program is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
GNU General Public License for more details.

You should have received a copy of the GNU General Public License
along with this program; if not, write to the Free Software
Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
-}

{- |
   Module      : Text.Pandoc.Readers.Textile
   Copyright   : Copyright (C) 2010-2012 Paul Rivier and John MacFarlane
   License     : GNU GPL, version 2 or above 

   Maintainer  : Paul Rivier <paul*rivier#demotera*com>
   Stability   : alpha
   Portability : portable

Conversion from Textile to 'Pandoc' document, based on the spec
available at http://redcloth.org/textile.

Implemented and parsed:
 - Paragraphs
 - Code blocks
 - Lists
 - blockquote
 - Inlines : strong, emph, cite, code, deleted, superscript,
   subscript, links
 - footnotes

Implemented but discarded:
 - HTML-specific and CSS-specific attributes

Left to be implemented:
 - dimension sign
 - all caps
 - continued blocks (ex bq..)

TODO : refactor common patterns across readers :
 - autolink
 - smartPunctuation
 - more ...

-}


module Text.Pandoc.Readers.Textile ( readTextile) where

import Text.Pandoc.Definition
import Text.Pandoc.Shared
import Text.Pandoc.Options
import Text.Pandoc.Parsing
import Text.Pandoc.Readers.HTML ( htmlTag, isInlineTag, isBlockTag )
import Text.Pandoc.Readers.LaTeX ( rawLaTeXInline, rawLaTeXBlock )
import Text.HTML.TagSoup.Match
import Data.Char ( digitToInt, isUpper )
import Control.Monad ( guard, liftM )
import Control.Applicative ((<$>), (*>), (<*))

-- | Parse a Textile text and return a Pandoc document.
readTextile :: ReaderOptions -- ^ Reader options
            -> String       -- ^ String to parse (assuming @'\n'@ line endings)
            -> Pandoc
readTextile opts s =
  (readWith parseTextile) def{ stateOptions = opts } (s ++ "\n\n")


-- | Generate a Pandoc ADT from a textile document
parseTextile :: Parser [Char] ParserState Pandoc
parseTextile = do
  -- textile allows raw HTML and does smart punctuation by default
  oldOpts <- stateOptions `fmap` getState
  updateState $ \state -> state{ stateOptions =
                                   oldOpts{ readerSmart = True
                                          , readerParseRaw = True
                                          , readerOldDashes = True
                                          } }
  many blankline
  startPos <- getPosition
  -- go through once just to get list of reference keys and notes
  -- docMinusKeys is the raw document with blanks where the keys/notes were...
  let firstPassParser = noteBlock <|> lineClump
  manyTill firstPassParser eof >>= setInput . concat
  setPosition startPos
  st' <- getState
  let reversedNotes = stateNotes st'
  updateState $ \s -> s { stateNotes = reverse reversedNotes }
  -- now parse it for real...
  blocks <- parseBlocks
  return $ Pandoc (Meta [] [] []) blocks -- FIXME

noteMarker :: Parser [Char] ParserState [Char]
noteMarker = skipMany spaceChar >> string "fn" >> manyTill digit (char '.')

noteBlock :: Parser [Char] ParserState [Char]
noteBlock = try $ do
  startPos <- getPosition
  ref <- noteMarker
  optional blankline
  contents <- liftM unlines $ many1Till anyLine (blanklines <|> noteBlock)
  endPos <- getPosition
  let newnote = (ref, contents ++ "\n")
  st <- getState
  let oldnotes = stateNotes st
  updateState $ \s -> s { stateNotes = newnote : oldnotes }
  -- return blanks so line count isn't affected
  return $ replicate (sourceLine endPos - sourceLine startPos) '\n'

-- | Parse document blocks
parseBlocks :: Parser [Char] ParserState [Block]
parseBlocks = manyTill block eof

-- | Block parsers list tried in definition order
blockParsers :: [Parser [Char] ParserState Block]
blockParsers = [ codeBlock
               , header
               , blockQuote
               , hrule
               , anyList
               , rawHtmlBlock
               , rawLaTeXBlock'
               , maybeExplicitBlock "table" table
               , maybeExplicitBlock "p" para
               , nullBlock ]

-- | Any block in the order of definition of blockParsers
block :: Parser [Char] ParserState Block
block = choice blockParsers <?> "block"

codeBlock :: Parser [Char] ParserState Block
codeBlock = codeBlockBc <|> codeBlockPre

codeBlockBc :: Parser [Char] ParserState Block
codeBlockBc = try $ do
  string "bc. "
  contents <- manyTill anyLine blanklines
  return $ CodeBlock ("",[],[]) $ unlines contents

-- | Code Blocks in Textile are between <pre> and </pre>
codeBlockPre :: Parser [Char] ParserState Block
codeBlockPre = try $ do
  htmlTag (tagOpen (=="pre") null)
  result' <- manyTill anyChar (try $ htmlTag (tagClose (=="pre")) >> blockBreak)
  -- drop leading newline if any
  let result'' = case result' of
                      '\n':xs -> xs
                      _       -> result'
  -- drop trailing newline if any
  let result''' = case reverse result'' of
                       '\n':_ -> init result''
                       _      -> result''
  return $ CodeBlock ("",[],[]) result'''

-- | Header of the form "hN. content" with N in 1..6
header :: Parser [Char] ParserState Block
header = try $ do
  char 'h'
  level <- digitToInt <$> oneOf "123456"
  optional attributes >> char '.' >> whitespace
  name <- normalizeSpaces <$> manyTill inline blockBreak
  return $ Header level name

-- | Blockquote of the form "bq. content"
blockQuote :: Parser [Char] ParserState Block
blockQuote = try $ do
  string "bq" >> optional attributes >> char '.' >> whitespace
  BlockQuote . singleton <$> para

-- Horizontal rule

hrule :: Parser [Char] st Block
hrule = try $ do
  skipSpaces
  start <- oneOf "-*"
  count 2 (skipSpaces >> char start)
  skipMany (spaceChar <|> char start)
  newline
  optional blanklines
  return HorizontalRule

-- Lists handling

-- | Can be a bullet list or an ordered list. This implementation is
-- strict in the nesting, sublist must start at exactly "parent depth
-- plus one"
anyList :: Parser [Char] ParserState Block
anyList = try $ ( (anyListAtDepth 1) <* blanklines )

-- | This allow one type of list to be nested into an other type,
-- provided correct nesting
anyListAtDepth :: Int -> Parser [Char] ParserState Block
anyListAtDepth depth = choice [ bulletListAtDepth depth,
                                orderedListAtDepth depth,
                                definitionList ]

-- | Bullet List of given depth, depth being the number of leading '*'
bulletListAtDepth :: Int -> Parser [Char] ParserState Block
bulletListAtDepth depth = try $ BulletList <$> many1 (bulletListItemAtDepth depth)

-- | Bullet List Item of given depth, depth being the number of
-- leading '*'
bulletListItemAtDepth :: Int -> Parser [Char] ParserState [Block]
bulletListItemAtDepth = genericListItemAtDepth '*'

-- | Ordered List of given depth, depth being the number of
-- leading '#'
orderedListAtDepth :: Int -> Parser [Char] ParserState Block
orderedListAtDepth depth = try $ do
  items <- many1 (orderedListItemAtDepth depth)
  return (OrderedList (1, DefaultStyle, DefaultDelim) items)

-- | Ordered List Item of given depth, depth being the number of
-- leading '#'
orderedListItemAtDepth :: Int -> Parser [Char] ParserState [Block]
orderedListItemAtDepth = genericListItemAtDepth '#'

-- | Common implementation of list items
genericListItemAtDepth :: Char -> Int -> Parser [Char] ParserState [Block]
genericListItemAtDepth c depth = try $ do
  count depth (char c) >> optional attributes >> whitespace
  p <- inlines
  sublist <- option [] (singleton <$> anyListAtDepth (depth + 1))
  return ((Plain p):sublist)

-- | A definition list is a set of consecutive definition items
definitionList :: Parser [Char] ParserState Block  
definitionList = try $ DefinitionList <$> many1 definitionListItem
  
-- | A definition list item in textile begins with '- ', followed by
-- the term defined, then spaces and ":=". The definition follows, on
-- the same single line, or spaned on multiple line, after a line
-- break.
definitionListItem :: Parser [Char] ParserState ([Inline], [[Block]])
definitionListItem = try $ do
  string "- "
  term <- many1Till inline (try (whitespace >> string ":="))
  def' <- inlineDef <|> multilineDef
  return (term, def')
  where inlineDef :: Parser [Char] ParserState [[Block]]
        inlineDef = liftM (\d -> [[Plain d]]) $ try (whitespace >> inlines)
        multilineDef :: Parser [Char] ParserState [[Block]]
        multilineDef = try $ do
          optional whitespace >> newline
          s <- many1Till anyChar (try (string "=:" >> newline))
          -- this ++ "\n\n" does not look very good
          ds <- parseFromString parseBlocks (s ++ "\n\n")
          return [ds]

-- | This terminates a block such as a paragraph. Because of raw html
-- blocks support, we have to lookAhead for a rawHtmlBlock.
blockBreak :: Parser [Char] ParserState ()
blockBreak = try (newline >> blanklines >> return ()) <|>
              (lookAhead rawHtmlBlock >> return ())

-- raw content

-- | A raw Html Block, optionally followed by blanklines
rawHtmlBlock :: Parser [Char] ParserState Block
rawHtmlBlock = try $ do
  (_,b) <- htmlTag isBlockTag
  optional blanklines
  return $ RawBlock "html" b

-- | Raw block of LaTeX content
rawLaTeXBlock' :: Parser [Char] ParserState Block
rawLaTeXBlock' = do
  guardEnabled Ext_raw_tex
  RawBlock "latex" <$> (rawLaTeXBlock <* spaces)


-- | In textile, paragraphs are separated by blank lines.
para :: Parser [Char] ParserState Block
para = try $ Para . normalizeSpaces <$> manyTill inline blockBreak


-- Tables
  
-- | A table cell spans until a pipe |
tableCell :: Parser [Char] ParserState TableCell
tableCell = do
  c <- many1 (noneOf "|\n")
  content <- parseFromString (many1 inline) c
  return $ [ Plain $ normalizeSpaces content ]

-- | A table row is made of many table cells
tableRow :: Parser [Char] ParserState [TableCell]
tableRow = try $ ( char '|' *> (endBy1 tableCell (char '|')) <* newline)

-- | Many table rows
tableRows :: Parser [Char] ParserState [[TableCell]]
tableRows = many1 tableRow

-- | Table headers are made of cells separated by a tag "|_."
tableHeaders :: Parser [Char] ParserState [TableCell]
tableHeaders = let separator = (try $ string "|_.") in
  try $ ( separator *> (sepBy1 tableCell separator) <* char '|' <* newline )
  
-- | A table with an optional header. Current implementation can
-- handle tables with and without header, but will parse cells
-- alignment attributes as content.
table :: Parser [Char] ParserState Block
table = try $ do
  headers <- option [] tableHeaders
  rows <- tableRows
  blanklines
  let nbOfCols = max (length headers) (length $ head rows)
  return $ Table [] 
    (replicate nbOfCols AlignDefault)
    (replicate nbOfCols 0.0)
    headers
    rows
  

-- | Blocks like 'p' and 'table' do not need explicit block tag.
-- However, they can be used to set HTML/CSS attributes when needed.
maybeExplicitBlock :: String  -- ^ block tag name
                    -> Parser [Char] ParserState Block -- ^ implicit block
                    -> Parser [Char] ParserState Block
maybeExplicitBlock name blk = try $ do
  optional $ try $ string name >> optional attributes >> char '.' >> 
    ((try whitespace) <|> endline)
  blk



----------
-- Inlines
----------


-- | Any inline element
inline :: Parser [Char] ParserState Inline
inline = choice inlineParsers <?> "inline"

-- | List of consecutive inlines before a newline
inlines :: Parser [Char] ParserState [Inline]
inlines = manyTill inline newline

-- | Inline parsers tried in order
inlineParsers :: [Parser [Char] ParserState Inline]
inlineParsers = [ autoLink
                , str
                , whitespace
                , endline
                , code
                , escapedInline
                , htmlSpan
                , rawHtmlInline
                , rawLaTeXInline'
                , note
                , try $ (char '[' *> inlineMarkup <* char ']')
                , inlineMarkup
                , link
                , image
                , mark
                , smartPunctuation inline
                , symbol
                ]

-- | Inline markups
inlineMarkup :: Parser [Char] ParserState Inline
inlineMarkup = choice [ simpleInline (string "??") (Cite [])
                      , simpleInline (string "**") Strong
                      , simpleInline (string "__") Emph
                      , simpleInline (char '*') Strong
                      , simpleInline (char '_') Emph
                      , simpleInline (char '+') Emph  -- approximates underline
                      , simpleInline (char '-') Strikeout
                      , simpleInline (char '^') Superscript
                      , simpleInline (char '~') Subscript
                      ]

-- | Trademark, registered, copyright
mark :: Parser [Char] st Inline
mark = try $ char '(' >> (try tm <|> try reg <|> copy)

reg :: Parser [Char] st Inline
reg = do
  oneOf "Rr"
  char ')'
  return $ Str "\174"

tm :: Parser [Char] st Inline
tm = do
  oneOf "Tt"
  oneOf "Mm"
  char ')'
  return $ Str "\8482"

copy :: Parser [Char] st Inline
copy = do
  oneOf "Cc"
  char ')'
  return $ Str "\169"

note :: Parser [Char] ParserState Inline
note = try $ do
  ref <- (char '[' *> many1 digit <* char ']')
  notes <- stateNotes <$> getState
  case lookup ref notes of
    Nothing   -> fail "note not found"
    Just raw  -> liftM Note $ parseFromString parseBlocks raw

-- | Special chars 
markupChars :: [Char]
markupChars = "\\[]*#_@~-+^|%="

-- | Break strings on following chars. Space tab and newline break for
--  inlines breaking. Open paren breaks for mark. Quote, dash and dot
--  break for smart punctuation. Punctuation breaks for regular
--  punctuation. Double quote breaks for named links. > and < break
--  for inline html.
stringBreakers :: [Char]
stringBreakers = " \t\n('-.,:!?;\"<>"

wordBoundaries :: [Char]
wordBoundaries = markupChars ++ stringBreakers

-- | Parse a hyphened sequence of words
hyphenedWords :: Parser [Char] ParserState String
hyphenedWords = try $ do
  hd <- noneOf wordBoundaries
  tl <- many ( (noneOf wordBoundaries) <|> 
               try (oneOf markupChars <* lookAhead (noneOf wordBoundaries) ) )
  let wd = hd:tl
  option wd $ try $ 
    (\r -> concat [wd, "-", r]) <$> (char '-' *> hyphenedWords)

-- | Any string
str :: Parser [Char] ParserState Inline
str = do
  baseStr <- hyphenedWords
  -- RedCloth compliance : if parsed word is uppercase and immediatly
  -- followed by parens, parens content is unconditionally word acronym
  fullStr <- option baseStr $ try $ do
    guard $ all isUpper baseStr
    acro <- enclosed (char '(') (char ')') anyChar
    return $ concat [baseStr, " (", acro, ")"]
  updateLastStrPos
  return $ Str fullStr

-- | Textile allows HTML span infos, we discard them
htmlSpan :: Parser [Char] ParserState Inline
htmlSpan = try $ Str <$> ( char '%' *> attributes *> manyTill anyChar (char '%') )

-- | Some number of space chars
whitespace :: Parser [Char] ParserState Inline
whitespace = many1 spaceChar >> return Space <?> "whitespace"

-- | In Textile, an isolated endline character is a line break
endline :: Parser [Char] ParserState Inline
endline = try $ do
  newline >> notFollowedBy blankline
  return LineBreak

rawHtmlInline :: Parser [Char] ParserState Inline
rawHtmlInline = RawInline "html" . snd <$> htmlTag isInlineTag

-- | Raw LaTeX Inline 
rawLaTeXInline' :: Parser [Char] ParserState Inline
rawLaTeXInline' = try $ do
  guardEnabled Ext_raw_tex
  rawLaTeXInline

-- | Textile standard link syntax is "label":target. But we
-- can also have ["label":target].
link :: Parser [Char] ParserState Inline
link = linkB <|> linkNoB

linkNoB :: Parser [Char] ParserState Inline
linkNoB = try $ do
  name <- surrounded (char '"') inline
  char ':'
  let stopChars = "!.,;:"
  url <- manyTill nonspaceChar (lookAhead $ space <|> try (oneOf stopChars >> (space <|> newline)))
  return $ Link name (url, "")

linkB :: Parser [Char] ParserState Inline
linkB = try $ do
  char '['
  name <- surrounded (char '"') inline
  char ':'
  url <- manyTill nonspaceChar (char ']')
  return $ Link name (url, "")

-- | Detect plain links to http or email.
autoLink :: Parser [Char] ParserState Inline
autoLink = do
  (orig, src) <- (try uri <|> try emailAddress)
  return $ Link [Str orig] (src, "")

-- | image embedding
image :: Parser [Char] ParserState Inline
image = try $ do
  char '!' >> notFollowedBy space
  src <- manyTill anyChar (lookAhead $ oneOf "!(")
  alt <- option "" (try $ (char '(' >> manyTill anyChar (char ')')))
  char '!'
  return $ Image [Str alt] (src, alt)

escapedInline :: Parser [Char] ParserState Inline
escapedInline = escapedEqs <|> escapedTag

escapedEqs :: Parser [Char] ParserState Inline
escapedEqs = Str <$> (try $ string "==" *> manyTill anyChar (try $ string "=="))

-- | literal text escaped btw <notextile> tags
escapedTag :: Parser [Char] ParserState Inline
escapedTag = Str <$>
  (try $ string "<notextile>" *> manyTill anyChar (try $ string "</notextile>"))

-- | Any special symbol defined in wordBoundaries
symbol :: Parser [Char] ParserState Inline
symbol = Str . singleton <$> oneOf wordBoundaries

-- | Inline code
code :: Parser [Char] ParserState Inline
code = code1 <|> code2

code1 :: Parser [Char] ParserState Inline
code1 = Code nullAttr <$> surrounded (char '@') anyChar

code2 :: Parser [Char] ParserState Inline
code2 = do
  htmlTag (tagOpen (=="tt") null)
  Code nullAttr <$> manyTill anyChar (try $ htmlTag $ tagClose (=="tt"))

-- | Html / CSS attributes
attributes :: Parser [Char] ParserState String
attributes = choice [ enclosed (char '(') (char ')') anyChar,
                      enclosed (char '{') (char '}') anyChar,
                      enclosed (char '[') (char ']') anyChar]

-- | Parses material surrounded by a parser.
surrounded :: Parser [Char] st t   -- ^ surrounding parser
	    -> Parser [Char] st a    -- ^ content parser (to be used repeatedly)
	    -> Parser [Char] st [a]
surrounded border = enclosed border (try border)

-- | Inlines are most of the time of the same form
simpleInline :: Parser [Char] ParserState t           -- ^ surrounding parser
                -> ([Inline] -> Inline)       -- ^ Inline constructor
                -> Parser [Char] ParserState Inline   -- ^ content parser (to be used repeatedly)
simpleInline border construct = surrounded border (inlineWithAttribute) >>=
                                return . construct . normalizeSpaces
  where inlineWithAttribute = (try $ optional attributes) >> inline

-- | Create a singleton list
singleton :: a -> [a]
singleton x = [x]