aboutsummaryrefslogtreecommitdiff
path: root/src/Text/Pandoc/Readers/HTML.hs
blob: 552e8a251746b45756736130bd0318a2f1f597df (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
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
{-
Copyright (C) 2006-2014 John MacFarlane <jgm@berkeley.edu>

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.HTML
   Copyright   : Copyright (C) 2006-2014 John MacFarlane
   License     : GNU GPL, version 2 or above

   Maintainer  : John MacFarlane <jgm@berkeley.edu>
   Stability   : alpha
   Portability : portable

Conversion of HTML to 'Pandoc' document.
-}
module Text.Pandoc.Readers.HTML ( readHtml
                                , htmlTag
                                , htmlInBalanced
                                , isInlineTag
                                , isBlockTag
                                , isTextTag
                                , isCommentTag
                                ) where

import Text.HTML.TagSoup
import Text.HTML.TagSoup.Match
import Text.Pandoc.Definition
import qualified Text.Pandoc.Builder as B
import Text.Pandoc.Builder (Blocks, Inlines, trimInlines)
import Text.Pandoc.Shared
import Text.Pandoc.Options
import Text.Pandoc.Parsing
import Data.Maybe ( fromMaybe, isJust )
import Data.List ( intercalate )
import Data.Char ( isDigit )
import Control.Monad ( liftM, guard, when, mzero )
import Control.Applicative ( (<$>), (<$), (<*) )
import Data.Monoid
import Text.Printf (printf)
import Debug.Trace (trace)

isSpace :: Char -> Bool
isSpace ' '  = True
isSpace '\t' = True
isSpace '\n' = True
isSpace _    = False

-- | Convert HTML-formatted string to 'Pandoc' document.
readHtml :: ReaderOptions -- ^ Reader options
         -> String        -- ^ String to parse (assumes @'\n'@ line endings)
         -> Pandoc
readHtml opts inp =
  case runParser parseDoc def{ stateOptions = opts } "source" tags of
          Left err'    -> error $ "\nError at " ++ show  err'
          Right result -> result
    where tags = canonicalizeTags $
                   parseTagsOptions parseOptions{ optTagPosition = True } inp
          parseDoc = do
             blocks <- (fixPlains False) . mconcat <$> manyTill block eof
             meta <- stateMeta <$> getState
             return $ Pandoc meta (B.toList blocks)

type TagParser = Parser [Tag String] ParserState

pBody :: TagParser Blocks
pBody = pInTags "body" block

pHead :: TagParser Blocks
pHead = pInTags "head" $ pTitle <|> pMetaTag <|> (mempty <$ pAnyTag)
  where pTitle = pInTags "title" inline >>= setTitle . trimInlines
        setTitle t = mempty <$ (updateState $ B.setMeta "title" t)
        pMetaTag = do
          mt <- pSatisfy (~== TagOpen "meta" [])
          let name = fromAttrib "name" mt
          if null name
             then return mempty
             else do
               let content = fromAttrib "content" mt
               updateState $ B.setMeta name (B.text content)
               return mempty

block :: TagParser Blocks
block = do
  tr <- getOption readerTrace
  pos <- getPosition
  res <- choice
            [ pPara
            , pHeader
            , pBlockQuote
            , pCodeBlock
            , pList
            , pHrule
            , pTable
            , pHead
            , pBody
            , pPlain
            , pDiv
            , pRawHtmlBlock
            ]
  when tr $ trace (printf "line %d: %s" (sourceLine pos)
             (take 60 $ show $ B.toList res)) (return ())
  return res


pList :: TagParser Blocks
pList = pBulletList <|> pOrderedList <|> pDefinitionList

pBulletList :: TagParser Blocks
pBulletList = try $ do
  pSatisfy (~== TagOpen "ul" [])
  let nonItem = pSatisfy (\t ->
                  not (tagOpen (`elem` ["li","ol","ul","dl"]) (const True) t) &&
                  not (t ~== TagClose "ul"))
  -- note: if they have an <ol> or <ul> not in scope of a <li>,
  -- treat it as a list item, though it's not valid xhtml...
  skipMany nonItem
  items <- manyTill (pInTags "li" block >>~ skipMany nonItem) (pCloses "ul")
  return $ B.bulletList $ map (fixPlains True) items

pOrderedList :: TagParser Blocks
pOrderedList = try $ do
  TagOpen _ attribs <- pSatisfy (~== TagOpen "ol" [])
  let (start, style) = (sta', sty')
                       where sta = fromMaybe "1" $
                                   lookup "start" attribs
                             sta' = if all isDigit sta
                                       then read sta
                                       else 1
                             sty = fromMaybe (fromMaybe "" $
                                   lookup "style" attribs) $
                                   lookup "class" attribs
                             sty' = case sty of
                                     "lower-roman"  -> LowerRoman
                                     "upper-roman"  -> UpperRoman
                                     "lower-alpha"  -> LowerAlpha
                                     "upper-alpha"  -> UpperAlpha
                                     "decimal"      -> Decimal
                                     _              -> DefaultStyle
  let nonItem = pSatisfy (\t ->
                  not (tagOpen (`elem` ["li","ol","ul","dl"]) (const True) t) &&
                  not (t ~== TagClose "ol"))
  -- note: if they have an <ol> or <ul> not in scope of a <li>,
  -- treat it as a list item, though it's not valid xhtml...
  skipMany nonItem
  items <- manyTill (pInTags "li" block >>~ skipMany nonItem) (pCloses "ol")
  return $ B.orderedListWith (start, style, DefaultDelim) $ map (fixPlains True) items

pDefinitionList :: TagParser Blocks
pDefinitionList = try $ do
  pSatisfy (~== TagOpen "dl" [])
  items <- manyTill pDefListItem (pCloses "dl")
  return $ B.definitionList items

pDefListItem :: TagParser (Inlines, [Blocks])
pDefListItem = try $ do
  let nonItem = pSatisfy (\t -> not (t ~== TagOpen "dt" []) &&
                  not (t ~== TagOpen "dd" []) && not (t ~== TagClose "dl"))
  terms <- many1 (try $ skipMany nonItem >> pInTags "dt" inline)
  defs  <- many1 (try $ skipMany nonItem >> pInTags "dd" block)
  skipMany nonItem
  let term = foldl1 (\x y ->  x <> B.linebreak <> y) terms
  return (term, map (fixPlains True) defs)

fixPlains :: Bool -> Blocks -> Blocks
fixPlains inList bs = if any isParaish bs'
                         then B.fromList $ map plainToPara bs'
                         else bs
  where isParaish (Para _) = True
        isParaish (CodeBlock _ _) = True
        isParaish (Header _ _ _) = True
        isParaish (BlockQuote _) = True
        isParaish (BulletList _) = not inList
        isParaish (OrderedList _ _) = not inList
        isParaish (DefinitionList _) = not inList
        isParaish _        = False
        plainToPara (Plain xs) = Para xs
        plainToPara x = x
        bs' = B.toList bs

pRawTag :: TagParser String
pRawTag = do
  tag <- pAnyTag
  let ignorable x = x `elem` ["html","head","body"]
  if tagOpen ignorable (const True) tag || tagClose ignorable tag
     then return []
     else return $ renderTags' [tag]

pDiv :: TagParser Blocks
pDiv = try $ do
  getOption readerParseRaw >>= guard
  TagOpen _ attr <- lookAhead $ pSatisfy $ tagOpen (=="div") (const True)
  contents <- pInTags "div" block
  return $ B.divWith (mkAttr attr) contents

pRawHtmlBlock :: TagParser Blocks
pRawHtmlBlock = do
  raw <- pHtmlBlock "script" <|> pHtmlBlock "style" <|> pRawTag
  parseRaw <- getOption readerParseRaw
  if parseRaw && not (null raw)
     then return $ B.rawBlock "html" raw
     else return mempty

pHtmlBlock :: String -> TagParser String
pHtmlBlock t = try $ do
  open <- pSatisfy (~== TagOpen t [])
  contents <- manyTill pAnyTag (pSatisfy (~== TagClose t))
  return $ renderTags' $ [open] ++ contents ++ [TagClose t]

pHeader :: TagParser Blocks
pHeader = try $ do
  TagOpen tagtype attr <- pSatisfy $
                           tagOpen (`elem` ["h1","h2","h3","h4","h5","h6"])
                           (const True)
  let bodyTitle = TagOpen tagtype attr ~== TagOpen "h1" [("class","title")]
  let level = read (drop 1 tagtype)
  contents <- trimInlines . mconcat <$> manyTill inline (pCloses tagtype <|> eof)
  let ident = fromMaybe "" $ lookup "id" attr
  let classes = maybe [] words $ lookup "class" attr
  let keyvals = [(k,v) | (k,v) <- attr, k /= "class", k /= "id"]
  return $ if bodyTitle
              then mempty  -- skip a representation of the title in the body
              else B.headerWith (ident, classes, keyvals) level contents

pHrule :: TagParser Blocks
pHrule = do
  pSelfClosing (=="hr") (const True)
  return B.horizontalRule

pTable :: TagParser Blocks
pTable = try $ do
  TagOpen _ _ <- pSatisfy (~== TagOpen "table" [])
  skipMany pBlank
  caption <- option mempty $ pInTags "caption" inline >>~ skipMany pBlank
  -- TODO actually read these and take width information from them
  widths' <- pColgroup <|> many pCol
  head' <- option [] $ pOptInTag "thead" $ pInTags "tr" (pCell "th")
  skipMany pBlank
  rows <- pOptInTag "tbody"
          $ many1 $ try $ skipMany pBlank >> pInTags "tr" (pCell "td")
  skipMany pBlank
  TagClose _ <- pSatisfy (~== TagClose "table")
  let isSinglePlain x = case B.toList x of
                             [Plain _] -> True
                             _         -> False
  let isSimple = all isSinglePlain $ concat (head':rows)
  let cols = length $ if null head' then head rows else head'
  -- fail if there are colspans or rowspans
  guard $ all (\r -> length r == cols) rows
  let aligns = replicate cols AlignDefault
  let widths = if null widths'
                  then if isSimple
                       then replicate cols 0
                       else replicate cols (1.0 / fromIntegral cols)
                  else widths'
  return $ B.table caption (zip aligns widths) head' rows

pCol :: TagParser Double
pCol = try $ do
  TagOpen _ attribs <- pSatisfy (~== TagOpen "col" [])
  skipMany pBlank
  optional $ pSatisfy (~== TagClose "col")
  skipMany pBlank
  return $ case lookup "width" attribs of
           Just x | not (null x) && last x == '%' ->
             fromMaybe 0.0 $ safeRead ('0':'.':init x)
           _ -> 0.0

pColgroup :: TagParser [Double]
pColgroup = try $ do
  pSatisfy (~== TagOpen "colgroup" [])
  skipMany pBlank
  manyTill pCol (pCloses "colgroup" <|> eof) <* skipMany pBlank

pCell :: String -> TagParser [Blocks]
pCell celltype = try $ do
  skipMany pBlank
  res <- pInTags celltype block
  skipMany pBlank
  return [res]

pBlockQuote :: TagParser Blocks
pBlockQuote = do
  contents <- pInTags "blockquote" block
  return $ B.blockQuote $ fixPlains False contents

pPlain :: TagParser Blocks
pPlain = do
  contents <- trimInlines . mconcat <$> many1 inline
  if B.isNull contents
     then return mempty
     else return $ B.plain contents

pPara :: TagParser Blocks
pPara = do
  contents <- trimInlines <$> pInTags "p" inline
  return $ B.para contents

pCodeBlock :: TagParser Blocks
pCodeBlock = try $ do
  TagOpen _ attr <- pSatisfy (~== TagOpen "pre" [])
  contents <- manyTill pAnyTag (pCloses "pre" <|> eof)
  let rawText = concatMap fromTagText $ filter isTagText contents
  -- drop leading newline if any
  let result' = case rawText of
                     '\n':xs  -> xs
                     _        -> rawText
  -- drop trailing newline if any
  let result = case reverse result' of
                    '\n':_   -> init result'
                    _        -> result'
  return $ B.codeBlockWith (mkAttr attr) result

inline :: TagParser Inlines
inline = choice
           [ pTagText
           , pQ
           , pEmph
           , pStrong
           , pSuperscript
           , pSubscript
           , pStrikeout
           , pLineBreak
           , pLink
           , pImage
           , pCode
           , pSpan
           , pRawHtmlInline
           ]

pLocation :: TagParser ()
pLocation = do
  (TagPosition r c) <- pSat isTagPosition
  setPosition $ newPos "input" r c

pSat :: (Tag String -> Bool) -> TagParser (Tag String)
pSat f = do
  pos <- getPosition
  token show (const pos) (\x -> if f x then Just x else Nothing)

pSatisfy :: (Tag String -> Bool) -> TagParser (Tag String)
pSatisfy f = try $ optional pLocation >> pSat f

pAnyTag :: TagParser (Tag String)
pAnyTag = pSatisfy (const True)

pSelfClosing :: (String -> Bool) -> ([Attribute String] -> Bool)
             -> TagParser (Tag String)
pSelfClosing f g = do
  open <- pSatisfy (tagOpen f g)
  optional $ pSatisfy (tagClose f)
  return open

pQ :: TagParser Inlines
pQ = do
  quoteContext <- stateQuoteContext `fmap` getState
  let quoteType = case quoteContext of
                       InDoubleQuote -> SingleQuote
                       _             -> DoubleQuote
  let innerQuoteContext = if quoteType == SingleQuote
                             then InSingleQuote
                             else InDoubleQuote
  let constructor = case quoteType of
                            SingleQuote -> B.singleQuoted
                            DoubleQuote -> B.doubleQuoted
  withQuoteContext innerQuoteContext $
    pInlinesInTags "q" constructor

pEmph :: TagParser Inlines
pEmph = pInlinesInTags "em" B.emph <|> pInlinesInTags "i" B.emph

pStrong :: TagParser Inlines
pStrong = pInlinesInTags "strong" B.strong <|> pInlinesInTags "b" B.strong

pSuperscript :: TagParser Inlines
pSuperscript = pInlinesInTags "sup" B.superscript

pSubscript :: TagParser Inlines
pSubscript = pInlinesInTags "sub" B.subscript

pStrikeout :: TagParser Inlines
pStrikeout = do
  pInlinesInTags "s" B.strikeout <|>
    pInlinesInTags "strike" B.strikeout <|>
    pInlinesInTags "del" B.strikeout <|>
    try (do pSatisfy (~== TagOpen "span" [("class","strikeout")])
            contents <- mconcat <$> manyTill inline (pCloses "span")
            return $ B.strikeout contents)

pLineBreak :: TagParser Inlines
pLineBreak = do
  pSelfClosing (=="br") (const True)
  return B.linebreak

pLink :: TagParser Inlines
pLink = try $ do
  tag <- pSatisfy (tagOpenLit "a" (isJust . lookup "href"))
  let url = fromAttrib "href" tag
  let title = fromAttrib "title" tag
  lab <- trimInlines . mconcat <$> manyTill inline (pCloses "a")
  return $ B.link (escapeURI url) title lab

pImage :: TagParser Inlines
pImage = do
  tag <- pSelfClosing (=="img") (isJust . lookup "src")
  let url = fromAttrib "src" tag
  let title = fromAttrib "title" tag
  let alt = fromAttrib "alt" tag
  return $ B.image (escapeURI url) title (B.text alt)

pCode :: TagParser Inlines
pCode = try $ do
  (TagOpen open attr) <- pSatisfy $ tagOpen (`elem` ["code","tt"]) (const True)
  result <- manyTill pAnyTag (pCloses open)
  return $ B.codeWith (mkAttr attr) $ intercalate " " $ lines $ innerText result

pSpan :: TagParser Inlines
pSpan = try $ do
  getOption readerParseRaw >>= guard
  TagOpen _ attr <- lookAhead $ pSatisfy $ tagOpen (=="span") (const True)
  contents <- pInTags "span" inline
  return $ B.spanWith (mkAttr attr) contents

pRawHtmlInline :: TagParser Inlines
pRawHtmlInline = do
  result <- pSatisfy (tagComment (const True)) <|> pSatisfy isInlineTag
  parseRaw <- getOption readerParseRaw
  if parseRaw
     then return $ B.rawInline "html" $ renderTags' [result]
     else return mempty

pInlinesInTags :: String -> (Inlines -> Inlines)
               -> TagParser Inlines
pInlinesInTags tagtype f = extractSpaces f <$> pInTags tagtype inline

pInTags :: (Monoid a) => String -> TagParser a
        -> TagParser a
pInTags tagtype parser = try $ do
  pSatisfy (~== TagOpen tagtype [])
  mconcat <$> manyTill parser (pCloses tagtype <|> eof)

pOptInTag :: String -> TagParser a
          -> TagParser a
pOptInTag tagtype parser = try $ do
  open <- option False (pSatisfy (~== TagOpen tagtype []) >> return True)
  skipMany pBlank
  x <- parser
  skipMany pBlank
  when open $ pCloses tagtype
  return x

pCloses :: String -> TagParser ()
pCloses tagtype = try $ do
  t <- lookAhead $ pSatisfy $ \tag -> isTagClose tag || isTagOpen tag
  case t of
       (TagClose t') | t' == tagtype -> pAnyTag >> return ()
       (TagOpen t' _) | t' `closes` tagtype -> return ()
       (TagClose "ul") | tagtype == "li" -> return ()
       (TagClose "ol") | tagtype == "li" -> return ()
       (TagClose "dl") | tagtype == "li" -> return ()
       (TagClose "table") | tagtype == "td" -> return ()
       (TagClose "table") | tagtype == "tr" -> return ()
       _ -> mzero

pTagText :: TagParser Inlines
pTagText = try $ do
  (TagText str) <- pSatisfy isTagText
  st <- getState
  case runParser (many pTagContents) st "text" str of
       Left _        -> fail $ "Could not parse `" ++ str ++ "'"
       Right result  -> return $ mconcat result

pBlank :: TagParser ()
pBlank = try $ do
  (TagText str) <- pSatisfy isTagText
  guard $ all isSpace str

pTagContents :: Parser [Char] ParserState Inlines
pTagContents =
      B.displayMath <$> mathDisplay
  <|> B.math        <$> mathInline
  <|> pStr
  <|> pSpace
  <|> smartPunctuation pTagContents
  <|> pSymbol
  <|> pBad

pStr :: Parser [Char] ParserState Inlines
pStr = do
  result <- many1 $ satisfy $ \c ->
                     not (isSpace c) && not (isSpecial c) && not (isBad c)
  pos <- getPosition
  updateState $ \s -> s{ stateLastStrPos = Just pos }
  return $ B.str result

isSpecial :: Char -> Bool
isSpecial '"' = True
isSpecial '\'' = True
isSpecial '.' = True
isSpecial '-' = True
isSpecial '$' = True
isSpecial '\8216' = True
isSpecial '\8217' = True
isSpecial '\8220' = True
isSpecial '\8221' = True
isSpecial _ = False

pSymbol :: Parser [Char] ParserState Inlines
pSymbol = satisfy isSpecial >>= return . B.str . (:[])

isBad :: Char -> Bool
isBad c = c >= '\128' && c <= '\159' -- not allowed in HTML

pBad :: Parser [Char] ParserState Inlines
pBad = do
  c <- satisfy isBad
  let c' = case c of
                '\128' -> '\8364'
                '\130' -> '\8218'
                '\131' -> '\402'
                '\132' -> '\8222'
                '\133' -> '\8230'
                '\134' -> '\8224'
                '\135' -> '\8225'
                '\136' -> '\710'
                '\137' -> '\8240'
                '\138' -> '\352'
                '\139' -> '\8249'
                '\140' -> '\338'
                '\142' -> '\381'
                '\145' -> '\8216'
                '\146' -> '\8217'
                '\147' -> '\8220'
                '\148' -> '\8221'
                '\149' -> '\8226'
                '\150' -> '\8211'
                '\151' -> '\8212'
                '\152' -> '\732'
                '\153' -> '\8482'
                '\154' -> '\353'
                '\155' -> '\8250'
                '\156' -> '\339'
                '\158' -> '\382'
                '\159' -> '\376'
                _      -> '?'
  return $ B.str [c']

pSpace :: Parser [Char] ParserState Inlines
pSpace = many1 (satisfy isSpace) >> return B.space

--
-- Constants
--

eitherBlockOrInline :: [String]
eitherBlockOrInline = ["applet", "button", "del", "iframe", "ins",
                  "map", "area", "object"]

{-
inlineHtmlTags :: [[Char]]
inlineHtmlTags = ["a", "abbr", "acronym", "b", "basefont", "bdo", "big",
                  "br", "cite", "code", "dfn", "em", "font", "i", "img",
                  "input", "kbd", "label", "q", "s", "samp", "select",
                  "small", "span", "strike", "strong", "sub", "sup",
                  "textarea", "tt", "u", "var"]
-}

blockHtmlTags :: [String]
blockHtmlTags = ["address", "article", "aside", "blockquote", "body", "button", "canvas",
                 "caption", "center", "col", "colgroup", "dd", "dir", "div",
                 "dl", "dt", "embed", "fieldset", "figcaption", "figure", "footer",
                 "form", "h1", "h2", "h3", "h4",
                 "h5", "h6", "head", "header", "hgroup", "hr", "html", "isindex", "map", "menu",
                 "noframes", "noscript", "object", "ol", "output", "p", "pre", "progress",
                 "section", "table", "tbody", "textarea", "thead", "tfoot", "ul", "dd",
                 "dt", "frameset", "li", "tbody", "td", "tfoot",
                 "th", "thead", "tr", "script", "style", "svg", "video"]

-- We want to allow raw docbook in markdown documents, so we
-- include docbook block tags here too.
blockDocBookTags :: [String]
blockDocBookTags = ["calloutlist", "bibliolist", "glosslist", "itemizedlist",
                    "orderedlist", "segmentedlist", "simplelist",
                    "variablelist", "caution", "important", "note", "tip",
                    "warning", "address", "literallayout", "programlisting",
                    "programlistingco", "screen", "screenco", "screenshot",
                    "synopsis", "example", "informalexample", "figure",
                    "informalfigure", "table", "informaltable", "para",
                    "simpara", "formalpara", "equation", "informalequation",
                    "figure", "screenshot", "mediaobject", "qandaset",
                    "procedure", "task", "cmdsynopsis", "funcsynopsis",
                    "classsynopsis", "blockquote", "epigraph", "msgset",
                    "sidebar", "title"]

blockTags :: [String]
blockTags = blockHtmlTags ++ blockDocBookTags

isInlineTag :: Tag String -> Bool
isInlineTag t = tagOpen isInlineTagName (const True) t ||
                tagClose isInlineTagName t ||
                tagComment (const True) t
                 where isInlineTagName x = x `notElem` blockTags

isBlockTag :: Tag String -> Bool
isBlockTag t = tagOpen isBlockTagName (const True) t ||
               tagClose isBlockTagName t ||
               tagComment (const True) t
                 where isBlockTagName ('?':_) = True
                       isBlockTagName ('!':_) = True
                       isBlockTagName x       = x `elem` blockTags
                                             || x `elem` eitherBlockOrInline

isTextTag :: Tag String -> Bool
isTextTag = tagText (const True)

isCommentTag :: Tag String -> Bool
isCommentTag = tagComment (const True)

-- taken from HXT and extended
-- See http://www.w3.org/TR/html5/syntax.html sec 8.1.2.4 optional tags
closes :: String -> String -> Bool
_ `closes` "body" = False
_ `closes` "html" = False
"a" `closes` "a" = True
"li" `closes` "li" = True
"th" `closes` t | t `elem` ["th","td"] = True
"tr" `closes` t | t `elem` ["th","td","tr"] = True
"dd" `closes` t | t `elem` ["dt", "dd"] = True
"dt" `closes` t | t `elem` ["dt","dd"] = True
"rt" `closes` t | t `elem` ["rb", "rt", "rtc"] = True
"optgroup" `closes` "optgroup" = True
"optgroup" `closes` "option" = True
"option" `closes` "option" = True
-- http://www.w3.org/TR/html-markup/p.html
x `closes` "p" | x `elem` ["address", "article", "aside", "blockquote",
   "dir", "div", "dl", "fieldset", "footer", "form", "h1", "h2", "h3", "h4",
   "h5", "h6", "header", "hr", "menu", "nav", "ol", "p", "pre", "section",
   "table", "ul"] = True
"meta" `closes` "meta" = True
"form" `closes` "form" = True
"label" `closes` "label" = True
"map" `closes` "map" = True
"object" `closes` "object" = True
_ `closes` t | t `elem` ["option","style","script","textarea","title"] = True
t `closes` "select" | t /= "option" = True
"thead" `closes` t | t `elem` ["colgroup"] = True
"tfoot" `closes` t | t `elem` ["thead","colgroup"] = True
"tbody" `closes` t | t `elem` ["tbody","tfoot","thead","colgroup"] = True
t `closes` t2 |
   t `elem` ["h1","h2","h3","h4","h5","h6","dl","ol","ul","table","div","p"] &&
   t2 `elem` ["h1","h2","h3","h4","h5","h6","p" ] = True -- not "div"
t1 `closes` t2 |
   t1 `elem` blockTags &&
   t2 `notElem` (blockTags ++ eitherBlockOrInline) = True
_ `closes` _ = False

--- parsers for use in markdown, textile readers

-- | Matches a stretch of HTML in balanced tags.
htmlInBalanced :: (Tag String -> Bool) -> Parser [Char] ParserState String
htmlInBalanced f = try $ do
  (TagOpen t _, tag) <- htmlTag f
  guard $ '/' `notElem` tag      -- not a self-closing tag
  let stopper = htmlTag (~== TagClose t)
  let anytag = liftM snd $ htmlTag (const True)
  contents <- many $ notFollowedBy' stopper >>
                     (htmlInBalanced f <|> anytag <|> count 1 anyChar)
  endtag <- liftM snd stopper
  return $ tag ++ concat contents ++ endtag

-- | Matches a tag meeting a certain condition.
htmlTag :: (Tag String -> Bool) -> Parser [Char] st (Tag String, String)
htmlTag f = try $ do
  lookAhead $ char '<' >> (oneOf "/!?" <|> letter)
  (next : _) <- getInput >>= return . canonicalizeTags . parseTags
  guard $ f next
  -- advance the parser
  case next of
       TagComment s -> do
          count (length s + 4) anyChar
          skipMany (satisfy (/='>'))
          char '>'
          return (next, "<!--" ++ s ++ "-->")
       _            -> do
          rendered <- manyTill anyChar (char '>')
          return (next, rendered ++ ">")

mkAttr :: [(String, String)] -> Attr
mkAttr attr = (attribsId, attribsClasses, attribsKV)
  where attribsId = fromMaybe "" $ lookup "id" attr
        attribsClasses = words $ fromMaybe "" $ lookup "class" attr
        attribsKV = filter (\(k,_) -> k /= "class" && k /= "id") attr