aboutsummaryrefslogtreecommitdiff
path: root/src/Text/Pandoc/Readers/HTML.hs
blob: f9a738e94ab9a2035f8e717bde07521f46a7c9f9 (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
-- | Converts HTML to 'Pandoc' document.
module Text.Pandoc.Readers.HTML ( 
                                 readHtml, 
                                 rawHtmlInline, 
                                 rawHtmlBlock, 
                                 anyHtmlBlockTag, 
                                 anyHtmlInlineTag  
                                ) where

import Text.Regex ( matchRegex, mkRegex )
import Text.ParserCombinators.Parsec
import Text.ParserCombinators.Pandoc
import Text.Pandoc.Definition
import Text.Pandoc.Shared 
import Text.Pandoc.HtmlEntities ( decodeEntities, htmlEntityToChar )
import Maybe ( fromMaybe )
import Char ( toUpper, toLower )

-- | Convert HTML-formatted string to 'Pandoc' document.
readHtml :: ParserState   -- ^ Parser state
         -> String        -- ^ String to parse
         -> Pandoc
readHtml = readWith parseHtml

-- for testing 
testString :: String -> IO ()
testString = testStringWith parseHtml

--
-- Constants
--

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"]

--
-- HTML utility functions
--

-- | Read blocks until end tag.
blocksTilEnd tag = try (do
  blocks <- manyTill (do {b <- block; spaces; return b}) (htmlEndTag tag)
  return blocks)

-- | Read inlines until end tag.
inlinesTilEnd tag = try (do
  inlines <- manyTill inline (htmlEndTag tag)
  return inlines)

-- extract type from a tag:  e.g. br from <br>, < br >, </br>, etc.
extractTagType tag = case (matchRegex (mkRegex  "<[[:space:]]*/?([A-Za-z0-9]+)") tag) of
                       Just [match]   -> (map toLower match)
                       Nothing        -> ""

anyHtmlTag = try (do
  char '<'
  spaces
  tag <- many1 alphaNum
  attribs <- htmlAttributes  
  spaces
  ender <- option "" (string "/")
  let ender' = if (null ender) then "" else " /"
  spaces
  char '>'
  return ("<" ++ tag ++ attribs ++ ender' ++ ">"))

anyHtmlEndTag = try (do
  char '<'   
  spaces
  char '/'
  spaces
  tagType <- many1 alphaNum
  spaces
  char '>'
  return ("</" ++ tagType ++ ">"))

htmlTag :: String -> GenParser Char st (String, [(String, String)])
htmlTag tag = try (do
  char '<'
  spaces
  stringAnyCase tag
  attribs <- many htmlAttribute
  spaces
  option "" (string "/")
  spaces
  char '>'
  return (tag, (map (\(name, content, raw) -> (name, content)) attribs)))

-- parses a quoted html attribute value
quoted quoteChar = do
  result <- between (char quoteChar) (char quoteChar) (many (noneOf [quoteChar]))
  return (result, [quoteChar])

htmlAttributes = do
  attrList <- many htmlAttribute
  return (concatMap (\(name, content, raw) -> raw) attrList)

htmlAttribute = htmlRegularAttribute <|> htmlMinimizedAttribute

-- minimized boolean attribute (no = and value)
htmlMinimizedAttribute = try (do
  spaces
  name <- many1 (choice [letter, oneOf ".-_:"])
  spaces
  notFollowedBy (char '=')
  let content = name
  return (name, content, (" " ++ name)))

htmlRegularAttribute = try (do
  spaces
  name <- many1 (choice [letter, oneOf ".-_:"])
  spaces
  char '='
  spaces
  (content, quoteStr) <- choice [ (quoted '\''), 
                                  (quoted '"'), 
                                  (do{ a <- (many (alphaNum <|> (oneOf "-._:"))); 
                                       return (a,"")} ) ]
  return (name, content, (" " ++ name ++ "=" ++ quoteStr ++ content ++ quoteStr)))

htmlEndTag tag = try (do
  char '<'   
  spaces
  char '/'
  spaces
  stringAnyCase tag
  spaces
  char '>'
  return ("</" ++ tag ++ ">"))

-- | Returns @True@ if the tag is an inline tag.
isInline tag = (extractTagType tag) `elem` inlineHtmlTags

anyHtmlBlockTag = try (do
  tag <- choice [anyHtmlTag, anyHtmlEndTag]
  if isInline tag then
      fail "inline tag"
    else
      return tag)

anyHtmlInlineTag = try (do
  tag <- choice [ anyHtmlTag, anyHtmlEndTag ]
  if isInline tag then
      return tag
    else
      fail "not an inline tag")

-- scripts must be treated differently, because they can contain <> etc.
htmlScript = try (do
  open <- string "<script"
  rest <- manyTill anyChar (htmlEndTag "script")
  return (open ++ rest ++ "</script>"))

rawHtmlBlock = try (do
  notFollowedBy' (choice [htmlTag "/body", htmlTag "/html"])
  body <- choice [htmlScript, anyHtmlBlockTag, htmlComment, xmlDec, definition]
  sp <- (many space)
  state <- getState
  if stateParseRaw state then
      return (RawHtml (body ++ sp))
    else
      return Null)

htmlComment = try (do
  string "<!--"
  comment <- manyTill anyChar (try (string "-->"))
  return ("<!--" ++ comment ++ "-->"))

--
-- parsing documents
--

xmlDec = try (do
  string "<?"
  rest <- manyTill anyChar (char '>')
  return ("<?" ++ rest ++ ">"))

definition = try (do
  string "<!"
  rest <- manyTill anyChar (char '>')
  return ("<!" ++ rest ++ ">"))

nonTitleNonHead = try (do
  notFollowedBy' (htmlTag "title")
  notFollowedBy' (htmlTag "/head")
  result <- choice [do {rawHtmlBlock; return ' '}, anyChar]
  return result)

parseTitle = try (do
  (tag, attribs) <- htmlTag "title"
  contents <- inlinesTilEnd tag
  spaces
  return contents)

-- parse header and return meta-information (for now, just title)
parseHead = try (do
  htmlTag "head"
  spaces
  skipMany nonTitleNonHead
  contents <- option [] parseTitle
  skipMany nonTitleNonHead
  htmlTag "/head"
  return (contents, [], ""))

skipHtmlTag tag = option ("",[]) (htmlTag tag)

-- h1 class="title" representation of title in body
bodyTitle = try (do
  (tag, attribs) <- htmlTag "h1"  
  cl <- case (extractAttribute "class" attribs) of
          Just "title" -> do {return ""}
          otherwise    -> fail "not title"
  inlinesTilEnd "h1"
  return "")

parseHtml = do
  sepEndBy (choice [xmlDec, definition, htmlComment]) spaces
  skipHtmlTag "html"
  spaces
  (title, authors, date) <- option ([], [], "") parseHead 
  spaces
  skipHtmlTag "body"
  spaces
  option "" bodyTitle  -- skip title in body, because it's represented in meta
  blocks <- parseBlocks
  spaces
  option "" (htmlEndTag "body")
  spaces
  option "" (htmlEndTag "html")
  many anyChar -- ignore anything after </html>
  eof
  state <- getState
  let keyBlocks = stateKeyBlocks state 
  return (Pandoc (Meta title authors date) (blocks ++ (reverse keyBlocks)))

--
-- parsing blocks
--

parseBlocks = do
  spaces
  result <- sepEndBy block spaces
  return result

block = choice [ codeBlock, header, hrule, list, blockQuote, para, plain, 
                 rawHtmlBlock ] <?> "block"

--
-- header blocks
--

header = choice (map headerLevel (enumFromTo 1 5)) <?> "header"

headerLevel n = try (do
    let level = "h" ++ show n
    (tag, attribs) <- htmlTag level
    contents <- inlinesTilEnd level
    return (Header n (normalizeSpaces contents)))

--
-- hrule block
--

hrule = try (do
  (tag, attribs) <- htmlTag "hr"
  state <- getState
  if (not (null attribs)) && (stateParseRaw state) then
       unexpected "attributes in hr"  -- in this case we want to parse it as raw html
     else
       return HorizontalRule)

--
-- code blocks
--

codeBlock = choice [ preCodeBlock, bareCodeBlock ] <?> "code block"

preCodeBlock = try (do
    htmlTag "pre" 
    spaces
    htmlTag "code"
    result <- manyTill anyChar (htmlEndTag "code")
    spaces
    htmlEndTag "pre"
    return (CodeBlock (stripTrailingNewlines (decodeEntities result))))

bareCodeBlock = try (do
    htmlTag "code"
    result <- manyTill anyChar (htmlEndTag "code")
    return (CodeBlock (stripTrailingNewlines (decodeEntities result))))

--
-- block quotes
--

blockQuote = try (do
    tag <- htmlTag "blockquote"
    spaces
    blocks <- blocksTilEnd "blockquote"
    return (BlockQuote blocks))

--
-- list blocks
--

list = choice [ bulletList, orderedList ] <?> "list"

orderedList = try (do
    tag <- htmlTag "ol"
    spaces
    items <- sepEndBy1 listItem spaces
    htmlEndTag "ol"
    return (OrderedList items))

bulletList = try (do
    tag <- htmlTag "ul"
    spaces
    items <- sepEndBy1 listItem spaces
    htmlEndTag "ul"
    return (BulletList items))

listItem = try (do
    tag <- htmlTag "li"
    spaces
    blocks <- blocksTilEnd "li"
    return blocks)

--
-- paragraph block
--

para = try (do 
  tag <- htmlTag "p"
  result <- inlinesTilEnd "p"
  return (Para (normalizeSpaces result)))

-- 
-- plain block
--

plain = do
    result <- many1 inline
    return (Plain (normalizeSpaces result))

-- 
-- inline
--

inline = choice [ text, special ] <?> "inline"

text =  choice [ entity, strong, emph, code, str, linebreak, whitespace ] <?> "text"

special = choice [ link, image, rawHtmlInline ] <?> "link, inline html, or image"

entity = try (do
  char '&'
  body <- choice [(many1 letter), 
                  (try (do{ char '#'; num <- many1 digit; return ("#" ++ num)}))]
  char ';'
  return (Str [fromMaybe '?' (htmlEntityToChar ("&" ++ body ++ ";"))]))

code = try (do 
  htmlTag "code"
  result <- manyTill anyChar (htmlEndTag "code")
  -- remove internal line breaks, leading and trailing space, and decode entities
  let result' = decodeEntities $ removeLeadingTrailingSpace $ joinWithSep " " $ lines result 
  return (Code result'))

rawHtmlInline = do
  result <- choice [htmlScript, anyHtmlInlineTag]
  state <- getState
  if stateParseRaw state then
      return (HtmlInline result)
    else
      return (Str "")

betweenTags tag = try (do
  htmlTag tag
  result <- inlinesTilEnd tag
  return (normalizeSpaces result))

emph = try (do 
  result <- choice [betweenTags "em", betweenTags "it"]
  return (Emph result))

strong = try (do 
  result <- choice [betweenTags "b", betweenTags "strong"]
  return (Strong result))

whitespace = do
  many1 space
  return Space

-- hard line break
linebreak = do
  htmlTag "br"
  return LineBreak 

str = do 
  result <- many1 (noneOf "<& \t\n")
  return (Str (decodeEntities result))

--
-- links and images
--

-- extract contents of attribute (attribute names are case-insensitive)
extractAttribute name [] = Nothing
extractAttribute name ((attrName, contents):rest) = 
  let name' = map toLower name 
      attrName' = map toLower attrName in
  if (attrName' == name') then Just contents else extractAttribute name rest

link = try (do
  (tag, attributes) <- htmlTag "a"  
  url <- case (extractAttribute "href" attributes) of
           Just url -> do {return url}
           Nothing  -> fail "no href"
  let title = fromMaybe "" (extractAttribute "title" attributes)
  label <- inlinesTilEnd "a"
  ref <- generateReference url title
  return (Link (normalizeSpaces label) ref))

image = try (do
  (tag, attributes) <- htmlTag "img" 
  url <- case (extractAttribute "src" attributes) of
           Just url -> do {return url}
           Nothing  -> fail "no src"
  let title = fromMaybe "" (extractAttribute "title" attributes)
  let alt = fromMaybe "" (extractAttribute "alt" attributes)
  ref <- generateReference url title
  return (Image [Str alt] ref))