aboutsummaryrefslogtreecommitdiff
path: root/src/Text/Pandoc/Writers/FB2.hs
blob: cf96393caf596fd9110a0f7bb3a1e985ee6048db (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
{-# LANGUAGE PatternGuards #-}

{-
Copyright (c) 2011-2012 Sergey Astanin
              2012-2017 John MacFarlane

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.Writers.FB2
Copyright   : Copyright (C) 2011-2012 Sergey Astanin
                            2012-2017 John MacFarlane
License     : GNU GPL, version 2 or above

Maintainer  : John MacFarlane
Stability   : alpha
Portability : portable

Conversion of 'Pandoc' documents to FB2 (FictionBook2) format.

FictionBook is an XML-based e-book format. For more information see:
<http://www.fictionbook.org/index.php/Eng:XML_Schema_Fictionbook_2.1>

-}
module Text.Pandoc.Writers.FB2 (writeFB2)  where

import Control.Monad (zipWithM)
import Control.Monad.Except (catchError)
import Control.Monad.State.Strict (StateT, evalStateT, get, lift, liftM, modify)
import Data.ByteString.Base64 (encode)
import qualified Data.ByteString.Char8 as B8
import Data.Char (isAscii, isControl, isSpace, toLower)
import Data.Either (lefts, rights)
import Data.List (intercalate, intersperse, isPrefixOf, stripPrefix)
import Data.Text (Text, pack)
import Network.HTTP (urlEncode)
import Text.XML.Light
import qualified Text.XML.Light as X
import qualified Text.XML.Light.Cursor as XC

import Text.Pandoc.Class (PandocMonad, report)
import qualified Text.Pandoc.Class as P
import Text.Pandoc.Definition
import Text.Pandoc.Logging
import Text.Pandoc.Options (HTMLMathMethod (..), WriterOptions (..), def)
import Text.Pandoc.Shared (capitalize, isHeaderBlock, isURI, orderedListMarkers)

-- | Data to be written at the end of the document:
-- (foot)notes, URLs, references, images.
data FbRenderState = FbRenderState
    { footnotes         :: [ (Int, String, [Content]) ]  -- ^ #, ID, text
    , imagesToFetch     :: [ (String, String) ]  -- ^ filename, URL or path
    , parentListMarker  :: String  -- ^ list marker of the parent ordered list
    , parentBulletLevel :: Int  -- ^ nesting level of the unordered list
    , writerOptions     :: WriterOptions
    } deriving (Show)

-- | FictionBook building monad.
type FBM m = StateT FbRenderState m

newFB :: FbRenderState
newFB = FbRenderState { footnotes = [], imagesToFetch = []
                      , parentListMarker = "", parentBulletLevel = 0
                      , writerOptions = def }

data ImageMode = NormalImage | InlineImage deriving (Eq)
instance Show ImageMode where
    show NormalImage = "imageType"
    show InlineImage = "inlineImageType"

-- | Produce an FB2 document from a 'Pandoc' document.
writeFB2 :: PandocMonad m
         => WriterOptions    -- ^ conversion options
         -> Pandoc           -- ^ document to convert
         -> m Text           -- ^ FictionBook2 document (not encoded yet)
writeFB2 opts doc = flip evalStateT newFB $ pandocToFB2 opts doc

pandocToFB2 :: PandocMonad m
            => WriterOptions
            -> Pandoc
            -> FBM m Text
pandocToFB2 opts (Pandoc meta blocks) = do
     modify (\s -> s { writerOptions = opts })
     desc <- description meta
     fp <- frontpage meta
     secs <- renderSections 1 blocks
     let body = el "body" $ fp ++ secs
     notes <- renderFootnotes
     (imgs,missing) <- fmap imagesToFetch get >>= \s -> lift (fetchImages s)
     let body' = replaceImagesWithAlt missing body
     let fb2_xml = el "FictionBook" (fb2_attrs, [desc, body'] ++ notes ++ imgs)
     return $ pack $ xml_head ++ showContent fb2_xml ++ "\n"
  where
  xml_head = "<?xml version=\"1.0\" encoding=\"UTF-8\"?>\n"
  fb2_attrs =
      let xmlns = "http://www.gribuser.ru/xml/fictionbook/2.0"
          xlink = "http://www.w3.org/1999/xlink"
      in  [ uattr "xmlns" xmlns
          , attr ("xmlns", "l") xlink ]

frontpage :: PandocMonad m => Meta -> FBM m [Content]
frontpage meta' = do
  t <- cMapM toXml . docTitle $ meta'
  return
    [ el "title" (el "p" t)
    , el "annotation" (map (el "p" . cMap plain)
                       (docAuthors meta' ++ [docDate meta']))
    ]

description :: PandocMonad m => Meta -> FBM m Content
description meta' = do
  bt <- booktitle meta'
  let as = authors meta'
  dd <- docdate meta'
  let lang = case lookupMeta "lang" meta' of
               Just (MetaInlines [Str s]) -> [el "lang" $ iso639 s]
               Just (MetaString s)        -> [el "lang" $ iso639 s]
               _                          -> []
             where iso639 = takeWhile (/= '-') -- Convert BCP 47 to ISO 639
  return $ el "description"
    [ el "title-info" (bt ++ as ++ dd ++ lang)
    , el "document-info" [ el "program-used" "pandoc" ] -- FIXME: +version
    ]

booktitle :: PandocMonad m => Meta -> FBM m [Content]
booktitle meta' = do
  t <- cMapM toXml . docTitle $ meta'
  return $ if null t
           then []
           else [ el "book-title" t ]

authors :: Meta -> [Content]
authors meta' = cMap author (docAuthors meta')

author :: [Inline] -> [Content]
author ss =
  let ws = words . cMap plain $ ss
      email = el "email" <$> take 1 (filter ('@' `elem`) ws)
      ws' = filter ('@' `notElem`) ws
      names = case ws' of
                [nickname] -> [ el "nickname" nickname ]
                [fname, lname] -> [ el "first-name" fname
                                    , el "last-name" lname ]
                (fname:rest) -> [ el "first-name" fname
                                , el "middle-name" (concat . init $ rest)
                                , el "last-name" (last rest) ]
                [] -> []
  in  list $ el "author" (names ++ email)

docdate :: PandocMonad m => Meta -> FBM m [Content]
docdate meta' = do
  let ss = docDate meta'
  d <- cMapM toXml ss
  return $ if null d
           then []
           else [el "date" d]

-- | Divide the stream of blocks into sections and convert to XML
-- representation.
renderSections :: PandocMonad m => Int -> [Block] -> FBM m [Content]
renderSections level blocks = do
    let secs = splitSections level blocks
    mapM (renderSection level) secs

renderSection :: PandocMonad m => Int -> ([Inline], [Block]) -> FBM m Content
renderSection level (ttl, body) = do
    title <- if null ttl
            then return []
            else return . list . el "title" . formatTitle $ ttl
    content <- if hasSubsections body
               then renderSections (level + 1) body
               else cMapM blockToXml body
    return $ el "section" (title ++ content)
  where
    hasSubsections = any isHeaderBlock

-- | Only <p> and <empty-line> are allowed within <title> in FB2.
formatTitle :: [Inline] -> [Content]
formatTitle inlines =
  let lns = split isLineBreak inlines
      lns' = map (el "p" . cMap plain) lns
  in  intersperse (el "empty-line" ()) lns'

split :: (a -> Bool) -> [a] -> [[a]]
split _ [] = []
split cond xs = let (b,a) = break cond xs
                in  (b:split cond (drop 1 a))

isLineBreak :: Inline -> Bool
isLineBreak LineBreak = True
isLineBreak _         = False

-- | Divide the stream of block elements into sections: [(title, blocks)].
splitSections :: Int -> [Block] -> [([Inline], [Block])]
splitSections level blocks = reverse $ revSplit (reverse blocks)
  where
  revSplit [] = []
  revSplit rblocks =
    let (lastsec, before) = break sameLevel rblocks
        (header, prevblocks) =
            case before of
              (Header n _ title:prevblocks') ->
                  if n == level
                     then (title, prevblocks')
                     else ([], before)
              _ -> ([], before)
    in (header, reverse lastsec) : revSplit prevblocks
  sameLevel (Header n _ _) = n == level
  sameLevel _              = False

-- | Make another FictionBook body with footnotes.
renderFootnotes :: PandocMonad m => FBM m [Content]
renderFootnotes = do
  fns <- footnotes `liftM` get
  if null fns
    then return []  -- no footnotes
    else return . list $
         el "body" ([uattr "name" "notes"], map renderFN (reverse fns))
  where
    renderFN (n, idstr, cs) =
        let fn_texts = el "title" (el "p" (show n)) : cs
        in  el "section" ([uattr "id" idstr], fn_texts)

-- | Fetch images and encode them for the FictionBook XML.
-- Return image data and a list of hrefs of the missing images.
fetchImages :: PandocMonad m => [(String,String)] -> m ([Content],[String])
fetchImages links = do
    imgs <- mapM (uncurry fetchImage) links
    return (rights imgs, lefts imgs)

-- | Fetch image data from disk or from network and make a <binary> XML section.
-- Return either (Left hrefOfMissingImage) or (Right xmlContent).
fetchImage :: PandocMonad m => String -> String -> m (Either String Content)
fetchImage href link = do
  mbimg <-
      case (isURI link, readDataURI link) of
       (True, Just (mime,_,True,base64)) ->
           let mime' = map toLower mime
           in if mime' == "image/png" || mime' == "image/jpeg"
              then return (Just (mime',base64))
              else return Nothing
       (True, Just _) -> return Nothing  -- not base64-encoded
       _               ->
         catchError (do (bs, mbmime) <- P.fetchItem link
                        case mbmime of
                             Nothing -> do
                               report $ CouldNotDetermineMimeType link
                               return Nothing
                             Just mime -> return $ Just (mime,
                                                      B8.unpack $ encode bs))
                    (\e ->
                       do report $ CouldNotFetchResource link (show e)
                          return Nothing)
  case mbimg of
    Just (imgtype, imgdata) ->
        return . Right $ el "binary"
                   ( [uattr "id" href
                     , uattr "content-type" imgtype]
                   , txt imgdata )
    _ -> return (Left ('#':href))


-- | Extract mime type and encoded data from the Data URI.
readDataURI :: String -- ^ URI
            -> Maybe (String,String,Bool,String)
               -- ^ Maybe (mime,charset,isBase64,data)
readDataURI uri =
  case stripPrefix "data:" uri of
    Nothing   -> Nothing
    Just rest ->
      let meta = takeWhile (/= ',') rest  -- without trailing ','
          uridata = drop (length meta + 1) rest
          parts = split (== ';') meta
          (mime,cs,enc)=foldr upd ("text/plain","US-ASCII",False) parts
      in  Just (mime,cs,enc,uridata)

 where
   upd str m@(mime,cs,enc)
       | isMimeType str                          = (str,cs,enc)
       | Just str' <- stripPrefix "charset=" str = (mime,str',enc)
       | str ==  "base64"                        = (mime,cs,True)
       | otherwise                               = m

-- Without parameters like ;charset=...; see RFC 2045, 5.1
isMimeType :: String -> Bool
isMimeType s =
    case split (=='/') s of
      [mtype,msubtype] ->
          (map toLower mtype `elem` types
           || "x-" `isPrefixOf` map toLower mtype)
          && all valid mtype
          && all valid msubtype
      _ -> False
 where
   types =  ["text","image","audio","video","application","message","multipart"]
   valid c = isAscii c && not (isControl c) && not (isSpace c) &&
             c `notElem` "()<>@,;:\\\"/[]?="

footnoteID :: Int -> String
footnoteID i = "n" ++ show i

linkID :: Int -> String
linkID i = "l" ++ show i

-- | Convert a block-level Pandoc's element to FictionBook XML representation.
blockToXml :: PandocMonad m => Block -> FBM m [Content]
blockToXml (Plain ss) = cMapM toXml ss  -- FIXME: can lead to malformed FB2
blockToXml (Para [Math DisplayMath formula]) = insertMath NormalImage formula
-- title beginning with fig: indicates that the image is a figure
blockToXml (Para [Image atr alt (src,'f':'i':'g':':':tit)]) =
  insertImage NormalImage (Image atr alt (src,tit))
blockToXml (Para ss) = (list . el "p") <$> cMapM toXml ss
blockToXml (CodeBlock _ s) = return . spaceBeforeAfter .
                             map (el "p" . el "code") . lines $ s
blockToXml b@(RawBlock _ _) = do
  report $ BlockNotRendered b
  return []
blockToXml (Div _ bs) = cMapM blockToXml bs
blockToXml (BlockQuote bs) = (list . el "cite") <$> cMapM blockToXml bs
blockToXml (LineBlock lns) =
  (list . el "poem") <$> mapM stanza (split null lns)
  where
    v xs = el "v" <$> cMapM toXml xs
    stanza xs = el "stanza" <$> mapM v xs
blockToXml (OrderedList a bss) = do
    state <- get
    let pmrk = parentListMarker state
    let markers = map (pmrk ++) $ orderedListMarkers a
    let mkitem mrk bs = do
          modify (\s -> s { parentListMarker = mrk ++ " "})
          item <- cMapM blockToXml $ plainToPara $ indentBlocks (mrk ++ " ") bs
          modify (\s -> s { parentListMarker = pmrk }) -- old parent marker
          return item
    concat <$> zipWithM mkitem markers bss
blockToXml (BulletList bss) = do
    state <- get
    let level = parentBulletLevel state
    let pmrk = parentListMarker state
    let prefix = replicate (length pmrk) ' '
    let bullets = ["\x2022", "\x25e6", "*", "\x2043", "\x2023"]
    let mrk = prefix ++ bullets !! (level `mod` length bullets)
    let mkitem bs = do
          modify (\s -> s { parentBulletLevel = level+1 })
          item <- cMapM blockToXml $ plainToPara $ indentBlocks (mrk ++ " ") bs
          modify (\s -> s { parentBulletLevel = level }) -- restore bullet level
          return item
    cMapM mkitem bss
blockToXml (DefinitionList defs) =
    cMapM mkdef defs
    where
      mkdef (term, bss) = do
          def' <- cMapM (cMapM blockToXml . sep . paraToPlain . map indent) bss
          t <- wrap "strong" term
          return [ el "p" t, el "p" def' ]
      sep blocks =
          if all needsBreak blocks then
              blocks ++ [Plain [LineBreak]]
          else
              blocks
      needsBreak (Para _)    = False
      needsBreak (Plain ins) = LineBreak `notElem` ins
      needsBreak _           = True
blockToXml h@Header{} = do
  -- should not occur after hierarchicalize, except inside lists/blockquotes
  report $ BlockNotRendered h
  return []
blockToXml HorizontalRule = return
                            [ el "empty-line" ()
                            , el "p" (txt (replicate 10 '—'))
                            , el "empty-line" () ]
blockToXml (Table caption aligns _ headers rows) = do
    hd <- mkrow "th" headers aligns
    bd <- mapM (\r -> mkrow "td" r aligns) rows
    c <- el "emphasis" <$> cMapM toXml caption
    return [el "table" (hd : bd), el "p" c]
    where
      mkrow :: PandocMonad m => String -> [TableCell] -> [Alignment] -> FBM m Content
      mkrow tag cells aligns' =
        el "tr" <$> mapM (mkcell tag) (zip cells aligns')
      --
      mkcell :: PandocMonad m => String -> (TableCell, Alignment) -> FBM m Content
      mkcell tag (cell, align) = do
        cblocks <- cMapM blockToXml cell
        return $ el tag ([align_attr align], cblocks)
      --
      align_attr a = Attr (QName "align" Nothing Nothing) (align_str a)
      align_str AlignLeft    = "left"
      align_str AlignCenter  = "center"
      align_str AlignRight   = "right"
      align_str AlignDefault = "left"
blockToXml Null = return []

-- Replace paragraphs with plain text and line break.
-- Necessary to simulate multi-paragraph lists in FB2.
paraToPlain :: [Block] -> [Block]
paraToPlain [] = []
paraToPlain (Para inlines : rest) =
    Plain inlines : Plain [LineBreak] : paraToPlain rest
paraToPlain (p:rest) = p : paraToPlain rest

-- Replace plain text with paragraphs and add line break after paragraphs.
-- It is used to convert plain text from tight list items to paragraphs.
plainToPara :: [Block] -> [Block]
plainToPara [] = []
plainToPara (Plain inlines : rest) =
    Para inlines : plainToPara rest
plainToPara (Para inlines : rest) =
    Para inlines : Plain [LineBreak] : plainToPara rest
plainToPara (p:rest) = p : plainToPara rest

-- Simulate increased indentation level. Will not really work
-- for multi-line paragraphs.
indentPrefix :: String -> Block -> Block
indentPrefix spacer = indentBlock
  where
  indentBlock (Plain ins) = Plain (Str spacer:ins)
  indentBlock (Para ins) = Para (Str spacer:ins)
  indentBlock (CodeBlock a s) =
    let s' = unlines . map (spacer++) . lines $ s
    in  CodeBlock a s'
  indentBlock (BlockQuote bs) = BlockQuote (map indent bs)
  indentBlock (Header l attr' ins) = Header l attr' (indentLines ins)
  indentBlock everythingElse = everythingElse
  -- indent every (explicit) line
  indentLines :: [Inline] -> [Inline]
  indentLines ins = let lns = split isLineBreak ins :: [[Inline]]
                    in  intercalate [LineBreak] $ map (Str spacer:) lns

indent :: Block -> Block
indent = indentPrefix spacer
  where
  -- indentation space
  spacer :: String
  spacer = replicate 4 ' '

indentBlocks :: String -> [Block] -> [Block]
indentBlocks _ [] = []
indentBlocks prefix (x:xs) = indentPrefix prefix x : map (indentPrefix $ replicate (length prefix) ' ') xs

-- | Convert a Pandoc's Inline element to FictionBook XML representation.
toXml :: PandocMonad m => Inline -> FBM m [Content]
toXml (Str s) = return [txt s]
toXml (Span _ ils) = cMapM toXml ils
toXml (Emph ss) = list `liftM` wrap "emphasis" ss
toXml (Strong ss) = list `liftM` wrap "strong" ss
toXml (Strikeout ss) = list `liftM` wrap "strikethrough" ss
toXml (Superscript ss) = list `liftM` wrap "sup" ss
toXml (Subscript ss) = list `liftM` wrap "sub" ss
toXml (SmallCaps ss) = cMapM toXml $ capitalize ss
toXml (Quoted SingleQuote ss) = do  -- FIXME: should be language-specific
  inner <- cMapM toXml ss
  return $ [txt "‘"] ++ inner ++ [txt "’"]
toXml (Quoted DoubleQuote ss) = do
  inner <- cMapM toXml ss
  return $ [txt "“"] ++ inner ++ [txt "”"]
toXml (Cite _ ss) = cMapM toXml ss  -- FIXME: support citation styles
toXml (Code _ s) = return [el "code" s]
toXml Space = return [txt " "]
toXml SoftBreak = return [txt " "]
toXml LineBreak = return [el "empty-line" ()]
toXml (Math _ formula) = insertMath InlineImage formula
toXml il@(RawInline _ _) = do
  report $ InlineNotRendered il
  return []  -- raw TeX and raw HTML are suppressed
toXml (Link _ text (url,ttl)) = do
  fns <- footnotes `liftM` get
  let n = 1 + length fns
  let ln_id = linkID n
  let ln_ref = list . el "sup" . txt $ "[" ++ show n ++ "]"
  ln_text <- cMapM toXml text
  let ln_desc =
          let ttl' = dropWhile isSpace ttl
          in if null ttl'
             then list . el "p" $ el "code" url
             else list . el "p" $ [ txt (ttl' ++ ": "), el "code" url ]
  modify (\s -> s { footnotes = (n, ln_id, ln_desc) : fns })
  return $ ln_text ++
         [ el "a"
                  ( [ attr ("l","href") ('#':ln_id)
                    , uattr "type" "note" ]
                  , ln_ref) ]
toXml img@Image{} = insertImage InlineImage img
toXml (Note bs) = do
  fns <- footnotes `liftM` get
  let n = 1 + length fns
  let fn_id = footnoteID n
  fn_desc <- cMapM blockToXml bs
  modify (\s -> s { footnotes = (n, fn_id, fn_desc) : fns })
  let fn_ref = el "sup" . txt $ "[" ++ show n ++ "]"
  return . list $ el "a" ( [ attr ("l","href") ('#':fn_id)
                           , uattr "type" "note" ]
                         , fn_ref )

insertMath :: PandocMonad m => ImageMode -> String -> FBM m [Content]
insertMath immode formula = do
  htmlMath <- fmap (writerHTMLMathMethod . writerOptions) get
  case htmlMath of
    WebTeX url -> do
       let alt = [Code nullAttr formula]
       let imgurl = url ++ urlEncode formula
       let img = Image nullAttr alt (imgurl, "")
       insertImage immode img
    _ -> return [el "code" formula]

insertImage :: PandocMonad m => ImageMode -> Inline -> FBM m [Content]
insertImage immode (Image _ alt (url,ttl)) = do
  images <- imagesToFetch `liftM` get
  let n = 1 + length images
  let fname = "image" ++ show n
  modify (\s -> s { imagesToFetch = (fname, url) : images })
  let ttlattr = case (immode, null ttl) of
                  (NormalImage, False) -> [ uattr "title" ttl ]
                  _                    -> []
  return . list $
         el "image" $
            [ attr ("l","href") ('#':fname)
            , attr ("l","type") (show immode)
            , uattr "alt" (cMap plain alt) ]
            ++ ttlattr
insertImage _ _ = error "unexpected inline instead of image"

replaceImagesWithAlt :: [String] -> Content -> Content
replaceImagesWithAlt missingHrefs body =
  let cur = XC.fromContent body
      cur' = replaceAll cur
  in  XC.toTree . XC.root $ cur'
  where
  --
    replaceAll :: XC.Cursor -> XC.Cursor
    replaceAll c =
        let n = XC.current c
            c' = if isImage n && isMissing n
                 then XC.modifyContent replaceNode c
                 else c
        in  case XC.nextDF c' of
              (Just cnext) -> replaceAll cnext
              Nothing      -> c'  -- end of document
  --
    isImage :: Content -> Bool
    isImage (Elem e) = elName e == uname "image"
    isImage _        = False
  --
    isMissing (Elem img@Element{}) =
        let imgAttrs = elAttribs img
            badAttrs = map (attr ("l","href")) missingHrefs
        in  any (`elem` imgAttrs) badAttrs
    isMissing _ = False
  --
    replaceNode :: Content -> Content
    replaceNode n@(Elem img@Element{}) =
        let attrs = elAttribs img
            alt = getAttrVal attrs (uname "alt")
            imtype = getAttrVal attrs (qname "l" "type")
        in case (alt, imtype) of
             (Just alt', Just imtype') ->
                 if imtype' == show NormalImage
                 then el "p" alt'
                 else txt alt'
             (Just alt', Nothing) -> txt alt'  -- no type attribute
             _ -> n   -- don't replace if alt text is not found
    replaceNode n = n
  --
    getAttrVal :: [X.Attr] -> QName -> Maybe String
    getAttrVal attrs name =
        case filter ((name ==) . attrKey) attrs of
           (a:_) -> Just (attrVal a)
           _     -> Nothing


-- | Wrap all inlines with an XML tag (given its unqualified name).
wrap :: PandocMonad m => String -> [Inline] -> FBM m Content
wrap tagname inlines = el tagname `liftM` cMapM toXml inlines

-- " Create a singleton list.
list :: a -> [a]
list = (:[])

-- | Convert an 'Inline' to plaintext.
plain :: Inline -> String
plain (Str s)               = s
plain (Emph ss)             = cMap plain ss
plain (Span _ ss)           = cMap plain ss
plain (Strong ss)           = cMap plain ss
plain (Strikeout ss)        = cMap plain ss
plain (Superscript ss)      = cMap plain ss
plain (Subscript ss)        = cMap plain ss
plain (SmallCaps ss)        = cMap plain ss
plain (Quoted _ ss)         = cMap plain ss
plain (Cite _ ss)           = cMap plain ss  -- FIXME
plain (Code _ s)            = s
plain Space                 = " "
plain SoftBreak             = " "
plain LineBreak             = "\n"
plain (Math _ s)            = s
plain (RawInline _ _)       = ""
plain (Link _ text (url,_)) = concat (map plain text ++ [" <", url, ">"])
plain (Image _ alt _)       = cMap plain alt
plain (Note _)              = ""  -- FIXME

-- | Create an XML element.
el :: (Node t)
   => String   -- ^ unqualified element name
   -> t        -- ^ node contents
   -> Content  -- ^ XML content
el name cs = Elem $ unode name cs

-- | Put empty lines around content
spaceBeforeAfter :: [Content] -> [Content]
spaceBeforeAfter cs =
    let emptyline = el "empty-line" ()
    in  [emptyline] ++ cs ++ [emptyline]

-- | Create a plain-text XML content.
txt :: String -> Content
txt s = Text $ CData CDataText s Nothing

-- | Create an XML attribute with an unqualified name.
uattr :: String -> String -> Text.XML.Light.Attr
uattr name = Attr (uname name)

-- | Create an XML attribute with a qualified name from given namespace.
attr :: (String, String) -> String -> Text.XML.Light.Attr
attr (ns, name) = Attr (qname ns name)

-- | Unqualified name
uname :: String -> QName
uname name = QName name Nothing Nothing

-- | Qualified name
qname :: String -> String -> QName
qname ns name = QName name Nothing (Just ns)

-- | Abbreviation for 'concatMap'.
cMap :: (a -> [b]) -> [a] -> [b]
cMap = concatMap

-- | Monadic equivalent of 'concatMap'.
cMapM :: (Monad m) => (a -> m [b]) -> [a] -> m [b]
cMapM f xs = concat `liftM` mapM f xs