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
|
{-
Copyright (C) 2007-2015 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.Writers.Ms
Copyright : Copyright (C) 2007-2015 John MacFarlane
License : GNU GPL, version 2 or above
Maintainer : John MacFarlane <jgm@berkeley.edu>
Stability : alpha
Portability : portable
Conversion of 'Pandoc' documents to groff man page format.
TODO:
[ ] is there a better way to do strikeout?
[ ] options for hyperlink rendering (currently footnote)
[ ] can we get prettier output using .B, etc. instead of
the inline forms?
[ ] tight/loose list distinction
[ ] internal hyperlinks (this seems to be possible since
they exist in the groff manual PDF version)
[ ] better template, with configurable page number, table of contents,
columns, etc.
[ ] support for images? gropdf (and maybe pdfroff) supports the tag
\X'pdf: pdfpic file alignment width height line-length'
and also seems to support bookmarks.
-}
module Text.Pandoc.Writers.Ms ( writeMs ) where
import Text.Pandoc.Definition
import Text.Pandoc.Templates
import Text.Pandoc.Shared
import Text.Pandoc.Writers.Shared
import Text.Pandoc.Options
import Text.Pandoc.Writers.Math
import Text.Printf ( printf )
import qualified Data.Map as Map
import Data.List ( stripPrefix, intersperse, intercalate, sort )
import Data.Maybe (fromMaybe)
import Text.Pandoc.Pretty
import Text.Pandoc.Builder (deleteMeta)
import Text.Pandoc.Class (PandocMonad, report)
import Text.Pandoc.Logging
import Control.Monad.State
import Data.Char ( isDigit, isLower, isUpper, toUpper )
import Text.TeXMath (writeEqn)
data WriterState = WriterState { stHasInlineMath :: Bool
, stNotes :: [Note]
, stInNote :: Bool
, stSmallCaps :: Bool
, stFontFeatures :: Map.Map Char Bool
}
defaultWriterState :: WriterState
defaultWriterState = WriterState{ stHasInlineMath = False
, stNotes = []
, stInNote = False
, stSmallCaps = False
, stFontFeatures = Map.fromList [
('I',False)
, ('B',False)
, ('C',False)
]
}
type Note = [Block]
type MS = StateT WriterState
-- | Convert Pandoc to Ms.
writeMs :: PandocMonad m => WriterOptions -> Pandoc -> m String
writeMs opts document =
evalStateT (pandocToMs opts document) defaultWriterState
-- | Return groff man representation of document.
pandocToMs :: PandocMonad m => WriterOptions -> Pandoc -> MS m String
pandocToMs opts (Pandoc meta blocks) = do
let colwidth = if writerWrapText opts == WrapAuto
then Just $ writerColumns opts
else Nothing
let render' = render colwidth
titleText <- inlineListToMs' opts $ docTitle meta
let title' = render' titleText
let setFieldsFromTitle =
case break (== ' ') title' of
(cmdName, rest) -> case reverse cmdName of
(')':d:'(':xs) | isDigit d ->
defField "title" (reverse xs) .
defField "section" [d] .
case splitBy (=='|') rest of
(ft:hds) ->
defField "footer" (trim ft) .
defField "header"
(trim $ concat hds)
[] -> id
_ -> defField "title" title'
metadata <- metaToJSON opts
(fmap (render colwidth) . blockListToMs opts)
(fmap (render colwidth) . inlineListToMs' opts)
$ deleteMeta "title" meta
body <- blockListToMs opts blocks
let main = render' body
hasInlineMath <- gets stHasInlineMath
let context = defField "body" main
$ setFieldsFromTitle
$ defField "has-inline-math" hasInlineMath
$ defField "hyphenate" True
$ defField "pandoc-version" pandocVersion
$ metadata
case writerTemplate opts of
Nothing -> return main
Just tpl -> return $ renderTemplate' tpl context
-- | Association list of characters to escape.
manEscapes :: Map.Map Char String
manEscapes = Map.fromList $
[ ('\160', "\\ ")
, ('\'', "\\[aq]")
, ('’', "'")
, ('"', "\\\"")
, ('\x2014', "\\[em]")
, ('\x2013', "\\[en]")
, ('\x2026', "\\&...")
, ('|', "\\[u007C]") -- because we use | for inline math
, ('-', "\\-")
, ('@', "\\@")
, ('\\', "\\\\")
]
escapeChar :: Char -> String
escapeChar c = case Map.lookup c manEscapes of
Just s -> s
Nothing -> [c]
-- | Escape | character, used to mark inline math, inside math.
escapeBar :: String -> String
escapeBar = concatMap go
where go '|' = "\\[u007C]"
go c = [c]
-- | Escape special characters for Ms.
escapeString :: String -> String
escapeString = concatMap escapeChar
toSmallCaps :: String -> String
toSmallCaps [] = []
toSmallCaps (c:cs)
| isLower c = let (lowers,rest) = span isLower (c:cs)
in "\\s-2" ++ escapeString (map toUpper lowers) ++
"\\s0" ++ toSmallCaps rest
| isUpper c = let (uppers,rest) = span isUpper (c:cs)
in escapeString uppers ++ toSmallCaps rest
| otherwise = escapeChar c ++ toSmallCaps cs
-- | Escape a literal (code) section for Ms.
escapeCode :: String -> String
escapeCode = concat . intersperse "\n" . map escapeLine . lines
where escapeCodeChar ' ' = "\\ "
escapeCodeChar '\t' = "\\\t"
escapeCodeChar c = escapeChar c
escapeLine codeline =
case concatMap escapeCodeChar codeline of
a@('.':_) -> "\\&" ++ a
b -> b
-- We split inline lists into sentences, and print one sentence per
-- line. groff/troff treats the line-ending period differently.
-- See http://code.google.com/p/pandoc/issues/detail?id=148.
-- | Returns the first sentence in a list of inlines, and the rest.
breakSentence :: [Inline] -> ([Inline], [Inline])
breakSentence [] = ([],[])
breakSentence xs =
let isSentenceEndInline (Str ys@(_:_)) | last ys == '.' = True
isSentenceEndInline (Str ys@(_:_)) | last ys == '?' = True
isSentenceEndInline (LineBreak) = True
isSentenceEndInline _ = False
(as, bs) = break isSentenceEndInline xs
in case bs of
[] -> (as, [])
[c] -> (as ++ [c], [])
(c:Space:cs) -> (as ++ [c], cs)
(c:SoftBreak:cs) -> (as ++ [c], cs)
(Str ".":Str (')':ys):cs) -> (as ++ [Str ".", Str (')':ys)], cs)
(x@(Str ('.':')':_)):cs) -> (as ++ [x], cs)
(LineBreak:x@(Str ('.':_)):cs) -> (as ++[LineBreak], x:cs)
(c:cs) -> (as ++ [c] ++ ds, es)
where (ds, es) = breakSentence cs
-- | Split a list of inlines into sentences.
splitSentences :: [Inline] -> [[Inline]]
splitSentences xs =
let (sent, rest) = breakSentence xs
in if null rest then [sent] else sent : splitSentences rest
blockToMs :: PandocMonad m
=> WriterOptions -- ^ Options
-> Block -- ^ Block element
-> MS m Doc
blockToMs _ Null = return empty
blockToMs opts (Div _ bs) = blockListToMs opts bs
blockToMs opts (Plain inlines) =
liftM vcat $ mapM (inlineListToMs' opts) $ splitSentences inlines
blockToMs opts (Para inlines) = do
contents <- liftM vcat $ mapM (inlineListToMs' opts) $
splitSentences inlines
return $ text ".LP" $$ contents
blockToMs _ b@(RawBlock f str)
| f == Format "man" = return $ text str
| otherwise = do
report $ BlockNotRendered b
return empty
blockToMs _ HorizontalRule = return $ text ".PP" $$ text " * * * * *"
blockToMs opts (Header level _ inlines) = do
contents <- inlineListToMs' opts inlines
let heading = if writerNumberSections opts
then ".NH"
else ".SH"
return $ text heading <> space <> text (show level) $$ contents
blockToMs _ (CodeBlock _ str) = return $
text ".IP" $$
text ".nf" $$
text "\\f[C]" $$
text (escapeCode str) $$
text "\\f[]" $$
text ".fi"
blockToMs opts (LineBlock ls) = do
blockToMs opts $ Para $ intercalate [LineBreak] ls
blockToMs opts (BlockQuote blocks) = do
contents <- blockListToMs opts blocks
return $ text ".RS" $$ contents $$ text ".RE"
blockToMs opts (Table caption alignments widths headers rows) =
let aligncode AlignLeft = "l"
aligncode AlignRight = "r"
aligncode AlignCenter = "c"
aligncode AlignDefault = "l"
in do
caption' <- inlineListToMs' opts caption
let iwidths = if all (== 0) widths
then repeat ""
else map (printf "w(%0.1fn)" . (70 *)) widths
-- 78n default width - 8n indent = 70n
let coldescriptions = text $ intercalate " "
(zipWith (\align width -> aligncode align ++ width)
alignments iwidths) ++ "."
colheadings <- mapM (blockListToMs opts) headers
let makeRow cols = text "T{" $$
(vcat $ intersperse (text "T}@T{") cols) $$
text "T}"
let colheadings' = if all null headers
then empty
else makeRow colheadings $$ char '_'
body <- mapM (\row -> do
cols <- mapM (blockListToMs opts) row
return $ makeRow cols) rows
return $ text ".PP" $$ caption' $$
text ".TS" $$ text "tab(@);" $$ coldescriptions $$
colheadings' $$ vcat body $$ text ".TE"
blockToMs opts (BulletList items) = do
contents <- mapM (bulletListItemToMs opts) items
return (vcat contents)
blockToMs opts (OrderedList attribs items) = do
let markers = take (length items) $ orderedListMarkers attribs
let indent = 1 + (maximum $ map length markers)
contents <- mapM (\(num, item) -> orderedListItemToMs opts num indent item) $
zip markers items
return (vcat contents)
blockToMs opts (DefinitionList items) = do
contents <- mapM (definitionListItemToMs opts) items
return (vcat contents)
-- | Convert bullet list item (list of blocks) to man.
bulletListItemToMs :: PandocMonad m => WriterOptions -> [Block] -> MS m Doc
bulletListItemToMs _ [] = return empty
bulletListItemToMs opts ((Para first):rest) =
bulletListItemToMs opts ((Plain first):rest)
bulletListItemToMs opts ((Plain first):rest) = do
first' <- blockToMs opts (Plain first)
rest' <- blockListToMs opts rest
let first'' = text ".IP \\[bu] 2" $$ first'
let rest'' = if null rest
then empty
else text ".RS 2" $$ rest' $$ text ".RE"
return (first'' $$ rest'')
bulletListItemToMs opts (first:rest) = do
first' <- blockToMs opts first
rest' <- blockListToMs opts rest
return $ text "\\[bu] .RS 2" $$ first' $$ rest' $$ text ".RE"
-- | Convert ordered list item (a list of blocks) to man.
orderedListItemToMs :: PandocMonad m
=> WriterOptions -- ^ options
-> String -- ^ order marker for list item
-> Int -- ^ number of spaces to indent
-> [Block] -- ^ list item (list of blocks)
-> MS m Doc
orderedListItemToMs _ _ _ [] = return empty
orderedListItemToMs opts num indent ((Para first):rest) =
orderedListItemToMs opts num indent ((Plain first):rest)
orderedListItemToMs opts num indent (first:rest) = do
first' <- blockToMs opts first
rest' <- blockListToMs opts rest
let num' = printf ("%" ++ show (indent - 1) ++ "s") num
let first'' = text (".IP \"" ++ num' ++ "\" " ++ show indent) $$ first'
let rest'' = if null rest
then empty
else text ".RS 4" $$ rest' $$ text ".RE"
return $ first'' $$ rest''
-- | Convert definition list item (label, list of blocks) to man.
definitionListItemToMs :: PandocMonad m
=> WriterOptions
-> ([Inline],[[Block]])
-> MS m Doc
definitionListItemToMs opts (label, defs) = do
labelText <- inlineListToMs' opts label
contents <- if null defs
then return empty
else liftM vcat $ forM defs $ \blocks -> do
let (first, rest) = case blocks of
((Para x):y) -> (Plain x,y)
(x:y) -> (x,y)
[] -> error "blocks is null"
rest' <- liftM vcat $
mapM (\item -> blockToMs opts item) rest
first' <- blockToMs opts first
return $ first' $$ text ".RS" $$ rest' $$ text ".RE"
return $ nowrap (text ".IP \"" <> labelText <> text "\"") $$ contents
-- | Convert list of Pandoc block elements to man.
blockListToMs :: PandocMonad m
=> WriterOptions -- ^ Options
-> [Block] -- ^ List of block elements
-> MS m Doc
blockListToMs opts blocks =
mapM (blockToMs opts) blocks >>= (return . vcat)
-- | Convert list of Pandoc inline elements to ms.
inlineListToMs :: PandocMonad m => WriterOptions -> [Inline] -> MS m Doc
-- if list starts with ., insert a zero-width character \& so it
-- won't be interpreted as markup if it falls at the beginning of a line.
inlineListToMs opts lst@(Str ('.':_) : _) = mapM (inlineToMs opts) lst >>=
(return . (text "\\&" <>) . hcat)
inlineListToMs opts lst = hcat <$> mapM (inlineToMs opts) lst
-- This version to be used when there is no further inline content;
-- forces a note at the end.
inlineListToMs' :: PandocMonad m => WriterOptions -> [Inline] -> MS m Doc
inlineListToMs' opts lst = do
x <- hcat <$> mapM (inlineToMs opts) lst
y <- handleNotes opts empty
return $ x <> y
-- | Convert Pandoc inline element to man.
inlineToMs :: PandocMonad m => WriterOptions -> Inline -> MS m Doc
inlineToMs opts (Span _ ils) = inlineListToMs opts ils
inlineToMs opts (Emph lst) =
withFontFeature 'I' (inlineListToMs opts lst)
inlineToMs opts (Strong lst) =
withFontFeature 'B' (inlineListToMs opts lst)
inlineToMs opts (Strikeout lst) = do
contents <- inlineListToMs opts lst
return $ text "[STRIKEOUT:" <> contents <> char ']'
inlineToMs opts (Superscript lst) = do
contents <- inlineListToMs opts lst
return $ text "\\*{" <> contents <> text "\\*}"
inlineToMs opts (Subscript lst) = do
contents <- inlineListToMs opts lst
return $ text "\\*<" <> contents <> text "\\*>"
inlineToMs opts (SmallCaps lst) = do
-- see https://lists.gnu.org/archive/html/groff/2015-01/msg00016.html
modify $ \st -> st{ stSmallCaps = not (stSmallCaps st) }
res <- inlineListToMs opts lst
modify $ \st -> st{ stSmallCaps = not (stSmallCaps st) }
return res
inlineToMs opts (Quoted SingleQuote lst) = do
contents <- inlineListToMs opts lst
return $ char '`' <> contents <> char '\''
inlineToMs opts (Quoted DoubleQuote lst) = do
contents <- inlineListToMs opts lst
return $ text "\\[lq]" <> contents <> text "\\[rq]"
inlineToMs opts (Cite _ lst) =
inlineListToMs opts lst
inlineToMs _ (Code _ str) =
withFontFeature 'C' (return $ text $ escapeCode str)
inlineToMs _ (Str str) = do
smallcaps <- gets stSmallCaps
if smallcaps
then return $ text $ toSmallCaps str
else return $ text $ escapeString str
inlineToMs opts (Math InlineMath str) = do
modify $ \st -> st{ stHasInlineMath = True }
res <- convertMath writeEqn InlineMath str
case res of
Left il -> inlineToMs opts il
Right r -> return $ text "|" <> text (escapeBar r) <> text "|"
inlineToMs opts (Math DisplayMath str) = do
res <- convertMath writeEqn InlineMath str
case res of
Left il -> do
contents <- inlineToMs opts il
return $ cr <> text ".RS" $$ contents $$ text ".RE"
Right r -> return $
cr <> text ".EQ" $$ text (escapeBar r) $$ text ".EN"
inlineToMs _ il@(RawInline f str)
| f == Format "man" = return $ text str
| otherwise = do
report $ InlineNotRendered il
return empty
inlineToMs _ (LineBreak) = return $ cr <> text ".br" <> cr
inlineToMs opts SoftBreak = handleNotes opts cr
inlineToMs opts Space = handleNotes opts space
inlineToMs opts (Link _ txt (src, _)) = do
let srcSuffix = fromMaybe src (stripPrefix "mailto:" src)
inNote <- gets stInNote
case txt of
[Str s]
| escapeURI s == srcSuffix ->
return $ text (escapeString srcSuffix)
_ | inNote -> do
-- avoid a note in a note!
contents <- inlineListToMs opts txt
return $ contents <> space <> char '(' <>
text (escapeString src) <> char ')'
| otherwise -> do
let linknote = [Plain [Str src]]
inlineListToMs opts (txt ++ [Note linknote])
inlineToMs opts (Image attr alternate (source, tit)) = do
let txt = if (null alternate) || (alternate == [Str ""]) ||
(alternate == [Str source]) -- to prevent autolinks
then [Str "image"]
else alternate
linkPart <- inlineToMs opts (Link attr txt (source, tit))
return $ char '[' <> text "IMAGE: " <> linkPart <> char ']'
inlineToMs _ (Note contents) = do
modify $ \st -> st{ stNotes = contents : stNotes st }
return $ text "\\**"
handleNotes :: PandocMonad m => WriterOptions -> Doc -> MS m Doc
handleNotes opts fallback = do
notes <- gets stNotes
if null notes
then return fallback
else do
modify $ \st -> st{ stNotes = [], stInNote = True }
res <- vcat <$> mapM (handleNote opts) notes
modify $ \st -> st{ stInNote = False }
return res
handleNote :: PandocMonad m => WriterOptions -> Note -> MS m Doc
handleNote opts bs = do
-- don't start with Paragraph or we'll get a spurious blank
-- line after the note ref:
let bs' = case bs of
(Para ils : rest) -> Plain ils : rest
_ -> bs
contents <- blockListToMs opts bs'
return $ cr <> text ".FS" $$ contents $$ text ".FE" <> cr
fontChange :: PandocMonad m => MS m Doc
fontChange = do
features <- gets stFontFeatures
let filling = sort [c | (c,True) <- Map.toList features]
return $ text $ "\\f[" ++ filling ++ "]"
withFontFeature :: PandocMonad m => Char -> MS m Doc -> MS m Doc
withFontFeature c action = do
modify $ \st -> st{ stFontFeatures = Map.adjust not c $ stFontFeatures st }
begin <- fontChange
d <- action
modify $ \st -> st{ stFontFeatures = Map.adjust not c $ stFontFeatures st }
end <- fontChange
return $ begin <> d <> end
|