aboutsummaryrefslogtreecommitdiff
path: root/src/Text/Pandoc/Readers/Org/Meta.hs
blob: ccb6744e71d13606c8fbead96e016096d2ae025f (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
{-# LANGUAGE FlexibleContexts  #-}
{-# LANGUAGE LambdaCase        #-}
{-# LANGUAGE OverloadedStrings #-}
{- |
   Module      : Text.Pandoc.Readers.Org.Meta
   Copyright   : Copyright (C) 2014-2021 Albert Krewinkel
   License     : GNU GPL, version 2 or above

   Maintainer  : Albert Krewinkel <tarleb+pandoc@moltkeplatz.de>

Parsers for Org-mode meta declarations.
-}
module Text.Pandoc.Readers.Org.Meta
  ( metaExport
  , metaKey
  , metaLine
  ) where

import Text.Pandoc.Readers.Org.BlockStarts
import Text.Pandoc.Readers.Org.ExportSettings (exportSettings)
import Text.Pandoc.Readers.Org.Inlines
import Text.Pandoc.Readers.Org.ParserState
import Text.Pandoc.Readers.Org.Parsing

import Text.Pandoc.Builder (Blocks, Inlines)
import qualified Text.Pandoc.Builder as B
import Text.Pandoc.Class.PandocMonad (PandocMonad)
import Text.Pandoc.Definition
import Text.Pandoc.Shared (blocksToInlines, safeRead)
import Text.Pandoc.Network.HTTP (urlEncode)

import Control.Monad (mzero, void)
import Data.List (intercalate, intersperse)
import Data.Map (Map)
import Data.Maybe (fromMaybe)
import Data.Text (Text)
import qualified Data.Map as Map
import qualified Data.Set as Set
import qualified Data.Text as T

-- | Returns the current meta, respecting export options.
metaExport :: Monad m => OrgParser m (F Meta)
metaExport = do
  st <- getState
  let settings = orgStateExportSettings st
  return $ (if exportWithAuthor  settings then id else removeMeta "author")
         . (if exportWithCreator settings then id else removeMeta "creator")
         . (if exportWithEmail   settings then id else removeMeta "email")
        <$> orgStateMeta st

removeMeta :: Text -> Meta -> Meta
removeMeta key meta' =
  let metaMap = unMeta meta'
  in Meta $ Map.delete key metaMap

-- | Parse and handle a single line containing meta information
-- The order, in which blocks are tried, makes sure that we're not looking at
-- the beginning of a block, so we don't need to check for it
metaLine :: PandocMonad m => OrgParser m Blocks
metaLine = try $ mempty <$ metaLineStart <* keywordLine

keywordLine :: PandocMonad m => OrgParser m ()
keywordLine = try $ do
  key   <- T.toLower <$> metaKey
  case Map.lookup key keywordHandlers of
    Nothing -> fail $ "Unknown keyword: " ++ T.unpack key
    Just hd -> hd

metaKey :: Monad m => OrgParser m Text
metaKey = T.toLower <$> many1Char (noneOf ": \n\r")
                    <*  char ':'
                    <*  skipSpaces

infix 0 ~~>
(~~>) :: a -> b -> (a, b)
a ~~> b = (a, b)

keywordHandlers :: PandocMonad m => Map Text (OrgParser m ())
keywordHandlers = Map.fromList
  [ "author" ~~> lineOfInlines `parseThen` collectLines "author"
  , "creator" ~~> fmap pure anyLine `parseThen` B.setMeta "creator"
  , "date" ~~> lineOfInlines `parseThen` B.setMeta "date"
  , "description" ~~> lineOfInlines `parseThen` collectLines "description"
  , "email" ~~> fmap pure anyLine `parseThen` B.setMeta "email"
  , "exclude_tags" ~~> tagList >>= updateState . setExcludedTags
  , "header-includes" ~~>
    lineOfInlines `parseThen` collectLines "header-includes"
  -- HTML-specifix export settings
  , "html_head" ~~>
    metaExportSnippet "html" `parseThen` collectAsList "header-includes"
  , "html_head_extra" ~~>
    metaExportSnippet "html" `parseThen` collectAsList "header-includes"
  , "institute" ~~> lineOfInlines `parseThen` collectLines "institute"
  -- topic keywords
  , "keywords" ~~> lineOfInlines `parseThen` collectLines "keywords"
  -- document language
  , "language" ~~> fmap pure anyLine `parseThen` B.setMeta "lang"
  -- LaTeX-specific export settings
  , "latex_class" ~~> fmap pure anyLine `parseThen` B.setMeta "documentclass"
  , "latex_class_options" ~~>
      (pure . T.filter (`notElem` ("[]" :: String)) <$> anyLine)
      `parseThen` B.setMeta "classoption"
  , "latex_header" ~~>
      metaExportSnippet "latex" `parseThen` collectAsList "header-includes"
  , "latex_header_extra" ~~>
      metaExportSnippet "latex" `parseThen` collectAsList "header-includes"
  -- link and macro
  , "link" ~~> addLinkFormatter
  , "macro" ~~> macroDefinition >>= updateState . registerMacro
  -- pandoc-specific way to include references in the bibliography
  , "nocite" ~~> lineOfInlines `parseThen` collectLines "nocite"
  -- compact way to set export settings
  , "options"  ~~> exportSettings
  -- pandoc-specific way to configure emphasis recognition
  , "pandoc-emphasis-post" ~~> emphChars >>= updateState . setEmphasisPostChar
  , "pandoc-emphasis-pre" ~~> emphChars >>= updateState . setEmphasisPreChar
  -- result markers (ignored)
  , "result" ~~> void anyLine
  , "select_tags" ~~> tagList >>= updateState . setSelectedTags
  , "seq_todo" ~~> todoSequence >>= updateState . registerTodoSequence
  , "subtitle" ~~> lineOfInlines `parseThen` collectLines "subtitle"
  , "title" ~~> lineOfInlines `parseThen` collectLines "title"
  , "todo" ~~> todoSequence >>= updateState . registerTodoSequence
  , "typ_todo" ~~> todoSequence >>= updateState . registerTodoSequence
  ]

parseThen :: PandocMonad m
          => OrgParser m (F a)
          -> (a -> Meta -> Meta)
          -> OrgParser m ()
parseThen p modMeta = do
  value <- p
  meta  <- orgStateMeta <$> getState
  updateState (\st -> st { orgStateMeta = modMeta <$> value <*> meta })

collectLines :: Text -> Inlines -> Meta -> Meta
collectLines key value meta =
  let value' = appendValue meta (B.toList value)
  in B.setMeta key value' meta
 where
  appendValue :: Meta -> [Inline] -> MetaValue
  appendValue m v = MetaInlines $ curInlines m <> v

  curInlines m = case collectInlines <$> lookupMeta key m of
    Nothing -> []
    Just [] -> []
    Just xs -> xs <> [B.SoftBreak]

  collectInlines :: MetaValue -> [Inline]
  collectInlines = \case
    MetaInlines inlns -> inlns
    MetaList ml       -> intercalate [B.SoftBreak] $ map collectInlines ml
    MetaString s      -> [B.Str s]
    MetaBlocks blks   -> blocksToInlines blks
    MetaMap _map      -> []
    MetaBool _bool    -> []

-- | Accumulate the result as a MetaList under the given key.
collectAsList :: Text -> Inlines -> Meta -> Meta
collectAsList key value meta =
  let value' = metaListAppend meta (B.toMetaValue value)
  in B.setMeta key value' meta
 where
  metaListAppend m v = MetaList (curList m ++ [v])
  curList m = case lookupMeta key m of
                Just (MetaList ms) -> ms
                Just x             -> [x]
                _                  -> []

-- | Read an format specific meta definition
metaExportSnippet :: Monad m => Text -> OrgParser m (F Inlines)
metaExportSnippet format = pure . B.rawInline format <$> anyLine

-- | Parse a link type definition (like @wp https://en.wikipedia.org/wiki/@).
addLinkFormatter :: Monad m => OrgParser m ()
addLinkFormatter = try $ do
  linkType <- T.cons <$> letter <*> manyChar (alphaNum <|> oneOf "-_") <* skipSpaces
  formatter <- parseFormat
  updateState $ \s ->
    let fs = orgStateLinkFormatters s
    in s{ orgStateLinkFormatters = Map.insert linkType formatter fs }

-- | An ad-hoc, single-argument-only implementation of a printf-style format
-- parser.
parseFormat :: Monad m => OrgParser m (Text -> Text)
parseFormat = try $ replacePlain <|> replaceUrl <|> justAppend
 where
   -- inefficient
   replacePlain = try $ (\x -> T.concat . flip intersperse x)
                     <$> sequence [tillSpecifier 's', rest]
   replaceUrl   = try $ (\x -> T.concat . flip intersperse x . urlEncode)
                     <$> sequence [tillSpecifier 'h', rest]
   justAppend   = try $ (<>) <$> rest

   rest            = manyTillChar anyChar         (eof <|> () <$ oneOf "\n\r")
   tillSpecifier c = manyTillChar (noneOf "\n\r") (try $ string ('%':c:""))

tagList :: Monad m => OrgParser m [Tag]
tagList = do
  skipSpaces
  map Tag <$> many (orgTagWord <* skipSpaces) <* newline

setExcludedTags :: [Tag] -> OrgParserState -> OrgParserState
setExcludedTags tags st =
  let finalSet = if orgStateExcludeTagsChanged st
                   then foldr Set.insert (orgStateExcludeTags st) tags
                   else Set.fromList tags
  in st { orgStateExcludeTags = finalSet, orgStateExcludeTagsChanged = True }

setSelectedTags :: [Tag] -> OrgParserState -> OrgParserState
setSelectedTags tags st =
  let finalSet = if orgStateSelectTagsChanged st
                   then foldr Set.insert (orgStateSelectTags st) tags
                   else Set.fromList tags
  in st { orgStateSelectTags = finalSet, orgStateSelectTagsChanged = True }

setEmphasisPreChar :: Maybe [Char] -> OrgParserState -> OrgParserState
setEmphasisPreChar csMb st =
  let preChars = fromMaybe (orgStateEmphasisPostChars defaultOrgParserState) csMb
  in st { orgStateEmphasisPreChars = preChars }

setEmphasisPostChar :: Maybe [Char] -> OrgParserState -> OrgParserState
setEmphasisPostChar csMb st =
  let postChars = fromMaybe (orgStateEmphasisPostChars defaultOrgParserState) csMb
  in st { orgStateEmphasisPostChars = postChars }

-- | Parses emphasis border character like @".,?!"@
emphChars :: Monad m => OrgParser m (Maybe [Char])
emphChars = do
  skipSpaces
  safeRead <$> anyLine

lineOfInlines :: PandocMonad m => OrgParser m (F Inlines)
lineOfInlines = do
  updateLastPreCharPos
  trimInlinesF . mconcat <$> manyTill inline newline

-- | Parses ToDo sequences / keywords like @TODO DOING | DONE@.
todoSequence :: Monad m => OrgParser m TodoSequence
todoSequence = try $ do
  todoKws <- todoKeywords
  doneKws <- optionMaybe $ todoDoneSep *> doneKeywords
  newline
  -- There must be at least one DONE keyword. The last TODO keyword is
  -- taken if necessary.
  case doneKws of
    Just done  -> return $ keywordsToSequence todoKws done
    Nothing    -> case reverse todoKws of
                    []     -> mzero  -- no keywords present
                    (x:xs) -> return $ keywordsToSequence (reverse xs) [x]

 where
   todoKeyword :: Monad m => OrgParser m Text
   todoKeyword = many1Char nonspaceChar <* skipSpaces

   todoKeywords :: Monad m => OrgParser m [Text]
   todoKeywords = try $
     let endOfKeywords = todoDoneSep <|> void newline
     in manyTill todoKeyword (lookAhead endOfKeywords)

   doneKeywords :: Monad m => OrgParser m [Text]
   doneKeywords = try $
     manyTill (todoKeyword <* optional todoDoneSep) (lookAhead newline)

   todoDoneSep :: Monad m => OrgParser m ()
   todoDoneSep = void . try $ skipSpaces *> char '|' <* skipSpaces1

   keywordsToSequence :: [Text] -> [Text] -> TodoSequence
   keywordsToSequence todo done =
     let todoMarkers = map (TodoMarker Todo) todo
         doneMarkers = map (TodoMarker Done) done
     in todoMarkers ++ doneMarkers

macroDefinition :: Monad m => OrgParser m (Text, [Text] -> Text)
macroDefinition = try $ do
  macroName <- many1Char nonspaceChar <* skipSpaces
  firstPart <- expansionPart
  (elemOrder, parts) <- unzip <$> many ((,) <$> placeholder <*> expansionPart)
  let expander = mconcat . alternate (firstPart:parts) . reorder elemOrder
  return (macroName, expander)
 where
  placeholder :: Monad m => OrgParser m Int
  placeholder = try . fmap (fromMaybe 1 . safeRead) $ char '$' *> many1Char digit

  expansionPart :: Monad m => OrgParser m Text
  expansionPart = try $ manyChar (notFollowedBy placeholder *> noneOf "\n\r")

  alternate :: [a] -> [a] -> [a]
  alternate []     ys     = ys
  alternate xs     []     = xs
  alternate (x:xs) (y:ys) = x : y : alternate xs ys

  reorder :: [Int] -> [Text] -> [Text]
  reorder perm xs =
    let element n = take 1 $ drop (n - 1) xs
    in concatMap element perm