c389211 · typeable/pandoc

git.iko.soy/typeable/pandoc

Support `yaml_metadata_block` extension form commonmark, gfm.

This is a bit more limited than with markdown, as documented in the manual: - The YAML block must be the first thing in the input. - The leaf notes are parsed in isolation from the rest of the document. So, for example, you can't use reference links if the references are defined later in the document. Closes #6537.

commitc389211e2f678d9327cd2c008d54a1e438f07a07 authorJohn MacFarlane <jgm@berkeley.edu> dateSat 20 Mar 2021, 15:45 (6 years ago) parent2274eb8

4 files changed+50−1

compare with master

unifiedsplit

MANUAL.txt+16 src/Text/Pandoc/Extensions.hs+3 src/Text/Pandoc/Parsing.hs+1−1 src/Text/Pandoc/Readers/CommonMark.hs+30

MANUAL.txt+16

@@ -4240,6 +4240,22 @@ will be interpreted as markdown. For example:

42404240 \renewcommand{\section}[1]{\clearpage\oldsection{#1}}

42414241 ```

42424242

4243+Note: the `yaml_metadata_block` extension works with

4244+`commonmark` as well as `markdown` (and it is enabled by default

4245+in `gfm` and `commonmark_x`). However, in these formats the

4246+following restrictions apply:

4247+

4248+- The YAML metadata block must occur at the beginning of the

4249+ document (and there can be only one). If multiple files are

4250+ given as arguments to pandoc, only the first can be a YAML

4251+ metadata block.

4252+

4253+- The leaf nodes of the YAML structure are parsed in isolation from

4254+ each other and from the rest of the document. So, for

4255+ example, you can't use a reference link in these contexts

4256+ if the link definition is somewhere else in the document.

4257+

4258+

42434259## Backslash escapes

42444260

42454261#### Extension: `all_symbols_escapable` ####

src/Text/Pandoc/Extensions.hs+3

@@ -354,6 +354,7 @@ getDefaultExtensions "gfm" = extensionsFromList

354354 , Ext_strikeout

355355 , Ext_task_lists

356356 , Ext_emoji

357+ , Ext_yaml_metadata_block

357358 ]

358359getDefaultExtensions "commonmark" = extensionsFromList

359360 [Ext_raw_html]

@@ -379,6 +380,7 @@ getDefaultExtensions "commonmark_x" = extensionsFromList

379380 , Ext_raw_attribute

380381 , Ext_implicit_header_references

381382 , Ext_attributes

383+ , Ext_yaml_metadata_block

382384 ]

383385getDefaultExtensions "org" = extensionsFromList

384386 [Ext_citations,

@@ -511,6 +513,7 @@ getAllExtensions f = universalExtensions <> getAll f

511513 , Ext_implicit_header_references

512514 , Ext_attributes

513515 , Ext_sourcepos

516+ , Ext_yaml_metadata_block

514517 ]

515518 getAll "commonmark_x" = getAll "commonmark"

516519 getAll "org" = autoIdExtensions <>

src/Text/Pandoc/Parsing.hs+1−1

@@ -517,7 +517,7 @@ parseFromString :: (Stream s m Char, IsString s)

517517 -> ParserT s st m r

518518parseFromString parser str = do

519519 oldPos <- getPosition

520− setPosition $ initialPos "chunk"

520+ setPosition $ initialPos " chunk"

521521 oldInput <- getInput

522522 setInput $ fromString $ T.unpack str

523523 result <- parser

src/Text/Pandoc/Readers/CommonMark.hs+30

@@ -26,13 +26,43 @@ import Text.Pandoc.Definition

2626import Text.Pandoc.Builder as B

2727import Text.Pandoc.Options

2828import Text.Pandoc.Error

29+import Text.Pandoc.Readers.Metadata (yamlMetaBlock)

2930import Control.Monad.Except

3031import Data.Functor.Identity (runIdentity)

3132import Data.Typeable

33+import Text.Pandoc.Parsing (runParserT, getPosition, sourceLine,

34+ runF, defaultParserState, take1WhileP, option)

35+import qualified Data.Text as T

3236

3337-- | Parse a CommonMark formatted string into a 'Pandoc' structure.

3438readCommonMark :: PandocMonad m => ReaderOptions -> Text -> m Pandoc

3539readCommonMark opts s

40+ | isEnabled Ext_yaml_metadata_block opts

41+ , "---" `T.isPrefixOf` s = do

42+ let metaValueParser = do

43+ inp <- option "" $ take1WhileP (const True)

44+ case runIdentity

45+ (commonmarkWith (specFor opts) "metadata value" inp) of

46+ Left _ -> mzero

47+ Right (Cm bls :: Cm () Blocks)

48+ -> return $ return $ B.toMetaValue bls

49+ res <- runParserT (do meta <- yamlMetaBlock metaValueParser

50+ pos <- getPosition

51+ return (meta, pos))

52+ defaultParserState "YAML metadata" s

53+ case res of

54+ Left _ -> readCommonMarkBody opts s

55+ Right (meta, pos) -> do

56+ let dropLines 0 = id

57+ dropLines n = dropLines (n - 1) . T.drop 1 . T.dropWhile (/='\n')

58+ let metaLines = sourceLine pos - 1

59+ let body = T.replicate metaLines "\n" <> dropLines metaLines s

60+ Pandoc _ bs <- readCommonMarkBody opts body

61+ return $ Pandoc (runF meta defaultParserState) bs

62+ | otherwise = readCommonMarkBody opts s

63+

64+readCommonMarkBody :: PandocMonad m => ReaderOptions -> Text -> m Pandoc

65+readCommonMarkBody opts s

3666 | isEnabled Ext_sourcepos opts =

3767 case runIdentity (commonmarkWith (specFor opts) "" s) of

3868 Left err -> throwError $ PandocParsecError s err