Tissevert/pandoc
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions 1

HTML reader: pay attention to lang attributes on body.

Browse source 
These (as well as lang attributes on html) should update
lang in metadata. See #6938.
This commit is contained in:
John MacFarlane 2020-12-10 15:51:20 -08:00
parent 0a502e5ff5
commit fcd0658189
1 changed files with 6 additions and 3 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

9
src/Text/Pandoc/Readers/HTML.hs
View file

@ -25,7 +25,6 @@ module Text.Pandoc.Readers.HTML ( readHtml
) where
import Control.Applicative ((<|>))
import Control.Arrow (first)
import Control.Monad (guard, msum, mzero, unless, void)
import Control.Monad.Except (throwError)
import Control.Monad.Reader (ask, asks, lift, local, runReaderT)
@ -121,14 +120,18 @@ setInPlain :: PandocMonad m => HTMLParser m s a -> HTMLParser m s a
setInPlain = local (\s -> s {inPlain = True})
pHtml :: PandocMonad m => TagParser m Blocks
pHtml = try $ do
pHtml = do
(TagOpen "html" attr) <- lookAhead pAny
for_ (lookup "lang" attr <|> lookup "xml:lang" attr) $
updateState . B.setMeta "lang" . B.text
pInTags "html" block
pBody :: PandocMonad m => TagParser m Blocks
pBody = pInTags "body" block
pBody = do
(TagOpen "body" attr) <- lookAhead pAny
for_ (lookup "lang" attr <|> lookup "xml:lang" attr) $
updateState . B.setMeta "lang" . B.text
pInTags "body" block
pHead :: PandocMonad m => TagParser m Blocks
pHead = pInTags "head" $ pTitle <|> pMetaTag <|> pBaseTag <|> (mempty <$ pAny)