From 0a502e5ff52b251bbf3da69fd1f9a88d5e0fe92c Mon Sep 17 00:00:00 2001
From: John MacFarlane <jgm@berkeley.edu>
Date: Thu, 10 Dec 2020 15:44:10 -0800
Subject: HTML reader: retain attribute prefixes and avoid duplicates.
Previously we stripped attribute prefixes, reading
`xml:lang` as `lang` for example. This resulted in
two duplicate `lang` attributes when `xml:lang` and
`lang` were both used. This commit causes the prefixes
to be retained, and also avoids invald duplicate
attributes.
Closes #6938.
---
src/Text/Pandoc/Readers/HTML/Parsing.hs | 20 +++++++++++++-------
1 file changed, 13 insertions(+), 7 deletions(-)
(limited to 'src/Text/Pandoc/Readers/HTML')
diff --git a/src/Text/Pandoc/Readers/HTML/Parsing.hs b/src/Text/Pandoc/Readers/HTML/Parsing.hs
index 2d58319da..e28ebe77b 100644
--- a/src/Text/Pandoc/Readers/HTML/Parsing.hs
+++ b/src/Text/Pandoc/Readers/HTML/Parsing.hs
@@ -193,14 +193,20 @@ t1 `closes` t2 |
_ `closes` _ = False
toStringAttr :: [(Text, Text)] -> [(Text, Text)]
-toStringAttr = map go
+toStringAttr = foldr go []
where
- go (x,y) =
- case T.stripPrefix "data-" x of
- Just x' | x' `Set.notMember` (html5Attributes <>
- html4Attributes <> rdfaAttributes)
- -> (x',y)
- _ -> (x,y)
+ go :: (Text, Text) -> [(Text, Text)] -> [(Text, Text)]
+ -- treat xml:lang as lang
+ go ("xml:lang",y) ats = go ("lang",y) ats
+ -- prevent duplicate attributes
+ go (x,y) ats
+ | any (\(x',_) -> x == x') ats = ats
+ | otherwise =
+ case T.stripPrefix "data-" x of
+ Just x' | x' `Set.notMember` (html5Attributes <>
+ html4Attributes <> rdfaAttributes)
+ -> go (x',y) ats
+ _ -> (x,y):ats
-- Unlike fromAttrib from tagsoup, this distinguishes
-- between a missing attribute and an attribute with empty content.
--
cgit v1.2.3