From acf932825bfe40d9a18046c9d304f4f14363a88a Mon Sep 17 00:00:00 2001
From: Albert Krewinkel <albert@zeitkraut.de>
Date: Sat, 5 Dec 2020 22:05:37 +0100
Subject: [PATCH] Org reader: preserve targets of spurious links
Links with (internal) targets that the reader doesn't know about are
converted into emphasized text. Information on the link target is now
preserved by wrapping the text in a Span of class `spurious-link`, with
an attribute `target` set to the link's original target. This allows to
recover and fix broken or unknown links with filters.
See: #6916
---
src/Text/Pandoc/Readers/Org/Inlines.hs | 9 ++++-----
test/Tests/Readers/Org/Meta.hs | 6 ++++--
2 files changed, 8 insertions(+), 7 deletions(-)
diff --git a/src/Text/Pandoc/Readers/Org/Inlines.hs b/src/Text/Pandoc/Readers/Org/Inlines.hs
index b234bee58..0330cf55f 100644
--- a/src/Text/Pandoc/Readers/Org/Inlines.hs
+++ b/src/Text/Pandoc/Readers/Org/Inlines.hs
@@ -477,17 +477,17 @@ linkToInlinesF linkStr =
internalLink :: Text -> Inlines -> F Inlines
internalLink link title = do
- anchorB <- (link `elem`) <$> asksF orgStateAnchorIds
- if anchorB
+ ids <- asksF orgStateAnchorIds
+ if link `elem` ids
then return $ B.link ("#" <> link) "" title
- else return $ B.emph title
+ else let attr' = ("", ["spurious-link"] , [("target", link)])
+ in return $ B.spanWith attr' (B.emph title)
-- | Parse an anchor like @<<anchor-id>>@ and return an empty span with
-- @anchor-id@ set as id. Legal anchors in org-mode are defined through
-- @org-target-regexp@, which is fairly liberal. Since no link is created if
-- @anchor-id@ contains spaces, we are more restrictive in what is accepted as
-- an anchor.
-
anchor :: PandocMonad m => OrgParser m (F Inlines)
anchor = try $ do
anchorId <- parseAnchor
@@ -501,7 +501,6 @@ anchor = try $ do
-- | Replace every char but [a-zA-Z0-9_.-:] with a hyphen '-'. This mirrors
-- the org function @org-export-solidify-link-text@.
-
solidify :: Text -> Text
solidify = T.map replaceSpecialChar
where replaceSpecialChar c
diff --git a/test/Tests/Readers/Org/Meta.hs b/test/Tests/Readers/Org/Meta.hs
index 041016f64..bc167f2a5 100644
--- a/test/Tests/Readers/Org/Meta.hs
+++ b/test/Tests/Readers/Org/Meta.hs
@@ -270,7 +270,8 @@ tests =
, "Search links are read as emph" =:
"[[Wally][Where's Wally?]]" =?>
- para (emph $ "Where's" <> space <> "Wally?")
+ para (spanWith ("", ["spurious-link"], [("target", "Wally")])
+ (emph $ "Where's" <> space <> "Wally?"))
, "Link to nonexistent anchor" =:
T.unlines [ "<<link-here>> Target."
@@ -278,5 +279,6 @@ tests =
, "[[link$here][See here!]]"
] =?>
(para (spanWith ("link-here", [], []) mempty <> "Target.") <>
- para (emph ("See" <> space <> "here!")))
+ para (spanWith ("", ["spurious-link"], [("target", "link$here")])
+ (emph ("See" <> space <> "here!"))))
]