SelfContained: Handle url() inside material retrieved from url().

This can happen e.g. with an @import of a google web font.
(What is imported is some CSS which contains an url reference
to the font itself.)

Also, allow unescaped pipe (|) in URL.

This is intended to help with #3629, but it doesn't seem to
work.

1 files changed, 12 insertions, 3 deletions

diff --git a/src/Text/Pandoc/SelfContained.hs b/src/Text/Pandoc/SelfContained.hs
index 53cb4a4b5..6391ef0e0 100644
--- a/src/Text/Pandoc/SelfContained.hs
+++ b/src/Text/Pandoc/SelfContained.hs
@@ -178,12 +178,21 @@ pCSSUrl sourceURL d = P.try $ do
   P.char ')'
   let fallback = B.pack ("url(" ++ maybe "" (:[]) quote ++ trim url ++
                             maybe "" (:[]) quote ++ ")")
-  case trim url of
+  -- pipes are used in URLs provided by Google Code fonts
+  -- but parseURI doesn't like them, so we escape them:
+  case escapeURIString (/='|') (trim url) of
       '#':_ -> return fallback
       'd':'a':'t':'a':':':_ -> return fallback
       u ->  do let url' = if isURI u then u else d </> u
-               enc <- lift $ getDataURI sourceURL "" url'
-               return (B.pack $ "url(" ++ enc ++ ")")
+               res <- lift $ getData sourceURL "" url'
+               case res of
+                    Left uri -> return (B.pack $ "url(" ++ uri ++ ")")
+                    Right (mt, raw) -> do
+                      -- note that the downloaded content may
+                      -- itself contain url(...).
+                      raw' <- cssURLs sourceURL d raw
+                      let enc = makeDataURI (mt, raw')
+                      return (B.pack $ "url(" ++ enc ++ ")")
 
 getDataURI :: PandocMonad m => Maybe String -> MimeType -> String -> m String
 getDataURI sourceURL mimetype src = do