3fb5499 · typeable/pandoc

git.iko.soy/typeable/pandoc

insertMediaBag: ensure we get sane mediaPath for URLs.

Long URLs cannot be treated as mediaPaths, but System.FilePath's `isRelative` often returns True for them. So we add a check for an absolute URL. We also ensure that extensions are derived only from the path portion of URLs (previously a following query was being included). Closes #7391.

commit3fb5499dd638dae7156fba63ba5c1522bed5e46d authorJohn MacFarlane <jgm@berkeley.edu> dateFri 18 Jun 2021, 12:06 (5 years ago) parent9612684

1 file changed+10−5

compare with master

unifiedsplit

src/Text/Pandoc/MediaBag.hs+10−5

src/Text/Pandoc/MediaBag.hs+10−5

@@ -26,13 +26,14 @@ module Text.Pandoc.MediaBag (

2626import qualified Data.ByteString.Lazy as BL

2727import Data.Data (Data)

2828import qualified Data.Map as M

29−import Data.Maybe (fromMaybe)

29+import Data.Maybe (fromMaybe, isNothing)

3030import Data.Typeable (Typeable)

3131import System.FilePath

3232import Text.Pandoc.MIME (MimeType, getMimeTypeDef, extensionFromMimeType)

3333import Data.Text (Text)

3434import qualified Data.Text as T

3535import Data.Digest.Pure.SHA (sha1, showDigest)

36+import Network.URI (URI (..), parseURI)

3637

3738data MediaItem =

3839 MediaItem

@@ -76,16 +77,20 @@ insertMedia fp mbMime contents (MediaBag mediamap) =

7677 , mediaContents = contents

7778 , mediaMimeType = mt }

7879 fp' = canonicalize fp

79− newpath = if isRelative fp && ".." `notElem` splitPath fp

80+ uri = parseURI fp

81+ newpath = if isRelative fp

82+ && isNothing uri

83+ && ".." `notElem` splitPath fp

8084 then T.unpack fp'

8185 else showDigest (sha1 contents) <> "." <> ext

8286 fallback = case takeExtension fp of

8387 ".gz" -> getMimeTypeDef $ dropExtension fp

8488 _ -> getMimeTypeDef fp

8589 mt = fromMaybe fallback mbMime

86− ext = case takeExtension fp of

87− '.':e -> e

88− _ -> maybe "" T.unpack $ extensionFromMimeType mt

90+ path = maybe fp uriPath uri

91+ ext = case takeExtension path of

92+ '.':e -> e

93+ _ -> maybe "" T.unpack $ extensionFromMimeType mt

8994

9095

9196-- | Lookup a media item in a 'MediaBag', returning mime type and contents.