Text.Pandoc.Format: change formatFromFilePaths...

so that it is smarter about URLs.  URLs are parsed, and we take the
format from the path component, if present.

This means that `https://emacs.org/` will be treated as HTML, while
`https://emacs.org/sample.org` will be treated as Org.

Closes #10141.

2 files changed, 7 insertions, 3 deletions

diff --git a/src/Text/Pandoc/App.hs b/src/Text/Pandoc/App.hs
index 4ae62f401..cdfcf7bbe 100644
--- a/src/Text/Pandoc/App.hs
+++ b/src/Text/Pandoc/App.hs
@@ -67,7 +67,6 @@ import Text.Pandoc.PDF (makePDF)
 import Text.Pandoc.Scripting (ScriptingEngine (..), CustomComponents(..))
 import Text.Pandoc.SelfContained (makeSelfContained)
 import Text.Pandoc.Shared (tshow)
-import Text.Pandoc.URI (isURI)
 import Text.Pandoc.Writers.Shared (lookupMetaString)
 import Text.Pandoc.Readers.Markdown (yamlToMeta)
 import qualified Text.Pandoc.UTF8 as UTF8
@@ -150,7 +149,6 @@ convertWithOpts' scriptingEngine istty datadir opts = do
       Nothing -> case Format.formatFromFilePaths sources of
         Just f' -> return f'
         Nothing | sources == ["-"] -> return $ defFlavor "markdown"
-                | any (isURI . T.pack) sources -> return $ defFlavor "html"
                 | otherwise -> do
                     report $ CouldNotDeduceFormat
                       (map (T.pack . takeExtension) sources) "markdown"
diff --git a/src/Text/Pandoc/Format.hs b/src/Text/Pandoc/Format.hs
index 1f0f86988..3e1f3da88 100644
--- a/src/Text/Pandoc/Format.hs
+++ b/src/Text/Pandoc/Format.hs
@@ -38,6 +38,7 @@ import Text.Pandoc.Extensions
   , showExtension
   , readExtension
   )
+import Network.URI (URI (..), parseURI)
 import Text.Pandoc.Parsing
 import qualified Data.Text as T
 
@@ -170,7 +171,7 @@ formatFromFilePaths = asum . map formatFromFilePath
 -- | Determines format based on file extension.
 formatFromFilePath :: FilePath -> Maybe FlavoredFormat
 formatFromFilePath x =
-  case takeExtension (map toLower x) of
+  case takeExtension (map toLower fpath) of
     ".Rmd"      -> defFlavor "markdown"
     ".adoc"     -> defFlavor "asciidoc"
     ".asciidoc" -> defFlavor "asciidoc"
@@ -233,3 +234,8 @@ formatFromFilePath x =
   withExtension Nothing _ = Nothing
   withExtension (Just (FlavoredFormat f ed)) ext = Just $
     FlavoredFormat f (ed <> ExtensionsDiff (extensionsFromList [ext]) mempty)
+  fpath = case parseURI x of
+            Nothing -> x
+            Just URI{ uriPath = "" } -> "index.html"
+            Just URI{ uriPath = "/" } -> "index.html"
+            Just URI{ uriPath = up } -> up