Ipnyb: allow lossless round-tripping of markdown cell content.

The reader now parses the contents of the markdown cell to a Pandoc
structure, but *also* stores the raw markdown in a `source`
attribute on the cell Div.  When we convert back to markdown,
this attribute is stripped off and the original source is used.
When we convert to other formats, the attribute is usually
ignored (though it will come through in HTML as a `data-source`
attribute, not unhelpfully).

I'll note some potential drawbacks of this approach:

- It makes it impossible to use pandoc to clean up or
  change the contents of markdown cells, e.g.
  going from `+smart` to `-smart`.

- There may be formats where the addition of the `source`
  attribute is problematic.  I can't think of any, though.

Closes #5408.

3 files changed, 10 insertions, 6 deletions

diff --git a/src/Text/Pandoc/Readers/Ipynb.hs b/src/Text/Pandoc/Readers/Ipynb.hs
index a245bdad3..056fa431c 100644
--- a/src/Text/Pandoc/Readers/Ipynb.hs
+++ b/src/Text/Pandoc/Readers/Ipynb.hs
@@ -79,7 +79,8 @@ cellToBlocks opts lang c = do
   case cellType c of
     Ipynb.Markdown -> do
       Pandoc _ bs <- walk fixImage <$> readMarkdown opts source
-      return $ B.divWith ("",["cell","markdown"],kvs)
+      let kvs' = ("source", source) : kvs
+      return $ B.divWith ("",["cell","markdown"],kvs')
              $ B.fromList bs
     Ipynb.Heading lev -> do
       Pandoc _ bs <- readMarkdown opts
diff --git a/src/Text/Pandoc/Writers/Ipynb.hs b/src/Text/Pandoc/Writers/Ipynb.hs
index d01d5a7e5..86e2abbdf 100644
--- a/src/Text/Pandoc/Writers/Ipynb.hs
+++ b/src/Text/Pandoc/Writers/Ipynb.hs
@@ -102,10 +102,13 @@ extractCells _ [] = return []
 extractCells opts (Div (_id,classes,kvs) xs : bs)
   | "cell" `elem` classes
   , "markdown" `elem` classes = do
-      let meta = pairsToJSONMeta kvs
+      let meta = pairsToJSONMeta [(k,v) | (k,v) <- kvs, k /= "source"]
       (newdoc, attachments) <-
         runStateT (walkM addAttachment (Pandoc nullMeta xs)) mempty
-      source <- writeMarkdown opts{ writerTemplate = Nothing } newdoc
+      source <- case lookup "source" kvs of
+                  Just s  -> return s
+                  Nothing -> writeMarkdown opts{ writerTemplate = Nothing }
+                              newdoc
       (Ipynb.Cell{
           cellType = Markdown
         , cellSource = Source $ breakLines $ T.stripEnd source
diff --git a/test/ipynb/simple.out.native b/test/ipynb/simple.out.native
index 4c060b82f..960230894 100644
--- a/test/ipynb/simple.out.native
+++ b/test/ipynb/simple.out.native
@@ -1,15 +1,15 @@
 Pandoc (Meta {unMeta = fromList [("jupyter",MetaMap (fromList [("nbformat",MetaString "4"),("nbformat_minor",MetaString "5")]))]})
-[Div ("",["cell","markdown"],[])
+[Div ("",["cell","markdown"],[("source","Lorem ipsum\n===========\n\n**Lorem ipsum** dolor sit amet, consectetur adipiscing elit. Nunc luctus\nbibendum felis dictum sodales.")])
  [Header 1 ("lorem-ipsum",[],[]) [Str "Lorem",Space,Str "ipsum"]
  ,Para [Strong [Str "Lorem",Space,Str "ipsum"],Space,Str "dolor",Space,Str "sit",Space,Str "amet,",Space,Str "consectetur",Space,Str "adipiscing",Space,Str "elit.",Space,Str "Nunc",Space,Str "luctus",SoftBreak,Str "bibendum",Space,Str "felis",Space,Str "dictum",Space,Str "sodales."]]
 ,Div ("",["cell","code"],[])
  [CodeBlock ("",["python"],[]) "print(\"hello\")"]
-,Div ("",["cell","markdown"],[])
+,Div ("",["cell","markdown"],[("source","Pyout\n-----")])
  [Header 2 ("pyout",[],[]) [Str "Pyout"]]
 ,Div ("",["cell","code"],[("execution_count","2")])
  [CodeBlock ("",["python"],[]) "from IPython.display import HTML\nHTML(\"\"\"\n<script>\nconsole.log(\"hello\");\n</script>\n<b>HTML</b>\n\"\"\")"
  ,Div ("",["output","execute_result"],[("execution_count","2")])
   [RawBlock (Format "html") "<script>\nconsole.log(\"hello\");\n</script>\n<b>HTML</b>\nhello"]]
-,Div ("",["cell","markdown"],[("tags","[\"foo\",\"bar\"]")])
+,Div ("",["cell","markdown"],[("source","Image\n-----\n\nThis image ![the moon](attachment:lalune.jpg) will be included as a cell\nattachment."),("tags","[\"foo\",\"bar\"]")])
  [Header 2 ("image",[],[]) [Str "Image"]
  ,Para [Str "This",Space,Str "image",Space,Image ("",[],[]) [Str "the",Space,Str "moon"] ("lalune.jpg",""),Space,Str "will",Space,Str "be",Space,Str "included",Space,Str "as",Space,Str "a",Space,Str "cell",SoftBreak,Str "attachment."]]]