I am trying out Scrivener, and I am interested in using it to write multimarkdown (so i can do all the formatting at the end, but still have some minimal styles in place). When I compile a sample document with the Markdown->RTF settings, and I have selected do Straighten Smart Quotes, convert elipses, convert em-dashes … the output file still has smart quotes and elipses inside (and this is true even after I have eliminated smart quotes in the Scrivner document itself).
Unfortunately, MS-Word 2008 does not read the smart characters correctly (perhaps there is something weird in the encoding settings of the RTF file that is created), and so I have been trying to figure out how to make sure they aren’t created in the first place.
I believe smart quotes and suchlike are added by the XSLT used to process the MMD file, so you would need to use a modified XSLT. However, I am no expert on MMD and could be way off - Ioa is the expert on this, so hopefully he’ll be able to set you straight. I do know that this is dealt by the MMD -> HTML -> RTF process, though, which is out of Scrivener’s hands, and that the MMD files will need playing with to alter this behaviour. As I say, Ioa will be able to explain it.
All the best,
Keith
That is essentially correct. Modern MMD doesn’t use XSLT by default, it uses a native converter written in C and is thus much, much faster than it used to be. Where Keith is correct is that this is in fact handled after everything Scrivener does is complete. MMD is designed to take straight quotes and automatically “smarten” them, so even if you specifically straighten them in Scrivener, you’ll still get smart quotes.
I’ve never actually tried to disable this before, so I don’t know how one would go about doing so in the new system. You might have better luck posting this query to MMD’s Github forum.
As an aside, I would consider using “Flat XML” instead of RTF. This .fodt file can be opened in LibreOffice (as well as Nisus Writer Pro) and saved as RTF from there. The results will be much superior to the result you get going straight to RTF. It will be prepared with logical stylesheet assignments (heading 1, body, quotation, etc), and use proper semantics where appropriate for things like cross-referencing and footnotes. The RTF is just brute-force converted from HTML (which of course isn’t much of a word processor format itself), via the [b]textutil[/b] UNIX utility installed on all Macs. The quality level leaves much to be desired, however it is fast, and doesn’t involve LibreOffice, so it is useful for proofing. For any serious word processor migration though, it pays to go with a robust word processor output. That may solve the quote problem for you, by treating the quotes properly in the first place. I don’t have Word 2008 for the Mac to test this theory however (I did test using Office 2010 for the PC, but both the straight RTF and FODT->RTF converted files worked fine).
