I write in MMD format the majority of the time. I was just compiling something using bold tags of ** on either side of the text. One time I got Italics with one * on either side of the text. When I put a space after **, I just get ** around the text with a space before and after, just the way I wrote it.
** should be a valid MMD bold tag, as far as I’ve read. Is the MMD compiler just not using those tags for bold? I’ll need to relearn which I use for bold it seems
Edit: Hmm… I tried __ as well, and that didn’t work. Strange. Which Markdown compiler is being used in the background?
We use an embedded version of MultiMarkdown for the conversion—I don’t think that should be the problem here though, unless you are using bold in a way that isn’t compatible with the CommonMark specs (which MMD works to adhere to).
What I would do is run a test compile to plain MultiMarkdown and examine the output. This is the text Scrivener generates before handing it off to the MMD exe for conversion, and thus will reveal any issues exclusive to its output, versus any issues generated by the Markdown conversion engine to the end format.
This may be a manifestation of this bug, which it looks like they haven’t fixed yet (there is a simple workaround, described there).
I ran the file with Default Markdown, and the Bold areas look correct. I was doing Markdown -> ODT, so you are probably right that it’s related to the other bug.
I think this is something different. The bug I referred you to would have erased asterisks with plain MMD output even. Looking at it a little more closely, my guess is that this is a manifestation of an older bug (or perhaps just evidence of it being under construction), which I’ve now tagged the thread as. I observed:
- Compile to MMD doesn’t damage the content (at least with a few simple tests).
- Compile to ODT strips asterisks and escapes other punctuation (tested block quote), turning the markdown into raw visible text.
- Compile to HTML strips the asterisks but doesn’t escape other punctuation, so some stuff works.
If you’re trying to actually export instead of just testing, I think the best thing to do is use plain MMD and process it yourself with multimarkdown.exe on the command prompt.
If you’re trying to test, I wouldn’t say it is quite ready for that yet, given the above. It’s pretty heavily under construction if each of the different converters has a different output! There are some internal things that change of course, but for the most part the only important thing that should be changing between HTML and ODT is the command-line flags being used. We shouldn’t be seeing evidence of entirely different temp Markdown files being sent.