Scrivener + LaTeX = UTF8 issues

hi, I wanted to see how I could combine the powers of Scrivener and LaTeX.

I started by compiling the tutorial that comes with Scrivener to LaTeX, then I ran pdflatex on the resulting Tutorial.tex.

I was met with the error ! Package inputenc Error: Unicode char \u8:▪ not set up for use with LaTeX.

Searching the forums I found some earlier posts regarding UTF8 issues and LaTeX here:

which suggested the usage of \usepackage[utf-8]{inputenc} however that gave rise to other errors. It seems to me Scrivener should be able to output usable LaTeX immediately without special configs.

Am I doing something wrong?

The tutorial is not designed to work with LaTeX out of the box. You’d have to do some substantial formatting revisions to get it working. The LaTeX output from Scrivener is generated via MultiMarkdown (Markdown on steroids, basically). It’s actually the piece of technology that does all of the conversion. So if the original document is not formatted to work with MMD, the output result will in most cases fail.

If you want to see a Scrivener project that is set up to compile to LaTeX, you’d need to download the user manual sample project. It’s a bit out of date at this point. I need to refresh that, and it would also be worth pointing out that it uses MMD2, not MMD3 that ships with Scrivener. You’d have to download MMD2 from Github, and use the compatibility flag in the compile options. This is what I do when I create versions of the user manual PDF.

but isn’t the whole point of dong a compile into something is that it will work?

Like, compiling to PDF, HTML, .doc, etc there’s no way users would be happy if it didn’t output a functioning version of those. So why should the expectation for valid LaTeX be less?

example, tutorial will compile to PDF, html, and .doc, etc

I’m not sure if I follow your question. The compilation selection states “MultiMarkdown -> LaTeX”. If you aren’t starting with a MultiMarkdown formatted document, then why expect a valid LaTeX document?