pandoc MD->Tex Wrapper - Githubissues

hupili / evermd

Ever MarkDown! Solution Stacks for Markdown Usage

5 stars 1 forks source link

pandoc MD->Tex Wrapper #1

Open hupili opened 12 years ago

hupili commented 12 years ago

substitute \href with \url, which seems more wider deployed.
substitute verbatim with Verbatim, which reserves indent better.

hupili commented 12 years ago

Add default preambles to make the output compile-able.
Pre-processing figures.

hupili commented 12 years ago

The "\href" is a more powerful notation. Use package, "\usepackage{hyperref}".

pandoc leave an extra "\" at the end of item list.

hupili commented 12 years ago

Two more flaws detected:

If the list notation is supplemented by more than 4 blanks, pandoc will translate it to verbatim. e.g.

   *      <- there are more than 4 blanks before words.

The ordered list is output as

%\begin{enumerate}[1.]

The "[1.]" seems extra and just do not compile at my side, by default. I don't want to waste time tweaking it. It seems pandoc is not a good baseline to start. The HTML output of markdown translator looks well structured. Probably translate it to latex is easier.

hupili commented 12 years ago

More trials:

There are some general parsing issues. By imposing some writing conventions, they can be avoided. e.g. 3rd header + immediate code embedding.
pandoc can convert md to pdf smoothly at most times. The flawed intermediate result (.tex) is annoying...

hupili commented 12 years ago

pandoc can automatically deal with formulae! Depending on the output, it uses different strategies. e.g. for pdf, fully parsing using tex facilities; for html, use special character set if possible, o/w show it literally.

hupili commented 12 years ago

pandoc:

Can not handle remote figures.
Try to parse texts in html tags.