lyx2web is a standalone Python script to convert a Lyx or a .tex document to a set of blog posts. This little tool is suitable if you maintain notes in one Lyx document and want to publish them as blog posts.

• For each section (\section in Latex), the tool will write two files: a markdown (.md) and a Latex file. The markdown file is suitable for publishing on the web if you use a static site generator (e.g., Pelican, Jekyll). If your static site generator is setup properly, the equation outputs in the markdown file can be rendered directly with MathJax.

• The generated .tex file from each section is compilable with pdflatex. The Latex preamble is copied from the input Lyx (or .tex) file.

For Pelican setup that can directly support the markdown output from this tool, see the source code of my blog for example.

The tool is loosely tested on Ubuntu 14.04 with Python 2.7.11.

Require

• Python 2.x.
• lyx (used for converting the input Lyx file into .tex)
• pandoc (used for converting to markdown)
• iconv (used for converting the tex source file to UTF8 encoding)

usage: lyx2web.py [-h] src dest

Convert a .tex or .lyx file into a sequence of markdown files. Save them into
the specified dest directory.

positional arguments:
  src         Path to source file to be converted.
  dest        Destination directory to save exported markdown files.

optional arguments:
  -h, --help  show this help message and exit

The repository contains test.lyx and test.tex (of identical content) for you to test. There are three sections in the document.

python lyx2web.py test.lyx export 

will create a new folder export/ that contains six files: (.tex, .md) files for each section.

Many static site generators require each post to have metadata at the beginning of its markdown file. For example, in Pelican, the metadata look like

Title: some title here
Date: 2014-06-29 
Tags: probability, statistics
Slug: prob_gaussian_ball

To specify this, first define a new command \postmeta in the Latex preamble with

\newcommand{\postmeta}[1]{}

At the beginning of each section, write

\postmeta{
Title: some title here
Date: 2014-06-29 
Tags: probability, statistics
Slug: prob_gaussian_ball
}

This metadata will not show up in the compiled pdf as the command \postmeta is defined to do nothing. The only purpose is so that the tool can copy to the beginning of the generated markdown file. The value of Slug will be used as the generated file name.

If \postmeta{..} is unspecified, the text specified in \section{...} will be used as the slug (after normalizing and removing unsuitable characters), today's date will be used for the date, and tags will be set to empty.

This tool is written in a few hours and obviously nowhere near complete.

Not supported (yet)

• Latex bibliography.

Supported

• Latex's \href (more suitable for blog posts than Latex bibliography).
• \footnote.