Changed the blog engine to nikola

I do enjoy using one general tool for many tasks - I like having a consistent environment for different tasks. For example, I chose Python (a general purpose language) instead of domain specific languages such as R and MATLAB for my programming needs.

I have been looking for something like that for my writing as well, and since I used Python a lot, reStructuredText was a good candidate. With reStructuredText, I could generate output in multiple formats (html, pdf, slides), and I like the grammar.

With reStructuredText, I started using Emacs more for writing. Emacs' rst-mode is very well done, and I have been using org-mode for a while as well, even though for a while I mainly used it for the TODO list management and some outline generation only.

Then this year I started teaching, and I had multiple sections, and I started looking for ways to minimize the redundancy in course contents files. For example, syllabi for all sections of classes share many common parts and if I maintain separate files for each class, I have to modify all the files every time I make some changes on one of the common parts.

So I started writing lecture slides in reStructuredText in org-mode, putting reStructuredText code in code blocks and using org-babel to tangle the source code into output files. In that way, I just write the common parts once, and I could include them in multiple files. For slide generation, I used hieroglyph package which was great. Also I liked that I can generate document-style lecture note for distribution from the same source.

But quickly I started feeling that putting reStructuredText code inside code blocks cumbersome. In addition, I really missed the easy navigation of tree structure of org-mode. So I looked for a way to just use org-mode to write everything instead of putting reStructuredText code inside code-blocks.

Then I learned about org-export which can export an org-mode document to multiple output formats. Since then I started using org-mode to write my course notes. I had to learn a few tricks and also install org 8.3beta, but now the results has been very satisfactory. Now I write lecture note in org-mode, , and use ox-beamer to generate beamer slides for presentations, and use ox-latex to generate lecture notes from the same source. org-mode is just a pleasure to use. Now I believe I found the tool-chain for all my writing needs: Emacs and org-mode.

Naturally I started searching for a blogging engine which enables me to use org-mode. I have been using tinkerer for a while, and it is great, but there are a couple of steps you have to go through to use org-mode to write posts. Exporting a subtree to a .rst file is easy, but there are some manual steps involved - basically you have to add a new post to the sphinx index file either by manually or by using tinkerer.

Then I found nikola. Unlike tinkerer, nikola automatically reads files in the designated posts folder and generates index - so you don't have to manually add anything to the index. In addition, there is a tool org2nikola which generates a post from a org subtree, so I can put all my posts into one .org file.

You need to add the .wp file extensions to the POSTS and PAGES tuples, and also to markdown compilers in your conf.py:

POSTS = (
    ("posts/*.rst", "posts", "post.tmpl"),
    ("posts/*.txt", "posts", "post.tmpl"),
    ("posts/*.wp", "posts", "post.tmpl"),
)
PAGES = (
    ("stories/*.rst", "stories", "story.tmpl"),
    ("stories/*.txt", "stories", "story.tmpl"),
    ("stories/*.wp", "stories", "story.tmpl"),
)

COMPILERS = {
    "rest": ('.rst', '.txt'),
    "markdown": ('.md', '.mdown', '.markdown', '.wp'),
    "textile": ('.textile',),
    "txt2tags": ('.t2t',),
    "bbcode": ('.bb',),
    "wiki": ('.wiki',),
    "ipynb": ('.ipynb',),
    "html": ('.html', '.htm'),
    # PHP files are rendered the usual way (i.e. with the full templates).
    # The resulting files have .php extensions, making it possible to run
    # them without reconfiguring your server to recognize them.
    "php": ('.php',),
    # Pandoc detects the input from the source filename
    # but is disabled by default as it would conflict
    # with many of the others.
    # "pandoc": ('.rst', '.md', '.txt'),
}

Just call org2nikola-export-subtree in the subtree that you want to export as a nikola post. It will ask you where is the root directory for your nikola blog. It seems it just uses the top level heading.