If you’re reading this you might have noticed that I have a shiny new blog! I had been pretty disappointed in Blogger practically since I started using it. But it was free, I was a poor college student, etc. I finally managed to get myself a domain name and set up a more proper blog!
I was between a few different pieces of blog software (namely WordPress, DotClear and Jekyll) and couldn’t really decide. I loved the idea of the simplicity of Jekyll, but it’s Ruby. (Who’s got time for that?) I wanted something I could read the code of and understand if necessary. (And it has been necessary!) My main requirements for blog software were:
After a quick Google search for “jekyll python“, I ended up on Hyde. It seemed alright, but no one seemed to use it. Farther down the page I came across a blog talking about moving to Pelican. I was hooked. (Note that I don’t necessarily agree with all the ideas in that post, it just introduced me to Pelican.)
(Since I dislike writing my own descriptions:)
“Pelican is a static site generator, written in Python, that requires no database or server-side logic.”
Cool! Sounds real simple to use! And it was. Mostly.
I pretty much followed the quickstart:
mkdir -p blog/blog # The first blog is to store other repos, etc. cd blog/blog git init && git checkout -b source # Explained later on. brew install pandoc # If you're not on Mac you'll need to do something else. mkvirtualenv blog pip install pelican markdown Fabric ghp_import feedparser pelican-quickstart
I’ll let you read the rest of the quickstart guide, but it was super quick to get up and running! (I also referenced another blog post I found that had some good information!) I, of course, had no content…but wait I did, it was just in Blogger!
Pelican does not directly support importing from Blogger (Edit: Seems someone just put up a `pull request`_ to support Blogger natively!), but it supports importing from a RSS feed. The first time I did this it missed a couple of articles (which I didn’t notice right away), so make sure you bump up the max amount in the URL like so (this also outputs in the folder “content”):
pelican-import --feed http://clokep.blogspot.com/feeds/posts/default\?alt\=rss\&max-results\=240 -o content
No import is ever perfect and I had to clean up a bunch of stuff by hand including:
I probably spent a couple of hours cleaning all the reStructuredText content up, but now I feel that I have a portable set of all of my blog posts, which I’m pretty happy about!
I tried a few different themes for Pelican, but eventually settled on just using and modifying the default theme. I, frankly, haven’t tried it on too many different systems, so hopefully it doesn’t totally break on small screen sizes or something. I’m not HTML expert, so I’d rather talk about the other stuff I modified. (Although, if you’re curious, the main elements I did modify are adding the sidebar to the left and the organization of the archives page.)
Blogger has a concept of “labels”, Pelican has a concept of “category” and “tags”. I hate this. What’s the difference? Anyway, I wanted to eradicate the concept of a “category” (and “authors” since I’m the only one on my blog!), so I added a few things to my pelicanconf.py:
# Disable categories. DISPLAY_CATEGORIES_ON_MENU = False DISPLAY_CATEGORY_ON_ARTICLE = False CATEGORY_FEED_ATOM = None CATEGORY_SAVE_AS = '' CATEGORIES_SAVE_AS = '' # Disable author pages. AUTHOR_SAVE_AS = '' AUTHORS_SAVE_AS = ''
Note that DISPLAY_CATEGORY_ON_ARTICLE is actually a variable I added and used in the template to not show categories above the list of tags on each article.
This is getting pretty long so I’ll leave how I’m actually serving this content to my next article!