Tweet

I Made A Blog!

1 Nov 2013

I wanted a place to publish articles I write; which would also hopefully inspire me to write more consistently, if not often.

I wanted to find something that would do everything that I want in a blog platform without much effort. And, I wanted some very specific things, which I thought others would have already developed; so I didn't have to.

What I Want(ed)

Let me explain what I was looking for - and eventually - did not find, and so wrote it myself. Maybe someone else will find it useful and use it themselves. Comments and criticisms always welcome.

  1. Simple
  2. Markdown Content

Simple

I did not want to deal with installing, configuring, connecting, and maintaining a Database so I opted out of using one all together. More advanced features could be enabled by using a DB such as: user authentication and authorization, proper tagging and categorization of content, scheduled publishing, and on and on. I did not - and will not, for a while - require any of these features.

I wanted to use as few external dependencies as possible. I am a JavaScript developer so I wanted to write as much of the system as I could stand to do; once I finally decided to stop looking for something existing and do it myself. I have in the past written a templating library and Markdown-like parser as experiments yet opted to not use those for this project. I could have but I erred on the side of community standards.

Markdown Content

When I write I find that I enjoy not being distracted by the editor. Markdown is great for that since it is very streamlined and to the point. So the rule I set was that all content editing would be done in Markdown files and rendered through a standard parser; nothing custom.

What I Created

I like using stuff I can understand. I didn't want to learn a whole lot to get a "working" site up and available. So I worked on reducing everything to bare necessities; what is absolutely needed.

  1. Router
  2. Server
  3. Templates
  4. Tag Filtering

Router

I needed a router to make sure that I could reliably handle requests. So I wrote a simple one for myself. It was kind of fun, a single function which performs double duty registering and returning routes. Add in some error checking and it's all done.

Server

The server was the next most important part. Once I created a way to handle a request I needed something to do all of that. Included in the server is some content caching for the styles and scripts - basically anything in the "public" folder (for right now) - and the articles. In the future I think that I would like to break this file up into at least two or three to make them more modular and replace-able or refactor-able.

Templating

I used the first templating library I found that had template inheritance since I thought that would be handy; and it was. Swig was the choice but I am sure there are plenty of others that would fit the bill.

Tagging & Filtering

The tags in the article files are plain text. They might not be robust but they are easy. The templates create a DOM location for the tags to be picked up by JavaScript and make useful for filtering in the page.

Conclusion

Obviously some shortcuts were taken and therefore the long-term scalability of it as a large blog is highly suspect, but it is a start and will work for a while. Time will tell how bad - or good - a thing I have created.

Tags

comments powered by Disqus