1. 2

I was bored the other day so I played with the following idea:

  1. Parse the Git commit log and group together commit messages from each author
  2. Do a sentiment analysis (or other text analysis) on the commit message content to generate one or more descriptive words for each author that summarize the overarching tone of their commit message history
  3. For each author, send the resulting words to a GIF api to grab a (hopefully funny) GIF per author
  4. Email the GIFs to the dev team for a good laugh

I wrote a script that does this, but most commit messages are pretty mundane and don’t have much of a “tone” that can be identified by sentiment analysis engines… Any fun ideas to make this more interesting?

  1. 2

    Also look at add/modify/delete lines ratios, but I don’t know how to add this in a meaningful way to the gif query.

    Look at comments added to the code.