Skip to content

Latest commit

 

History

History
40 lines (31 loc) · 2.19 KB

README.md

File metadata and controls

40 lines (31 loc) · 2.19 KB

Paco's Version of BeerSnobSays

This is a fork of BeerSnobSays, a Twitter bot written by @gjreda. Follow the bot at @cigibot and follow me at @pacohope.

It's using the text of @cigitalgem's book Software Security and the text of the BSIMM 6 report. It has produced a few fun gems.

So far, I've made a few small enhancements:

  1. I made it randomly hashtag words. Right now, it looks to see if a word is longer than 6 characters, then it rolls a die. 20% of the time, it will hashtag the word.
  2. It capitalises the first letter of the tweet, whether that word was originally capitalised or not. (there's a silly bug: if the first word of the tweet gets hashtagged, then it doesn't get capitalized)
  3. It delays a random amount of time (between 0 and 1200 seconds) just so the tweets aren't bang on the minute every single time.

It's built with Python and a bunch of modules. In my case, I had to do the following to get my environment running on FreeBSD:

  • sudo pkg install git readline python
  • Make a link to /usr/local/bin/bash in /bin because Python's readline library won't compile correctly unless you do. sudo ln -s /usr/local/bin/bash /bin
  • curl -O https://bootstrap.pypa.io/get-pip.py
  • sudo python get-pip.py
  • sudo pip install Twython ipython gnureadline
  • create a config.py file

You invoke it by calling it with a file name of text that you want it to generate from. E.g.,:

/usr/local/bin/python markov.py input.txt

I took a bunch of word docs I had for a book and converted them to text files so it could use them. On MacOS that looks like:

for i in *.doc
  do
    /Applications/LibreOffice.app/Contents/MacOS/soffice --headless --convert-to txt:Text "$i"
done

I invoke it from cron like so:

6 7,9,11,13,15,17,19,21,23 * * Mon,Tue,Wed,Thu,Fri,Sat /usr/local/bin/python /path/to/markov.py /path/to/cigibot.txt

You can follow the original @beersnobsays and read his original blog post.