Here’s another little Bash script I whipped up. If you’re hoping to take the GRE, you’ll know that it includes two essays or analytical writing tasks – one ‘argument’ task and one ‘issue’ task. ETS, the maker of the GRE, lists the entire pool of topics on their website. This script picks a random essay topic from the pool of GRE argument and issue topics.

First I saved the two webpages and extracted the content from them. I converted the resultant files into Markdown with judicious use of regex substitutions. (I could have used pandoc, but that’s just occurring to me now as I write this…)

Normally, you can use shuf to get a random subset of lines from a file. But here, all the topics have multiple lines/paragraphs. I replaced the horizontal rules between topics with --- in the resultant Markdown, so I decided to split each topic into its own file using:

$ awk '{print $0 ""> "issue" NR}' RS='---'  ../issue\ topics.md
$ awk '{print $0 ""> "argument" NR}' RS='---'  ../argument\ topics.md

(These two commands are in their own respective directories, for clarity.)

So now, we can use shuf to get a random file from each directory!

I wrote a quick script that picks an essay prompt for each. View it here.

Let me explain a few lines.

datestr=$(date '+%Y-%m-%d')
filename=$datestr"_AWA_topics.md"
touch $filename

This creates a file with today’s date, e.g. 2019_12_28_AWA_topics.md.

argpromptfile=$(ls /home/tesserakt/College/Fall_2019/GRE/arg_pool | shuf -n 1)
issuepromptfile=$(ls /home/tesserakt/College/Fall_2019/GRE/issue_pool | shuf -n 1)

This lists the files in each directory – remember, each file contains a single topic – and run shuf on those, to get one topic for each. The chosen filename is stored in the variable.

  1. After that, we cat the files and append them to our initial file. (>> appends, whereas > overwrites.)

   sed ':a;N;$!ba;s/\n\n\n/\n\n/g'
   $filename > $filename.tmp
   mv $filename.tmp $filename
   cat $filename

The sed looks cryptic – basically, we replace triple newlines \n\n\n with double newlines \n\n. This preserves spacing between paragraphs while removing excess whitepace. (Check out the link below for a more detailed explanation of this command.)

Finally, we view the file with our two topics. Et voilà!

References: