Filter out duplicate fortunes

[inguin]

Remove two completely redundant files from off/unrotated and a large number of exact duplicates (identified by script).

[shlomif]

@inguin : hi! Thanks for the pull-request! It looks fine at first glance, but I
need to write a script to validate it.

[inguin]

Makes sense. This is the hacky Python script I used: https://gist.github.com/inguin/f4e8e5f5fb80c447527f596295edaf5a

It just reads all files into a dictionary (text -> locations), then prints those with multiple locations. Might be worth implementing as a CI test, but my Perl skills are a bit rusty.

[inguin]

Just realized that my script can be easily tweaked to check that I didn't accidentally remove too much. Just print out all unique fortunes into a merged and sorted file with:

for text in sorted(locations_by_text.keys()):
    print(text, end="%\n")

If I do that before and after my changes I see that the following are gone:

• Empty fortunes (files beginning with %, or consecutive % lines)
• Three redundant fortunes with typos (in favor of the fixed ones)
• "Shit Happens." (in favor of "Shit happens." with different capitalization)

[shlomif]

@inguin : thanks again. I'll try to take a look

[shlomif]

@inguin : hi! I used a tweaked version of that python3 script to create my own filtering-own changeset. can you please license your code under MITL or similar ( https://github.com/shlomif/Freenode-programming-channel-FAQ/blob/master/FAQ_with_ToC__generated.md#i-want-to-release-my-code---which-open-source-licence-should-i-use ).

commit: 810a89e

thanks!

[inguin]

can you please license your code under MITL or similar

Sure, please feel free to use the script under whichever license is best for the project.

Is that sufficient or would you like me to make a commit adding a copyright header?