Programmer's Nightmare – Ruby Prank
github.comCritique:
1) no LICENSE file
2) not configurable
3) not idiomatic ruby (e.g. camelCase not snake_case)
4) overtly complicated. A better, and shorter version, is:
bad_whitespace = " "
ARGF.inplace_mode = ""
ARGF.each_line do |line|
print line.gsub(" ", bad_whitespace)
end
This is a sign of a college coder who had some class where the teacher said "you're required to have at least this much commenting", or a beginner programmer.
And finally, this is obviously a dumb toy program.
This isn't useful to show-hn. This isn't anything novel, or even interesting. It's not hard to do. It's poorly implemented and not portable (really, why require the user to edit the filename? ARGF exists for a reason).
This was also submitted by a new hn account mere hours after it was created (clearly by the author). I have no problem with showing off your own work, but at least let it cool down to see if it's a good idea; see that other people think it's interesting or useful in its own right.
I consider presenting an essentially 5-line program that is not generally useable or insightful as being basically spam.
I do recommend keeping on programming and, once you've got a project you're proud of that has seen some use besides yourself, seen at least one other happy user, and relates to HN's interests, post away!
If he's a beginner your comment might be a bit harsh.
If it's not interesting to you don't upvote the post but saying it's "dumb" and "ARGF exists for a reason" is just being rude.
Out of curiosity (haven't done ruby for awhile): is your line-by-line implementation safe? Is it guaranteed that the lines are identical in the output file besides the character you're replacing? I'm thinking different CR+LR combinations. What if the final line is not terminated by newline? Doesn't print always add it?
You're thinking of 'puts' which appends a newline.
'print' does not
The code does handle/preserve all the cases you mentioned (\r\n, final or no final terminating newline, etc).
Even worse, we actually had a regression at work caused by a zero width space sneaking into some code. Most editors won't even display anything for that character, even the code review tool we used didn't!
In the end I configured Emacs to highlight all unicode characters, trailing whitespace and tabs bright red. I also had it render zero width characters as normal spaces, otherwise they were still invisible. https://github.com/kzar/emacs.d/blob/master/init.el#L181-L19...
Sounds like some evil plan StackOverflow could use: inserting those characters randomly in the answers' codes to deter copy-pasting.
Along the same vein: Things to commit just before leaving your job - https://gist.github.com/aras-p/6224951
Why would you want to do that with ruby script instead of your favorite editor, or tr/sed on the commandline? And why would you write a "Close file" comment before file.close?
git checkout - uf