Fix regexp to use case-insensitivity correctly #889
Conversation
The regexp /[A-z]/ matches all alphabets, but also six non-alphabetic
characters, as demonstrated by ("A".."z").to_a.join
=> "ABCDEFGHIJKLMNOPQRSTUVWXYZ[\\]^_`abcdefghijklmnopqrstuvwxyz"
|
Good catch, thank you. There's a failure in the integration test ( Also, we might want a unit test to protect against regressions here. What do you think, @rkh? Too paranoid? |
|
The test failure from Travis seems unrelated to my changes. I copied the relevant part here, these two warnings were the cause:
Also, the tests ran successfully on my own machine. No idea what causes those... |
|
Warning seems unrelated to this patch. |
|
I have no idea where this issue comes from all of a sudden. Test would handy, indeed. Unless I'm missing something, the main difference will be that |
I noticed a faulty regexp in Sinatra's uri():
The regexp /[A-z]/ matches all alphabets, but also six non-alphabetic characters, as demonstrated by ("A".."z").to_a.join
=> "ABCDEFGHIJKLMNOPQRSTUVWXYZ[]^_`abcdefghijklmnopqrstuvwxyz"