What are the usecases?

Also, I'd call it swap_case. It's not the same kind of name as upcase.

Mmh, I haven't used swapcase in Ruby yet and never missed in Crystal, so I'm curious about its usecases :)

Then again, even Python has a swapcase method (named like this too), so it's OK

I was just porting some Ruby code to Crystal that used swapcase to randomize character cases in randomly-generated strings. Thought it would be nice to have it for compatibility.

The name is for Ruby equivalence (http://ruby-doc.org/core-2.2.0/String.html#method-i-swapcase).

Can you link/show the Ruby code?

Original gist: https://gist.github.com/lpar/1031933

Well, there's no need for swapcase in that example. The source text is guaranteed to be lowercase and the comment right above it even says that the intent is to upcase a letter, so using upcase is just fine here.

Yes, obviously the linked code isn't great code, that's not really the point. There are so many different ways to write that, we could go on about it all day.

In porting the concept over I just happened to notice that a simple Ruby function wasn't available in Crystal and took the few minutes to add it. People are coming to Crystal from Ruby and I thought it would be nice to have that additional singular point of compatibility.

Feel free to accept or reject, it's your language, I don't feel strongly enough either way to continue this.

I also can't find a strong use case for this. Searching in Google and GitHub I found two use cases:

1. Matching a password as is and with swapcase, in case the user had caps lock on (well, let's say a "use case")
2. Use in competitions that ask a user to implement a swapcase function. In Ruby and Python one uses swapcase and that's it, problem solved.

Given that the implementation is trivial (at least seeing the string as ASCII) and that it's unlikely to change in the future, I wouldn't mind including it in the standard library. At least it doesn't bother.

A note: the linked code mutates a String, so that will anyway fail to compile and you'll have to do it in a different way... though maybe it would be nice to have a String method to replace a char or a string at a given position/range with another char/string/object. Ruby uses []= for that, maybe we can add a replace_at(index, value) and replace_at(range, value), or maybe just reuse the sub name?

sub(Int, Char) and sub(Range, String) should be okay.

👍

Since #3560 is merged this PR should take care of passing Unicode::CaseOptions.

I don't think there's a strong use case for this, so I'm closing this.