UTF normalization middleware

[ghost]

used unicode_utils

[sferik]

Looks good to me. Arguably, we should be normalizing in Faraday itself.

[mislav]

It could use some documentation, at least on the class level. For instance, it could say what it actually does, because I don't know what "unicode normalization" is or when do I need it

[ghost]

The first links in Google gives a complete understanding of the subject

http://www.unicode.org/faq/normalization.html
http://unicode.org/reports/tr15/

[mislav]

So, when is normalization needed?

It doesn't seem to me like the answer is "all the time". If I'm not comparing strings, I guess I don't care if they are normalized or not?

[ghost]

Not normalized utf can bring many surprises when regexp matching, or if the unnormalized faraday response store in the database, then the selections can be incorrect result (the same person's name can be composed of different characters UTF), all of them can be reduced to a "string comparison", but this is not quite correct.
I needed this functionality to remove "junk" characters from the text.
Hardly need all this functionality, but it's pretty useful.

[ghost]

extracted this functional in a separate gem faraday_utf