News for hackers. Stuff that matters.
Re:Unicode Dicks by rogerbinns on Wednesday October 31, 2018 @03:09PM (#18348439)
> no knowledge of how languages are formed

I recommend reading this article. The problem he is solving is representing text in ASCII. For example an e with an accent can be replaced with a plain e. But what about languages not using the base Roman alphabet? You get a fascinating tour of Divehi and the final result - the unidecode module - works rather well.

http://interglacial.com/tpj/22/

Re:Unicode Dicks by logfromblammo on Wednesday October 31, 2018 @04:15PM (#18348918)
> "I am rather sure, however, that it cannot be done with a mere regular expression, and that is not something I say lightly!"

This sounds like a challenge.