My WIP on Unicode

[Isaac Bennetch]

Hi,

> On 2/10/15 7:17 AM, Nicolas Sebrecht wrote:
>  Hello,
> 
> Supporting Unicode is much harder than what it seems. The main reasons
> are:
> - Python 2.X sucks with Unicode;

Thanks for your research and work. Does Python 3 handle this better;
would it be easier to implement if you only targeted Python 3? If that's
true, I would consider that a reasonable compromise in order to add
decent Unicode support, if it makes development that much easier. 

Anyway, thanks so much for your work!

> - our codebase has too much enterlaced variables over objects and
>  modules;
> - each library (if not each module of libraries) handles Unicode
>  differently.
> 
> In the past weeks, I could not come with a fully working Unicode
> version. My first attempt has led to issues very hard to debug.  So,
> I've changed of paradigm but it led to a lot of breakages.  With so many
> breakages I've even started a new module to help debugging: I could not
> find any debugging tool out there which helps much for the purpose.
> 
> I believe they worth a look. I'll polish and publish both unicode
> attempts in a near future for review and comments. I'll try to finish
> and publish the debugging module, too.
> 
> Then, we'll have to decide if this is actually something we want and
> what strategy look better.
>