On Thu, 2008-04-10 at 20:26 +0200, Gábor Farkas wrote: [...] > the point is,that in my opinion, 99% of all developers want to have > their unicode-data normalized, before they process it. (maybe they > don't know yet that they want it, but they want it :-)
87.35% of all statistics are just made up on the spot. :-) A large bunch of the time you just won't care. If you're just serving back the data that was entered and you don't need to search over that data, or you know that your input data is very unlikely to carry anything in the ambiguous encoding sections, you might well choose to forgo the extra processing overhead. Normalisation isn't a free operation. It involves a reasonable amount of table lookups and, ultimately, another linear pass through all the input data for decomposition and recomposition. We need should be careful about the extra overhead being introduced. I think it's not a bad idea to add this as something that is possible, but it's probably a bit costly to do on every request (based on some simple timings I've done just now -- although I want to look at it further). Don't blow things out of proportion by trying to claim it's a no-brainer, though. That just hurts your argument. Definitely worth looking at as an option on an HttpRequest, though. Regards, Malcolm -- Telepath required. You know where to apply... http://www.pointy-stick.com/blog/ --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "Django developers" group. To post to this group, send email to [email protected] To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/django-developers?hl=en -~----------~----~----~----~------~----~------~--~---
