On Tue, 2006-09-19 at 18:33 +0000, Pupeno wrote: > On Tuesday 19 September 2006 02:58, Malcolm Tredinnick wrote: > > That's true. We aren't using unicode universally at the moment. There is > > work in progress to fix this, but it has no completion date at the > > moment. > > So, I can't store Unicode strings on PostgreSQL and expect it to work under > any conditions with any possible workarounds ?
No, that isn't correct. You can store an encoding of unicode strings in your database (e.g. UTF-8), providing the database is set up to handle it. When you read it back you will get back a string of bytes. If you need to convert that string to a Python unicode instance, you will need to know something about the encoding used. What we aren't doing is transparently converting everything to unicode instances yet, and your original question was asking about that, since you said SimpleTAL was expecting only unicode instances and not string instances. This distinction hinges off the difference between an encoding of a unicode string (which is what you store in the database or send back to a client) and the unicode string itself, independent of any encoding. Although I have not used non-ISO-8859-1 storage much at all, I know that there are some people around who have been using it for a while now quite successfully, so almost all of the problems are easily solvable with current Django code. One day I hope to find time to play some more with this so that I'll know all the tricks, but at the moment I'm not the expert on the "how" side of things. Regards, Malcolm --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "Django users" group. To post to this group, send email to django-users@googlegroups.com To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/django-users -~----------~----~----~----~------~----~------~--~---
