On 19 Sep 2009, at 17:19 , Joshua Russo wrote:
>> ... in fact using utf-8 string literals can cause problems in other
>> places
>> with code that assumes another encoding (e.g. ascii) for byte
>> strings.
>>
>
> Could you expand on this? I know that the Unicode string object has
> different methods than standard String, but are there other
> scenarios where
> a unicode literal could cause problems?
What Karen is saying is not about unicode literals, it's about
bytestring literals ('foo' not u'foo'): when libraries encounter
bytestrings they *might* want to decode them (depending on their needs).
Now to decode a bytestring, you need to provide a codec (an encoding),
so usually libraries either use a default (potentially overridable,
potentially not, through the call site or through a conf file) or use
the system's default encoding.
In the second case and if your bytestring encoding is different than
the system's default, then the library will blow up (note: there are
other similar cases, this is just an example) whereas with a unicode
string, the library wouldn't have needed to decode it (as it's already
decoded).
That's the kind of problems a non-ascii bytestring can generate.
--~--~---------~--~----~------------~-------~--~----~
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
django-users+unsubscr...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/django-users?hl=en
-~----------~----~----~----~------~----~------~--~---
