Andreas Reuleaux wrote:
Instead of defining individual columns
to hold unicode data in sqlobject
class Mytable(SQLObject)
mycol=UnicodeCol()
...
I would rather use normal string columns
class Mytable(SQLObject)
mycol=StringCol()
...
and declare the whole database to use utf-8.
I can do this in mysql
alter database mydb character set utf8
and then I can fill in unicode strings from python mysqldb
>>> ustr=someunicodestring
>>> import MySQLdb
>>> con=MySQLdb.connect(...)
>>> c=con.cursor()
>>> c.execute("insert into mytable (mycol) values (%s)", ustr)
but I can't from sqlobject:
# mycol=StringCol() as above
from model import Mytable
m=Mytable()
mytable.mycol=ustr
File
"/opt/tg/lib/python2.4/site-packages/SQLObject-0.7.0-py2.4.egg/sqlobject/main.py",
line 1031, in _SO_setValue
dbValue = from_python(value, self._SO_validatorState)
File
"/opt/tg/lib/python2.4/site-packages/SQLObject-0.7.0-py2.4.egg/sqlobject/col.py",
line 498, in from_python
return value.encode("ascii")
UnicodeEncodeError: 'ascii' codec can't encode character u'\xfc' in position 1:
ordinal not in range(128)
Any advice?
try "mytable.mycol=ustr.encode('utf8')".
if i understand your problem correctly, you told sqlobject that that
field contains non-unicode string. so when you pass sqlobject an unicode
string, he tries to convert it to a non-unicode string.
and it does it using the "default" charset for python. which is defined
as "ascii". and of course for most of the unicode characters, there's no
way to represent them in 'ascii'. that's why it fails.
so it's better to encode to utf8 before using sqlobject.
(assuming that really this was the problem ;)
gabor
