[sqlalchemy] Re: PickleType too small

2007-01-24 Thread milena
I have tried simple examples with PickleType and modified MyPickle (as suggested) and it works just fine. But, when I try it my segment of code, where I read row in a simple way and than put it in another structure, it didn't wanted to work (like unpickling wasn't working at all). So I did

[sqlalchemy] Re: PickleType too small

2007-01-22 Thread milena
Now that I have succeeded to write in my Pickle type into a database I have another problem. When I read it from the database how do I convert it to my original data (since I get pickle-file-like data after reading a row)? --~--~-~--~~~---~--~~ You received this

[sqlalchemy] Re: PickleType too small

2007-01-21 Thread milena
I have tried adding max_allowed_packet=16M to my.ini under [mysqld] but it didn't work. As far as I have seen BLOB type is used when mapping PickleType to MySQL. I would like to use LONGBLOB, so I did: class MyPickle(PickleType): impl = mysql.MSLongBlob and Column('data',MyPickle()) in a

[sqlalchemy] Re: PickleType too small

2007-01-21 Thread Michael Bayer
there is no MSLongBlob at the moment. there is an MSMediumBlob which comes out as MEDIUMBLOB, mysql says they are 16 megs in size. or try doing this: class MSLongBlob(MSBinary): def get_col_spec(self): return LONGBLOB On Jan 21, 2007, at 5:03 PM, milena wrote: I have tried