If we prefer it to be characters, then we should probably use
CHAR_LENGTH instead of DATA_LENGHT when reflecting, and add the CHAR
specifier to the column generators. I can put together a patch if that
helps.




On Aug 27, 10:02 am, jek <[email protected]> wrote:
> On Aug 26, 6:01 pm, Michael Bayer <[email protected]> wrote:
>
>
>
> > On Aug 26, 2009, at 8:53 PM, chris e wrote:
>
> > > I just checked the trunk, it the same reflection code is in place, as
> > > far as the column length is concerned.
>
> > > To me the question is, should sqlalchemy be aware of Char vs Byte
> > > storage?
>
> > > Is VARCHAR2(400) the same as VARCHAR2(100 CHAR), by storage size it
> > > is, but do we want storage size or number of characters?
>
> > > I'm not 100% sure. Anyone else out there using oracle, and have an
> > > opinion.
>
> > oh , I hadn't realized the VARCHAR2(100 CHAR) syntax.  in that case we  
> > should stick with data length.   the point is that a round trip  
> > reflection back to CREATE TABLE creates the same data type.
>
> > although a thornier issue is, should the number we send to String()  
> > attempt to resolve itself in terms of character length instead of data  
> > length.   For that I'd love for Jason to chime in but we might have to  
> > put out the bat signal on the roof for that to happen.
>
> IIRC SQL specifies the length of VARCHARs as characters rather than
> bytes, and I'd imagine that's what we're currently getting via
> reflection & DDL generation on most backends.  I guess I'd vote for
> the oracle dialect to do the CHAR annotation for generation for parity
> with String(...) on other backends.  The specter of doing char set
> detection & width math for reflection is pretty icky though.
--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups 
"sqlalchemy" 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/sqlalchemy?hl=en
-~----------~----~----~----~------~----~------~--~---

Reply via email to