Re: ANN: Alpha release of new MS SQL Server backend

2007-08-06 Thread mamcx
Thanks for the info. I test it and tommorrow in the next 2 days do another release... --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups "Django developers" group. To post to this group, send email to django-developer

Re: ANN: Alpha release of new MS SQL Server backend

2007-08-06 Thread pk11
hey, sorry, forgot one thing: in django/db/models/fields 222 # mssql can not handle more than two 99-s 223 return ['%s-01-01 00:00:00' % value, '%s-12-31 23:59:59.99' % value] On Aug 3, 6:34 pm, mamcx <[EMAIL PROTECTED]> wrote: > Excelent. Please send me your patch so I

Re: ANN: Alpha release of new MS SQL Server backend

2007-08-06 Thread pk11
mamcx, I sent you my mods. Jeremy, I tested the admin + pymssql backend again without my modification (i.e. changing object_id's type), and now seems to be working just fine. do not really know what was causing the issue before though. anyway, thanks for the information. Peter On Aug 4, 9:21

Re: ANN: Alpha release of new MS SQL Server backend

2007-08-04 Thread Jeremy Dunck
On 8/3/07, pk11 <[EMAIL PROTECTED]> wrote: > the issue with object_id is that mssql can not insert an integer PK > into a TEXT field. > I guess I'm confused. Admin LogEntryManager forces object_id to a string before saving. Do you see somewhere that's directly filling LogEntry.object_id with

Re: ANN: Alpha release of new MS SQL Server backend

2007-08-03 Thread pk11
hey, i will be able to send you my patch on monday. the generic idea is to attach LIMIT as if the backend supported it and at query execution time replace it with TOP. OFFSET can be handled by nested queries but i did not have enough time to finish that. (i.e. http://groups.google.com/group/djan

Re: ANN: Alpha release of new MS SQL Server backend

2007-08-03 Thread pk11
Hi Jeremy, thanks for the info, I meant mssql, sorry. the issue with object_id is that mssql can not insert an integer PK into a TEXT field. so how should it be handled then if the field type can not be changed? peter On Aug 3, 9:02 pm, "Jeremy Dunck" <[EMAIL PROTECTED]> wrote: > On 8/3/07

Re: ANN: Alpha release of new MS SQL Server backend

2007-08-03 Thread Jeremy Dunck
On 8/3/07, pk11 <[EMAIL PROTECTED]> wrote: > - django's admin log model object_id's type is changed from text to > integer (is there a reason why object_id is declared as textfield?) object_id is whatever the arbitrary model's PK is, and you can't assume it's an integer in the general case. The

Re: ANN: Alpha release of new MS SQL Server backend

2007-08-03 Thread mamcx
Excelent. Please send me your patch so I take a look at it. I finish today a implementation of LIMIT so I'm interested in how you solve it. --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups "Django developers" group

Re: ANN: Alpha release of new MS SQL Server backend

2007-08-03 Thread pk11
hey all, we are also using django with pymssql/mssql, in addition to your work, we changed: - django's admin log model object_id's type is changed from text to integer (is there a reason why object_id is declared as textfield?) - pymssql is changed to allow unicode strings - simulate limit with

Re: ANN: Alpha release of new MS SQL Server backend

2007-08-03 Thread mamcx
Yes... I'm waitting to see if the actual pathc for pymssql is enough or still need more work for do a single submit... --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups "Django developers" group. To post to this grou

Re: ANN: Alpha release of new MS SQL Server backend

2007-08-03 Thread Jacob Kaplan-Moss
Hey Mamcx -- On 8/2/07, mamcx <[EMAIL PROTECTED]> wrote: > I submit a ticket with patchs for enable a new backend, so we can have > a clean direction from the ado_mssql confusion of tickets. Lookin' good -- keep up the good work! I know you're very much in the early stages of this, so I'll hold

ANN: Alpha release of new MS SQL Server backend

2007-08-02 Thread mamcx
I submit a ticket with patchs for enable a new backend, so we can have a clean direction from the ado_mssql confusion of tickets. Is at http://code.djangoproject.com/ticket/5062. I run my own test and the standar django ones (however I'm very intrigated by the fact I found only 2 problems here..