The core of the issue is that this: ub.book = book
places UserBook into the Session via cascades. Usually this is convenient. However, it can be tailored to behave more accurately. WIthout any changes, for this use case I usually just construct association objects with all their state at once: UserBook(book=mybook, user=myuser) Another method is to attach "book" after user.reads has been accessed: user.reads.append(ub) ub.book = book In some cases, when I have a lot going on that i need autoflush to not get in the way, I'll turn it off temporarily: session.autoflush = False <do stuff> session.autoflush = True some recipes for that are at http://www.sqlalchemy.org/trac/wiki/UsageRecipes/DisableAutoflush Then finally, you can make "ub.book = book" not actually place UserBook in the session by setting cascade_backrefs=False, docs for that at http://www.sqlalchemy.org/docs/orm/session.html#controlling-cascade-on-backrefs On Dec 18, 2011, at 7:05 AM, Manav Goel wrote: > Thanks for the answer. > Yes this comes in trace > File "/Library/Frameworks/Python.framework/Versions/2.7/lib/python2.7/ > site-packages/sqlalchemy/orm/query.py", line 2031, in __iter__ > self.session._autoflush() > > What is the workaround this? > > On Dec 17, 10:32 pm, Michael Bayer <mike...@zzzcomputing.com> wrote: >> On Dec 17, 2011, at 6:26 AM,ManavGoelwrote: >> >> >> >> >> >> >> >> >> >>> Hi >>> I created a many to many relation between two tables as shown >>> in the tutorial using Association object as I have extra fields in >>> connecting table. >>> I have table User(user_id), Book(book_id) and UserBook(user_id, >>> book_id) >> >>> This code gave me Integrity error of UserBook.user_id of being null. >>> Basically it is not user id of user automatically to UserBook object : >> >>> def con(user, mybook): >>> ub = UserBook() >>> ub.book = mybook >>> user.reads.append(ub) >>> session.commit() >> >>> But this code worked : >>> def con(user, mybook): >>> ub = UserBook() >>> ub.book = mybook >>> ub.user = user >>> user.reads.append(ub) >>> session.commit() >> >>> Why the first one is not working? >> >> It's not certain without a full example + stack trace, however often this >> occurs because of autoflush (perhaps when user.reads hits the DB to load >> itself first). The stack trace would illustrate if the integrity error is >> due to a premature autoflush (you'd see _autoflush() in the trace), and >> there are ways to get around that issue. > > -- > You received this message because you are subscribed to the Google Groups > "sqlalchemy" group. > To post to this group, send email to sqlalchemy@googlegroups.com. > To unsubscribe from this group, send email to > sqlalchemy+unsubscr...@googlegroups.com. > For more options, visit this group at > http://groups.google.com/group/sqlalchemy?hl=en. > -- You received this message because you are subscribed to the Google Groups "sqlalchemy" group. To post to this group, send email to sqlalchemy@googlegroups.com. To unsubscribe from this group, send email to sqlalchemy+unsubscr...@googlegroups.com. For more options, visit this group at http://groups.google.com/group/sqlalchemy?hl=en.
