This is the exact code that is running now:
log.info("Creating proxies for file %s" % job.file)
job = session.merge(job)
if agent.transcoder.run(job.local_filename):
log.info("Transcoding finished for file %s" % job.local_filename)
else:
log.error("Occurred a problem while transcoding the file. Please, verify
the logs.")
job.error_status = "Occurred a problem while transcoding the file. Please,
verify the logs."
(The problem is occurring on the log.info line)
As you can see, i *am* merging the job on the current session - so, the
exception shouldn't be happening.
Also, i use the local_filename variable before, so it's not expired (afaik).
On Nov 17, 2009, at 1:35 PM, Michael Bayer wrote:
>
> Fernando Takai wrote:
>>
>> I have changed my code to be like this:
>>
>> job = session.merge(job)
>> # Merge docs says that object does not get into the session
>> session.add(job)
>>
>> log.info("Job finished! %s" % job.file)
>>
>> When using latest SQLAlchemy trunk and 0.5.6 sometimes i get
>> UnboundExecutionError.
>>
>> Also, i can assure that all my attributes are loaded - i use all of them
>> before and no error occurs.
>
> but they may be expired at some point. If their host session is rolled
> back or committed, for example. The stack trace will illustrate exactly
> what action is initiating the load operation. For example, in the stack
> trace you posted, the "job.local_filename" attribute is unloaded or
> expired. You can test for this by seeing that "local_filename" is not
> present in job.__dict__.
>
>
>>
>> Maybe there's something wrong with the way i'm getting the session:
>>
>> session = Session.object_session(job)
>> if not session:
>> session = Session()
>
> there's nothing "wrong" with it per se except it appears to be guessing as
> to what session should be used - "well lets see if this object has a
> session from somewhere, or otherwise we'll just make one, maybe not".
>
> Sessions can be created as much as you like but they work best when they
> are in charge of their scope, i.e.:
>
> def do_something():
> # build the one session we care about for this thread/operation
> sess = session()
>
> # ensure everyone from outside is merged into our one session
> o1 = sess.merge(o1)
> o2 = sess.merge(o2)
> o3 = sess.merge(o3)
>
> # commit with authority
> sess.commit()
>
>
> as opposed to:
>
> def do_something():
> s1 = object_session(o1) or Session()
> s2 = object_session(o2) or Session()
> s3 = object_session(o3) or Session()
>
> s3.commit()
> s2.commit() ? or no ? where did this session come from ?
>
>
>
>
>
>
> --~--~---------~--~----~------------~-------~--~----~
> 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
> -~----------~----~----~----~------~----~------~--~---
>
--
Fernando Takai
http://twitter.com/fernando_takai
--
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.
