Re: [sqlalchemy] Re: Strange issue with SQLAlchemy History and generating unnecessary updates

2016-09-16 Thread 'Nicholas A Fries' via sqlalchemy
Hi Mike, Thanks for the detailed explanation and example code. I suspect this is exactly what is happening (that the deserialization library is using add instead of merge) - I will look over the code and confirm. Thanks! On Fri, Sep 16, 2016 at 7:05 AM, Mike Bayer

Re: [sqlalchemy] Re: Strange issue with SQLAlchemy History and generating unnecessary updates

2016-09-16 Thread Mike Bayer
On 09/15/2016 11:22 PM, 'Nicholas A Fries' via sqlalchemy wrote: Hi Mike. Thanks for the reply and clarification on how the history system is implemented. I will investigate further and review the code you mentioned. Right now, I can see that get_history() is showing changes for one of the

Re: [sqlalchemy] Re: Strange issue with SQLAlchemy History and generating unnecessary updates

2016-09-15 Thread 'Nicholas A Fries' via sqlalchemy
Hi Mike. Thanks for the reply and clarification on how the history system is implemented. I will investigate further and review the code you mentioned. Right now, I can see that get_history() is showing changes for one of the relationships (an add and a delete of an object with the same type

Re: [sqlalchemy] Re: Strange issue with SQLAlchemy History and generating unnecessary updates

2016-09-15 Thread Mike Bayer
On 09/15/2016 08:04 PM, 'Nicholas A Fries' via sqlalchemy wrote: I have a typo for item #7 in my post - this is unexpected. Cases 7 - 10 are the problem. We don't expect there to be changes showing for data that is the same as in the database. having a hard time following this. In your

[sqlalchemy] Re: Strange issue with SQLAlchemy History and generating unnecessary updates

2016-09-15 Thread 'Nicholas A Fries' via sqlalchemy
I have a typo for item #7 in my post - this is unexpected. Cases 7 - 10 are the problem. We don't expect there to be changes showing for data that is the same as in the database. On Thursday, September 15, 2016 at 5:01:40 PM UTC-7, Nicholas A Fries wrote: > > Hi guys, > > The history system in