On Jan 28, 2006, at 2:52 PM, Michael Bayer wrote:
What this is doing is taking the usual dependency sort at the table level, detecting if there are cycles, and if so breaking it down to a dependency sort at the object instance level. So you still need to ultimately not have a circular relationship between two individual rows; i havent figured out how that situation should be handled yet. do you see your tables having such cycles within them, i.e. two rows pointing to each other ?
just to be clear, in my actual application i currently only have self-refs in a single table - the question regarding multiple-table cycles just came up as i was skimming topological.py trying to get a sense of how the circular case was handled. however, if i'm following what you're doing correctly then this doesn't matter anymore? it sounds like you're doing general cycle detection at the table/class level, and dropping back to row/instance topo-sorting if you detect a cycle? (an aside - do the tasks get partitioned based on connected-components? i.e., if there was one cyclic task and a bunch of non-cyclic ones, is the "extra work" isolated to just the cyclic task? ... i still don't entirely get how the task hierarchy is constructed). in any case, to answer your question, within my actual single self-ref table, i may occasionally need to create row-cycles. (such a cycle is "wrong" based on the semantics of our table, but sometimes we have to store incorrect data for later analysis ... ) finally, i seem to have mostly worked out what i need to integrate sqlalchemy into my main application, so i'll probably not be dropping by this list as much, but i did want to say that i really appreciated the consistent support that i got in response to my rather numerous questions over the past couple of weeks. thanks! d ------------------------------------------------------- This SF.net email is sponsored by: Splunk Inc. Do you grep through log files for problems? Stop! Download the new AJAX search engine that makes searching your log files as easy as surfing the web. DOWNLOAD SPLUNK! http://sel.as-us.falkag.net/sel?cmd=lnk&kid=103432&bid=230486&dat=121642 _______________________________________________ Sqlalchemy-users mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/sqlalchemy-users
