what gets me, is that after 15 iterations, it there are still conditions that keep the other files from loading. you would think that with the PK off and/or dummied that atleast it would match up again.
David E Jones sent the following on 11/26/2007 8:29 PM: > > The point of the graph traversal is that they are NOT a solution (but > rather proof that there is no solution down that road). That's why we > have the other two things I mentioned. > > -David > > > On Nov 26, 2007, at 9:23 PM, BJ Freeman wrote: > >> I did check my ML from 2005 to now before writing this. >> did try the current webtools import feature as well. >> will read up on thegraph traversal algorithms >> >> David E Jones sent the following on 11/26/2007 8:04 PM: >>> >>> On Nov 26, 2007, at 8:58 PM, BJ Freeman wrote: >>> >>>> I have been fumbling with this for a while. >>>> I made a reader reload to reload exported data as a transfer ability. >>>> >>>> so my question, from the architects. >>>> is there a sequence of all the entities that will allow them to load? >>> >>> There have been approximately 2000 mailing list messages on this topic. >>> >>> A single ordered list of entities by foreign key dependencies is >>> impossible, there are loops in them. If you want to research theory on >>> this check out graph traversal algorithms and the like. >>> >>> Use one of the various tools to get around FKs, like dummy fks or the >>> check/update database page that has tools to turn them on/off. >>> >>> -David >>> >
