I am interested in these discussions but it seems to me that far too much is being written and I apologise if I haven't read all the references which have been proposed.
The fact is that Hierarchical data does not fit at all well into a RDBMS. The main issue is therefore to decide what 'kludge' to use. The best kludge will vary case by case so the ORM would have to be incredibly sophisticated to always choose an optimised kludge without the developer specifying exactly what he wants. I do not know how SQLA achieves this, but I cannot imagine the DAL being developed to cover all cases in the best possible way. Nevertheless it would indeed be nice to cover at least one common use-case of hierarchical data and I would suggest that someone should propose a specific case. Here is a simple example: geographical areas. I would propose either a nested set (or perhaps simply a materialised path, which I like very much for its simplicity). I think such an example would be very useful. A few simple functions to manipulate this included in the DAL would be really great. I believe it is only with a specific proposal in mind that this discussion will progress. A theoretical discussion will remain just that and nothing practical will result. There are truckloads of stuff already written on this topic and I cannot imagine that this thread would add anything new to that. -- --- You received this message because you are subscribed to the Google Groups "web2py-users" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. For more options, visit https://groups.google.com/groups/opt_out.
