Hi Massimo, I thank you for adressing this thread, and apologize that it became so long and verbose, and that you are tackled with this ORM-stuff "yet again"... I imagine it might feel tedious by now, considering the amount of discussion that has already been done over this issue in the past.
However, I still think it should be re-visited. I had already gone through the link you posted, before - it had appeared in many past-discussion that I had read in this group when searching for 'ORM' and SQLAlchemy" keywords. I was disappointed to discover that this example is proorly-formed. The comparison ther, (as well as in other places, like this: http://web2py.com/examples/static/sqla2.html) feels somewhere between ill-informed to disingenuous. It completely overlooks the actual "usage" of the design in SQLA that would be manifested once this structure is defined there. The ORM features are not even mentioned - it onl compared the DAL's "syntax" to the SQLA's SQL-Expression DDL/DML syntax - there is no reference to semantic-meaning that pertain to the usage of SQLA's ORM afterwards... The reason, as it seems (and as is evident in the link I gave here), is that ther IS NO comparable semantics in the DAL, or in web2py in general - it simply does not exist there. Now, I don't imply that there was any malicious-intent on your-part, to misdirect the reader - I can't know that - but I AM mentioning the feeling I got when I read it, just so you know - it felt like a bad rep for web2py, and so was disappointing. I would have expected a more revealing description of what web2py is missing in comparison - it might not have given web2py points in feature-support, but would have given it many points in integrity and sincerity. But that may just be my personal interpretation... I think you should be proud of web2py for it's achievements, and be confident enough about it's capabilities to admit it's limitations front and center. I think that would help a lot for web2py's rep. anyways, I was not looking for a SQLA-bridge - as I said, I don't like their SQL-Expression syntax, and would much rather still use web2p's DAL. Hell, we have thousands of lines of code using it... It would be a nightmare to switch-gears at this point. No. What I AM asking for, is extra tools in-and-around the DAL, to help and facilitate the construction of custom ORM classes around it. As for your usage of the DAL, that is actually exactly what I had in mind when I was thinking about a back-end for a GUI-Schema-designer. This is a very declarative style of defining schema-metadata, so it could very well be serialized and stored in a declarative format, that could then be used in a GUI-Schema-Builder. It can obviously also be used to separate the concerns and build ORM classes around. It's the same kind of separation that I would expect between ORM classes and business logic classes - having a single-source-of-truth that is framework-neutral. Very cool stuff... -- --- 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.
