Forum OpenACS Development: Re: Functions in .xql files...

Collapse
Posted by Don Baccus on
Jun's saying something I was arguing a long time ago when Richard Li and friends were convinced I was an asshole and an idiot (they were half right :)  More Tcl API abstraction is what is wanted and needed, no doubt about it.  Tcl is the level we work at when we write applications on top of the datamodel.

Janine ... C++ includes the "inline" function modifier so the kind of efficiency hit one sees in SQL need not be there.  So one can choose to be religious without paying for it in cycles.  Not true in our space (I'm just saying that the approach overall makes more sense in C++ and no sense at all in our world, but "more sense" doesn't necessarily mean "sensible" :)  If you force people to do it in C++ it doesn't cost them execution speed due to "inline" being available (unless they're too stupid to use it) and does leave the door open to later more tightly controlling who changes the data.

But in our SQL world you get no control and just loose efficiency.  No gain, pain.  Hey, that's not the prescription for building a stronger body!  Damn!