DJ Delorie <[EMAIL PROTECTED]> wrote: > >> For the time being, however, I don't see any reason not to stick to >> using some sort of external helper programs for actually accessing >> the database, as long as we can get the UI and basic Scheme API >> nailed down. > > I've argued in the past that gattrib should be responsible for > attributes, not gschem.
Yes and no. Gatrib is good to add/change attributes *after* you have a schematic design. It is a nice tool, but I think gschem too must be capable to add attributes. That should be the primary attribute adder tool. > And, there's no reason why the attributes need to be in the schematic! > We can store them in the BOM. There's a pleasing symmetry to it: > > gschem uses *.sym files to make *.sch > gattrib uses db files to make *.bom > pcb uses *.fp files to make *.pcb > > The only interesting change is to gsch2pcb, which needs to read a > *.bom file in addition to the schematics. IMHO this just adds complexity. What we need is an abstract layer on the top of symbols, not between schematic and pcb, in a sense when you producing pcbs, not simulating a model. >> > The symbol files should be as light as possible, and we should make them >> > heavy by adding information coming from the database. >> >> Totally agree. > > I agree symbols should be as light as possible, but "adding to them" > isn't a requirement - we can store them in the BOM. If we store the > attributes there then we also have a clue about which attributes came > from which programs, and the bom can store more than just a name/value > pair - it can store an "origin" value (how the value got chosen), for > example. That is true. But why do we need this? Why should pcb know that the attribute "value=LM2596" came from a database, or from manual input, script generated? -- Levente http://web.interware.hu/lekovacs _______________________________________________ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user