Hi Peter, On Dienstag, 7. Oktober 2008, Peter Clifton wrote: > On Tue, 2008-10-07 at 20:26 +0100, Peter Clifton wrote: > > Well, I guess we could convert the component selector to opening > > the symbol with some equivelant to o_complex_new (), then grabbing > > them as real attributes. It would make the code handling it more > > generic with the rest of gschem's attribute handling. > > Having said this, since you're not after the same list as the > promotable attributes, there is no point contriving to re-use that > function. > > The implementation you linked to is pretty succinct anyway, so it is > probably useful to iterate over things the way you've done it.
Yes but I'm iterating over the whole object list multiple times. (once for each string in the filter list). Thus filtering out all attributes from the symbol once and then just loop over this short list is much better. (2 commits of attr2 branch) http://repo.or.cz/w/geda-gaf/werner.git?a=shortlog;h=refs/heads/attr2 > BTW.. Is there any case you know of where you actually _want_ to > filter out any of the toplevel attributes? The filter list sorts the attributes, too. This is even the main reason why I've created the filter list. Maybe I can add a special filter list that catches all attributes: (component-dialog-attributes '(*)) > The only case I can think of, would be if its presence invoked some > other action. Like, for example, you had a http hyperlink to the > documentation URL outside of the attributes tree. > The only differentiation I might consider making would be to pick out > the attributes which would be promoted as bold, for example. The > function which identifies that on a per-attribute basis is > o_complex_is_eligible_attribute (). (Currently a static function in > o_complex.c) Nice ideas. Regards Werner _______________________________________________ geda-dev mailing list [email protected] http://www.seul.org/cgi-bin/mailman/listinfo/geda-dev
