> You didn't CC p5ee maillist. Intentionally? No, it was an oversight. I forwarded the message when I noticed I'd forgotten.
> > I haven't looked at the Attribute::* modules yet, so they may > > follow this > > same philosophy, but if you find yourself having to "use > > Attribute::*;" in > > your module in order to take advantage of it, then it falls into > > the runtime trap. > > Yes, that's the case. All those modules have runtime overhead and we > need to live with it. That's not critical for webservices where you > need this information anyway, but it may be not feasible for other > cases, when you don't interested in description. My point is that I don't believe we do have to live with it. I gave this a lot of thought before I plunged into ModInfo, and the "must evaluate to inspect" problem is one of the major factors that drove me to the "sidekick document" solution. Modules can potentially do undesired things at runtime, not the least of which is load even more modules that chew up more RAM. It's just not economical to require evaluation when there's an alternative. BTW: Is this thread off-topic, or is it a philosophical discussion in regards to the documentation/introspection facilities of p5ee? I'll defer to the opinions of the list-members. jpt
