Andreas Hartmann wrote: > Michael Wechner schrieb: >> Andreas Hartmann wrote: >> >>> Hi Lenya devs, >>> >>> ATM we build the javadocs for all modules separately. This wouldn't >>> be necessary if all modules used a package convention, e.g. >>> >>> API: org.apache.lenya.modules.foo >>> org.apache.lenya.modules.foo.util >>> org.apache.lenya.modules.foo.xyz >>> >>> Impl: org.apache.lenya.modules.foo.impl >>> org.apache.lenya.modules.foo.util.impl >>> org.apache.lenya.modules.foo.xyz.impl >> >> >> what about custom packages? I guess you still would have to build them >> separately? > > What do you mean with "custom packages"? > > IMO the convention makes sense independently from the javadocs. For > instance, you can tell from the declaration of a sitemap component which > module it belongs to. > That is true. Today I often have to search for a component using find -name ;-)
> >>> I guess we'd have to postpone this to the next major release, though, >>> because it would break backwards compatibility. >> >> break backwards compatibility just because of javadocs? > > Like I said, the change doesn't make sense in the stable branch. > I think backward compatibility should only be broken due to really good reasons and I am not sure whether the above convention really justify it. If it is decided to break backwards compatibility we should at least present a migration path. Jann -- Jann Forrer Informatikdienste Universität Zürich Winterthurerstr. 190 CH-8057 Zürich oooO mail: [EMAIL PROTECTED] ( ) phone: +41 44 63 56772 \ ( fax: +41 44 63 54505 \_) http://www.id.unizh.ch --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
