On Tue, Mar 16, 2010 at 02:21:44PM -0700, Jordan Brown wrote: > Nicolas Williams wrote: > >With extended templates svccfg(1M) is actually fairly high-level. > > It doesn't respect the hidden attribute, and it sounds like its > respect for "required" is marginal. I haven't checked its respect > for the other constraints.
Sounds like a bug. > >It's still limited in that complex "business logic" is still beyond SMF's > >grasp (hey now, don't go adding an extension programming language to > >SMF). > > Weeeellll... maybe. Programming language, no. Way to hook in > dynamically loaded validators? Maybe. The more that validation can > be embedded in SMF, the less that has to be implemented in the > various user interfaces, the less that gets duplicated between the > various user interfaces, and the more that those user interfaces are > consistent. I think the argument would be that if you need much more than what SMF extended templates give you then SMF should not be your project's admin UI. That said, I'm not opposed to external validators that svccfg(1M) can invoke. Executables or shared objects? The latter can be used in more contexts, but the former can be shell scripts. > Look at it like you might look at a DBMS: best practice is to put > as much validation and business logic in the schema as possible. Yes, but SMF is not an RDBMS (granted that it has SQLite2 under the covers). Nico --
