"Leopold Toetsch" <[EMAIL PROTECTED]> wrote > What about the following (also considering, that we might not like one > A dynamic PMC library has one init function returning a dyn_pmc_info > structure:
In the past, I've found the "init-fn returns struct" model can become a burden when, in a few years time, you want to change the info that's returned. Having the init function call a callback for each PMC that it wishes to register tends to be more future-proof (because you can define a different callback for the init functions that know about the extended registration info). Dave.
