This is working as designed. As with anything else, optimize as necessary, cache as necessary, etc. "As necessary" is the key word here. You are correct we must perform the work in mount twice, but you also gain other benefits so the performance concerns are more nuanced. For example, the data you fetch and hang onto on mount (current user, preferences, roles, etc) will not have to be fetch again for user interactions, which are very frequent. So on the whole, you may find your LV use reduces the load on parts of your system. If you do run into performance walls, then standard rules apply – introduce cache, pool, etc, but don't worry about it until you actually run into a bottleneck. When in doubt, measure :)
> --
> You received this message because you are subscribed to the Google Groups "phoenix-core" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to
phoenix-core...@googlegroups.com.
> To view this discussion on the web visit
https://groups.google.com/d/msgid/phoenix-core/m2mu2bx6u0.fsf%40achempion.com.