Why would they have to run phing tasks? Wouldn't *you* compile the configs and distribute them as part of the installation package that you send to the client? --Bob
On 11/24/05, Markus Lervik <[EMAIL PROTECTED]> wrote: > Veikko Mäkinen wrote: > > >> Proposal: I back this DONT_COMPILE_TEH_CONFIGS_YOU_RETARD flag out, > >> and we > >> add such a functionality together with the phing task in 0.11. I could > >> also > >> imagine this: with AG_DEBUG set to off, config checks aren't performed, > >> nothing is compiled etc, and you'd have to call the task manually. I > >> think > >> this is even better, isn't it? In production environments, configs change > >> almost never, so it would be fine to do it this way. > > > So: > > - I'm in favor of keeping the current behaviour (debug=on/off) and only > > adding more choices (cache time-to-live and/or cache "freezing"). > > I agree on this one; I'd rather go for what me and Veikko talked about > a few nights ago; having an AG_CACHE_TTL(?) setting where -1 would mean > never compile the configs, 0 would mean always compile the config and > anything else would mean compile the cache after a certain amount of > seconds/minutes, ie. AG_CACHE_TTL = 3600 would mean recompile the > cache once per hour. > > I certainly wouldn't want to require my clients to pop up a terminal, > ssh into the server and start running phing tasks. > > > - markus > > -- > Markus Lervik, CTO | "In God we Trust - all others > Necora Systems Ltd | must submit an X.509 certificate" > http://www.necora.fi | - Charles Forsythe > [EMAIL PROTECTED] | (X.509 certificate available at > +358-40-832 6709 | http://www.necora.fi/markus/x509.crt) > > > _______________________________________________ > agavi-dev mailing list > [email protected] > http://labworkz.com/cgi-bin/mailman/listinfo/agavi-dev > > > > _______________________________________________ agavi-dev mailing list [email protected] http://labworkz.com/cgi-bin/mailman/listinfo/agavi-dev
