Maybe we can view these two things separately as they have not really anything to do with each other.
1) Removing some log priorities I guess that isn't too bad, but nonetheless I'd like to see some reasoning why you want to drop individual log levels. Otherwise it's just based on some feeling, and imo that's not good enough for potentially breaking third party code at this point. We are in a stabilizing phase so these BC breaks should already be kept to a minimum. Fabien has turned down things for this reason before, so it just seems consequential that we need to discuss this a bit more thoroughly before making this change. 2) Changing method names I don't think that it makes sense to prefix methods with "add". First, logging is always about adding. There is no reason to delete, or edit a log entry; there will never be methods like "removeDebug", "editDebug", etc. So, the prefix is not necessary, it's always implied if you call such a method. Second, this would be a departure from the syntax that many other libraries have for no good reason. To sum this up, I think 1) isn't so bad if you can substantiate your claims that they are not necessary (but I think a change here shouldn't be made lightly at this point). For 2) I don't think that it's necessary for the reasons stated above, and in fact might falsely suggest that you can perform other actions than "add". If you want consistency with Monolog, then please change it in Monolog. Kind regards, Johannes On Thu, Apr 21, 2011 at 9:23 AM, Jordi Boggiano <[email protected]> wrote: > On 21.04.2011 08:48, Johannes Schmitt wrote: > > Did you do any investigation what method names/log levels other > > frameworks use for logging? The current ones are not only close to > > syslog, or Zend, but also close to log4j, to Logbook (after which you > > have designed Monolog), and I'll probably find more if I search further. > > > > I feel that this change is a bit rushed at this point, and IMO it should > > only be made if there is a real benefit, or if the current system is > > broken which I fail to see right now. Why would we want to invent our > > own conventions now? > > First I'm not inventing conventions I'm adjusting to the conventions of > the framework. Second, Logbook only has CRITICAL, ERROR, WARNING, INFO, > DEBUG, it doesn't have the full stack of syslog levels, and they don't > support custom levels either (unlike ZF and log4j). > > ZF took the classic/academic approach (8 syslog-like levels), and I > don't think it makes any sense in the web context. > > > To address any possible confusion what the priorities mean, this mainly > > seems like a documentation issue to me. > > We're web developers, and I seriously couldn't find a good way to > explain where you should use ALERT over CRIT or ERR etc. To me > error/warning/info is more than enough, debug I just kept for a tiny bit > more flexibility into low level logging. > > Cheers > > -- > Jordi Boggiano > @seldaek :: http://seld.be/ > > -- > If you want to report a vulnerability issue on symfony, please send it to > security at symfony-project.com > > You received this message because you are subscribed to the Google > Groups "symfony developers" group. > To post to this group, send email to [email protected] > To unsubscribe from this group, send email to > [email protected] > For more options, visit this group at > http://groups.google.com/group/symfony-devs?hl=en > -- If you want to report a vulnerability issue on symfony, please send it to security at symfony-project.com You received this message because you are subscribed to the Google Groups "symfony developers" group. To post to this group, send email to [email protected] To unsubscribe from this group, send email to [email protected] For more options, visit this group at http://groups.google.com/group/symfony-devs?hl=en
