On Tue, 25 Aug 2009, Stanislav Malyshev wrote: > > Considering that there shouldn't be any errors in the first place, this > > You must be kidding me. Any fopen of non-existing file, and loading of > non-perfect XML from third party produces errors.
You can test for the first case, and the second one can be prevented by using the libxml_errors functionality. > There is absolutely no way > you can write code that does anything useful that would guaranteedly never > produce errors. So we should fix that instead. > They won't be fixed. I am getting an impression nobody's actually > reading the comments, only writing. Once again: I'm not talking about > people that are looking for errors in their code to debug, I am > talking about setups disabling errors ON PURPOSE, because they DO NOT > WANT them. Different use case. And the suggestion? Yet another ini setting (some even advocate making it INI_SYSTEM, which is obviously totally wacko)... This idea is an ostrich method. > BTW we discussed that in Chicago meetup and agreed we want to do it. > Of course, not that agreeing on anything ever would stand in a way of > wasting another bunch of time on repeating same stuff over and over > and nobody listening and blocking another useful feature. I don't find it useful, and obviously not everybody agreed with all the things in Chicago either. I didn't find this back in the notes either (http://wiki.php.net/summits/pdmnotesmay09). regards, Derick -- PHP Internals - PHP Runtime Development Mailing List To unsubscribe, visit: http://www.php.net/unsub.php
