Re: cvs: php4 /ext/standard incomplete_class.c php_incomplete_class.h type.c
| From: | Yasuo Ohgaki | Date: | Wed, 24 Jul 2002 11:20:00 +0000 |
| Subject: | Re: cvs: php4 /ext/standard incomplete_class.c php_incomplete_class.h type.c | ||
| References: | 1 2 | Groups: | php.cvs php.dev |
| Request: | Send a blank email to php-dev+get-86154@lists.php.net to get a copy of this message | ||
derick@php.net wrote:
On Wed, 24 Jul 2002, Yasuo Ohgaki wrote:I thought we've finished since there is no reply within 24H from last post.yohgaki Wed Jul 24 05:55:11 2002 EDTPlease discuss this first, I objected to this, and I still think this isModified files: /php4/ext/standard incomplete_class.c php_incomplete_class.h type.c Log:is_object() returns FALSE if object is a "incomplete object". Raise E_NOTICE, instead of E_ERROR, for setting/getting properties to/from a "incomplete object".
not a good idea to be able to set and get properties from half-baken objects. (And I'm not tired this time :)Yes, it's not good thing to do, but it's perfectly worked with older versions w/o any warnings nor errors. Besides, PHP4 does not have property protection. Raising fatal error for accessing incomplete object property is too much. It _BREAKS_ scripting take advantage of PHP4's object behavior. I don't think breaking compatibility is not needed now. However, I'm not against to raise fatal error in PHP5. PHP5 breaks scripts in many ways. Why don't we save it for PHP5? It's not fun to rewrite scripts many times. Let's break them at once and rewrite at once. :) -- Yasuo Ohgaki