TS>>No, you don't have to go that far. For starters, one could allow function TS>>(and possibly operator) overloading, based on type hints. The following is TS>>legal PHP5:
That will already open the can of worms. And make each function call to go through all the hoops of signature matching. TS>>That's it. That's all you need for function overloading. You can safely keep Then you couldn't make operator+ that adds an integer to complex (and make it different from one adding complex to complex ) - because typehints do not include integers. TS>>sure we'd find a way to deal with any "worms". Sure. Just make another C++ and you are OK. The problem is we don't want another C++. TS>>overload resolution. However, how much on an impact this would have on TS>>execution speed remains to be seen. Exactly. Do you have any base to claim the speed impact would not be serious? Bring it forward. TS>>It was a reaction to the oft-mentioned argument that essentially says: "We TS>>don't need more advanced features." I got that feeling when you said "Using Nobody ever said "we don't need more advanced features". You know perfectly it is not true. What was said is that we probably don't need *this particular* feature. This has no relation to discussion of any other feature or discussion about having features in general. Having discussion in family-fight style along the lines of "You never listen to me!" rarely results in anything useful. -- Stanislav Malyshev, Zend Products Engineer [EMAIL PROTECTED] http://www.zend.com/ +972-3-6139665 ext.115 -- PHP Internals - PHP Runtime Development Mailing List To unsubscribe, visit: http://www.php.net/unsub.php
