RE: [PHP-DEV] Re: new construct
| From: | John Coggeshall | Date: | Tue, 25 Feb 2003 16:58:16 +0000 |
| Subject: | RE: [PHP-DEV] Re: new construct | ||
| References: | 1 | Groups: | php.dev |
| Request: | Send a blank email to php-dev+get-94565@lists.php.net to get a copy of this message | ||
I was asking myself -- I had assumed that __construct() would be
searched for first. I was /am under the impression __construct() is a
special function that the engine wouldn't allow you to use in PHP5 in
any other context than its intended purpose.
I don't know what Zeev plans on doing with this, but perhaps an option
would be to issue a notice or warning if both A::A() and __construct()
exist in a single class A and call __construct() in those cases...
Just a thought.
John
>Perhaps i may be mistaken, but it seems logical to search for
>the old-style constructor first in order to be backwards
>compatible with old-style scripts. Putting __construct in old
>style classes would yield a classname of '__construct', an
>ugly and unlikely name for a class. Seeing as A::A() would be
>much nicer, i can see the logic behind parsing the old-style
>first, rather than the new-style.
>
>That's just my two cents, feel free to correct me otherwise..
>
>~ Andrew Heebner
>
>>
>> >__construct is the new method of defining a constructor, but the
>> >'bug' you suspect is not a bug. The parser will search for a
>> >function of the same name in the class as the constructor for
>> >backwards compatibility with Older scripts, etc...
>>
>> But shouldn't __construct() be searched for and used prior to the
>> old-style constructor? Intuitively it seems it should.
>>
>> John
>>
>
>
>
>--
>PHP Development Mailing List <http://www.php.net/>
>To unsubscribe, visit: http://www.php.net/unsub.php
>
>