Hi Vesselin,
I see the point and objections against "quick and dirty", but on the other
hand the discussion about the namespaces started long time ago - two years
already?
Longer than that - they were thrown out in 5.0 too.
If for two years there wasn't an agreement how they have to be
implemented (or even whether to add them at all! because I see many
comments
in this sense) what are the reasons to think that in the next 5-6 months,
when php6 will reach its "release stage" with the first alphas, this will
be
done?
The difference between 5.0 and now is that there wasn't an agreed base
implementation in 5.0. That much, we have now. The problem (which was only
defined clearly within the last few weeks) is that there's no consensus
about the extent of namespace support, ie whether or not to go beyond
classes as namespace elements. We have the choice of saying this will never
happen, or providing the means to make it possible for it to happen, and
that's where the discussion is now. If we provide the means to make
namespaced functions/constants possible at this stage, we won't be able to
change the chosen approach at a later stage even if it becomes apparent that
the approach is wrong. Also if we don't provide for it, we won't have the
option to change our collective mind later once namespace support is in PHP.
Even the namespaces dont go in php5.3 I think a solution/agreement has to
be
found quickly because it looks like php 6.0 could be in the same situation
after few months.
Moving it to PHP 6 allows more room for experimentation. We can try the ->
thing and see if it works well for people or if we need a new symbol to
express namespace elements, or even if the whole thing of function/constant
support is a Bad Thing for PHP. We don't have time to do all that before
5.3.0.
- Steph
--
PHP Internals - PHP Runtime Development Mailing List
To unsubscribe, visit: http://www.php.net/unsub.php
