exactlyOriginally Posted by Archbob
| SitePoint Sponsor |
exactly
Maybe I am missing something, but why is Unicode support such a big leap?
Becuase alot of stuff needed to be changed to acheive it.
There aren't going to be too many changes to OO or anything as far as I can tell from the internals list. We might finnaly get some sort of namespace support however.
Im glad to see an opcode cache included by default(And of course to see register_globals and magic_quotes dissappear)
~ Eric
Eric Coleman
We're consentratin' on fallin' apart
We were contenders, now throwin' the fight
I just wanna believe, I just wanna believe in us
Well yes I realise a lot had to be changed, but why is it so important is more of what I was asking? I am still a little confused about what kind of new functionality it will offer.Becuase alot of stuff needed to be changed to acheive it.
There aren't going to be too many changes to OO or anything as far as I can tell from the internals list. We might finnaly get some sort of namespace support however.
Im glad to see an opcode cache included by default
~ Eric
In my opinion, supporting Unicode is major goal. PHP is not so nice once you create multi-language systems thanks to the messed up world of character encoding standards. I suspect supporting Unicode has been a huge change to the code.
Not really sure but if they plan to add unicode support to existing string functions then that would probably be a big change in the internals.
This is pretty good because i read somewhere that one of the features of .NET that makes you think twice between it and RoR is it's well thought out unicode support and transaction support. If PHP gets released with good unicode support and namespaces it would add a lot in making it "Enterprise Ready" for the eyes of outsiders.
As far as I know, php is the only one out of the perl and python with little to no unicode support...
Im glad to see them making progress, and I am really glad to see safe_mode being dropped.
- Eric
Eric Coleman
We're consentratin' on fallin' apart
We were contenders, now throwin' the fight
I just wanna believe, I just wanna believe in us
PHP is already Enterprise Ready, more so with PHP5 which offers the developer more control. If someone determines that PHP is not yet, already enterprise ready, then they're either ignorant, or speaking horse manure (putting it politely, that is)Enterprise Ready
Btw, a lot of people who develop with other technology do in fact, speak horse manure, not all people of course, but a lot do... So don't believe everything you read or hear, determine your own point of view and learn PHP and then learn how to use it in a large scale applicaiton environment.
You'll therefore find that there is more to just a technology or the language, to Enterprise Ready. By that I mean you have Unit Testing for example, which plays an integrated and important role, but there are other things to take into account as well
Isn't everyone jumping the gun a bit?
Just because they're calling the development trunk PHP 6 doesn't mean a release is imminent. It might take months before it is officially released. That's the impression I got from reading the blog link in the original post at least.
It's a bit like thinking: oh no! I've just upgraded Wordpress to 1.5.2, I'll have to upgrade again soon to 1.6 because it's in the development trunk! Quite an interesting comparison since 1.6 is a fair way off yet, as PHP 6 is no doubt.
yeah although i kind of miss namespace supportPHP is already Enterprise Ready, more so with PHP5 which offers the developer more control. If someone determines that PHP is not yet, already enterprise ready, then they're either ignorant, or speaking horse manure (putting it politely, that is)
Yup what actually hurts is when the client pushes you to make it in .NET for example (not so much the other developers). There are just a lot of people who believe that .NET is better (of course i don't believe that else i would have jumped in their bandwagon). Unfortunately some of us can't really pick clients either you have them or you don'tBtw, a lot of people who develop with other technology do in fact, speak horse manure, not all people of course, but a lot do... So don't believe everything you read or hear, determine your own point of view and learn PHP and then learn how to use it in a large scale applicaiton environment.
Yes in fact a lot of the people developing in .NET here (in our country) don't even use unit testing and fail to abide by proper layering of their apps.You'll therefore find that there is more to just a technology or the language, to Enterprise Ready. By that I mean you have Unit Testing for example, which plays an integrated and important role, but there are other things to take into account as well
Anyway it just feels good when the programming language that you love just gets better and better.
Although removing magic quotes and register globals is good from a developer point of view, I don't know if its such a good idea for the general public. All my scripts are no longer dependant on either one of them(although I will have to take out one line in the header) but this change will make alot of older script break beyond repair. This may hinder transition.
Its nice to be backwards compliant, but at the same time, the things they are removing backwards compliance from, should never have been there to begin with. True it will break lots of old scripts if their hosts were to run php6, but those people will either have to move there website to another host or upgrade there scripts.. I'm guessings hosts that use php6 will have an option of running php5 instead(or maybe even 4). Backwards compatibility is good, but I believe there comes a time when you just have those old bad habits had to be let go..
I think this is one step in the right direction for them.
But i'm guessing 2-3 years at least before we see any large hosts running php6 (assuming php6 has been officially released by then)
Well with magic_quotes_gpc and register_globals taken out, I think it will help PHP grow even if it breaks backwards compatability. Those two "features" account for a good majority of the questions in the PHP forum, and by having them out, people will hopefully be prompted to upgrade quicker. Also, if a script still wont work without either of them on, I wouldnt want to be running anywhere near it on account it probably has a few security holes in it as well.
If magic quotes and register globals is removed I can see web hosts having trouble, even years down the line. Suppose they could offer both versions with different file extensions like what happened with version 3 and 4. There's also a lot of complex bespoke systems out there written with both magic quotes and register globals assumed, so that might place pressure on the continued maintainance of a long-running legacy version of PHP for security fixes.
A majority of my personal programming is influenced by what hosts support. If I'm developing a web app, it needs to use a language that's widely supported.
Right now I'm focusing on PHP4 and Python--PHP5 and Ruby (on Rails) just don't have very wide support yet. PHP will be doing a disservice to themselves if they release PHP6 before getting PHP5 in use.
Another problem is backwards compatibility.. it's decent with PHP5 but a lot of free scripts will break. If a host adds PHP5, they can't even remove PHP4 yet. A simple ini_set option to parse as PHP4 would do a world of good, IMHO.
The following is completely opinion based and does not reflect the views of SitePoint or its members. Reader’s discretion is advised.
For the sake of progression, should we not get rid of PHP4 once PHP6 is released? If you have a script built on PHP4 and does not work on PHP6, rewrite it. If we keep this mentality about backwards compatibility, progression in this industry will be stifled.
I believe that is the plan, hence the new version number.
Hay, perhaps hosts will just skip PHP 5 and go straight to PHP 6. You know that programming is built on even numbers
We might even get real support for __toString, which would be great for OO work.
Douglas
Hello World
>> A majority of my personal programming is influenced by what hosts support.
Amen to that. Those of us who make our living by programming web apps must (usually) program in what we can reach the largest audience with. For us PHP guys, that means we're mostly still coding PHP4 apps. Otherwise, we don't eat
As long as they don't make debug_mode being on mandatory, I'll be ok if they get rid of magic quotes and register globals since generally I don't declare any variables before using them in PHP.
naughty :P
Eric Coleman
We're consentratin' on fallin' apart
We were contenders, now throwin' the fight
I just wanna believe, I just wanna believe in us
Actually taking away register globals will void any reason to actually declare variables in php before using them.
Undeclared variables are initialized as empty strings. This will fail.
if($x === null)
{
echo "blah";
}
Always, in every programming language, no matter how much it might seem like it's unnecessary, initilize your variables. It's just good practice.
Unicode support is huge because it makes developing *applications* much easier. PHP is "enterprise ready" in the sense that it can / does power enterprise-class websites, but it's not "enterprise ready" from the "application" standpoint. Yes, it's possible to do many great things in PHP, but not a day goes by that I don't wish I had some feature from C++ or Java when developing applications when developing. Of course, it has many other advantages over those languages, but several major shortcomings:
1.) No unicode support.
2.) Huge inconsistencies, from function naming conventions to parameter order (needle, haystack? haystack,needle?), from half the language being case-sensitive and the other half not, throwing fatal errors instead of exceptions, etc.
3.) Lack of namespaces. If you don't see this as "important", I wager that you've never had to work with code from a different vendor that you didn't have frequent contact with. Try integrating wikipedia with another php project some time -- just watch out for microtime_float()...
Overall, the language just isn't that polished. Those of us who know better understand that it doesn't matter, but unfortunately we're not the ones who make the decision. Java and C# are extremely polished languages, each backed by huge, powerful corporations. PHP is viewed in the same league as Perl or Python -- "just a scripting language". This is the kind of stuff that can change that perception.
I like the sound of those features, and having the language as a whole revised. I think it is important to make PHP attractive to both developers and enterprise, making life easier for both. There is big problems in PHP, including not having control over things (ie. things defined in php.ini). While it can be useful to have the language configuable, it does add alot of work for developers to maintain portability. If it is unneccessary, remove it. Register globals, magic quotes, etc, can be emulated anyhow.
As for PHP6 being adopted by hosts: if they do good QA and promotion, maybe it won't take too long... (I'm just being optimistic). I say we are a long way off until PHP6 is out, anyway.
Good Luck
I would expect timelines to be in years. Looking at how long PHP5 took, I think you're right when you suggest there's some gun-jumping going on in this thread
I agree, largely, and I think that that time is coming for PHP. But in practice there's a lot of old code out there - all those forums for example, and I know at work we have stacks of legacy code running with register_globals on.
So maybe this is not the time to get het up about an imminent relase of PHP6, rather it's the time to be cleaning up all that nasty old code, and improving our programming habits at the same time, so when the day comes, we'll all be laughing.
Well, I have to say that php 5.5 is the release of php that had to be php 1.0, Unicode, suport had to be there from the start!!
Posting Permissions
Bookmarks