On Wed, 05 Dec 2001, Maxx wrote: > Well, it seems to me, guys, that we're falling in some kind of politics > forgetting about programming. Someone who made something in PHP on the same > topic, thinks that his/her product will lose its popularity after appearing > of standard PHP module. I'd say this is wrong. There's a place for every > project and there are things that this extension will never do because > they're not have been even planned for this module and won't be.
Every project, including Open Source ones, involves politics, like it or not. > I thought PHP has something to do with Open Source Community, but it's more > like a Personal Closed Project. Where did you get this idea from? > I don't complain. If you read my messages carefully, you won't find a > word like "include my module to the main CVS tree now!" or anything > like that. I'm NOT asking for nothing. I have just put in public a > piece of my code for everyone could use it. Do I have to write the > previous sentence in upper case for everyone understood this? :) No, and I know you didn't say "include my code with PHP". Other people starting saying it, though. > Should I say something about existing template engines? I wouldn't even > start to write this extension if I was satisfied with their quality. Ones > are written in a very bad manner, others are too large and require some kind > of Zend Cache to be ran in the proper way. But this will cost $$$ and is not > full Open Source. Most Open Source projects are started to "scratch an itch". Yours in no exception, I guess, and I don't see any problem with you making this extension. I do have a problem with those who want to create ext/template, for reasons that I have already explained. > There's no any secret that the code $str_val = "sometext $str_variable > sometext"; will be executed _much_ slower than the $str_val = > "sometext".$str_variable."sometext"; but the latter structure is being used > in almost 50% of template engines! Thankfully, mine doesn't. :) > That would be OK to use it in any class of software but extreme one. > Under the term "extreme software" I assume the code which should be > executed in the fastest possible manner and should consume less > possible resources. Any template engine is an extreme software and > should be really fast. This is the rule. That's the idea. Monte and I are always thinking about ways to make Smarty faster. Even now, as we speak, we are discussing a way to make it so that the main chunk of Smarty code is not loaded if you need to display a cached file. > Another issue is the PHP code. Everytime we call a page, we need to parse > all PHP libraries it includes. It requires some time. So, less PHP code we > have to parse, faster page we have. Isn't that right? With internal > extension you have no code to parse, so it will be faster than anything > written in PHP apriory. And again don't tell me about Zend Cache! PHP should > have its own possibilities to work faster. C extensions are not a panacea. I seriously think that if Smarty was initially written in C, it would be much less successful than right now. Why? Because it is much easier to prototype and develop things in PHP than in C, and because you get more people who understand the code and can contribute patches and enhancements. If it was in C, then I'd pretty much be the only developer. That is not to say that certain portions of Smarty couldn't be rewritten in C for optimization, perhaps they could. > Template caching engines? Please, tell me which of two things will be > faster: > 1. Read file, parse PHP code in it, replace some tags, print output > 2. Read file, replace some tags, print output > Anything else to be said? What does this have to do with template caching engines? > So, why noone wants to participate in creating something really good and > fast? I have started this project, but that doesn't mean you can not > contribute to it making it better. The current API is good. There's no need > to change it in future. Please. The history is all too full of sayings like, "640k should be enough for everybody". Don't make the same mistake. > So, we only could add new things to it, and improve what is already > done. Let's make it better and better untill we can say "Yes, sir. > This module could be called a standard". That is the problem, different people will want different things to be standard. As you've probably heard: the great thing about standards is that there are so many to choose from. > In conclusion, my purpose is to create a really fast template engine and I > will do this. I am not a "new school" 15 y/o kid who has just read a book > about PHP yesterday and today starts to send his code everywhere, neither > you are. I've got enough years of experience to be sure that I will finish > the job and will get a better solution. If you want to use it too, you know > the place where you can get it. It's free. I earn $$$ on other things. No need to get defensive, you haven't been called names and your programming skills haven't been called into question. You simply touched a sensitive topic. Regards, -Andrei * E Pluribus UNIX. * -- PHP Development Mailing List <http://www.php.net/> To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] To contact the list administrators, e-mail: [EMAIL PROTECTED]
