Muted Template <Review/Critique/Suggestions>

M2tM

Hello everyone! I'm new to this forum, I found it when searching for a place specifically to get feedback on a project of mine. This is "Muted Template" which is yet another templating language built over top of PHP which is itself a templating language. Some extra stuff is included because this is going to serve as the basis for a CMS I am developing.

The language itself is built with extensibility and full Utf-8 support at its core. This means that basic language constructs such as if, for, set and more are actually "plug-ins" and you can develop any other fundamental language tags provided you register them with the engine with MutedTemplate->createTag($TagName, $ClassName) and the class defines its own parsing rules appropriately. The only thing which is built-in to the template parsing class is variable output tags such as {$value[index]}.

The size of the project may seem daunting, however, there is a unit test folder where you can view a number of examples and there is a brief comment on how to use each tag in the Code/Parse/Tags folder. If you're feeling adventurous you can even try your hand at making a new tag. The best way to approach this would probably be to find a tag that does something similar to what you want and copy/modify that.

I have a demo here where you can take the language for a spin:
http://www.michaelhamilton.com/MutedTemplate/demo.php

The PHP source code is here:
http://www.michaelhamilton.com/Files/MutedTemplate.zip

If you would like to use this in your own projects please let me know via e-mail maxmike@gmail.com, mostly I just like knowing that people find it useful. If you make new tags which may be useful, again, let me know and I may include them in the core distribution!

But more importantly this project is in its infancy and I am looking for comments on both the PHP code running the project and also the language I have designed.

This project is PHP 4 and PHP 5 compatible.

I have posted this in another PHP developer forum for critique as well, I encourage you to skim it for my comments (though much of the discussion seems to be pretty off-topic, my posts include examples, descriptions and more. Part of the reason I post it here as well is due to the large volume of off-topic discussion.)
http://forums.devnetwork.net/viewtopic.php?f=50&t=113476&start=0

Thanks for your time and comments!

M2tM

Hello again. This is basically a bump as I am still interested in feedback.

halojoy

I have had a quick look at your code
as well as readsome of the discussions at the other forum.

Think you have done an impressive work with your template language.
Some of the most advanced in its kind.
What will be used and liked by website programmers round the internet
is mainly about to get a go.

1. To submit to places where your code will be seen.
It is a bit like Google SEO. The more hits and views, the better!

Your application should be a plugin or part of some leading software.
By the size of phpBB, Wordpress, Joomla etc.

Only you classes and library is not enough to attract normal users.
You need at least 2-3 useful attached application examples, that can show the good.
This application should be included in download and ready to run out-of-the-box.

I have written a tiny Template Engine of my own.
Who hasnt 😃 ???
But my template class is of another kind
and more the simple assign and replace macros, like {{replace this with something}}

I am like you a little disappointed you do not get any Reply and comments here.
I dont know why.
But good luck anyway, from me 🙂
/halojoy - done his duty

M2tM

Thank you for your comments! Yes, getting this out into the world is something I would love to do. Your suggestions are all good. I am currently working on a CMS which will be using this system, so hopefully if I can get that out and it becomes popular other people will see the language itself and begin rolling new projects with that.

It's still very much in development though, before it becomes widely adopted I want to make sure it works for its purposes, so I'm not at the evangelizing stage yet and I am still playing with the entire system (recently I re-wrote the UTF8 string handling code.) 😉

Your praise means a lot though!

sohguanh

M2tM;10946907 wrote:
Thank you for your comments! Yes, getting this out into the world is something I would love to do. Your suggestions are all good. I am currently working on a CMS which will be using this system, so hopefully if I can get that out and it becomes popular other people will see the language itself and begin rolling new projects with that.

It's still very much in development though, before it becomes widely adopted I want to make sure it works for its purposes, so I'm not at the evangelizing stage yet and I am still playing with the entire system (recently I re-wrote the UTF8 string handling code.) 😉

Your praise means a lot though!

Usually when developers look for Open Source packages we tend to go to more "well-known" sites. For Perl it is CPAN. I guess for PHP it seem to be PEAR and PECL.

So instead of re-write from scratch in your case, the current stable of HTML_Template_* packages at PEAR cannot serve your needs at all? Or why not combine your effort with one of the existing packages developer at PEAR?

What I see in PEAR and PECL is something that had happened at CPAN Perl for the past decade. There are so many packages overlap and duplication that when the dust settles, only a few worthy are left standing.

In general, the ones left standing are those with easy to use API and offer good performance. Some packages are great in API design but perform horrendously slow and usually such packages fall into disfavor soon.

On the contrary some packages has horrible API design but developers are forced to use it as we want the performance it offers. My personal case would be Perl SOAP::Lite. I guess we can't have the best of both worlds isn't it ? 😉

Hope to see your effort at PEAR or PECL 🙂

Thanks.

M2tM

It is obvious you haven't even looked at the project I have up if you are comparing it to PEAR's template class. Evangelize other projects in other places.

This is the type of response I have contended in some of the other forums. The fact is there are -many- template packages and I have not looked at all of them. The ones I have looked at do not have many of the language features mine has.

PEAR's template system specifically looks to only have basic variable and loop support. I could not find any quick tutorials that presented more interesting features. Muted Template is a complete language which supports basic functions, control structures, variables, loops, and it has an extremely powerful plug-in system. Muted Template has more in common with Smarty honestly... But Smarty isn't the project up for discussion either.

I'm looking for comments on -this- project.

M2tM

I will say that performance concerns will be addressed when I get around to porting this to C++ and make an extension out of the project. Right now this project has acceptable performance, but it doesn't output PHP as Smarty's system does and some of the flexibility of the system actually makes that difficult.

A C++ PHP plug-in would provide pretty fast execution and maximum flexibility with the language.

big_nerd

M2tM,

Although i'm at work and have a serious hard time looking into cool things like this, I liked what I saw to start.

I'd like to inspect the code a little more (security reasons) and assuming everything is great I'm thinking about giving it a real-word try.

I can definitely see potential in this project (assuming everything you've said to be true).

Good Work!

M2tM

That's great! I've done some major overhauling of the UTF8 string handling features, and have slightly improved the performance of the template system. There have been a few other improvements and new unit tests as well.

Feel free to download the project again at http://www.michaelhamilton.net/Files/MutedTemplate.zip

Also, feel free to e-mail me if you have any questions or need assistance. mike@m2tm.net

Weedpacket

I've given it a bit of a once-over; like everyone else (I suspect) I haven't got the leisure to follow everything down to the architectural principles, but a couple of questions of policy did occur to me:

Why

if(!defined('_MV_FOO_')){
	define('_MV_FOO_', true);
	// ...blahblahblah...
	}

include 'foo.php';

Instead of

//...blahblahblah...

include_once 'foo.php';

In short, why use C-style include guards instead of the runtime's list of included files?

Why continue to support PHP 4? Surely an acceptable minimum these days would be 5.2.

M2tM

1) You cannot enforce everyone who uses this will be use include_once. Many people just use include and call it a day. I figured the include guards are the most certain way to ensure the correct inclusion regardless of how someone decides to include something.

2) By supporting PHP 4 as well I ensure this can be run by pretty much anyone. I can only find usage statistics for 2008 where 50% used PHP4, I imagine that number has fallen to maybe around 20% or so, this is a wild guess but I can't find accurate information. The cost of supporting PHP 4 is not very high, it's pretty easy to write backward compatible code, I program -into- the language, not in it (I simulate private member functions with _, and impose restrictions on organization/naming to keep things sane, I override stupid language defaults present in both PHP5 and PHP4 like magic_quotes where possible, unit testing to ensure things pass in the specific environment you are running it in, and I do my best to present and use clean interfaces.)

So potentially 20% more people can use this tool, many of whom don't control their development environments (possibly they have a large number of legacy projects and haven't investigated installing parallel PHP versions on their server, or they do not have that option as it is a virtual server with limited support) and I am not running into any walls of impossible problems to solve in PHP4 syntax yet.

I guess my question is, do you see anything I've done which could be improved significantly by migrating to PHP5? If so, how is it an improvement and why does it offset a potential 1/5th of the PHP user population?

M2tM

I should also say that just because internally I support PHP4 does not mean anyone who includes this has to even really think about that. This is not a framework which wraps around a user's project like CakePHP or Code Igniter. This is more a library which can be included and as such it doesn't really pose any restrictions of use on the person including it.

So where a framework provider's choice of PHP5 support would influence the people who use the framework in a big way, the same is not true for this as it is really just a drop-in tool.

The only place it will affect users is when they develop plug-ins for the parser, and that is a pretty limited impact.

M2tM

I have a question for Big.Nerd, I don't see any obvious way to get in contact with him directly. Did you ever perform an evaluation of Muted Template? If so, what were the results?