As I'm never a big fan of relying solely on the fact that technologies
have been "used", I'd compare feature sets. Here's what I like from GXP:
- Pre-compile for performance
- Type safe
- Injectable
- White space reduction
- l10n and i18n support (although difficult to determine exactly how
they work)
- Componentized
- Auto-escaping
Here's what I don't like
- XML syntax yet again
- Somewhat complex
- Lacking sufficient documentation for all features
I haven't yet, but I would be comparing feature sets rather than usage.
At the same time though, they can both be used for results. I'd
probably only want to use one of them for tag files to reduce
complexity, but anyone interested in writing a pluggable tag system
could easily support both for tag files as well.
-bp
On Aug 1, 2008, at 6:35 AM, Joe Hudson wrote:
Thank you for taking the time to evaluate zipscript.
This is not currently being used in any real world apps because it is
new. That is certainly a valid concern. It is kind of a catch 22
because how can you ever be used in a real world app if people only
want an engine that is being used in a real world app?
Anyway, I'd like to address my assumption of your main concerns
regarding real world app usage:
1) Is it reliable?
All of the functionality is unit tested before a release is performed
and there are many unit tests which can be seen here:
- Unit Tests: http://code.google.com/p/zipscript/source/browse/#svn/
trunk/core/src/main/test/test/hudson/zipscript
- Referenced Templates and Results:
http://code.google.com/p/zipscript/source/browse/#svn/trunk/core/src/
main/test/templates
2) How does it perform?
There is a unit test that merges a template consisting of multiple
macro references (called PerformanceTest)
http://code.google.com/p/zipscript/source/browse/trunk/core/src/main/test/test/hudson/zipscript/PerformanceTestCase.java
This test, when performed on my laptop, can perform 1000 merges in
about 1 second.
As to the other issues of porting to other languages. That isn't
currently in the plan now but certainly would be something on the
plate if this language got a foothold.
Joe
On Fri, Aug 1, 2008 at 6:20 AM, Ted Husted <[EMAIL PROTECTED]> wrote:
For me, it's the "real world apps", *plus* the interest in porting
GXP
to other platforms, like C++, and Ruby. Personally, I'd like to add
C#
to that list. ASPX, like JSF, has its charms, but many .NET
developers
(including me) would like alternatives.
I do find the zipscript YUI examples interesting, since we are also
looking for an alternative approach to supporting Ajax with the
Struts
tags.
-Ted.
On Fri, Aug 1, 2008 at 1:30 AM, Al Sutton <[EMAIL PROTECTED]>
wrote:
Joe,
My main reason for backing GXP is that it's been used in high
volume "real
world" apps by Google and so has been proven to scale.
Does ZipScript have the same real world testing, and if so where
is it used?
Al.
Joe Hudson wrote:
Hello,
With all the talk about GXP, I wanted to let you know about another
template engine called ZipScript (zipscript.org).
Ok, you're probably saying... another one??? but this has some
distinct advantages and I would love for you to give it a try - I
really think you'll be happy you did.
1) it can be used to create complex UI components
2) it has a powerful non XML-based syntax (which is helpful since
the
most common output is XML)
3) it can also be used to get more than just text from expressions
(like booleans or objects) - great for having smarter config
properties
Here are some quick links:
- homepage:
http://zipscript.googlecode.com
- Yahoo! UI macro examples:
http://code.google.com/p/zipscript/wiki/YUIMacros
- Language Syntax Examples
http://code.google.com/p/zipscript/wiki/SyntaxExamples
Thanks
Joe Hudson
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
