[Laszlo-dev] CommonJS effort sets JavaScript on path for world domination

[Rami Ojares / AMG Oy]

What is the compiler currently doing?
My quess is something like this.

1) XML parser is used to build some sort of javascript source from lzx files
2) Then this is combined with the core laszlo files and the combination 
is some sort of version of ecma edition 4.
3) Using some compiler compiler like antlr or similar AST is created.
4) And using various techniques (mostly manual hacks :) from this AST 
the source/bytecode file(s) are generated for the chosen runtime.

Was this quess in the right direction?
What is the compiler compiler you are using?

- rami

Max Carlson wrote:
> Well...  Yes!  Rhino is much less of a hack than Jython.  And much 
> better supported!
> 
> P T Withington wrote:
>> Ha! After all the work I did to rip out Jython you want me to cast the 
>> compiler back into another dynamic language?
>>
>> On 2009-12-02, at 13:27, Max Carlson wrote:
>>
>>> I don't know about JVM vs JavaScriptCore/V8 performance.  But I do 
>>> know that Rhino has a lot of overhead when moving stuff across the 
>>> Java/JavaScript divide, which may have been a pain point.
>>>
>>> Last night I was fantasizing about embedding Rhino in the compiler so 
>>> we could hack the AST with something more dynamic than Java.  Not 
>>> sure how practical that is, but it could be cool to play with when 
>>> you're done with all the compiler refactoring!
>>>
>>> P T Withington wrote:
>>>> Which is kinda funny, since it is compiling to Java byte codes.
>>>> Does this mean that Javascript VM's are out-performing Java VM's 
>>>> these days?
>>>> On 2009-12-02, at 12:57, Max Carlson wrote:
>>>>> Yeah, I believe the original webtop server used Rhino.  Not sure 
>>>>> why that was scrapped in favor of 'Pure Java' - probably for 
>>>>> performance. Rhino is no where near some of the more modern JS 
>>>>> engines in terms of performance...
>>>>>
>>>>> P T Withington wrote:
>>>>>> They never would have slipped that by the Times editors!
>>>>>> I thought Adam's initial wt work used Javascript on the server?
>>>>>> On 2009-12-01, at 22:53, Henry Minsky wrote:
>>>>>>> That is a thought-provoking article. And we do need some kind of 
>>>>>>> a standard
>>>>>>> module system...
>>>>>>>
>>>>>>> I liked this quote
>>>>>>>
>>>>>>>
>>>>>>> "In these post-QBasic times, HTML, CSS, and JavaScript have 
>>>>>>> become for every
>>>>>>> ungainly, bespectacled troglodyte the gateway from social exile 
>>>>>>> to deeper
>>>>>>> social exile."
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>> On Tue, Dec 1, 2009 at 10:37 PM, Max Carlson <max at openlaszlo.org> 
>>>>>>> wrote:
>>>>>>>
>>>>>>>> http://arstechnica.com/web/news/2009/12/commonjs-effort-sets-javascript-on-path-for-world-domination.ars?utm_source=rss&utm_medium=rss&utm_campaign=rss 
>>>>>>>>
>>>>>>>>
>>>>>>>> Interesting!  I'd love to be able to develop server-side code in
>>>>>>>> JavaScript.  I was just chatting with a friend (and OL 
>>>>>>>> developer) the other
>>>>>>>> day about being able to describe server-side behavior as part of 
>>>>>>>> LZX, and
>>>>>>>> have part of the application run on the server and part on the 
>>>>>>>> client.
>>>>>>>>
>>>>>>>> So many possibilities!
>>>>>>>>
>>>>>>>> -- 
>>>>>>>> Regards,
>>>>>>>> Max Carlson
>>>>>>>> OpenLaszlo.org
>>>>>>>>
>>>>>>> -- 
>>>>>>> Henry Minsky
>>>>>>> Software Architect
>>>>>>> hminsky at laszlosystems.com
>>>>> -- 
>>>>> Regards,
>>>>> Max Carlson
>>>>> OpenLaszlo.org
>>> -- 
>>> Regards,
>>> Max Carlson
>>> OpenLaszlo.org
>>
>