On Fri, 2012-02-17 at 10:25 +0100, Roland Zwaga wrote:
> By the way,
>
> here is the Java port of the goldparser engine:
>
> https://github.com/ridencww/goldengine
>
> it might be an idea to try out both ANTLR and goldparser
> and compare the results, I'd say the fastest one we'd take ;)
How abo
>
> Michael Schmalle already created jasblocks, a Java port of his asblocks
> engine
> which is able to parse AS3.
> https://github.com/teotigraphix/as3-commons-jasblocks/
> Since he's already created grammars for AS3, E4X, regex and ASDoc we might
> be able to stand on the shoulders of giants :)
>
>
> By the way,
>
> here is the Java port of the goldparser engine:
>
> https://github.com/ridencww/goldengine
>
> it might be an idea to try out both ANTLR and goldparser
> and compare the results, I'd say the fastest one we'd take ;)
>
Michael Schmalle already created jasblocks, a Java port of h
By the way,
here is the Java port of the goldparser engine:
https://github.com/ridencww/goldengine
it might be an idea to try out both ANTLR and goldparser
and compare the results, I'd say the fastest one we'd take ;)
cheers,
Roland
> If a 3rd party really needs to release a new .swc, then t
I really like the ideas being forth here. I still want to continue to push
Adobe to finish Falcon work in an open development model, and I hope they
respond as such. However, assuming not only that Falcon will retain its closed
development model, but also that it is likely to slip past the end
Very nice. Have you considered RST [1] (or another markup language)
for the design document formatting? It's readable as plain text,
github displays some of these natively (I think at least RST and
Markdown), and RST can be "compiled" to PDF or HTML (e.g., Python uses
it for their documentation and
>
> If a 3rd party really needs to release a new .swc, then they can still
> use the compc compiler as I think we are supposed to be getting that
> too. By requiring them to use compc, we make it clear that they are
> using old technology to do things the old way.
>
> Of course, having said all tha
On Thu, 2012-02-16 at 12:10 -0800, Om wrote:
> David, from your pdf (yeah, I read it ;-) )
> This sounds like a serious limitation - especially for Flex based library
> projects which is a pretty important feature. Can you please elaborate on
> the workaround you hinted in your doc?
What I was thi
>
> "To facilitate supporting multiple targets, Goshawk will not generate SWC
> library files as
> they can only be used by a compiler targetting SWFs. Instead, the
> language-agnostic parsetree
> generated by the ANTLR front-end will be used as a library file format.
> This may be
> stored in XML,
David, from your pdf (yeah, I read it ;-) )
"To facilitate supporting multiple targets, Goshawk will not generate SWC
library files as
they can only be used by a compiler targetting SWFs. Instead, the
language-agnostic parsetree
generated by the ANTLR front-end will be used as a library file forma
On Thu, 2012-02-16 at 20:48 +0100, Roland Zwaga wrote:
> not to throw a spanner in the works immediately but I was wondering if you
> could
> consider a different parser from ANTLR. There is a project called
> Goldparser:
> http://goldparser.org/
> From what I understand this thing is the fastest o
On 17/02/2012 03:01, David Arno wrote:
Obviously, as per the mentors' request recently, whilst it makes sense
to experiment with this compiler on Github, we need to keep discussions
happening here. Perhaps we could prefix topics with [Goshawk]? That way
those that wish to concentrate on making th
>
> As there was some positive feedback to the idea of creating our own
> compiler, I have gone ahead and started a new project on Github and
> created an initial design document on how I think it might work. I've
> taken the liberty of giving it the code name "Goshawk." If you are
> curious as to
On Thursday, February 16, 2012, David Arno wrote:
...
> Obviously, as per the mentors' request recently, whilst it makes sense
> to experiment with this compiler on Github, we need to keep discussions
> happening here...
Yes, that would be great.
> ..Perhaps we could prefix topics with [Goshawk]
14 matches
Mail list logo
