success using spidermonkey / base2 / packer
Diego Perini
just to let you know that the size savings obtained with your latest
Packer 3.1 are awesome.
I was able to successfully execute "base2.js" and "packer.js" under
latest SpiderMonkey release (1.8.0rc1).
No errors, no warnings, a compression speed boost of about 16x over
the same process with Rhino 1.7R2.
YUI Compressor is the past ! Your Packer is the future !
The effectiveness and compatibility of your scripts and coding style
have no equals.
--
Diego
Dean Edwards
> just to let you know that the size savings obtained with your latest
> Packer 3.1 are awesome.
Are you using base62 compression or standard minification? It's just
that there is an overhead for decompression in the client if you use
base62 compression. Packer 3.1 is much faster at decompression than 3.0
though.
> I was able to successfully execute "base2.js" and "packer.js" under
> latest SpiderMonkey release (1.8.0rc1).
>
> No errors, no warnings, a compression speed boost of about 16x over
> the same process with Rhino 1.7R2.
Good to hear. I've run base2 on quite a few platforms. If you find any
warnings or errors then please let me know. I want base2 to be usable in
as many JS runtimes as possible.
> YUI Compressor is the past ! Your Packer is the future !
Ha ha. OK. :)
YUI Compressor has been kicking packer's butt lately. Packer is still
very popular but most people associate packer with base62 encoding.
Packer is also a very efficient minifier. I think it wins on pretty much
every library on compressorrater.
> The effectiveness and compatibility of your scripts and coding style
> have no equals.
>
Dude, you are pretty good yourself. I've been looking at NWMatcher and
NWEvents recently. Awesome stuff. It seems that we do very similar work.
If you fancy helping out on base2 then I could really do with some help
testing. :) I'd also be happy to chat about the ideas behind your code,
NWMatcher particularly.
-dean
Diego Perini
On 26 Ago, 20:58, Dean Edwards <dean.edwa...@gmail.com> wrote:
> On 26/08/2009 18:19, Diego Perini wrote:
>
> > just to let you know that the size savings obtained with your latest
> > Packer 3.1 are awesome.
>
> Are you using base62 compression or standard minification? It's just
> that there is an overhead for decompression in the client if you use
> base62 compression. Packer 3.1 is much faster at decompression than 3.0
> though.
>
"shirnk" + "privateVars".
> > I was able to successfully execute "base2.js" and "packer.js" under
> > latest SpiderMonkey release (1.8.0rc1).
>
> > No errors, no warnings, a compression speed boost of about 16x over
> > the same process with Rhino 1.7R2.
>
> Good to hear. I've run base2 on quite a few platforms. If you find any
> warnings or errors then please let me know. I want base2 to be usable in
> as many JS runtimes as possible.
>
errors / warnings that I could see.
I didn't load the complete base2, just base2.s is needed by the
"packer.js" to minify code.
I tried loading some other base2 modules with no errors except the
missing window/document structure.
> > YUI Compressor is the past ! Your Packer is the future !
>
> Ha ha. OK. :)
>
> YUI Compressor has been kicking packer's butt lately. Packer is still
> very popular but most people associate packer with base62 encoding.
> Packer is also a very efficient minifier. I think it wins on pretty much
> every library on compressorrater.
>
> > The effectiveness and compatibility of your scripts and coding style
> > have no equals.
>
> Dude, you are pretty good yourself. I've been looking at NWMatcher and
> NWEvents recently. Awesome stuff. It seems that we do very similar work.
> If you fancy helping out on base2 then I could really do with some help
> testing. :) I'd also be happy to chat about the ideas behind your code,
> NWMatcher particularly.
>
> -dean
parts of them.
I have already translated some of my tests with the "events" part of
base2 / JSB, will send you the results.
--
Diego