Re: [hybi] Restarting IESG review on permessage-deflate

For PHP, soywiz has worked out that the "$encoding" parameter allows you to
set the number of window bits--see the comment at
http://www.php.net/manual/en/function.zlib-encode.php

Interestingly, the problem with PHP seems to be not supporting
no_context_takeover, but supporting the default context_takeover behaviour.
zlib_decode() creates a new zlib context every time.

There might be a way to abuse the gz* functions to do it, I have not
explored them in depth.

It is always possible to support no_context_takeover in any language by
creating a new compression context for each message.

On 5 October 2013 02:41, Joakim Erdfelt <joakim@intalio.com> wrote:

> Ok, add to the b) side of the equation:
>
> scala - same classlib and jvm as java
> jython - same classlib and jvm as java, no native zlib support here
> php - http://www.php.net/manual/en/ref.zlib.php (cannot support
> max_window_bits, no_context_takeover might be possible)
> c# -
> http://msdn.microsoft.com/en-us/library/system.io.compression.deflatestream_methods.aspx (cannot
> support max_window_bits, no_context_takeover might be possible)
>
>
> --
> Joakim Erdfelt <joakim@intalio.com>
> webtide.com <http://www.webtide.com/> - intalio.com/jetty
> Expert advice, services and support from from the Jetty & CometD experts
> eclipse.org/jetty - cometd.org
>
>
> On Fri, Oct 4, 2013 at 10:11 AM, Tobias Oberstein <
> tobias.oberstein@tavendo.de> wrote:
>
>> >>Well, I'm sorry for the Java community not being able to fix this in
>> 10+ years;)
>> >Yet, oddly, with web technologies, this hasn't been a problem since our
>> initial use of Oak in 1995 (yes, Jetty has existed since 1995)
>>
>> Jetty being around since 1995 is obviously awesome!
>>
>> As a strange coincidence, zlib has also be around since 1 May 1995 (I
>> haven't checked whether the respective options have been in the first
>> version though).
>>
>> > The capabilities of one set of libraries (c & python) should not be
>> forced on all environments as well.
>>
>> Lets collect a list: a) stuff that exposes the complete zlib vs b) stuff
>> that does not;)
>>
>> a) C, C++, Python, NodeJS, Perl, Ruby, PHP, Erlang, Haskell
>> b) Java
>>
>> Caveat: I (obviously) have not used all of above myself. Just had a quick
>> glance at their docs.
>>
>> Pointers:
>> http://perldoc.perl.org/Compress/Zlib.html#Deflate-Interface
>> http://www.ruby-doc.org/stdlib-2.0.0/libdoc/zlib/rdoc/Zlib/Deflate.html
>> http://www.php.net/manual/de/function.gzdeflate.php
>> http://erlang.org/doc/man/zlib.html
>>
>> http://hackage.haskell.org/package/zlib-0.5.4.1/docs/Codec-Compression-Zlib.html
>>
>> I stop here;)
>>
>> /Tobias
>>
>
>
> _______________________________________________
> hybi mailing list
> hybi@ietf.org
> https://www.ietf.org/mailman/listinfo/hybi
>
>