[haiku-inc] Re: Contract proposal: website & R1B1 work
- From: "Adrien Destugues" <pulkomandy@xxxxxxxxxxxxx>
- To: haiku-inc@xxxxxxxxxxxxx
- Date: Thu, 28 May 2015 16:13:01 +0000
28 mai 2015 17:44 "Augustin Cavalier" <waddlesplash@xxxxxxxxx> a écrit:
On Thu, May 28, 2015 at 10:09 AM, Axel Dörfler <axeld@xxxxxxxxxxxxxxxx> wrote:
Before you start, I'd like to work out a detailed plan of action, though, so
that we
a) have a good mutual understanding of what you are working on exactly, and
I'll start by poking at Alex's buildserver and see what I think of it, and
then investigating other
scripting languages to use instead of Ruby.
A quick investigation of io.js revealed that the only thing really blocking
us from porting it is
the awful Gyp-based build system, but that shouldn't be *too* hard to tweak.
JS engines tend to be
pretty similar in their OS requirements, and with JavaScriptCore's JIT
humming along nicely in Web+
(after some kernel changes by Ingo and Adrien), I see no reason why V8js
wouldn't work just as
well.
I'm not sure it's that simple. Porting of Javascriptcore started in 2008 or so
and it was a lot of work to get it running as it is now. Your contract is going
to be rather short, so maybe it would be better to start with a
language/interpreter already known to work in Haiku, at least for the slave
part. If your contract starts with porting V8 and getting it to run, there may
not be much time left to actually write code using it.
It's your choice ultimately, but it would be nice to know why io.js would be
better than alternatives (I can think of perl, python, lua, bash, ruby 1.8,
yab, C, C++ which are known to work in Haiku - there are possibly more, and not
all of them would be good choices). Why would io.js be a better choice? Has it
big enough advantages for this project that justify spending time on getting it
to run?
--
Adrien.
Other related posts:
