Re: gcj and jikes

[Joerg Brunsmann <joerg_brunsmann at yahoo dot de>]

 --- Ranjit Mathew wrote: 

> Have you guys seen the Pizza compiler?
> ..
> and of special interest, its "daughter compiler" GJ:
> 
>     http://www.research.avayalabs.com/user/wadler/gj/index.html
> ...

Thanks for your mail. At the moment there's only minmal consensus about 
the fact that it's worth to switch the front end. It has been questioned 
whether jikes is the right choice because of missing generics support and 
it has been suggested that it's worth to wait for a complete generics 
implementation in one of the available front ends. At the end of the day 
I always get the impression that jikes will be the only viable alternative 
because of it's widespread use, it's proven reliability compiling big java 
projects and the convenience of attaching it to gcj, see:

  http://gcc.gnu.org/ml/java/2003-04/msg00383.html

Do others share this impression?

> Jikes is released under the simple "IBM Public License":
> 
> http://oss.software.ibm.com/developerworks/opensource/license10.html
> 
> IANAL, but that license allows you to create derivative works as
> you wish and imposes itself (IPL) only if you distribute the
> original program - not the derivative works.

Sounds great, considering http://gcc.gnu.org/ml/java/2003-04/msg00391.html

> As a layman, I certainly find the Jikes source code to be far
> more readable and understandable than the parse.y in the GCJ
> front-end, but that could simply be because I haven't had any
> experience in writing parsers beyond my undergraduate days.

You're not alone. That's exactly one of the reasons why it does
make sense to switch the front end.

Jörg

__________________________________________________________________

Gesendet von Yahoo! Mail - http://mail.yahoo.de
Bis zu 100 MB Speicher bei http://premiummail.yahoo.de