This is the mail archive of the
java-discuss@sourceware.cygnus.com
mailing list for the Java project.
Re: jc1 segfaults compiling volano
- To: Andrew Haley <aph@pasanda.cygnus.co.uk>
- Subject: Re: jc1 segfaults compiling volano
- From: Per Bothner <per@bothner.com>
- Date: 22 Jun 1999 09:09:29 -0700
- Cc: java-discuss@sourceware.cygnus.com
- References: <19990622093934.2200.qmail@pasanda.cygnus.co.uk>
Andrew Haley <aph@pasanda.cygnus.co.uk> writes:
> Sigh. I'm beginning to wonder if the bytecode compiler should just
> give up after failing to find a class. After all, the compilation is
> bound to fail, so what's the point of trying to continue? I know that
> the C++ and C compilers continue checking syntax and try to recover
> (sometimes successfully) but I don't know if it makes any sense to do
> so with a bytecode compiler. Comments welcome.
Well, in some cases we should be able to continue safely: Making a
call to a method in a missing class or referencing a field in a
missing class or both things we can do without having te actual
class. We can also cast to or from a missing class (by letting
it always succeed).
It is not a very important "feature", but it is a nice "quality of
implementation" bonus.
--
--Per Bothner
bothner@pacbell.net http://home.pacbell.net/bothner/