|Please add these C# compilers to the list of open source compilers email@example.com (Eric) (2006-05-15)|
|Re: Please add these C# compilers to the list of open source compilers firstname.lastname@example.org (Paolo Molaro) (2006-05-16)|
|Re: Please add these C# compilers to the list of open source compilers email@example.com (Eric) (2006-05-18)|
|Re: Please add these C# compilers to the list of open source compilers firstname.lastname@example.org (Paolo Molaro) (2006-05-22)|
|From:||Paolo Molaro <email@example.com>|
|Date:||22 May 2006 02:09:31 -0400|
|References:||06-05-054 06-05-055 06-05-062|
|Posted-Date:||22 May 2006 02:09:31 EDT|
On 2006-05-19, Eric <firstname.lastname@example.org> wrote:
>>> Blue (public domain - no restriction):
>>According to the webpage it's quite incomplete, so I'd include it only
>>on a list of toy compilers.
> True, but it's pretty good for being 100% public domain. There aren't
> many compilers with a liberal license like that. It's also got a hand
> written recursive descent parser, and that may be interesting to study
> for many who read this forum.
I'd agree, though I'll point out that there are plenty of recursive
descent parser sources on the net to study from. If someone wants to
study a C#-specific parser, he better study one that actually implements
the spec and solves some of the hard issues that arise only when
implementing the full specification, otherwise it is not much more
useful than studying a simple expression parser from a undergrad book.
Maybe someone with a bit of time could try to run the Blue compiler
against the Mono C# compiler test suite and see how many tests it can
>>The compiler is quite limited, no 2.0 support and a few things missings
>>from 1.x support, too.
> A lot of code was written using 1.x, and it's still a reasonable
> platform. This compiler is also interesting because it's written in C,
> not C#. Also, there's an effort underway to write a Java bytecode
> generator for this compiler.
The compiler has been basically unchanged for the last 20 months or so.
The java bytecode generator has never been a real effort: generating
java bytecode from C# code has some of the same complexities as
generating it from C code, hence this is a 'feature' only useful for very
simple tests (if it actually works).
email@example.com Monkeys do it better
Return to the
Search the comp.compilers archives again.