[antlr-interest] [C target] [3.1.3] several bugs in the generated parser

[Tomas Potrusil]

I've tried the fixes and nothing works :-)

1. There is still the invalid cast from the pANTLR3_STRING to
pANTLR3_UINT8. Temporarily I've solved that by writing:
function:       ^(Function ^(DatabaseObject Nil COUNT) countArgument)
-> ^(Function[$COUNT->getText($COUNT)->chars] countArgument);
instead of:
function:       ^(Function ^(DatabaseObject Nil COUNT) countArgument)
-> ^(Function[$COUNT.text] countArgument);

2. rewrite = true still produces invalid code. The errors are
different than before and there are still some.

3. When there is a wildcard in the grammar, it can be compiled now,
but the resulting code contains some undeclared variables.

Tomas

2009/6/11 Jim Idle <jimi at temporal-wave.com>:
> Tomas Potrusil wrote:
>>
>> All those problems remain there also for the latest builds (from May
>> 21, version 3.1.4-SNAPSHOT)
>>
>> add 1) Is there any hope that it will be fixed soon? Or can I write
>> that grammar any other way in order to achieve the similar
>> functionality?
>
> OK - I looked in the issues you reported and I have fixed them all. The
> caveat is that I have not tested the fixes, so I am afraid you will have to
> be the guinea-pig.
>
> Hudson has built the new runtime and you can get it from there or from the
> www.antlr.org/download/C directory. However, this also requires a new antlr
> tool jar. Hudson has built the standalone jar, so you can get it from the
> last successful build link, but it is still building the complete jar, which
> takes a while because it runs the regression tests twice. Maybe about an
> hour from now. Visit: www.antlr.org/hudson.
>
> Let me know if those fixes work for you.
>
> Jim
>