ah. that would do it! I'll try now. that would explain all known
behavior...i get weird errors as if it's generating code for the wrong
grammar. protected but my mojo is stronger than java's; howz this for
deviant behavior:
public void setUp() throws Exception {
ErrorManager.resetErrorState();
// force reset of static caches
new StringTemplateGroup("") {
{
StringTemplateGroup.nameToGroupMap =
Collections.synchronizedMap(new HashMap());;
StringTemplateGroup.nameToInterfaceMap =
Collections.synchronizedMap(new HashMap());;
}
};
}
Shit. no effect on the failing tests.
Ter
On Nov 11, 2008, at 11:55 AM, Sam Harwell wrote:
> Try resetting the StringTemplate group cache in the test init
> (assuming
> the test init runs before each test and not just once at the
> beginning).
> That was my best guess before.
>
> Sam
>
> -----Original Message-----
> From: Terence Parr [mailto:[EMAIL PROTECTED]
> Sent: Tuesday, November 11, 2008 1:49 PM
> To: Sam Harwell
> Cc: Jim Idle; ANTLR-dev Dev
> Subject: Re: [antlr-dev] Unit test differences between Java and C#
> port
>
>
> On Nov 11, 2008, at 11:44 AM, Sam Harwell wrote:
>
>> I only had this occur when StringTemplate tests ran in the same
>> session
>> as, but following, the ANTLR tests. Is this what you are seeing?
>
> interesting...could be ST related...mostly thinking antlr though.
> weird stuff.
>
> Ter
_______________________________________________
antlr-dev mailing list
[email protected]
http://www.antlr.org:8080/mailman/listinfo/antlr-dev
