[antlr-interest] What are channels and what are they used for?

[Robin Davies]

Are there ways to attach a parser to a channel other than default?

I'm working on a grammar that has need of a javaoc-like facilities, and a 
C#-style preprocessor (i.e. very simple #if's and #defines); and if channels 
actually do anything right now, I'm pretty certain I could make use of it. 
Currently, I'm pushing token text for ("#'  (~LF)* LF ) out to a secondary 
parser. But I can't help feeling that there's a better way to do this.

(Absolutely loving ANTLR 3, btw. After all the grim alternatives -- LL(k), 
LALR(1) and/or hopelessly license-encumbered -- ANTLR v3 is a marvelous 
change. In my opinion, the release of ANTLR v3 is  one of the great 
landmarks in the recent history of software development. Very exciting! 
Congratulations on your release.).

----- Original Message ----- 
From: "Terence Parr" <parrt at cs.usfca.edu>
To: "ANTR Interest" <antlr-interest at antlr.org>
Sent: Tuesday, June 05, 2007 12:01 PM
Subject: Re: [antlr-interest] What are channels and what are they used for?


> Hi. :)  token channels are a way to avoid discarding tokens (still  send 
> tokens to the parser) w/o having the parser see them.  So, you  could send 
> whitespace on one channel, javadoc on another, and the  java tokens on the 
> default channel.  Then actions can access the  hidden channels for 
> translation.
>
> Ter
>
> On Jun 5, 2007, at 6:58 AM, Luke A. Guest wrote:
>
>> On Tue, 2007-06-05 at 13:29 +0200, Johannes Luber wrote:
>>> Luke A. Guest wrote:
>>>> Hi,
>>>>
>>>> I think the subject says it all really, I have come across it,  but 
>>>> don't
>>>> know what it's all about? Can anyone enlighten me?
>>>
>>> The channels are supposed to make tokens invisible so one doesn't  have
>>> to deal e.g. with whitespace in the parser. It is possible to access
>>> those hidden tokens, but I didn't figure out how to to do that  exactly.
>>
>> Is that all it's used for? I would've thought there'd be more to it  than
>> that.
>>
>> Thanks,
>> Luke.
>>
>>
>