[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: mapping token identifier to value

From: r0ller
Subject: Re: mapping token identifier to value
Date: Mon, 17 Dec 2018 09:46:09 +0100 (CET)

Hi Akim,

Ok, thanks for the feedback! I haven't found any better either so I implemented 
it and it seems to work but need to test it thoroughfully. Fortunately, as the 
mapping is generated as well, it's not tedious at all:)

Best regards,

-------- Eredeti levél --------
Feladó: Akim Demaille < address@hidden (Link -> mailto:address@hidden) >
Dátum: 2018 december 17 06:28:28
Tárgy: Re: mapping token identifier to value
Címzett: r0ller < address@hidden (Link -> mailto:address@hidden) >
> Le 14 déc. 2018 à 14:21, r0ller <address@hidden> a écrit :
> Hi Akim,
> I'm trying to get rid of numbering tokens and let bison do that but it seems 
> to be more difficult than I thought. Although, I managed to come up with an 
> idea as I'm anyway generating the bison source I could generate a map like:
> std::map<std::string, unsigned int> symbol2token={<!-- -->
> { "t_Con", yy::parser::token::t_Con },
> ...
> };
> and could then get the values simply by:
> symbol2token.find("t_Con")->second;
> However, still wanted to ask if there's any better solution maybe one that I 
> overlooked and is provided by bison itself;)
I personally would go for this solution. I agree it is tedious,
but that's anyway what happens in a typically scanner in Lex. And
that's probably something you already do somewhere, but with your
own token numbers assigned by hand, so it's probably not very

reply via email to

[Prev in Thread] Current Thread [Next in Thread]