K-user mailing list

[Radu Mereuta <headness13 AT gmail.com>]

Hi Gary,

At the moment comments are hard coded in the C style.

In rules we combine the syntax of the language with the K syntax, and handling different styles of layout would considerably complicate rule parsing.

We couldn't find any quick solution so we kind of avoided this problem for now.

Radu Mereuta

On Wed, Jun 21, 2017 at 1:33 PM, Christos Kloukinas <c.kloukinas AT gmail.com> wrote:

Hi,

Checking file k-distribution/samples/bf/bf.k (K 4.0), I can see that it uses the following:

  syntax Ignore ::= Token{~[\>\<\+\-\.\,\[\]\ \n\r\t]+}
                   [regex("[^\\>\\<\\+\\-\\.\\,\\[\\]\\ \n\r\t]+"), token, onlyLabel]
And then below eats up all such tokens:

  rule I:Ignore => .K

Best

On 21/06/2017 04:25, Gary Leavens wrote:

Hi,

In defining the syntax of a new language in K, how do I specify the comment convention for that language?  For example, suppose I want comments to start with 

-- and continue to the end of the line, as in Haskell. It seems like something along the lines of

syntax Comment ::= Token{"--" ~[\n\r]*[\n\r]}

should work, but then how do I get the parser to ignore these Comment tokens?

        Regards,

        Gary T. Leavens

        437D Harris Center (Bldg. 116)

        Computer Science, University of Central Florida

        4000 Central Florida Blvd., Orlando, FL 32816-2362 USA

        http://www.cs.ucf.edu/~leavens  phone: +1-407-823-4758

        leavens AT cs.ucf.edu