[antlr-interest] Curiosity question

Michael Bedward michael.bedward at gmail.com
Thu May 12 18:50:38 PDT 2011


Judging by those examples I think regex matching is your best bet.


On 13 May 2011 00:22, Alain Perry <alain.perry at gmail.com> wrote:
> Hi there,
> I'm an absolute newbie when it comes to lexers and parsers. I might
> have fiddled some with lex- and yacc-style tools while I was a CS
> student, but not much of it has been left in my volatile brain.
> The thing is, I'm curious if a tool like antlr is the right one for a
> particular job: I need (surprise) to parse some strings that contain
> tokens of interest. Thing is, these tokens are incredibly loosely
> defined. Here are some examples where I need to detect the part
> between braces (the braces aren't there in the real world):
> [123456] this is some sample text
> [12:34] this is some more sample text
> [123456] some more...
> [12:34:56^2] another example
> [12:34:56 at bleh] still another
> [12:34:56^1 at blah] and a last one, just for the fun of it.
> On the contrary, the strings [12345], [1234567], [12:34:5] etc. should
> not be detected as a token of interest.
> So as I said above, my question is: should I be using antlr to detect
> these things ? Should I stick with regex'es ?
> To be honest and since I'm a playful type of guy, I started writing a
> grammar definition for my problem. I might post it here to treat you
> all knowledgeable people to a good laugh. But I might be interested
> first in any online (and free, for the time being) resource you would
> point me to in order to not make to much of a fool of myself and get
> to know more about this stuff in a practical and not too theoretical
> kind of way.
> Now, let me just thank you for reading till this point.
> Regards,
> --
> Alain Perry
> List: http://www.antlr.org/mailman/listinfo/antlr-interest
> Unsubscribe: http://www.antlr.org/mailman/options/antlr-interest/your-email-address

More information about the antlr-interest mailing list