jakarta-regexp-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Walid Joseph Gedeon" <wged...@nerim.net>
Subject Project enhancement proposal
Date Thu, 04 Dec 2003 11:57:22 GMT
Hello all,

    Euh, ping! (Am I on, or is it really a low traffic list? :)

    I was working on an interesting project and found the need to extend
the Regexp functionality by defining a meta-structure that combines
regexps together... and was wondering if there would be interest to
include such an expansion.

    The need is to be able to extract matched values from an input text,
for example, say an input is:

        key is = thefirstkey, value is = thefirstvalue

then we would like to say that (actually, this is the format that I have
implemented):

        (^key is = ){param1=([:alphanum:]*)}(, value is =
){value1=([:alphanum:]*)}($)

would match it and return a map with ["param1"->"thefirstkey",
"param2"->"thesecondkey"]

    Of course, there's a lot omitted here I would layout in more details
if this is interesting (such as dealing with optional parts).


    I have currently implemented a
depth-first-search,first-good-match-returning simple algorithm, but I
am sure it can be improved to be faster and to heuristically compare
matches and choose to carry instead of stopping at the first one.

    Have I been wasting time implementing something that already exists?
(My good friend google reported nothing of the sort, or I might have
asked her the wrong question)
    If this is a good idea, how do we go about submitting it for general
code review and the sort? (I'm going through the docs about this that
are on the site)

    Any comments, suggestions are welcome :)

-jo (wgedeon at nerim dot net)

---------------------------------------------------------------------
To unsubscribe, e-mail: regexp-dev-unsubscribe@jakarta.apache.org
For additional commands, e-mail: regexp-dev-help@jakarta.apache.org


Mime
View raw message