caml-list - the Caml user's mailing list
 help / color / mirror / Atom feed
From: Xavier Leroy <xavier.leroy@inria.fr>
To: caml-list@inria.fr
Subject: [Caml-list] Re: Regular expression library: a poll on features
Date: Mon, 15 Jul 2002 17:52:44 +0200	[thread overview]
Message-ID: <20020715175244.B15691@pauillac.inria.fr> (raw)
In-Reply-To: <20020705161302.C20462@pauillac.inria.fr>; from xleroy@pauillac.inria.fr on Fri, Jul 05, 2002 at 04:13:02PM +0200

As promised, here is a summary of the replies I got concerning the
"feature poll" for OCaml's regexp library.

Feature 1: back-references in regexps (e.g. "([a-z]+)\1", meaning any sequence
of lowercase letters followed by another occurrence of the same
sequence)

     use often                  4
     use occasionally           3
     no use                     5

Feature 2: partial string matching as per Str.string_partial_match, i.e.
the ability to recognize that a string is a prefix of a string that
match a regexp.

     has already used           0
     could use in some cases    6
     no use                     8

A few more remarks on these two features, and why I asked about them
in particular.  

Feature 1 is standard in many regexp packages because it's trivial to
implement in a backtracking regexp matcher.  However, it's essentially
impossible to implement if the regexp is compiled down to a
deterministic automata.  Thus, supporting feature 1 precludes
DFA-based implementations such as Jérôme Vouillon's RE library.

Feature 2 is unusual and I haven't heard from anyone that uses it :-)
I got two replies suggesting one plausible scenario where partial
matching could come handy: find delimiters in a piece of text that
is being read block by block.  However, I'm not sure
Str.string_partial_match is adequate here, it looks like a
"search forward for a partial match" operation is needed, which Str
doesn't provide...

It was also suggested to me that the effect of partial matching
against a regexp R can be achieved by exact matching against a regexp
R' derived from R.  This is true for "textbook regexps", e.g. if R is
"ab*c", then R' would be "epsilon|a(epsilon|b*(epsilon|c))",
but doesn't work for more complex regexps languages, especially if
back-references are supported.  (Consider R = "(a+)\1".)

Thanks for your input,

- Xavier Leroy
-------------------
To unsubscribe, mail caml-list-request@inria.fr Archives: http://caml.inria.fr
Bug reports: http://caml.inria.fr/bin/caml-bugs FAQ: http://caml.inria.fr/FAQ/
Beginner's list: http://groups.yahoo.com/group/ocaml_beginners


  parent reply	other threads:[~2002-07-15 15:52 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2002-07-05 14:13 [Caml-list] " Xavier Leroy
2002-07-05 15:56 ` Pixel
2002-07-15 15:52 ` Xavier Leroy [this message]
2002-07-15 18:41   ` [Caml-list] " Matt Armstrong

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20020715175244.B15691@pauillac.inria.fr \
    --to=xavier.leroy@inria.fr \
    --cc=caml-list@inria.fr \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).