MIME-Version: 1.0
Content-Type: multipart/mixed; boundary=f46d04462f8a39e90a04d64e16e2
Message-ID:  <CACKZBPVSd-j1wAbZDkcsKWO6_i2mupfuPE7VUok=qLmpysRH3g@mail.gmail.com>
Date:         Fri, 22 Feb 2013 11:59:15 +0100
Reply-To: Mailing list for the LaTeX3 project
              <LATEX-L@LISTSERV.UNI-HEIDELBERG.DE>
Sender: Mailing list for the LaTeX3 project <LATEX-L@LISTSERV.UNI-HEIDELBERG.DE>
From: Michiel Helvensteijn <mhelvens+latex@GMAIL.COM>
Subject: Some ideas for the standard LaTeX3 modules (first: scanning functions)
To: LATEX-L@LISTSERV.UNI-HEIDELBERG.DE
Precedence: list
Envelope-To: <rainer.schoepf@GMX.NET>
Status: R

--f46d04462f8a39e90a04d64e16e2
Content-Type: text/plain; charset=ISO-8859-1

Hi all!

I'm writing a large package in LaTeX3 right now (a rewrite of
'concepts' actually). I keep feeling the need to write some general
purpose functions. For now I'm putting them as private functions in my
package but I was wondering if they might not be a nice addition to
the standard modules.

However, even the example idea I mentioned in TeX.se chat seemed to be
controversial, so I'll start with just that one. It's a small set of
functions I wrote to scan ahead in the input stream to find modifiers
and various delimited optional arguments without having to write
auxiliary functions.

First there's `\scan_tokens:nnw`, which scans ahead for any tokens
that are in `#1`, grabs them and calls the (literal) function `#2`
with the resulting token list as an argument. In my package I use them
to grab digits and modifiers. Example:

------------------------------------------------------------

\ExplSyntaxOn
    \NewDocumentCommand \GrabDigits {} {
        \scan_tokens:nnw {1234567890} {
            (##1)
        }
    }

    \NewDocumentCommand \GrabModifiers {} {
        \scan_tokens:nnw {!*^} {
            (\tl_to_str:n{##1})
        }
    }
\ExplSyntaxOff

\GrabDigits 76543xyz   % yields (76543)xyz

\GrabModifiers ^!^*$x$   % yields (^!^*)$x$

------------------------------------------------------------

I've also written several related functions, like `\scan_tokens:nnFw`
which calls `F` if no tokens were found. Also `\scan_delimited:NNnw`
and `\scan_delimited:NNnFw` which are essentially an inline wrapper
for an xparse `r` argument.

I've attached a MWE. The implementation could be more efficient. I
just felt like doing a recursive version. It would have been
expandable if only `\peek_meaning_remove:NTF` was. :-)

Please tell me what you think!

-- 
www.mhelvens.net

--f46d04462f8a39e90a04d64e16e2
Content-Type: application/x-tex; name="test.tex"
Content-Disposition: attachment; filename="test.tex"
Content-Transfer-Encoding: base64
X-Attachment-Id: f_hdh7xnnh0
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--f46d04462f8a39e90a04d64e16e2--