Path: csiph.com!v102.xanadu-bbs.net!xanadu-bbs.net!news.glorb.com!dotsrc.org!filter.dotsrc.org!news.dotsrc.org!not-for-mail
From: Lars Brinkhoff <lars.spam@nocrew.org>
Newsgroups: comp.lang.forth
Subject: Re: Recognizers and DT-TOKEN:
Organization: nocrew
References: <oe3i48$qc2$1@dont-email.me> <f13713b2-c31a-4967-83c7-d70113a92b8d@googlegroups.com> <dJidncMN9rGNPpTEnZ2dnUU7-anNnZ2d@supernews.com> <2017May4.160014@mips.complang.tuwien.ac.at> <ZOKdnZcItf2M_pbEnZ2dnUU7-UHNnZ2d@supernews.com> <oej7j8$j6c$2@dont-email.me> <BdKdnezS8sdOfZPEnZ2dnUU7-bfNnZ2d@supernews.com> <oeoatm$3p2$4@dont-email.me> <ef09e180-51e7-4d36-90cc-88cb810554dc@googlegroups.com> <2017May8.134427@mips.complang.tuwien.ac.at> <1e139225-3735-458b-b8d2-fde65fed0bc9@googlegroups.com> <2017May8.193009@mips.complang.tuwien.ac.at> <768e8435-91e6-4ad9-9f80-2566a46d4c75@googlegroups.com> <oeusb5$hdm$1@dont-email.me> <oeuva3$r9u$1@dont-email.me> <86bmr0amdi.fsf@molnjunk.nocrew.org> <oev4kj$dvv$1@dont-email.me>
Date: Wed, 10 May 2017 16:08:00 +0200
Message-ID: <8637ccajpb.fsf@molnjunk.nocrew.org>
User-Agent: Gnus/5.13 (Gnus v5.13) Emacs/24.3 (gnu/linux)
Cancel-Lock: sha1:XCTjPNofE5YLM/GMKDWbNey82Ig=
MIME-Version: 1.0
Content-Type: text/plain
Lines: 21
NNTP-Posting-Host: 178.18.91.250
X-Trace: news.sunsite.dk DXC=WJ[]ESj41hQ5LoVIWDNeH\YSB=nbEKnk[:02ZWQ`i[jTnY:6X7^ZBNWI57e>YZ9SO[1HBfORCSCcSD5J_h8H;EFSI2Ic:1WDKBS
X-Complaints-To: staff@sunsite.dk
Xref: csiph.com comp.lang.forth:58984

Alex wrote:
>> I have PARSED right before the dictionary search.
>> And NUMBER right after a failed directionary search.
>
> Do PARSED and NUMBER point at the input buffer, or to a copy (for
> example in a counted string from WORD, that's perhaps been case
> adjusted to make FIND easier)?

Interpreting a line looks much like this:

  : interpret ( -- ) begin parse-name dup while parsed repeat 2drop ;

PARSE-NAME returns ( a u ) pointing into the input buffer.  I don't
accomodate FIND, because I don't have that in the kernel.  I have
FIND-NAME ( a u -- a u 0 | xt +-1 ).

The default PARSED searches the dictionary and executes/compiles xts.
If not found, NUMBER is called.

Note: I'm not trying to argue much of anything.  In particular, I'm not
advocating for or against recognizers.