Path: csiph.com!x330-a1.tempe.blueboxinc.net!usenet.pasdenom.info!weretis.net!feeder1.news.weretis.net!news.imp.ch!de-l.enfer-du-nord.net!feeder1.enfer-du-nord.net!news.weisnix.org!newsfeed.ision.net!newsfeed2.easynews.net!ision!newsfeed3.dallas1.level3.net!newsfeed2.dallas1.level3.net!news.level3.com!postnews.google.com!glegroupsg2000goo.googlegroups.com!not-for-mail From: JennyB Newsgroups: comp.lang.forth Subject: Re: Is this use of DOES> according to the standard - and portable? Date: Wed, 30 Nov 2011 06:19:40 -0800 (PST) Organization: http://groups.google.com Lines: 31 Message-ID: <4368110.745.1322662780927.JavaMail.geo-discussion-forums@vbjs5> References: <4ed4169f$0$6849$e4fe514c@news2.news.xs4all.nl><04b41ba9-e69d-45e8-8f19-93b4061d6e30@u1g2000prl.googlegroups.com> <4ed47263$0$6562$9b4e6d93@newsspool4.arcor-online.net> <6ab92624-e5b9-419b-a837-02c72a495102@v29g2000yqv.googlegroups.com> <4ed5e4c1$0$6852$e4fe514c@news2.news.xs4all.nl> Reply-To: comp.lang.forth@googlegroups.com NNTP-Posting-Host: 212.159.59.80 Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable X-Trace: posting.google.com 1322662786 29942 127.0.0.1 (30 Nov 2011 14:19:46 GMT) X-Complaints-To: groups-abuse@google.com NNTP-Posting-Date: Wed, 30 Nov 2011 14:19:46 +0000 (UTC) Cc: hansoft@xs4all.nl In-Reply-To: <4ed5e4c1$0$6852$e4fe514c@news2.news.xs4all.nl> Complaints-To: groups-abuse@google.com Injection-Info: glegroupsg2000goo.googlegroups.com; posting-host=212.159.59.80; posting-account=pe0d8AkAAADONuicQESwvPXuIt5O5Ll1 User-Agent: G2/1.0 X-Google-Web-Client: true Xref: x330-a1.tempe.blueboxinc.net comp.lang.forth:7641 On Wednesday, 30 November 2011 08:10:39 UTC, Hans Bezemer wrote: > Ed wrote: > > .RNO didn't CREATE the current definition so all bets are off. > >=20 > > 6.1.1250 DOES> > > ... > > Compilation: ( C: colon-sys1 -- colon-sys2 ) > >=20 > > Append the run-time semantics below to the current definition. ... > Depends on that you define as "current definition". If this is the same > as "most recent definition" then I don't see any problems. The TC doesn't > go any further than "typical use" - it doesn't elaborate and did away wit= h > SMUDGE. So how many options do implementors have? >=20 > Hans Bezemer The "current definition" in this case is ".RNO". It has to be. This happens= when DOES> is compiled, so any child words that the definition creates whe= n it is executed do not yet exist. The run-time semantics appended by DOES>= can be compared to IMMEDIATE - both modify the most recent definition. I d= on't see any reason not to use the same mechanism to find the most recent d= efinition in each case, though it doesn't make much sense for DOES> to modi= fy anything that has not been defined by CREATE.=20 I suppose it is just possible that a DOES> which assumes that a CREATE has = been previously compile in the current definition could work by modifying t= hat code at compile time, but I don't see how or why that might be done.=20 In any case, the fact that : CREATE ... DOES> ... DOES> ... : is Standard = shows that : ... DOES> ... ; must also be.