X-Received: by 2002:a0c:f807:0:b0:66d:a22c:1dab with SMTP id r7-20020a0cf807000000b0066da22c1dabmr113888qvn.2.1698526795987;
        Sat, 28 Oct 2023 13:59:55 -0700 (PDT)
X-Received: by 2002:a05:6870:4d02:b0:1dc:7909:91fa with SMTP id
 pn2-20020a0568704d0200b001dc790991famr4212191oab.2.1698526795726; Sat, 28 Oct
 2023 13:59:55 -0700 (PDT)
Path: ...!3.us.feeder.erje.net!1.us.feeder.erje.net!feeder.erje.net!usenet.blueworldhosting.com!diablo1.usenet.blueworldhosting.com!peer02.iad!feed-me.highwinds-media.com!news.highwinds-media.com!news-out.google.com!nntp.google.com!postnews.google.com!google-groups.googlegroups.com!not-for-mail
Newsgroups: fr.comp.text.tex
Date: Sat, 28 Oct 2023 13:59:55 -0700 (PDT)
In-Reply-To: <ugej1s$3v7g2$1@dont-email.me>
Injection-Info: google-groups.googlegroups.com; posting-host=89.149.22.17; posting-account=dtL1OgoAAAABPdH-AXByPRqcqzbeD8au
NNTP-Posting-Host: 89.149.22.17
References: <564ae975-a378-42f5-8048-691605fdeea5n@googlegroups.com>
 <ufs001$2dann$1@dont-email.me> <dca2cab4-21f2-4003-bccf-5da7a33dbc81n@googlegroups.com>
 <a9880a47-f6ea-4097-ad43-5044b830f49fn@googlegroups.com> <9b7345d6-74ba-4cad-9ca4-260737d96c3bn@googlegroups.com>
 <ugej1s$3v7g2$1@dont-email.me>
User-Agent: G2/1.0
MIME-Version: 1.0
Message-ID: <0f492a94-b10c-4658-8d3f-303a785b1b76n@googlegroups.com>
Subject: Re: \advance\toks 41{...}
From: Florent CHERVET <florestan.chervet@gmail.com>
Injection-Date: Sat, 28 Oct 2023 20:59:55 +0000
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
X-Received-Bytes: 3754
Bytes: 4008
Lines: 58

Le samedi 14 octobre 2023 =C3=A0 19:27:28 UTC+2, Texboy a =C3=A9crit=C2=A0:
> Le 13/10/2023 =C3=A0 15:13, G=C3=A9rard Lemenn a =C3=A9crit :=20
>=20
> >> Pour =C3=AAtre plus pr=C3=A9cis :=20
> >> \toks 0=3D\toks 10=20
> >>=20
> >> est ultra rapide, car alors \toks 0 pointe sur la m=C3=AAme chaine que=
 \toks 10=20
> >> Il n'y a pas de copie du contenu de \toks 10 dans \toks 0.
> Ah ?=20
>=20
> \toks0=3D{A} \toks10=3D{B}=20
> \toks0=3D\toks10=20
> \toks10=3D{C}=20
> \showthe\toks0 % donne B et non pas C
> > Et on a besoin d'un token auxilliaire pour stocker la chaine =C3=A0 ajo=
uter.=20
> > % ------------------------------------------=20
> > \def\toks@advance {\expandafter \toks@adva \romannumeral-`\q }%=20
> > \def\toks@adva {\afterassignment \toks@advb \let\@let@token =3D}%=20
> > \def\toks@advb {\ifx \toks\@let@token \begingroup=20
> > \afterassignment\toks@advc \count@ =3D%=20
> > \else \afterassignment\toks@advd \toks@advtok \fi=20
> > }% \toks@advb=20
> > \def\toks@advc {\expandafter \endgroup \expandafter \toksdef \expandaft=
er \@let@token \the\count@=20
> > \afterassignment\toks@advd \toks@advtok=20
> > }% \toks@advc=20
> > \def\toks@advd {\@let@token =3D\expandafter {\the\expandafter\@let@toke=
n \the\toks@advtok }}%=20
> > % -------------------------------------------
> Oui. Pour le sucre syntaxique et faire la distinction entre=20
> \toks<nombre> et \<toksname>, on peut aussi utiliser \meaning et all=C3=
=A9ger=20
> un peu le code. Ce qui, =C3=A0 mon avis, ne doit pas modifier de fa=C3=A7=
on=20
> significative la vitesse d'ex=C3=A9cution (pas test=C3=A9)...=20
>=20
> \newtoks\tmptoks \newcount\tmpcount=20
> \def\toksadd{\expandafter\toksaddA\meaning}=20
> \expandafter\def\expandafter\toksaddA\string\toks{\afterassignment\toksad=
dB\tmpcount}=20
> \def\toksaddB{\afterassignment\toksaddC\tmptoks=3D }=20
> \def\toksaddC{\toks\tmpcount\expandafter{\the\toks\expandafter\tmpcount\t=
he\tmptoks}}

Mon code marcherait m=C3=AAme si on =C3=A9crit :
    \toks@advance \macro
o=C3=B9 \macro commence par : \toks@
Alors que votre code non.

Mais on peut bien s=C3=BBr passer par \meaning si on veut, mais il faut
le \romannumeral-`\q avant de s'occuper du \meaning.

Donc en fait, hormis \global impossible, mon code s'approche le plus de la =
syntaxe de TeX.

Bref, bon weekend !