Received: by nummer-3.proteosys id <01C19443.41927BFC@nummer-3.proteosys>; Thu, 3 Jan 2002 11:41:52 +0100 MIME-Version: 1.0 x-vm-v5-data: ([nil nil nil nil nil nil nil t nil][nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil]) Content-Type: multipart/alternative; boundary="----_=_NextPart_001_01C19443.41927BFC" X-MimeOLE: Produced By Microsoft Exchange V6.5 Content-class: urn:content-classes:message Subject: Attributes: a possible syntax Date: Mon, 22 Apr 1991 19:39:45 +0100 Message-ID: X-MS-Has-Attach: X-MS-TNEF-Correlator: From: Sender: "LaTeX-L Mailing list" To: "Rainer M. Schoepf" Reply-To: "LaTeX-L Mailing list" Status: R X-Status: X-Keywords: X-UID: 328 This is a multi-part message in MIME format. ------_=_NextPart_001_01C19443.41927BFC Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable In an earlier message to LaTeX-L, I made reference to the concept of = static and dynamic variables which formed a fundamental part of Ps-Fonts, and noted = in passing that the syntax was liable to change. I have now re-written = that aspect of Ps-Fonts, and wonder whether the new syntax (see example = below) might provide a possible basis for the proposed `attributes' feature of = LaTeX-3. The new syntax resembles (and indeed, is intended to resemble) the = concept known in Algol-68 as a `serial clause'; to wit, it is a `closed clause' = (in TeX terms, a closed group), separated by semi-colons. Two examples from = Ps-Fonts follow: \dynamic {\baselineskip =3D 1.2 \fontsize; \parskip =3D 0.5 \baselineskip; \strutheight =3D 0.85 \fontsize; \strutdepth =3D 0.35 \fontsize } \static {\abovedisplayskip =3D \parskip; \belowdisplayskip =3D \parskip; \abovedisplayshortskip =3D \parskip; \belowdisplayshortskip =3D \parskip; \topskip =3D 12 pt; \bigskipamount =3D \baselineskip; \medskipamount =3D 0.5 \baselineskip; \smallskipamount =3D 0.25 \baselineskip } The text is easy to parse (no special catcodeing of `;' is necessary), = and left- and right components are easily distinguishable. Furthermore it = is simple to ascertain whether the l-h-s is complex (implying an error). = This would allow Michael Downes token-list lookup, for example, to be easily = added, to allow different semantics based on the `type' of the l-h-s. I omit = the details of the parser in this initial submission, but of course I will = supply them if the idea is of interest. Philip Taylor Royal Holloway and Bedford New College, ``The University of London at Windsor'' ------_=_NextPart_001_01C19443.41927BFC Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Attributes: a possible syntax

In an earlier message to LaTeX-L, I made reference to = the concept of static and
dynamic variables which formed a fundamental part of = Ps-Fonts, and noted in
passing that the syntax was liable to change.  I = have now re-written that
aspect of Ps-Fonts, and wonder whether the new syntax = (see example below) might
provide a possible basis for the proposed = `attributes' feature of LaTeX-3.

The new syntax resembles (and indeed, is intended to = resemble) the concept
known in Algol-68 as a `serial clause'; to wit, it is = a `closed clause' (in TeX
terms, a closed group), separated by = semi-colons.  Two examples from Ps-Fonts
follow:

        \dynamic = {\baselineskip =3D 1.2 \fontsize;
          &nbs= p;       \parskip =3D 0.5 = \baselineskip;
          &nbs= p;       \strutheight =3D 0.85 = \fontsize;
          &nbs= p;       \strutdepth =3D 0.35 = \fontsize
          &nbs= p;      }

        \static = {\abovedisplayskip =3D \parskip;
          &nbs= p;      \belowdisplayskip =3D \parskip;
          &nbs= p;      \abovedisplayshortskip =3D = \parskip;
          &nbs= p;      \belowdisplayshortskip =3D = \parskip;
          &nbs= p;      \topskip =3D 12 pt;
          &nbs= p;      \bigskipamount =3D = \baselineskip;
          &nbs= p;      \medskipamount =3D 0.5 = \baselineskip;
          &nbs= p;      \smallskipamount =3D 0.25 = \baselineskip
          &nbs= p;     }

The text is easy to parse (no special catcodeing of = `;' is necessary), and
left- and right components are easily = distinguishable.  Furthermore it is
simple to ascertain whether the l-h-s is complex = (implying an error).  This
would allow Michael Downes token-list lookup, for = example, to be easily added,
to allow different semantics based on the `type' of = the l-h-s.  I omit the
details of the parser in this initial submission, but = of course I will supply
them if the idea is of interest.

        =         =         =         =         Philip = Taylor
        =         =             Royal Holloway and Bedford New = College,
        =         =             ``The University of London at = Windsor''

------_=_NextPart_001_01C19443.41927BFC--