From: Lawrence G. Mayka
Subject: Printing a symbol whose name is a potential number
Date: 
Message-ID: <LGM.93Jun20151631@excalibur.flw.att.com>
Fact 1: CLtL2 (p. 517, text unchanged in CLtL1 and the dpANS) says
that a potential number can have "number markers," but does not define
this latter term precisely.  For example, it is not clear whether
letters other than s, S, f, F, d, D, l, L, e, and E can ever be
"number markers" in a potential number, and if so, in what
circumstances.

Fact 2: CLtL2 (p. 552, text unchanged in CLtL1 but dropped in the
dpANS) says that the standard printer should add escape characters to
a symbol name when that name would otherwise be a potential number.

The net result is that implementation behaviors vary.  Given the
symbols |1.0E|, |1E|, and |1A|:

- At least 3 implementations use escape characters for all.

- At least 2 implementations use escape characters for none.

- At least 1 implementation uses escape characters for |1.0E| only.

- At least 1 implementation uses escape characters for |1.0E| and |1E|
only.

Any ideas on how this was meant to be?
--
        Lawrence G. Mayka
        AT&T Bell Laboratories
        ยทยทยท@iexist.att.com

Standard disclaimer.
From: Pekka P Pirinen
Subject: Re: Printing a symbol whose name is a potential number
Date: 
Message-ID: <PEKKA.93Jun21140706@gollum.harlqn.co.uk>
I don't think there is any serious doubt here about the committee's
intent:

1. All letters (I presume this means all alphabetic1 characters) are
number markers, unless they are adjacent to another letter, or happen
to be digits.

The text is somewhat ill-organized (the part about digits appears as a
separate paragraph lower down), but clearly, no other constraints on
number markers are intended.  Some of the examples of potential
numbers in CLtL2 (p.517, identical in CLtL1 and the dpANS) are
isomorphic to yours: 777777q to 1A, and 1.7J to 1.0E.

2. The printer should insert escape characters when printing a symbol
whose print name has the syntax of a potential number (when
*PRINT-ESCAPE* or *PRINT-READABLY* is true).

The sentence about potential numbers that was dropped from the dpANS
has been replaced by a whole chapter: 22.1.2.  Admittedly, that
chapter is written in an oblique way, talking about printing potential
numbers rather than printing symbols.  The chapter on printing symbols
now only says that escape characters are included "as required"
without actually stating any requirements, but then, we always took
the general description of print-read consistency -- vague as it is --
to lay down the requirements.

Pekka P. Pirinen
Harlequin Limited, Cambridge, England