Received: by nummer-3.proteosys id <01C19443.A7881C3C@nummer-3.proteosys>; Thu, 3 Jan 2002 11:44:43 +0100 Return-Path: MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----_=_NextPart_001_01C19443.A7881C3C" x-vm-v5-data: ([nil nil nil nil nil nil nil nil nil][nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil]) X-MimeOLE: Produced By Microsoft Exchange V6.5 Content-class: urn:content-classes:message Subject: Ignoring spaces before \index{} Date: Thu, 23 Apr 1992 15:12:37 +0100 Message-ID: X-MS-Has-Attach: X-MS-TNEF-Correlator: From: "Nelson H. F. Beebe" Sender: "Mailing list for the LaTeX3 project" To: "Rainer M. Schoepf" Reply-To: "Mailing list for the LaTeX3 project" Status: R X-Status: X-Keywords: X-UID: 679 This is a multi-part message in MIME format. ------_=_NextPart_001_01C19443.A7881C3C Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable [I'm away at a workshop this week, so I've had little time to respond to mail; thus the delay in responding]. Rainer Sch\"opf asks about my observation that multiple \index{} commands separated by newline could give different output, depending on whether they were written foo% \index{...} \index{...} or foo% \index{...}% \index{...} The only difference here is the addition of a single % following the first \index{}. In my comments, I was not thinking particularly of the case where a page break happens to fall between a word and its index entry, as in "foo \index{...}". That is a known problem that Leslie Lamport documented on p. 78 of the LaTeX User Guide and Reference Manual. The spacing problem that I was commenting on is more subtle. You can perhaps predict it from the discussion on p. 153 (particularly the footnote) of LUGRM. It has proved difficult to come up with an example that you'll be able to reproduce locally. I took a 57-page document with a 7-page index containing 1079 index entries, and prepared two versions: one with trailing percent on multiple \index entries (the second case above), and one without (the first case above). Comparing the PostScript output by dvialw for these two documents revealed 13 places where there are differences in vertical spacing. Most are hard to see in the output, with differences of 2 to 5 dots (at 300dpi) between successive lines; I never noticed them in numerous proofreadings of the document. However, one stands out glaringly; it arises from the LaTeX input \item Explicit representation of undefined values, called NaNs (Not A Number); they can be quiet (non-interrupting), or signalling (interrupting).% \index{floating-point!NaN}% \index{NaN|see{floating-point}} Without the percent following the first \index{}, there is an additional blank line in the output (40 vertical dots at 300dpi =3D=3D 9.6pt). The output is formatted something like this (from dvitty): >>... >> The most important features of IEEE 754 arithmetic are: >> >> o Explicit representation of signed and unsigned infinity. >> >> o Explicit representation of undefined values, called NaNs (Not A = Num- >> ber); they can be quiet (non-interrupting), or signalling = (interrupting). >> >> >> >> o Gradual underflow, so that below the normal underflow limit, = preci- >> sion is allowed to decrease to widen the floating-point range. >>... Notice the extra blank lines following the second bulleted item. With the addition of the single percent, the output looks like this (again from dvitty): >>... >> The most important features of IEEE 754 arithmetic are: >> >> o Explicit representation of signed and unsigned infinity. >> >> o Explicit representation of undefined values, called NaNs (Not A = Num- >> ber); they can be quiet (non-interrupting), or signalling = (interrupting). >> >> o Gradual underflow, so that below the normal underflow limit, = preci- >> sion is allowed to decrease to widen the floating-point range. >>... I tried to extract the LaTeX input fragment to incorporate as an example for you to run locally. Unfortunately, when I switch back to Computer Modern from the AP-TeX New Century Schoolbook font, which few of you are likely to have, the difference disappears. The sentence in the second bulleted item ends at the right margin with New Century Schoolbook, and a few points to the left of the right margin with Computer Modern, and the additional vertical space disappears in the Computer Modern output. With some more work, I could probably come up with an example that would show the difference in Computer Modern, but I think the point is clear: the typeset output from multiple adjacent \index{} entries may NOT be insensitive to intervening whitespace. I believe that this is an undesirable feature that should be fixed if it is possible to do so. =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D Nelson H.F. Beebe Center for Scientific Computing Department of Mathematics 220 South Physics Building University of Utah Salt Lake City, UT 84112 USA Tel: (801) 581-5254 FAX: (801) 581-4148 Internet: beebe@math.utah.edu =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D ------_=_NextPart_001_01C19443.A7881C3C Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Ignoring spaces before \index{}

[I'm away at a workshop this week, so I've had little = time to respond
to mail; thus the delay in responding].

Rainer Sch\"opf asks about my observation that = multiple \index{}
commands separated by newline could give different = output, depending
on whether they were written

foo%
\index{...}
\index{...}

or

foo%
\index{...}%
\index{...}

The only difference here is the addition of a single % = following the
first \index{}.

In my comments, I was not thinking particularly of the = case where a
page break happens to fall between a word and its = index entry, as in
"foo \index{...}".  That is a known = problem that Leslie Lamport
documented on p. 78 of the LaTeX User Guide and = Reference Manual.  The
spacing problem that I was commenting on is more = subtle.  You can
perhaps predict it from the discussion on p. 153 = (particularly the
footnote) of LUGRM.

It has proved difficult to come up with an example = that you'll be able
to reproduce locally.  I took a 57-page document = with a 7-page index
containing 1079 index entries, and prepared two = versions: one with
trailing percent on multiple \index entries (the = second case above),
and one without (the first case above).

Comparing the PostScript output by dvialw for these = two documents
revealed 13 places where there are differences in = vertical spacing.
Most are hard to see in the output, with differences = of 2 to 5 dots
(at 300dpi) between successive lines; I never noticed = them in numerous
proofreadings of the document.   However, = one stands out glaringly; it
arises from the LaTeX input

  \item Explicit representation of = undefined
        values, = called NaNs (Not A Number); they
        can be = quiet (non-interrupting), or
        signalling = (interrupting).%
\index{floating-point!NaN}%
\index{NaN|see{floating-point}}

Without the percent following the first \index{}, = there is an
additional blank line in the output (40 vertical dots = at 300dpi =3D=3D
9.6pt).  The output is formatted something like = this (from dvitty):


>>...
>>  The most important features of IEEE = 754 arithmetic are:
>>
>>  o Explicit representation of signed = and unsigned infinity.
>>
>>  o Explicit representation of undefined = values, called NaNs (Not A Num-
>>    ber); they can be quiet = (non-interrupting), or signalling (interrupting).
>>
>>
>>
>>  o Gradual underflow, so that below the = normal underflow limit, preci-
>>    sion is allowed to = decrease to widen the floating-point range.
>>...

Notice the extra blank lines following the second = bulleted item.

With the addition of the single percent, the output = looks like this
(again from dvitty):

>>...
>>  The most important features of IEEE = 754 arithmetic are:
>>
>>  o Explicit representation of signed = and unsigned infinity.
>>
>>  o Explicit representation of undefined = values, called NaNs (Not A Num-
>>    ber); they can be quiet = (non-interrupting), or signalling (interrupting).
>>
>>  o Gradual underflow, so that below the = normal underflow limit, preci-
>>    sion is allowed to = decrease to widen the floating-point range.
>>...

I tried to extract the LaTeX input fragment to = incorporate as an
example for you to run locally.  Unfortunately, = when I switch back to
Computer Modern from the AP-TeX New Century = Schoolbook font, which few
of you are likely to have, the difference = disappears.  The sentence in
the second bulleted item ends at the right margin = with New Century
Schoolbook, and a few points to the left of the right = margin with
Computer Modern, and the additional vertical space = disappears in the
Computer Modern output.

With some more work, I could probably come up with an = example that
would show the difference in Computer Modern, but I = think the point is
clear: the typeset output from multiple adjacent = \index{} entries may
NOT be insensitive to intervening whitespace.  I = believe that this is
an undesirable feature that should be fixed if it is = possible to do
so.


=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
Nelson H.F. Beebe
Center for Scientific Computing
Department of Mathematics
220 South Physics Building
University of Utah
Salt Lake City, UT 84112
USA
 Tel: (801) 581-5254
 FAX: (801) 581-4148
 Internet: beebe@math.utah.edu
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=

------_=_NextPart_001_01C19443.A7881C3C--