[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

general considerations about lout

From: Olaf Rogalsky
Subject: general considerations about lout
Date: Thu, 28 Dec 2000 18:27:51 +0100


while trying to incorporate a different font set for symbols
and greek letters into the equation package I was faced with
a problem in the design of lout.

Lout (horizontally) concatenate objects such that row marks
merge. The row mark of a letter is defined to be half of
XHeight (as defined in the metrics file of the corresponding
font, or 500 if not defined therein) above the origin of the
letters coordinate system.

The x axis of a letter in postscript fonts usually is chosen
to coincide with the base line of the font. This ensures, that
glyphs of different fonts and sizes printed next to each other
lie pleasantly on the same base line.

In contrast, with lout glyphs of different fonts or sizes usually
do not lie on the same base line. This is demonstrated in the
following two examples:

{Times  Base 48p} @Font A{Times Base 12p} @Font bc  #Example 1
{Symbol Base 48p} @Font h{Times Base 48p} @Font h   #Example 2

In the first example this is very clearly visible, whereas in the
second example the effect is rather smallish but nevertheless leads
to a wiggled appearance. It is a consequence of XHeight beeing 450
units in the Times-Roman font and 500 units (actually it is not
defined in the afm file) in the Symbol font.

For various reasons it is not desirable to adjust the XHeight in the
afm files for lout in order to solve this problem. Further more
there is no automatic way to get access to the current base line
except by escapeing to the backend with @Graphic, but then there is
no way to pass this knowledge back to lout.

A possible solution would be to introduce an addtional unit of
measurement, say 'x', which is the XHeight of the current font.
The second example then could be written as:

{Symbol Base 48p} @Font "+0.5x" @VShift h
{Times Base 48p}  @Font "+0.5x" @VShift h

A better solution would be to let lout's row mark coincide with the
x axis of the corresponding font. In this case the 'x' unit from
above still is needed for determining the "center" of a row.
Unfortunately this would require a major redesign and recoding of

Have a nice party, Olaf

PS: Of course I am aware of the possibility to @VShift things
    individually into the right position. But freeing me from
    such adjustments is what a text formating system is all

                                  (. .)
I Dipl. Phys. Olaf Rogalsky                 Institut f. Theo. Physik I I
I Tel.: 09131 8528440                       Univ. Erlangen-Nuernberg   I
I Fax.: 09131 8528444                       Staudtstrasse 7 B3         I
I address@hidden  D-91058 Erlangen           I

reply via email to

[Prev in Thread] Current Thread [Next in Thread]