Usenet 1994 January/usenetsourcesnewsgroupsinfomagicjanuary1994.iso/sources/unix/volume10/crc_plot/part03

Internet Message Format | 1987-07-07 | 46.1 KB

Path: uunet!rs
From: [e-mail] (Rich Salz)
Newsgroups: comp.sources.unix
Subject: v10i047:  CRC Plotting Package, Part03/06
Message-ID: <[e-mail]>
Date: 9 Jul 87 01:14:27 GMT
Organization: UUNET Communications Services, Arlington, VA
Lines: 2020
Approved: [e-mail]

Submitted-by: "Wombat" <[e-mail]>
Posting-Number: Volume 10, Issue 47
Archive-name: crc_plot/Part03



#    This is a shell archive.
#    Remove everything above and including the cut line.
#    Then run the rest of the file through sh.
#----cut here-----cut here-----cut here-----cut here----#
#!/bin/sh
# shar:    Shell Archiver
#    Run the following text with /bin/sh to create:
#    doc
mkdir doc
chdir doc
cat << \SHAR_EOF > alpha.3g
.TH ALPHA 3g "CRC Graphics Package"
.SH NAME
alpha - Return Tektronix to alpha-numeric mode
.SH SYNOPSIS
.nf
>From C:
    alpha()

>From F77:
    call alpha
.fi
.SH DESCRIPTION
The software leaves the Tektronix in the plot state after
each subrotuine call.
If the user writes to the terminal
without a proceeding the write with a call to
.I alpha,
the Tektronix will think that the alpha-numeric data
from the user is valid plot information.
Garbage graphics will result.
.PP
The terminal remains in an alpha-numeric state until
the next graphics call.
.PP
A call to
.I alpha
when plotting
on the HP plotter will cause the software
to flush its internal output buffers.
This allows for pseudo interactive plotting
on the HP plotter.
SHAR_EOF
cat << \SHAR_EOF > axis.3g
.TH AXIS 3g "CRC Graphics Package"
.SH NAME
axis - Draw axis
.SH SYNOPSIS
.nf
>From C:
    float   x,y,size,min,max;
    int     xy,flag;
    char    *label;

    axis(x,y,label,xy,size,min,max,flag);

>From F77:
    real    x,y,size,min,max,
    integer xy,flag
    character label()

    call axis(x,y,label,xy,size,min,max,flag)
.fi
.SH DESCRIPTION
To draw a labeled coordinate axis with numerically annotated
tic-marks at one unit intervals.
.PP
The following parameters are used to specify the details of the axis.
.TP 14
x, y
the coordinates (in plot units) of the starting point
of the axis.
.TP
label
a pointer to a zero terminated string which is to be used
as an axis label.
.TP
xy
One of the following four values indicate the direction and
orientation of the axis.
.br
0: plot axis at zero degrees, tic marks below the axis.
.br
1: plot axis at ninety degrees, tic marks left of the axis.
.br
2: plot axis at zero degrees, tic marks above the axis.
.br
3: plot axis at ninety degrees, tic marks right of the axis.
.TP
size
the length (in plot units) of the coordinate axis.
.TP
min
the numerical value corresponding to the first tic-mark
of the axis (at the point (x,y)), usually the same
as the value computed by 'scale'.
.TP
max
The numerical value corresponding to the last tic-mark of
the axis.
.TP
flag
if flag is positive, the numerical values calculated for
the tic-marks will be truncated to form integers.
If flag is zero, the annotations will be floating point.
If flag is negative, no numerical annotation will be done.
.SH REMARKS
The distance between tic-marks can be changed by setting
the variable
.I ticdis
to the desired distance using subroutine
.I axisv.
Ticdis will be adjusted so that it is an integer divisor
of size.
If ticdis is too small, the annotations at the tic-marks
will run over each other.
.PP
When integer annotations are desired, (
.I max - min
) should be
divisible by (
.I size / ticdis
) to prevent truncation errors.
.PP
When the tic numerical values become very small or very large (in
absolute value), the values will be converted into  base ten
mantissa and characteristic notation with the characteristic
appended to the label.
.PP
The axis itself starts at (x,y) and all the labeling appears
below, next to, or above it. Be sure to allow space for
the labeling.
.PP
The number of significant digits used in the annotations can
be changed by setting the variable
.I digits
to the number
of digits desired using the
.I axisv
subroutine.
SHAR_EOF
cat << \SHAR_EOF > axisv.3g
.TH AXISV 3g "CRC Graphics Package"
.SH NAME
axisv - change parameters of axis
.SH SYNOPSIS
.nf
>From C:
    float    ticdis;
    int    digits;

    axisv(ticdis,digits);

>From F77:
    real    ticdis
    integer digits

    call axisv(ticdis,digits)
.fi
.SH DESCRIPTION
To change the default distance between axis tic marks and the number
of significant digits in the numerical annotations.
The parameters are as follows
.TP 10
ticdis
the distance (in plot units) between tic-marks on the axis.
.TP
digits
the number of significant digits in the numerical annotation
.SH REMARKS
The default values for ticdis and digits are 1.0 and 6, respectively.
.PP
The values will be in effect until the next call to
.I axisv
for all
subsequent calls to
.I axis
and
.I laxis.
.PP
The
.I axis
subroutine will modify
.I ticdis
so that
the length of the axis divided by
.I ticdis
is
an integer.
.SH SEE ALSO
axis(3g),
laxis(3g)
SHAR_EOF
cat << \SHAR_EOF > crclabel.1g
.TH CRCLABEL 1 "CRC Graphics"
.SH NAME
crclabel - Label a CRC plot
.SH SYNOPSIS
crclabel [arguments]
.SH DESCRIPTION
.B Crclabel
is used to add additional labels to a CRC plot.
The
.I xp
and
.I yp
parameters are used to set the location of the
text specified by the
.I tl
option.
.I Qplot(1)
style arguements are used to specify the device where the text
is to be written.
.PP
.B Crcplot
generates a plot on
the default Printronix line printer as defined in
.B gplp
(1).
If ARPA is the host machine, the output will appear on
the Comtal graphics overlay bit-plane zero.
.PP
The following options are
available to modify the graph:

**** POSITIONING AND REDUCING OPTIONS ****
.TP 14
xp=r
The x starting coordinate of the text is moved to
position r.
The default value of r is zero.
.TP 14
yp=r
Same as the `xp=r' flag but the y origin of the text is moved to r.
.TP 14
scfac=r
The graph is expanded or reduced in size by r.
The
default value is 1.0.

.RS 0
**** DEVICE SELECTION OPTIONS ****
.TP 14
-c
The plot is displayed on a Comtal image instead of a
graphic overlay.
The image that is currently being
displayed is used unless the -n flag is used.
.TP 14
-g
Display the plot on a Comtal graphic overlay.
This flag
is set by default if ARPA is the host machine.
The
Comtal is connected to the ARPA machine.
.TP 14
-n
The plot is displayed on the Comtal's graphic overlay n
or Comtal image n.
N can be either 0,1, or 2.
.TP 14
-G
Display the plot on the Grinnel connected to the ARPA
machine.
.TP 14
-t
Display the plot on a Tektronix 4010 or 4014 display.
Standard output must end up on the specified terminal.
.TP 14
-T
Display the plot on a Retro-Graphics RG-512  in
conjunction with an ADM-3A terminal.
.TP 14
-h
Display the plot on an HP plotter.
.TP 14
pen=n
When using the HP plotter, the pen in bin n is used to
plot the graph.
The default bin is 1.
This invokes the `-h' flag.
.TP 14
speed=n
The HP plotter's pen movement velocity will be changed
to `n' cm/sec.
The valid range for n is [1,36].
The
default value is 36.
This option will invoke the `-h'
flag.
.TP 14
-p
The plot is sent to the Printronix line printer through
.B gplp
(1).
This is the default output option for all
host machines except for ARPA.
See
.B gplp
(1) to
determine to which network line printer the output
will be sent.
.TP 14
site=XX
The output is sent to the network line printer
specified by XX.
See
.B opr
(1) for more information
about the allowable printer names.
This option invokes
the `-p' flag.
.TP 14
-v
The plot is sent to the alternate Versatec through a
pipe to gp (1).
.TP 14
-o
The graph is written out to standard output.
.TP 14
g=file
The graph is written out to a file named `file'.
This
file can be sent to the Versatec or Printronix with
.B gp
(1) and
.B gplp
(1), respectively.
.TP 14
op=str
The string is appended to the call to
.B gplp, gp, hpd,
and
.B gd.
This provides a method to change any of
the options to the driver programs.
.TP 14
-b
The display device is not blanked before plotting.
This
option has no effect when used with the `h', `v', `V',
and `p' flags.
If the `g=file' option is used
and `file' exists in the local directory before plotting
begins, then the new graphics will overlay the graphics
contained in `file'.
.TP 14
-B
Turn off the -b flag.
This option is set by default.

.RS 0
**** LABEL OPTIONS ****
.TP 14
tl=str
The string will be used as a label.
The string will be placed at the position on the screen indicated
by the
.I xp
and
.I yp
parameters.
.PP
The parse routines internal to
.B crclabel
allow for two default
mechanisms to specify options.
The first method is to create a
file `.crclabelrc' in your HOME directory (see environ (5)).
In it,
one can place options and flags that
.B crclabel
will use before it
parses your command line.
The syntax is one command or flag
string per line.
.PP
As an example, consider the case where the user wishes to
always obtain output on the Versatec.
The correct format for
the file `.crclabelrc' would be:
.RS 1i
 -v
.RE
.PP
The second method to set default options is to set the
environment variable `CRCLABELARGS'. The format of `CRCLABELARGS' is the
same as the input command line.
.PP
The options set in the previous example can be set using the
following procedure:
.PP
For /bin/sh:
    $CRCLABELARGS='-v'
    $export CRCLABELARGS
.PP
For /bin/csh
    $setenv CRCLABELARGS '-v'
.PP
.B crclabel
will parse the file `.crclabel' first, if it exists.
Then
it will parse `CRCLABELARGS', if it exists.
Finally,
.B crclabel
will parse
the command line.
.SH SEE ALSO:
graphics/crc
qplot(1)
.SH AUTHOR
Hacked together by Malcolm
.SH FILES
.TP 14
/etc/cpu
Contains name of host
.SH BINARIES
This program has not been installed on the system.
Currently it resides in "~malcolm/bin/crclabel" on the
EC, ED, EE, EF and EG Vax computers.
.SH SUPPORT
This was a quick hack by Malcolm and will be supported on a time
available basis by the author.
All of the code was stolen directly from
.I qplot.
SHAR_EOF
cat << \SHAR_EOF > dline.3g
.TH DLINE 3g "CRC Graphics Package"
.SH NAME
dline - draw a parametric line
.SH SYNOPSIS
.nf
>From C:
    float   *x,*y,*dsh,*gap,lx,ly;
    int     n,m;

    dline(x,y,n,dsh,gap,m,lx,ly);

>From F77:
    real    x(),y(),dsh(),gap(),lx,ly
    integer n,m

    call dline(x,y,n,dsh,gap,m,lx,ly)
.fi
.SH DESCRIPTION
To draw a plot of a parametric relationship (x[i],y[i]),
i=0,2,...,n-1, with a dashed line.
The
.I i=n'th
and
.I i=n+1'th
elements of each array should represent the minimum and maximum values
of the points in that vector.
The user can call the routine
.I scale(3g)
to automatically find the minimum and maximum values.
All data values are scaled to these values.
.PP
The following parameters are used to specify the line.
.TP
x,y
pointers to arrays containing the x and y coordinates of a
parametric relationship (x[i],y[i]).
.TP
n
the number of points (x[i],y[i]) in the parametric relationship.
.TP
dsh
a pointer to an array of length m containing the lengths of
the sequence of dashes to be used in plotting the relationship.
.TP
gap
a pointer to an array of length m containing the lengths of
the sequence of  spaces between dashes to be used in plotting the relationship.
.TP
m
the number of dashes and gaps in the dash-gap sequence.
.TP
lx
the length (in plot units) that (xmax - xmin) is mapped to in page
coordinates.
Should be the same value used in the axis subroutine.
.TP
ly
the length (in plot units) that (ymax - ymin) is mapped to in
page coordinates. Should be the value used in the axis subroutine.
.SH REMARKS
Arbitrarily complex dash-gap sequences may be specified.
Dshline
first plots a dash of length dsh[0], then a gap of length
gap[0], then a dash of length dsh[1], etc. After plotting
gap[m-1], the sequence is repeated starting again with dsh[0].
.PP
If
.I m = 0, dsh
and
.I gap
are ignored and the curve is plotted as
a solid line curve with subroutine
.I line.
.PP
The arrays pointed to by
.I x
and
.I y
must be dimensioned at least
.I n+2,
and the minimum and maximum values of each
array must be stored in the
.I n+1st
and
.I n+2nd
array positions
respectively.
.PP
An error will occur if xmin = xmax or ymin = ymax.
SHAR_EOF
cat << \SHAR_EOF > examples.8g
.TH EXAMPLES 8g "CRC Graphics Package"
.SH NAME
examples - several examples using the CRC Grapics Package
.SH "EXAMPLE 1"
The first example will demonstrate the use of Qplot using
the output from a FORTRAN program compiled with `f77'.
Consider the following program:
.nf

    real y1(100),y2(100),x(100),y3(100)
    open(unit=2,file='y1',status='new',form='unformatted')
    open(unit=3,file='y2',status='new',form='unformatted')
    open(unit=7,file='y3',status='new',form='unformatted')
    open(unit=4,file='x',status='new',form='unformatted')
    do 10 i=1,100
    x(i) = (i - 1) * 2.0 * 3.14159 / 99.
    y1(i) = sin(x(i))
    y2(i) = cos(x(i)) * 2.0
    if(i .lt. 50)then
        y3(i) = i
    else
        y3(i) = 100 - i
    end if
10    continue
    write(2)y1
    write(3)y2
    write(7)y3
    write(4)x
    end

.fi
.ti +5
If the program is in a file `test.f', it should be compiled with:
.nf

    $f77. test.f

.fi
.ti +5
Execute the program with:
.nf

    $a.out

.fi
.PP
The program will generate four binary files in F77 binary format.
The file `y1' will contain a sine wave. `y2' will have a
cosine in it. The file `y3' will have a ramp in it. The last
file, `x', will contain the points in which the sine and cosine
were generated from. To convert the f77 binary format to standard
UNIX binary, the following commands should be run:
.nf

    $strip7 y1 y2 y3 x

.fi
.PP
The following Qplot/Gplp sequence will plot the three
curves on the Printronix line printer:
.nf

$qplot y=y1 x=x,4 digits=3 ylen=3.0  'xl=sin($wt)' g=g1
$qplot y=y2 x=,4 digits=3 ylen=3.0 xlen=3.0 yp=4.5 \\
    'yl=cos($wt)' xl=t g=g2
$qplot y=y3 xmin=-1.0 xmax=1.0 len=3.0 yp=4.5 xp=5.0 digits=3 -r \\
    xtic=1.5 "xl=ramp function" g=g3
$gplp g1 g2 g3
$rm g1 g2 g3

.fi
.PP
The output generated is listed on the following page.
.bp
.sp 6
.ce
Output from example 1
.bp
.SH "EXAMPLE 2"
.PP
This example demonstrates a simple user generated graphics
program in FORTRAN.
Consider the following program which writes it output on the HP plotter:
.nf

    call plots(7,0)
    call plot(1.0,1.0,-3)
    call plot(8.0,0.0,2)
    call plot(8.0,8.0,1)
    call plot(0.0,8.0,1)
    call plot(0.0,0.0,1)
    call symbol(.5,4.0,.3,"graphics",0.0)
    call plot(0.0,0.0,999)
    end

.fi
.PP
This program will plot a box with the word 'graphics'
in the middle of it.
It should be compiled with:
.nf

    $f77 prog.f -i -lG

.fi
.PP
To obtain the output use:
.nf

    $a.out

.fi
The output generated is:
.bp
.SH "EXAMPLE 3"
.PP
This example will present a more complicated graphics program
written in C.
Consider the following program written in C:
.nf

#include    <math.h>

main(){
    int    i;
    float    x[631],y[631];

    plots(69,0,"ver");

    for(i=0;i<630;i++){    /* compute data */
        x[i] = cos(3.0 * i / 100) + 1.0;
        y[i] = sin(4.0 * i / 100);
    }

    scale(y,629);    /* get scaling information */
    scale(x,629);

    plot(1.5,1.0,-3); /* move origin away from corner */
    axisv(2.0,3); /* change default ticdis and digits */
    axis(0.,0.,"x axis",0,8.,x[629],x[630],0);
    axis(0.,0.,"y axis",1,8.,y[629],y[630],0);
    plot(0.0,8.0,3);    /* draw border */
    plot(8.0,8.0,2);
    plot(8.0,0.0,2);
    symbol(2.5,8.5,.2,"lissajous figure",0.0);
    line(x,y,629,0,8.0,8.0);

    plot(0.0,0.0,999);    /* terminate plotting */
}

.fi
.PP
This should be compiled with:
.nf

    $cc. prog.c -i -lG -lm

.fi
.PP
To get the
output on the Versatec, use:
.nf

    $a.out

.fi
.PP
The graphics output is listed on the following page.
.bp
.sp 5
.ce
Output from example 3
SHAR_EOF
cat << \SHAR_EOF > factor.3g
.TH FACTOR 3g "CRC Graphics Package"
.SH NAME
factor - change scale factor
.SH SYNOPSIS
.nf
>From C:
    float   sf;

    factor(sf);

>From F77:
    real    sf

    call factor(sf)
.fi
.SH DESCRIPTION
To provide for changing the scale factor, initially 1.0, for
subsequent x and y coordinates.
The
.I sf
parameters is
the new scale factor to be used in scaling all subsequent
x and y coordinates unless and until factor is called again.
All plotting is scaled by this factor thus the user coordinates
(x,y) are scaled to the point (sf*x,sf*y).
SHAR_EOF
cat << \SHAR_EOF > fname.3g
.TH FNAME 3g "CRC Graphics Package"
.SH NAME
fname - change default file name
.SH SYNOPSIS
.nf
>From C:
    char *name;

    fname(name);

>From F77:
    character name()

    call fname(name)
.fi
.SH DESCRIPTION
To change the default file name when the plots (3g) device is 0.
The parameter
.I name
is a pointer to a zero terminated string containing the name of the file that
the final graphics output will be written.
If the name is "-", then the graphics output will be written out
to standard output.
.SH REMARKS
.I Fname
should be called before the call to
.I plots.
SHAR_EOF
cat << \SHAR_EOF > fontinfo.8g
.TH FONTINFO 8g "CRC Graphics Package"
.SH NAME
Character Font Information
.SH DESCRIPTION
The graphics package draws all of its own symbols.
This allows the user great flexibility in the size and type of characters.
The default character set is shown as an ADM-3A keyboard on the
next page.
The left hand side of each key is the character you would
normally see. If the character on the left is preceded by a `$'
then the character on the right is displayed.
(Online a list follows. The first column is the normal character
set and the second column is the alternate font.)
.PP
The `open sup' and `open sub' characters cause an effective
half line shift up and down, respectively. The `close sup'
and `close sub' negate the effects of the respective `open' commands.
The `bs' character will back up one character. This allows for
over-written characters. The `$:' character is a combination
of a `bs' and an overbar
and the `$;' character is a combination of a `bs' and an underbar.
.PP
The `$1' through `$0', `$,', `$*', and `$+' characters are all
on-center symbols used for marking points in space.
The rest of the characters assume that the lower left hand corner
is the start of the character.
.PP
The actual character descriptions are the file:
.nf

/usr/lib/graphics/font.5x7.
.fi
.sp
The routine
.I symbol (3g)
defaults to this file. The user can supply his/her own font file
by using the `fontint' subroutine call.
.PP
A program called
.I genfont
is available
to generate the font file from a set of ascii vector coordinates.
.bp
FONT CORRESPONDENCE
.nf

a alpha                    ; underscore
A ALPHA                    + center +
b beta                    @ @
B BETA                    ` `
c xi                    \\ \\
C XI                    | integral
d delta                    _ backspace
D DELTA                    , center + bs o
e epsilon                < <=
E EPSILON                . bold .
f phi                    > >=
F PHI                    / divide
g gamma                    ? ?
G GAMMA                    1 center ' bs O
h eta                    2 center DELTA bs '
H ETA                    3 center Y
i iota                    4 center X
I IOTA                    5 center diamond bs '
j sigma                    6 center up arrow
J SIGMA                    7 center X over/under bar
k kappa                    8 center Z
K KAPPA                    9 center X bs o
l lambda                0 center box bs '
L LAMBDA                : overbar
m mu                    * center *
M MU                    - +-
n nu                    = equivalent
N NU                    [ half space down
o omicron                { half space up
O OMICRON                ] half space up
p pi                    } half space down
P PI                    ^ angstrom
q theta                    ~ infinity
Q THETA                    ! sqrt
r rho                    " "
R RHO                    # approx
s sigma                    $ $
S SIGMA                    % not equal
t tau                    & mho
T TAU                    ' degree
u upsilon                ( (
U UPSILON                ) )
v partial
V gradient
w omega
W OMEGA
x chi
X CHI
y psi
Y PSI
z zeta
Z ZETA
.fi
SHAR_EOF
cat << \SHAR_EOF > gd.1g
.TH GD 1g "CRC Graphics Package"
.SH NAME
gd - Image Processing Graphics Driver
.SH SYNOPSIS
gd [-bcGio0123] [-mHOST] [if=stdin]
.SH DESCRIPTION
.I
Gd
is an interface program between the output of the
plotting subroutines and an image display device.
Currently,
.I gd
supports Comtal and Grinnell image displays or
graphics overlays.
.PP
The file:
.sp
    /usr/lib/graphics/gd.site
.sp
contains information about which network machine
has these devices.
The file contains `local'
if there are devices connected locally or the
name of the HOST that does have the display devices.
If
.I gd
can not run locally, then the graphics
information is sent to:
.sp
    HOST!/usr/lib/graphics/gd [-Gcio0123]
.sp
where HOST comes from either the file `gd.site' or from
the option `-mHOST'. The mode is determined
by the options the local Gd is invoked with.
.PP
If the
.I -b
flag is used the device is not cleared (blanked) before the graphical data
is written on the screen.
The
.I if=file
parameter specifies that
the graphics information should be read from `file'.
The
default input mode is standard input.
.PP
The device to write on is determined by a combination of the
default action and the flags specified on the command line.
The default action is to write the image on the Comtal Graphics
Overlay number 0.
In addition
on the Comtal it is possible to determine the current image number
being displayed so when writing to a Comtal image plane (-ci) the
default action is to write to that image.
These actions can be modified by using the following options.
.TP 10
-c
Write the data on the Comtal instead of the Grinnell.
.TP
-G
Write the data on the Grinnell instead of the Comtal.
.TP
-g
Write the data to a graphics overlay instead of an image plane.
.TP
-i
Write the data to an image plane instead of a graphics overlay.
.TP
-[n]
Write on either the image plane or graphics overlay labelled
.I n.
The digit
.I n
can be either 0, 1, 2 or 3.
(Note that not all of the image processing hardware is configured
with all 4 planes.)
.SH FILES
.TP 30
/usr/lib/graphics/gd.site
contains default site name
.TP
/etc/cpu
contains local host name
.TP
/dev/Gr/[0123]
Grinnell Image Planes
.TP
/dev/Gr/gov[0123]
Grinnell Graphics Overlays
.TP
/dev/ct[012]
Comtal images
.TP
/dev/ct/tc
Comtal status
.TP
/dev/ct/gov[012]
Comtal graphic overlays
.SH AUTHOR
Carl Crawford
and Malcolm Slaney
.SH SEE ALSO
graphics/crc
SHAR_EOF
cat << \SHAR_EOF > gp.1g
.TH GP 1g "CRC Graphics Package"
.SH NAME
gp - graphics plot to versatec
.SH SYNOPSIS
gp [-ahfi] [-mHOST] file1 file2 ... filen
.SH DESCRIPTION
This command is obsolete on all ECN machines except for
ARPA.
Users desiring output on other ECN versatec machines should
use
.I plot(5g)
format and send the plot to the versatec using
the
.I plot(1g)
command.
.PP
Gp copies graphic files to a Versatec plotter.
The
input files are `or'ed together to form one picture.
The -i flag will force the first file to be read from standard input.
.PP
Gp can be initiated from any machine on the network.
The output will
come out on the Versatec connected to the ARPA machine.
Also file names `0', `1', and `2'
refer to the Comtal graphics overlays 0, 1, and 2,
respectively.
.PP
The option -f produces 4:1 output only on the
ARPA Versatec. The option -h, which is a
default flag, produces 2:1 output.
.PP
The flag -s causes
.I gp
not to output the message given to the
user when
.I gp
finishes.
.PP
The -m flag is used to change the name of the host where the
output is sent to. The characters immediately after `m' form the name
of the new site.
As an example,
the command "gp -marpa file" will always send the output
to the Versatec on the ARPA machine.
.SH FILES
.TP 30
/usr/lib/graphics/gp.site
default site specifications
.TP
/etc/cpu
contains local host name
/dev/vc
Versatec
.SH AUTHOR
Carl Crawford
.SH SEE ALSO
gplp(1),
qplot(1),
graphics/crc
.SH NOTES
The file `gp.site' contains the name of the machine where the
output is routed. If the machine has a Versatec then the name
should be `local #', where # is the length of the Versatec
line buffer.
For the ARPA Versatec, # is 132.
SHAR_EOF
cat << \SHAR_EOF > gplp.1g
.TH GPLP 1g "CRC Graphics Package"
.SH NAME
gplp - graphics print on a line printer
.SH SYNOPSIS
gplp [-XX] [-i] file1 ... filen
.SH DESCRIPTION
.I Gplp
copies graphic files to a Printronix line
printer.
The input files are `or'ed together to form one picture.
The -i flag will force the first file to be read from standard input.
.PP
Depending on which host machine `gplp' is run from, the output will
go to various line printers on the network.
As of this date the following table
indicates which host's Printronix will be used for each originating
computer:
.ta 1i 3i
.nf
    ORIGINATING    OPR SITE
      MACHINE      -XX
       aa       ep
       ca       cp
       cb       cp
       ea       ep
       ec       ep
       ed       ep
       ee       ep
       ef       ep
       eg       em
       ga       ep
       gb       ep
       ka       ep
       ma       mp
       mb       mp
       pa       pp
       pb       pp
       pc       pp
       eeg       ep
      arpa       ep
.fi
.PP
The default site name can be changed using the
-XX option.
See
.I opr(1)
for more information about the
valid printer sites.
.SH FILES
/usr/lib/graphics/gplp.site    contains default site name
.SH AUTHOR
Carl Crawford
.SH SEE ALSO
qplot(1),
gp(1),
graphics/crc
.SH NOTES
The file `gplp.site' has the form `-XX<cr>', where XX is the
name of the default line printer.
SHAR_EOF
cat << \SHAR_EOF > hpd.1g
.TH HPD 1g "CRC Graphics Package"
.SH NAME
hpd - HP plotter driver
.SH SYNOPSIS
hpd [-mHOST] [device]
.SH DESCRIPTION
Hpd is an interface program between the output of the
plotting subroutines and the HP plotter.
It passes standard input to the plotter
specified by the `device' field.
Specification of a host with the `-m' flag
will cause:
.sp
           HOST!/usr/lib/graphics/hpd [device]
.sp
to be executed.
Thus, any plotter on the network can be reached.
.PP
The default HOST is contained in the
file:
.sp
           /usr/lib/graphics/hpd.site.
.sp
The file contains `local' if there is a plotter on the
machine or the name of a HOST that has
a plotter.
.PP
The program will time out after 3 minutes after the
plotter hangs.
.PP
This program unfortunately only knows about one plotter per machine.
The
.I plot(1g)
routine in conjunction with the
.I -P
option to
.I qplot
or
.I plot3d
can be used to send output to the other plotter(s).
.SH FILES
.TP 30
/usr/lib/graphics/hpd.site
contains default site name
.TP
/dev/plt0
default path to plotter
.TP
/etc/cpu
contains local host name
.SH AUTHOR
Carl Crawford
.SH SEE ALSO
graphics/crc
SHAR_EOF
cat << \SHAR_EOF > laxis.3g
.TH LAXIS 3g "CRC Graphics Package"
.SH NAME
laxis - draw a logarithmic axis
.SH SYNOPSIS
.nf
>From C:
    float x,y,size;
    int xy,logmin,logmax,flag;
    char *label;

    laxis(x,y,label,xy,size,logmin,logmax,flag);

>From F77:
    real x,y,size
    integer xy,logmin,logmax,flag
    character label()

    call laxis(x,y,label,xy,size,logmin,logmax,flag)
.fi
.SH DESCRIPTION
.I Laxis
draws
a labelled logarithmic coordinate axis with numerically
annotated tic marks.
The following parameters are used:
.TP 10
x,y
the coordinates of the starting point
of the axis.
.TP
label
a pointer to a zero terminated string which is to be used
as an axis label.
.TP
xy
0: plot the axis at zero degrees, tic marks below the axis.
.br
1: plot the axis at ninety degrees, tic marks to the left of the axis.
.br
2: plot the axis at zero degrees, tic marks above the axis.
.br
3: plot the axis at ninety degrees, tic marks to the right of the axis.
.TP
size
the length of the coordinate axis in plot units.
.TP
logmin
the exponent corresponding to the first tic mark of the axis (at the
point (x,y)), usually the floor of the min computed by 'scale'.
.TP
logmax
the exponent corresponding to the last tic mark of the axis, usually
the ceiling of the max computed by 'scale'.
.TP
flag
0: default, all logarithmic tic marks plotted.
.br
\+\/\- 1: no logarithmic tic marks plotted.
.br
\+\/\-\ \2: only the logarithmic tic marks \
corresponding to 2 and 5 are plotted.
.br
\+\/\- 3: all logarithmic tic marks are plotted.
.br
If flag is negative, no numerical annotation will be done.
.SH REMARKS
The user must supply
.I scale, line,
.I and
.I dline
with the log base 10
of the data to be plotted.
Also, the min and max supplied to
.I line
and
.I dline
must be logmin and logmax, respectively.
.SH AUTHOR
Jeffery L. Gray
SHAR_EOF
cat << \SHAR_EOF > libinfo.8g
.TH LIBINFO 8g "CRC Graphics Package"
.SH NAME
Library Information
.SH DESCRIPTION
The graphics library (/u/lib/libG.a) contains user callable
subroutines for generating graphics.
They all are callable from
a program written in C or F77.
The only restriction on the calling
sequence is that
.I plots
is called before any other plot calls
and `plot(0.0,0.0,999)' is the last call.
It should be noted that
after the plot is terminated,
.I plots(3g)
can be called again.
.PP
Here is a list
of the available routines:
.nf
.ta 1i
alpha    returns Tektronix to alpha-numeric mode
axis    plots numerically annotated axes
axisv    set environment variables for \fIaxis\fR and \fIlaxis\fR
dline    same as \fIline\fR but plots with a dashed line
factor    change the scale factor of the plot
fname    set output file name when graphics device is a file
laxis    plots numerically annotated logarithmic axes
line    plots a line through a set of coordinate pairs
newpen    changes the pen on the HP plotter
number    provides formatted numeric labelling
plot    moves pen and terminates plotting
plots    initializes the plotting devices
scale    computes scale values for \fRline\fR, \fIdline\fR, and \fIsline\fR
site    change the default line printer name in link to Gplp
sline    same as \fIline\fR but can put on-center symbols on the line
speed    changes the plotting speed for the HP plotter
symbol    puts strings on the graph
where    returns current pen position and scale values
.PP
Compile your program with
one of the following commands:
.sp
.nf
    From F77 (11's):
        $f77 prog.f -i -lG
    From F77 (Vax):
        $f77 prog.f -lG
    From C (11's):
        $cc prog.c -i -lG -lm
    From C (Vax):
        $cc prog.c -lG -lm
.fi
.SH SEE ALSO
See the CRC Graphics Introduction for more information.
SHAR_EOF
cat << \SHAR_EOF > line.3g
.TH LINE 3g "CRC Graphics Package"
.SH NAME
line - draw a parametric line
.SH SYNOPSIS
.nf
>From C:
    float   *x,*y,lx,ly;
    int     n,bar;

    line(x,y,n,bar,lx,ly);

>From F77:
    real    x(),y(),lx,ly
    integer n,bar

    call line(x,y,n,bar,lx,ly)
.fi
.SH DESCRIPTION
To draw a line of a parametric relationship (x[i],y[i]), i=0,2,...,n-1.
The
.I i=n'th
and
.I i=n+1'th
elements of each array should represent the minimum and maximum values
of the points in that vector.
The user can call the routine
.I scale(3g)
to automatically find the minimum and maximum values.
The following parameters are used:
.TP 10
x,y
pointers to arrays containing the x and y coordinates of the
parametric relationship (x[i],y[i]).
.TP
n
the number of points in the parametric relationship.
.TP
bar
normally a straight line is drawn between two points but if bar
is non-zero, zero order interpolation will be used between points.
.TP
lx
the length (in plot units) that (xmax - xmin) is mapped to in page
coordinates.
.TP
ly
the length (in plot units) that (ymax - ymin) is mapped to in
page coordinates.
.SH REMARKS
The Xi's and Yi's need not represent plot coordinates in inches.
.PP
The minimum and maximum values for the arrays must be stored
in the n+1st and n+2nd locations of the arrays respectively.
This can be easily done using the
routine
.I scale(3g).
.PP
The point (x[i],y[i]) will be plotted at the page coordinates:
.nf

    x = ((x[i] - x[n]) / (x[n+1] - x[n])) * lx
    y = ((y[i] - y[n]) / (y[n+1] - y[n])) * ly

.fi
relative to the current origin. Hence depending upon the location
of the current origin, the adjusted minima need not actually be
the minima of the data which is to be plotted so long as each point
(x[i],y[i]) has page coordinates which lie within the plotting area.
.PP
An error will occur if xmin = xmax or ymin = ymax.
.PP
The values of
.I lx
and
.I ly
are normally the values passed to
the axis subroutines.
SHAR_EOF
cat << \SHAR_EOF > newpen.3g
.TH NEWPEN 3g "CRC Graphics Package"
.SH NAME
newpen - change pens on the HP plotter
.SH SYNOPSIS
.nf
>From C:
    int     n;

    newpen(n);

>From F77:
    integer n

    call newpen(n)
.fi
.SH DESCRIPTION
To deselect the current pen and select one of the other
pens in the HP plotter.
The command is ignored if the HP plotter is not the selected
output device.
.I N
should be set to
the bin number of the new pen, n=1,2,3,4, which is to be
selected.
.SH REMARKS
Initially, pen 1 is selected.
When a pen is selected, the current pen is raised and the newly
selected pen is left in the up position.
.PP
If n is the number of the currently selected pen, it is raised
and left in the up position.
SHAR_EOF
cat << \SHAR_EOF > number.3g
.TH NUMBER 3g "CRC Graphics Package"
.SH NAME
number - provide numeric labelling on plots
.SH SYNOPSIS
.nf
>From C:
    float   x,y,height,angle;
    char    *format;
    (num type must conform with format)

    number(x,y,height,angle,format,num);

>From F77:
    real    x,y,height,angle
    character format()
    (num type must conform with format)

    call number(x,y,height,angle,format,num)
.fi
.SH DESCRIPTION
This routine plots a single number.
It is usually used as part of plotting the axis and the
following parameters are used.
.TP 10
x,y
the coordinates of the point where the
lower left corner of the first character is to be plotted.
.TP
height
the height (in plot units) of the alphanumeric character
string.
.TP
angle
the angle (in degrees) counterclockwise from the +x direction
at which the string is to be plotted.
.TP
format
a pointer to a zero terminated string which contains the `C format'
by which num is to be converted for plotting.
.TP
num
the number whose value is to be plotted.
.SH REMARKS
>From f77 integers must be printed out using the `c' format %ld, since fortran
integers are two bytes.
.PP
The width of the characters is four-sevenths of height and they
are spaced at intervals of six-sevenths of height. The space
between characters is two-sevenths of height.
.PP
A second set of characters can be accessed by preceeding a valid
character with a `$'. A `$' can be obtained using `$$'. The second
set contains Greek letters and some special mathematical
symbols.
See
.I fontinfo(8g)
for more information about this alternate
character font.
.PP
See the documentation for
.I printf(3)
for more information on the
syntax of the format statement
.I number
expects.
SHAR_EOF
cat << \SHAR_EOF > plot.3g
.TH PLOT 3g "CRC Graphics Package"
.SH NAME
plot - move the pen
.SH SYNOPSIS
.nf
>From C:
    float   x,y;
    int     i;

    plot(x,y,i);

>From F77:
    real    x,y
    integer i

    call plot(x,y,i)
.fi
.SH DESCRIPTION
To provide for moving the pen in a straight line from its
current position to a new position, and for terminating
the plotting subroutines.
The following parameters are used.
.TP 10
x,y
the coordinates of the point (in plot units) to which the
pen is to be moved relative to the current origin.
.TP
i
.nf
+/- 1: do not change vertical position of pen.
+/- 2: put pen into down position.
+/- 3: put pen into up position.
  999: terminate plot subroutines
.fi
.SH REMARKS
.PP
If
.I i
is +/- 1, +/- 2, or +/- 3, the pen is moved from
its current position to the point (x,y) along a straight
line with the vertical position of the pen as specified.
.PP
If
.I i
is -1, -2, or -3, the point (x,y) becomes the new
origin for subsequent plotting. Unless and until the origin
is again redefined, all future coordinates will specify
positions with respect to this point.
.PP
If
.I i
is 999, the values of x and y are disregarded and may as
well be `0.0'.
An error will occur if subsequent plot calls are made
prior to another call to
.I plots.
SHAR_EOF
cat << \SHAR_EOF > plots.3g
.TH PLOTS 3g "CRC Graphics Package"
.SH NAME
plots - initialize CRC Graphics Package
.SH SYNOPSIS
.nf
>From C:
    int     dev,overwrite;
    char    *options;

    plots(dev,overwrite,options);

>From F77:
    integer dev,overwrite
    char options()

    call plots(dev,overwrite,options)
.fi
.SH DESCRIPTION
Plots is used to allocate buffers and initialize the devices
before plotting can begin.
Except for calls to
.I site(3g)
and
.I fname(3g)
this subroutine must be called
before any other subroutines are called.
The following parameters should be passed to
.I plots.
.TP 10
dev
The device the plot should be sent to.  The following device
numbers are currently supported (add 64 to the number in the table
if the option string is used):
.ta .1i 1i
.nf
    0    file or standard output
    8    Versatec through gp(1g) (ARPA Machine Only!!!)
    16    Printronix through gplp (1g) and opr (1g)

    1    Comtal graphics overlay 0(*)
    9    Comtal graphics overlay 1(*)
    17    Comtal graphics overlay 2(*)

    2    Comtal image image displayed(*)
    10    Comtal image 0(*)
    18    Comtal image 1(*)
    26    Comtal image 2(*)

    3    Grinnell graphics overlay 0(*)
    11    Grinnell graphics overlay 1(*)
    19    Grinnell graphics overlay 2(*)
    27    Grinnell graphics overlay 3(*)

    4    Grinnell Image being Displayed (*)
    12    Grinnell Image Plane 0(*)
    20    Grinnell Image Plane 1(*)
    28    Grinnell Image Plane 2(*)
    36    Grinnell Image Plane 3(*)
    44    Grinnell Image Plane 4(*)

    5    Plot Subroutines through plot(I)

    6    Tektronix through standard output
    14    Retro-Graphics through standard output
    22    Tektronix 4113

    7    HP through /u/lib/graphics/hpd
.fi
The output to those devices marked (*) is sent via the
.I gd(1g)
program.
.TP
overwrite
If this parameters is zero then
the specified device will be cleared before plotting.
Otherwise (if overwrite is non zero) then the new plot will overwrite whatever
is currently being displayed.
.TP
options
A null terminated string that contains options for
.I gplp(1g), hpd(1g), gd(1g)
and
.I plot(1g).
The value passed as
.I dev
must be 64 plus the number found in the table above
for this string to be used.
When
.I dev
is equal to 69
then the
.I plot(5)
output is automatically
passed to
.I plot(1g).
The
.I options
string is appened to the string ``-T'' and the result
is passed as the first argument to
.I plot(1g).
Thus use
.I
plots(69,0,"ver")
\fRto use the Versatec through the
.I plot(1g)
program,
.I
plots(71,0,"eg")
\fRto send a plot to the a HP plotter on the `eg' machine.
.I plots(6,0)
is also a valid plots call when the option string isn't needed.
.SH REMARKS
The default file name when
.I dev
equals zero can be changed using the
subroutine
.I fname.
Also if the file
exists in the
local directory before plotting begins, then the
new graphics will overlay the graphics contained in `file'.
.PP
For dev=printronix(16) there exists a subroutine
.I site(3g)
to change the
name of the default line printer.
.PP
Space limitations occur on the PDP 11 computers
when the device selected is a
bit mode device (all dev numbers except for 3, 4, 11).
For these devices the software has to allocate
an 64K byte buffer. If your program has a lot of
text and data, then the graphics software can't get
its space. An error message will printed in these
situations. When you are on the 11's try compiling
your program with the loader option `-i' to get
more space for data.
.PP
The origin is set to (0,0) before plotting begins. This
coordinate is the lower left hand corner of the plot
device.
.PP
The
.I options
field is an optional parameter.
It is intended for advanced applications of the
package with special emphasis placed on the
network interface and future expansion of
device daemons.
.SH NOTE
If the CRC-Graphics routines
.I fname(3g)
and
.I site(3g)
are needed then they
must be called before the first call to
.I plots(3g).
SHAR_EOF
cat << \SHAR_EOF > scale.3g
.TH SCALE 3g "CRC Graphics Package"
.SH NAME
scale - find the minimum and maximum values of a vector
.SH SYNOPSIS
.nf
>From C:
    float   *a;
    int     n;

    scale(a,n);

>From F77:
    real    a()
    integer n

    call scale(a,n)
.fi
.SH DESCRIPTION
This routine is most often used to find the minimum and maximum values
of a vector.
The result is placed at the end of the vector in
locations
.I a[n]
and
.I a[n+1]
(here the array subscripting is assumed to start from zero as in C).
The first argument
.I a
is a pointer to an array
and
.I n
is the number of points in the array.
.SH REMARKS
The array must be dimensioned at least n+2. The minimum value
of the array is stored in a[n] and the maximum is stored in
a[n+1].
.PP
In Fortran arrays are referenced starting from 1 so
the minimum value is stored in a(n+1) and the maximum is stored
in a(n+2).
.PP
The minimum and maximum values are NOT adjusted to pleasing
values as they are with the CDC Calcomp routines.
This is because it
is difficult to define pleasing and the author prefers to see the
maximum and minimum values on the axis.
.PP
If the integer flag is used with the axis subroutine, then the
min and max values returned by scale should be
adjusted so that (max - min) is divisable by (size / ticdis).
SHAR_EOF
cat << \SHAR_EOF > site.3g
.TH SITE 3g "CRC Graphics Package"
.SH NAME
site - change default line printer site
.SH SYNOPSIS
.nf
>From C:
    char *s;

    site(s);

>From F77:
    character s()

    call site(s)
.fi
.SH DESCRIPTION
This routine is used to change the default line printer site for
.I gplp (1g).
The parameter,
.I s,
is a null terminated string of the form: ``-XX'', where
XX is the name of the printer.
.SH REMARKS
.I site
must be called before the call to
.I plots.
.PP
See
.I opr(1)
for more information about the line printer names.
.PP
The line printer must be a Printronix.
SHAR_EOF
cat << \SHAR_EOF > sline.3g
.TH SLINE 3g "CRC Graphics Package"
.SH NAME
sline - draw a parametric line with on-center symbols
.SH SYNOPSIS
.nf
>From C:
    float *x,*y,lx,ly;
    int n,j,sym;

    sline(x,y,n,lx,ly,j,sym);

>From F77:
    real x(),y(),lx,ly
    integer n,j,sym

    call sline(x,y,n,lx,ly,j,sym)
.fi
.SH DESCRIPTION
To draw a line of a parametric relationship (x[i],y[i]), i=0,2,...,n-1,
with on-center symbols plotted every jth point.
The
.I i=n'th
and
.I i=n+1'th
elements of each array should represent the minimum and maximum values
of the points in that vector.
The user can call the routine
.I scale(3g)
to automatically find the minimum and maximum values.
The following parameters are used:
.TP 10
x,y
Pointers to arrays containing the x and y coordinates of the
parametric relationship (x[i],y[i]).
.TP
n
The number of points in the parametric relationship.
.TP
lx
The length (in plot units) that (xmax - xmin) is mapped to in page
coordinates.
.TP
ly
The length (in plot units) that (ymax - ymin) is mapped to in
page coordinates.
.TP
j
Plot the on-center symbol specified by `sym` at every |j|th point.
if j is negative, no line is drawn between on-center symbols.
.TP
sym
Specifies which on-center symbol is to be used.
The following on-center symbols can be used.
.ta 1i 2i
.nf
    \fIsym    on-center symbol\fR

     0    no on-center symbol
     i    $'i-1'    i=1,2,...,10
    11           $*
    12           $+
    13           $,
.fi
.SH REMARKS
The Xi's and Yi's need not represent plot coordinates.
.PP
The minimun and maximum values for the arrays must be stored
in the n+1st and n+2nd locations of the arrays respectively.
.PP
The point (x[i],y[i]) will be plotted at the page coordinates:
.nf

    x = (x[i] - x[n]) / (x[n+1] - x[n]) * lx
    y = (y[i] - y[n]) / (y[n+1] - y[n]) * ly

.fi
relative to the current origin. Hence depending upon the location
of the current origin, the adjusted minima need not actually be
the minima of the data which is to be plotted so long as each point
(x[i],y[i]) has page coordinates which lie within the plotting area.
.PP
An error will occur if xmin = xmax or ymin = ymax.
SHAR_EOF
cat << \SHAR_EOF > speed.3g
.TH SPEED 3g "CRC Graphics Package"
.SH NAME
speed - change plotting speed of the HP plotter
.SH SYNOPSIS
.nf
>From C:
    int vel

    speed(vel);

>From F77:
    integer vel

    call speed(vel)
.fi
.SH DESCRIPTION
The quality of a plot on an HP plotter is dependent on the
quality of the pen and the speed at which it is moved
across the paper.
Generally as a pen ages and dries out a slower velocity can
be used to maintain the quality of the plot.
.PP
The single parameter
.I speed
is the velocity of the pen in cm/sec.
The valid range is [1,36].
The default velocity is 36 cm/sec.
.SH REMARKS
Higher quality output is obtained at lower velocities.
SHAR_EOF
cat << \SHAR_EOF > strip7.1g
.TH STRIP7 1g "CRC Graphics Package"
.SH NAME
strip7 - remove byte counts from f77 data files
.SH SYNOPSIS
strip7 [-r] file1 ...
.SH DESCRIPTION
Strip7 removes the byte counts that f77 puts
into its binary output files.
The 'stripped' files replace the original files.
The '-r' option causes the 'stripped' files
to be placed in a new files. The new file names are
the original file names with a ".s7" suffix.
.SH COMMENT
Plots will be obtained even if 'strip7' is not run before
calling
.I qplot (1g)
and
.I plot3d(1g).
The resulting graphics will be garbage
because
.I qplot
and
.I plot3d
will interpret the byte count as a floating point number and
it is usually much larger than any data in the user`s file..
.SH FILES
.TP 14
strip7.tmp
temp file
.SH AUTHOR
Carl Crawford
.SH BUGS
The maximum record length is only 4096 bytes.
SHAR_EOF
cat << \SHAR_EOF > symbol.3g
.TH SYMBOL 3g "CRC Graphics Package"
.SH NAME
symbol - add alphanumeric labelling to a plot
.SH SYNOPSIS
.nf
>From C:
    float x,y,height,angle;
    char *string;

    symbol(x,y,height,string,angle);

>From F77:
    real x,y,height,angle
    character string()

    call symbol(x,y,height,string,angle)
.fi
.SH DESCRIPTION
The
.I symbol
routine adds alphanumeric notation to a plot.
The location, size, angle and string to be plotted
are specified with the following parameters.
.TP 10
x,y
the coordinates of the point where the
lower left corner of the first character is to be plotted.
.TP
height
the height (in plot units) of the alphanumeric character
string.
.TP
string
a pointer to a zero terminated string.
.TP
angle
the angle (in degrees) counterclockwise from the +x direction
at which the string is to be plotted.
.SH REMARKS
The width of the characters is four-sevenths of height and they
are spaced at intervals of six-sevenths of height. The space
between characters is two-sevenths of height.
.PP
A second set of characters can be accessed by preceeding a valid
character with a `$'. (A `$' can be obtained using `$$'.)
The second
font contains Greek letters and some special mathematical
symbols.
See
.I fontinfo(8g)
for more information about this alternate character
font.
SHAR_EOF
cat << \SHAR_EOF > where.3g
.TH WHERE 3g "CRC Graphics Package"
.SH NAME
where - return current pen position
.SH SYNOPSIS
.nf
>From C:
    float   x,y,sf;

    where(&x,&y,&sf);

>From F77:
    real    x,y,sf

    call where(x,y,sf)
.fi
.SH DESCRIPTION
This routine aids in the optimization of plotting by returning the current
pen coordinates and scale factor to the calling program.
The current state is returned in the following variables:
.TP 10
x,y
variables for the return of the current pen coordinates
relative to the current origin.
.TP
sf
variable for the return of the current scale factor.
.SH REMARKS
Values of
.I x, y,
and
.I sf
when this subroutine is called are
disregarded.
.PP
Subroutine
.I where
can be used, for example, to determine which
direction a line between the points (x1,y1) and (x2,y2) should
be drawn in order to reduce pen movement when the pen position
is unknown.
SHAR_EOF
chdir ..
#    End of shell archive
exit 0

--

Rich $alz            "Anger is an energy"
Cronus Project, BBN Labs    [e-mail]
Moderator, comp.sources.unix    [e-mail]