Hi,
the man page for glFrustum (and several others; available
for example at
https://manpages.opensuse.org/Leap-16.0/Mesa-libGL-devel/glFrustum.3gl.en.gz
and attached for convenience) signal a "division by zero"
error when producing text output with groff 1.23.0:
| $ man ./glFrustum.3gl.en.gz > /dev/null
| troff:<standard input>:226: error: division by zero
| troff:<standard input>:272: error: division by zero
| $
AFAICT, this is caused by the second eqn formula (the first
being "delim $$"):
| .EQ
| down 150
| {left [ ~~~ {
| down 165 matrix {
| ccol { {{2 ~ "near_val"} over {"right" ~-~ "left"}} above 0 above 0 above
0 }
| ccol { 0 above {{2 ~ "near_val"} over {"top" ~-~ "bottom"}} ~ above 0
above 0 }
| ccol { A ~~~~ above B ~~~~ above C ~~~~ above -1 ~~~~}
| ccol { 0 above 0 above D above 0}
| }}
| ~~~ right ]}
| .EN
The text output is also mangled:
| DESCRIPTION
| glFrustum describes a perspective matrix that pro‐
| duces a perspective projection. The current matrix
| (see glMatrixMode) is multiplied by this matrix and
| the result replaces the current matrix, as if
| glMultMatrix were called with the following matrix
| as its argument:
|
| ⎣right_left 0 A 0⎦
| 0 top_bottom B 0
| 0 0 C D
| 0 0 −1 0
|
| A=right_left
|
| B=top_bottom
|
| C=−far_val_near_val
|
| D=−farrval_near_val
The PostScript output looks fine.
Is this an error in the man page? If so, how can one fix
it?
TIA,
Tim
'\" e
'\"! eqn | mmdoc
'\"macro stdmacro
.ds Vn Version 1.2
.ds Dt 24 September 1999
.ds Re Release 1.2.1
.ds Dp Jan 14 18:30
.ds Dm 01 frustum.g
.ds Xs 13962 6 frustum.gl
.TH GLFRUSTUM 3G
.SH NAME
.B "glFrustum
\- multiply the current matrix by a perspective matrix
.SH C SPECIFICATION
void \f3glFrustum\fP(
GLdouble \fIleft\fP,
.nf
.ta \w'\f3void \fPglFrustum( 'u
GLdouble \fIright\fP,
GLdouble \fIbottom\fP,
GLdouble \fItop\fP,
GLdouble \fInear_val\fP,
GLdouble \fIfar_val\fP )
.fi
.EQ
delim $$
.EN
.SH PARAMETERS
.TP \w'\f2left\fP\ \f2right\fP\ \ 'u
\f2left\fP, \f2right\fP
Specify the coordinates for the left and right vertical clipping planes.
.TP
\f2bottom\fP, \f2top\fP
Specify the coordinates for the bottom and top horizontal clipping planes.
.TP
\f2near_val\fP, \f2far_val\fP
Specify the distances to the near and far depth clipping planes.
Both distances must be positive.
.SH DESCRIPTION
\%\f3glFrustum\fP describes a perspective matrix that produces a perspective
projection.
The current matrix (see \%\f3glMatrixMode\fP) is multiplied by this matrix
and the result replaces the current matrix, as if
\%\f3glMultMatrix\fP were called with the following matrix
as its argument:
.P
.sp 5
.ce
.EQ
down 150
{left [ ~~~ {
down 165 matrix {
ccol { {{2 ~ "near_val"} over {"right" ~-~ "left"}} above 0 above 0 above 0 }
ccol { 0 above {{2 ~ "near_val"} over {"top" ~-~ "bottom"}} ~ above 0 above
0 }
ccol { A ~~~~ above B ~~~~ above C ~~~~ above -1 ~~~~}
ccol { 0 above 0 above D above 0}
}}
~~~ right ]}
.EN
.sp
.ce
.EQ
down 130
{A ~=~ {"right" ~+~ "left"} over {"right" ~-~ "left"}}
.EN
.sp
.ce
.EQ
down 130
{B ~=~ {"top" ~+~ "bottom"} over {"top" ~-~ "bottom"}}
.EN
.sp
.ce
.EQ
down 130
{C ~=~ -{{"far_val" ~+~ "near_val"} over {"far_val" ~-~ "near_val"}}}
.EN
.sp
.ce
.EQ
down 130
{D ~=~ -^{{2 ~ "far_val" ~ "near_val"} over {"far_val" ~-~ "near_val"}}}
.EN
.sp 2
.P
Typically, the matrix mode is \%\f3GL_PROJECTION\fP, and
(\f2left\fP, \f2bottom\fP, \-\f2near_val\fP) and (\f2right\fP, \f2top\fP,
\-\f2near_val\fP)
specify the points on the near clipping plane that are mapped
to the lower left and upper right corners of the window,
assuming that the eye is located at (0, 0, 0).
\-\f2far_val\fP specifies the location of the far clipping plane.
Both \f2near_val\fP and \f2far_val\fP must be positive.
.P
Use \%\f3glPushMatrix\fP and \%\f3glPopMatrix\fP to save and restore
the current matrix stack.
.SH NOTES
Depth buffer precision is affected by the values specified for
\f2near_val\fP and \f2far_val\fP.
The greater the ratio of \f2far_val\fP to \f2near_val\fP is,
the less effective the depth buffer will be at distinguishing between
surfaces that are near each other.
If
.sp
.ce
$r ~=~ "far_val" over "near_val"$
.sp
roughly $log sub 2 (r)$ bits of depth buffer precision are lost.
Because $r$ approaches infinity as \f2near_val\fP approaches 0,
\f2near_val\fP must never be set to 0.
.SH ERRORS
\%\f3GL_INVALID_VALUE\fP is generated if \f2near_val\fP or \f2far_val\fP is not
positive, or if \f2left\fP = \f2right\fP, or \f2bottom\fP = \f2top\fP.
.P
\%\f3GL_INVALID_OPERATION\fP is generated if \%\f3glFrustum\fP
is executed between the execution of \%\f3glBegin\fP
and the corresponding execution of \%\f3glEnd\fP.
.SH ASSOCIATED GETS
\%\f3glGet\fP with argument \%\f3GL_MATRIX_MODE\fP
.br
\%\f3glGet\fP with argument \%\f3GL_MODELVIEW_MATRIX\fP
.br
\%\f3glGet\fP with argument \%\f3GL_PROJECTION_MATRIX\fP
.br
\%\f3glGet\fP with argument \%\f3GL_TEXTURE_MATRIX\fP
.br
\%\f3glGet\fP with argument \%\f3GL_COLOR_MATRIX\fP
.SH SEE ALSO
\%\f3glOrtho(3G)\fP,
\%\f3glMatrixMode(3G)\fP,
\%\f3glMultMatrix(3G)\fP,
\%\f3glPushMatrix(3G)\fP,
\%\f3glViewport(3G)\fP
