=head1 NAME

RhumbSolve -- perform rhumb line calculations

=head1 SYNOPSIS

B<RhumbSolve> [ B<-i> | B<-L> I<lat1> I<lon1> I<azi12> ]
[ B<-e> I<a> I<f> ]
[ B<-d> | B<-:> ] [ B<-w> ] [ B<-p> I<prec> ] [ B<-s> ]
[ B<--comment-delimiter> I<commentdelim> ]
[ B<--version> | B<-h> | B<--help> ]
[ B<--input-file> I<infile> | B<--input-string> I<instring> ]
[ B<--line-separator> I<linesep> ]
[ B<--output-file> I<outfile> ]

=head1 DESCRIPTION

The path with constant heading between two points on the ellipsoid at
(I<lat1>, I<lon1>) and (I<lat2>, I<lon2>) is called the rhumb line or
loxodrome.  Its length is I<s12> and the rhumb line has a forward
azimuth I<azi12> along its length.  Also computed is I<S12> is the area
between the rhumb line from point 1 to point 2 and the equator; i.e., it
is the area, measured counter-clockwise, of the geodesic quadrilateral
with corners (I<lat1>,I<lon1>), (0,I<lon1>), (0,I<lon2>), and
(I<lat2>,I<lon2>).  A point at a pole is treated as a point a tiny
distance away from the pole on the given line of longitude.  The
longitude becomes indeterminate when a rhumb line passes through a pole,
and B<RhumbSolve> reports NaNs for the longitude and the area in this
case.

B<NOTE:> the rhumb line is B<not> the shortest path between two points;
that is the geodesic and it is calculated by GeodSolve(1).

B<RhumbSolve> operates in one of three modes:

=over

=item 1.

By default, B<RhumbSolve> accepts lines on the standard input containing
I<lat1> I<lon1> I<azi12> I<s12> and prints I<lat2> I<lon2> I<S12> on
standard output.  This is the direct calculation.

=item 2.

With the B<-i> command line argument, B<RhumbSolve> performs the inverse
calculation.  It reads lines containing I<lat1> I<lon1> I<lat2> I<lon2>
and prints the values of I<azi12> I<s12> I<S12> for the corresponding
shortest rhumb lines.  If the end points are on opposite meridians,
there are two shortest rhumb lines and the east-going one is chosen.

=item 3.

Command line arguments B<-L> I<lat1> I<lon1> I<azi12> specify a rhumb
line.  B<RhumbSolve> then accepts a sequence of I<s12> values (one per
line) on standard input and prints I<lat2> I<lon2> I<S12> for each.
This generates a sequence of points on a rhumb line.

=back

=head1 OPTIONS

=over

=item B<-i>

perform an inverse calculation (see 2 above).

=item B<-L> I<lat1> I<lon1> I<azi12>

line mode (see 3 above); generate a sequence of points along the rhumb
line specified by I<lat1> I<lon1> I<azi12>.  The B<-w> flag can be used
to swap the default order of the 2 geographic coordinates, provided that
it appears before B<-L>.  (B<-l> is an alternative, deprecated, spelling
of this flag.)

=item B<-e> I<a> I<f>

specify the ellipsoid via the equatorial radius, I<a> and
the flattening, I<f>.  Setting I<f> = 0 results in a sphere.  Specify
I<f> E<lt> 0 for a prolate ellipsoid.  A simple fraction, e.g., 1/297,
is allowed for I<f>.  By default, the WGS84 ellipsoid is used, I<a> =
6378137 m, I<f> = 1/298.257223563.

=item B<-d>

output angles as degrees, minutes, seconds instead of decimal degrees.

=item B<-:>

like B<-d>, except use : as a separator instead of the d, ', and "
delimiters.

=item B<-w>

on input and output, longitude precedes latitude (except that on input
this can be overridden by a hemisphere designator, I<N>, I<S>, I<E>,
I<W>).

=item B<-p> I<prec>

set the output precision to I<prec> (default 3); I<prec> is the
precision relative to 1 m.  See L</PRECISION>.

=item B<-s>

By default, the rhumb line calculations are carried out exactly in terms
of elliptic integrals.  This includes the use of the addition theorem
for elliptic integrals to compute the divided difference of the
isometric and rectifying latitudes.  If B<-s> is supplied this divided
difference is computed using Krueger series for the transverse Mercator
projection which is only accurate for |I<f>| E<lt> 0.01.  See
L</ACCURACY>.

=item B<--comment-delimiter> I<commentdelim>

set the comment delimiter to I<commentdelim> (e.g., "#" or "//").  If
set, the input lines will be scanned for this delimiter and, if found,
the delimiter and the rest of the line will be removed prior to
processing and subsequently appended to the output line (separated by a
space).

=item B<--version>

print version and exit.

=item B<-h>

print usage and exit.

=item B<--help>

print full documentation and exit.

=item B<--input-file> I<infile>

read input from the file I<infile> instead of from standard input; a file
name of "-" stands for standard input.

=item B<--input-string> I<instring>

read input from the string I<instring> instead of from standard input.
All occurrences of the line separator character (default is a semicolon)
in I<instring> are converted to newlines before the reading begins.

=item B<--line-separator> I<linesep>

set the line separator character to I<linesep>.  By default this is a
semicolon.

=item B<--output-file> I<outfile>

write output to the file I<outfile> instead of to standard output; a
file name of "-" stands for standard output.

=back

=head1 INPUT

B<RhumbSolve> measures all angles in degrees, all lengths (I<s12>) in
meters, and all areas (I<S12>) in meters^2.  On input angles (latitude,
longitude, azimuth, arc length) can be as decimal degrees or degrees,
minutes, seconds.  For example, C<40d30>, C<40d30'>, C<40:30>, C<40.5d>,
and C<40.5> are all equivalent.  By default, latitude precedes longitude
for each point (the B<-w> flag switches this convention); however on
input either may be given first by appending (or prepending) I<N> or
I<S> to the latitude and I<E> or I<W> to the longitude.  Azimuths are
measured clockwise from north; however this may be overridden with I<E>
or I<W>.

For details on the allowed formats for angles, see the C<GEOGRAPHIC
COORDINATES> section of GeoConvert(1).

=head1 PRECISION

I<prec> gives precision of the output with I<prec> = 0 giving 1 m
precision, I<prec> = 3 giving 1 mm precision, etc.  I<prec> is the
number of digits after the decimal point for lengths.  For decimal
degrees, the number of digits after the decimal point is I<prec> + 5.
For DMS (degree, minute, seconds) output, the number of digits after the
decimal point in the seconds component is I<prec> + 1.  The minimum
value of I<prec> is 0 and the maximum is 10.

=head1 ERRORS

An illegal line of input will print an error message to standard output
beginning with C<ERROR:> and causes B<RhumbSolve> to return an exit code
of 1.  However, an error does not cause B<RhumbSolve> to terminate;
following lines will be converted.

=head1 ACCURACY

The algorithm used by B<RhumbSolve> uses exact formulas for converting
between the latitude, rectifying latitude (I<mu>), and isometric
latitude (I<psi>).  These formulas are accurate for any value of the
flattening.  The computation of rhumb lines involves the ratio (I<psi1>
- I<psi2>) / (I<mu1> - I<mu2>) and this is subject to large round-off
errors if I<lat1> is close to I<lat2>.  So this ratio is computed using
divided differences using one of two methods: by default, this uses the
addition theorem for elliptic integrals (accurate for all values of
I<f>); however, with the B<-s> options, it is computed using the series
expansions used by TransverseMercatorProj(1) for the conversions between
rectifying and conformal latitudes (accurate for |I<f>| E<lt> 0.01).
For the WGS84 ellipsoid, the error is about 10 nanometers using either
method.

=head1 EXAMPLES

Route from JFK Airport to Singapore Changi Airport:

   echo 40:38:23N 073:46:44W 01:21:33N 103:59:22E |
   RhumbSolve -i -: -p 0

   103:34:58.2 18523563

N.B. This is B<not> the route typically taken by aircraft because it's
considerably longer than the geodesic given by GeodSolve(1).

Waypoints on the route at intervals of 2000km:

   for ((i = 0; i <= 20; i += 2)); do echo ${i}000000;done |
   RhumbSolve -L 40:38:23N 073:46:44W 103:34:58.2 -: -p 0

   40:38:23.0N 073:46:44.0W 0
   36:24:30.3N 051:28:26.4W 9817078307821
   32:10:26.8N 030:20:57.3W 18224745682005
   27:56:13.2N 010:10:54.2W 25358020327741
   23:41:50.1N 009:12:45.5E 31321269267102
   19:27:18.7N 027:59:22.1E 36195163180159
   15:12:40.2N 046:17:01.1E 40041499143669
   10:57:55.9N 064:12:52.8E 42906570007050
   06:43:07.3N 081:53:28.8E 44823504180200
   02:28:16.2N 099:24:54.5E 45813843358737
   01:46:36.0S 116:52:59.7E 45888525219677

=head1 SEE ALSO

GeoConvert(1), GeodSolve(1), TransverseMercatorProj(1).

An online version of this utility is availbable at
L<https://geographiclib.sourceforge.io/cgi-bin/RhumbSolve>.

The Wikipedia page, Rhumb line,
L<https://en.wikipedia.org/wiki/Rhumb_line>.

=head1 AUTHOR

B<RhumbSolve> was written by Charles Karney.

=head1 HISTORY

B<RhumbSolve> was added to GeographicLib,
L<https://geographiclib.sourceforge.io>, in version 1.37.