[brlcad-tracker] [ brlcad-Feature Requests-3152646 ] Coordinates Conversion by Bearing and Distance

[SourceForge.net <no...@so...>]

Feature Requests item #3152646, was opened at 2011-01-06 16:09
Message generated for change (Comment added) made by tbrowder2
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=640805&aid=3152646&group_id=105292

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: Geometry Conversion
Group: Minor Effort
Status: Open
Resolution: None
Priority: 3
Private: No
Submitted By: thardman (thardman)
Assigned to: Nobody/Anonymous (nobody)
Summary: Coordinates Conversion by Bearing and Distance

Initial Comment:
It would be useful to have a way to provide start coordinates and the specify bearing and distance and thus to generate end coordinates. 

This should provide options for "Great Circle" point-to-point on spherical surfaces, and should allow for various formats input and output such as "digital degrees", "degree:minute:second.sss", "degree:minute.mmm, etc. 

Ideally this would be a commandline application that can read a table of input data and output a processed table. 


----------------------------------------------------------------------

Comment By: Tom Browder (tbrowder2)
Date: 2011-02-11 21:53

Message:
While browsing Perl modules today I see that most (if not all) of the
functions needed are in the standard Math::Trig module.  On your Linux
system do "man Math::Trig" and see what I mean.  (Note that "standard"
means that the module is distributed as part of the base Perl package.)

----------------------------------------------------------------------

Comment By: thardman (thardman)
Date: 2011-01-13 17:54

Message:
Extended usage:

beardist -lat lat -lng lng -b bearing -d dist [-ell ellipsoid -ug
units-geometry -ud degree-units -ci CSV-input -co CSV-output -si
space-separated-input -so space-separated-output -a
append-output-to-copy-of-input]

Default usage:

beardist lat long bearing distance

For example 39.7765 -77.4477 315 500

(start at 39.7765 -77.4477 heading NW 500 feet)


Output:

39.77747 -77.44895

Note that all lat long should be in Decimal Degrees, input and output.

I should mention here that this would be, so far as I know, the first
open-source *NIX commandline tool to perform this function. It might become
more ubiquitous than you might think right now. ;)

Note that this should be WGS84 ellipsoid, and also note that the output as
specified needs incredibly little massaging to be in the Google Maps API
format. Five significant figures to the right of the decimal place should
be just fine.

Many thanks!



----------------------------------------------------------------------

Comment By: Sean Morrison (brlcad)
Date: 2011-01-13 15:44

Message:
Let's not get too crazy now.. :)  Implementing output formatting options
isn't usually warranted until a tool is in prevalent use.  I don't see this
tool ever getting to that frankly so the design criteria should just make
the output simple enough that it can be reformatted with other tools.

So that said, what would your ideal default usage and output format look
like?  Be exact.  CSV tables fed on input and produced on output?  What do
the columns contain exactly?  

Which parameters are absolutely essential just to be at all useful, and
which options are just convenience or nice to have?  We need to distill
this down to the smallest possible project that would be of value, and then
build UP capability from there depending on how complicated things get.  We
have to start small though.  Your comments are all over the place at the
moment... :)

Good start, though. 


----------------------------------------------------------------------

Comment By: thardman (thardman)
Date: 2011-01-07 18:21

Message:
Another Useful Option would be the ability to specify ordering of x,y or
y,x for either or both input and output. This also should be something that
can be either specified on the commandline or read from the environment
variable set.

For example

beardist -i0 -o0 [data]

would accept input and would output ordered as latitude longitude.

beardist -i1 -o0 [data]

would accept input as longitude latitude and output as latitude longitude.



----------------------------------------------------------------------

Comment By: thardman (thardman)
Date: 2011-01-07 09:50

Message:
Perhaps a usage could be like this: 

beardist -lat lat -lng lng -b bearing -d dist [-ell ellipsoid -ug
units-geometry -ud degree-units -ci CSV-input -co CSV-output -si
space-separated-input -so space-separated-output -a
append-output-to-copy-of-input]

(Anything in brackets is "special purpose" and could be specified ideally
both on command line or in environmental variables.)

Alternatively, the ellipsoid spec (WGS84, NAD27, whatever) could be an
environmental variable, as could the units of geometry (inches, feet,
yards, perches, miles, nm, mm, cm, m, km etc) and degree format (decimal
degrees, D:M:S.s, D:MM.mm etc).

If all of that was set in env vars, behavior assuming -lat -lng -b and -d
could exist, so that something like 

beardist 39.0000 -77.0000 180 1450

would be assumed to mean a start point at 39.0000 N, 77.0000 W, bearing
due south, 1450 feet.

With env vars set and default behavior, 

beardist < IN-TABLE > OUT-TABLE 

should be able to plow through a table of lat,long,bearing,distance and
write out to a table of end-points.

It might or might not make the most sense for it to default to accepting
input as CSV (comma separated values) for easy interaction with database
and spreadsheet import/export. "PHP friendly" for LAMP use might be a
subsidiary goal. It might make sense for the default to be the -a option,
which takes in a 4-column spec of x,y,bearing,distance and outputs 6
columns, x,y,bearing,distance,x2,y2. Perhaps a sub-option to -a to insert a
different separator between the 4 original columns and the result columns
might be useful, so that 'cut -f 2 -d "_"' could run nicely from 'bash'. 

See also the man page for "cs2cs" under OSGEO's "proj" project. It's
open-source and they probably don't mind sharing.
http://trac.osgeo.org/proj/wiki/man_cs2cs   and
http://trac.osgeo.org/proj/

Cheers, I hope I'm not being too wildly optimistic here. ;)


----------------------------------------------------------------------

Comment By: Sean Morrison (brlcad)
Date: 2011-01-07 00:12

Message:
So what would the exact usage of this tool ideally look like?  Can you
provide a mock input/output "screenshot" showing a few sample command line
uses?

For the tool to be more generally useful, it'd ideally have some option(s)
for specifying geoid/ellipsoid dimensions instead of assuming a perfect
sphere.  The usage should take that into account as well.

The calculators at http://williams.best.vwh.net/gccalc.htm look promising
and the javascript probably wouldn't be too hard to convert, but how those
convert to some sort of command-line tool could be tricky.  If you can sort
those details out, it would greatly simplify the implementation. 

----------------------------------------------------------------------

Comment By: Tom Browder (tbrowder2)
Date: 2011-01-06 18:48

Message:
In case this is approved, I have received permission from Mr. Ed Williams
to use data and code from his aviation formulary site
(http://williams.best.vwh.net/avform.htm), with suitable attribution:

On Thu, Jan 6, 2011 at 17:39, Ed Williams <ed...@ed...> wrote:
> At 02:26 PM 1/6/2011, you wrote:
>>
>> Mr. Williams, would you permit the use of your aviation formulary
>> information to be incorporated into Perl or other code for use on the
>> BRL-CAD web site (<http://brlcad.org>?
>>
>> Thanks for your consideration.
>>
>> Regards,
>>
>> -Tom
>>
>> Thomas M. Browder, Jr.
>> Niceville, Florida
>> USA
>
>  Absolutely.  Just give credit as necessary.

WILCO

Regards,

-Tom

----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=640805&aid=3152646&group_id=105292