Re: [find_orb] Problems compiling the github version

Bill Gray Apr 7, 2017

Hi Markus,

Hmmm... after compiling the 'lunar' code, did you run 'make install'?

Doing so should cause 'date.h', 'watdefs.h', and other include files
to be copied to ~/include, and libraries to ~/lib. If you did run
'make install', I would check to make sure that those directories exist
and that date.h and watdefs.h got copied to them.

If those files are indeed present in those directories, we'd have
to wonder why they aren't found when 'jpl_eph' is built. Some systems
are set up to look for include files in ~/include and libraries in ~/lib
by default. However, when you build jpl_eph, the very first command
to be executed ought to be

g++ -Wall -O3 -Wextra -pedantic -I ~/include -c asc2eph.cpp

...with the '-I ~/include' telling the compiler to search for date.h
and friends in that directory, and similarly for everything else that
gets compiled.

The other projects (jpl_eph, sat_code, find_orb) also all require
that you run both 'make' and 'make install'.

As it happens, I wrote up some installation instructions for the
Catalina Sky Survey last month, going into a good bit of detail on
getting Find_Orb (and its non-interactive sibling 'fo') installed
on Linux boxes. At some point, I'll modify this to be less CSS
specific. But it should be useful to others "as is", so... here
it is :

https://www.projectpluto.com/install.htm

It also describes some other tools that may (or may not) be of
interest to some on this list, such as the 'integrat' program for
numerically integrating MPCORB.DAT from one epoch to another, code
to download NEOCP astrometry (used for my on-line NEOCP list at

https://www.projectpluto.com/neocp2/summary.htm

), code to generate an index to MPECs (see

https://www.projectpluto.com/mpecs/2017.htm#the_end

for the resulting index), and so forth.

-- Bill

On 2017-04-07 01:45, Markus Gregor Kempf m.kempf@... [find_orb] wrote:
>
> Bill,
>
> I tried to compile the github versions of find_orb, jpl_eph, sat_code
> and lunar on a RHEL 7 64bit Linux system. Sat_code and lunar worked
> flawlessly. Jpl_eph has a missing date.h file. I tried to substitute
> another date.h, but this caused several other problems. find_orb itself
> has a missing watdefs.h in fo.cpp.
>
> Markus
>
>
>
> ------------------------------------
> Posted by: Markus Gregor Kempf <m.kempf@...>
> ------------------------------------
>
>
> ------------------------------------
>
> Yahoo Groups Links
>
>
>