Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > comp.graphics.apps.gnuplot > #3714 > unrolled thread

Ephemerides data

Back to article view | Back to comp.graphics.apps.gnuplot

Contents

  Ephemerides data robertvanpass@gmail.com - 2017-08-20 06:36 -0700
    Re: Ephemerides data Karl Ratzsch <mail.kfr@gmx.net> - 2017-08-20 22:56 -0300
      Re: Ephemerides data Karl Ratzsch <mail.kfr@gmx.net> - 2017-08-20 23:59 -0300
        Re: Ephemerides data robertvanpass@gmail.com - 2017-08-21 15:01 -0700
          Re: Ephemerides data Karl Ratzsch <mail.kfr@gmx.net> - 2017-08-22 03:59 +0200
            Re: Ephemerides data robertvanpass@gmail.com - 2017-08-22 07:57 -0700
              Re: Ephemerides data Karl Ratzsch <mail.kfr@gmx.net> - 2017-08-22 19:09 +0200
                Re: Ephemerides data robertvanpass@gmail.com - 2017-08-22 14:09 -0700
                  Re: Ephemerides data Karl Ratzsch <mail.kfr@gmx.net> - 2017-08-23 01:45 +0200
                    Re: Ephemerides data robertvanpass@gmail.com - 2017-08-23 13:09 -0700

#3714 — Ephemerides data

Hi,

I have difficulties to plot ephemerides data.
They are generated by SkytechX in the following format:

Date;Time;Local R.A.;Local Dec.;
8/20/2017;12:00:00 UTC+1.0;10h 35m 50.60s;+03° 47' 15.38";
8/21/2017;12:00:00 UTC+1.0;10h 33m 13.63s;+04° 02' 23.70";
8/22/2017;12:00:00 UTC+1.0;10h 30m 23.99s;+04° 20' 34.64";
...

The column "Local R.A." is in time format: Hours Minits Seconds,
should be converted to: Hours + Minits/60 + Seconds/3600.

Column "Local Dec." is in angle format: Degrees Arcminits Arcseconds,
should be converted to Degrees + Arcminits/60 + Arcseconds/3600.

So the first record should become:
8/20/2017;12:00:00 UTC+1.0;10.587606;3.787606";

How to implement this convertion to decimal formats in Gnuplot ?

Many thanks !

[toc] | [next] | [standalone]

#3715

Am 20.08.2017 um 10:36 schrieb robertvanpass@gmail.com:
> Hi,
> 
> I have difficulties to plot ephemerides data.
> They are generated by SkytechX in the following format:
> 
> Date;Time;Local R.A.;Local Dec.;
> 8/20/2017;12:00:00 UTC+1.0;10h 35m 50.60s;+03° 47' 15.38";
> 8/21/2017;12:00:00 UTC+1.0;10h 33m 13.63s;+04° 02' 23.70";
> 8/22/2017;12:00:00 UTC+1.0;10h 30m 23.99s;+04° 20' 34.64";
> ...
> 
> The column "Local R.A." is in time format: Hours Minits Seconds,
> should be converted to: Hours + Minits/60 + Seconds/3600.
> 
> Column "Local Dec." is in angle format: Degrees Arcminits Arcseconds,
> should be converted to Degrees + Arcminits/60 + Arcseconds/3600.
> 
> So the first record should become:
> 8/20/2017;12:00:00 UTC+1.0;10.587606;3.787606";

$dat << EOD
8/20/2017;12:00:00 UTC+1.0;10h 35m 50.60s;+03° 47' 15.38";
8/21/2017;12:00:00 UTC+1.0;10h 33m 13.63s;+04° 02' 23.70";
8/22/2017;12:00:00 UTC+1.0;10h 30m 23.99s;+04° 20' 34.64";
EOD

set dataf sep "; \"
set table

plot $dat us \
	(timecolumn(1,"%d/%m/%Y")):\
	(timecolumn(2,"%H:%M:%S")):\
	(timecolumn(4,"%Hh %Mm %Ss")/3600):\
	(($7+$8/60+$9/3600)) w table

gives

 1.53369e+009 43200 10.5974 3.78761
 1.53636e+009 43200 10.5538 4.03992
 1.53896e+009 43200 10.5067 4.34296

	
This ignores the timezone specifier, I'm not sure how to catch that.
The date is in unix time.

Best, Karl

[toc] | [prev] | [next] | [standalone]

#3716

Am 20.08.2017 um 22:56 schrieb Karl Ratzsch:
> Am 20.08.2017 um 10:36 schrieb robertvanpass@gmail.com:
>> Hi,
>>
>> I have difficulties to plot ephemerides data.
>> They are generated by SkytechX in the following format:
>>
>> Date;Time;Local R.A.;Local Dec.;
>> 8/20/2017;12:00:00 UTC+1.0;10h 35m 50.60s;+03° 47' 15.38";
>> 8/21/2017;12:00:00 UTC+1.0;10h 33m 13.63s;+04° 02' 23.70";
>> 8/22/2017;12:00:00 UTC+1.0;10h 30m 23.99s;+04° 20' 34.64";
>> ...
>>
>> The column "Local R.A." is in time format: Hours Minits Seconds,
>> should be converted to: Hours + Minits/60 + Seconds/3600.
>>
>> Column "Local Dec." is in angle format: Degrees Arcminits Arcseconds,
>> should be converted to Degrees + Arcminits/60 + Arcseconds/3600.
>>
>> So the first record should become:
>> 8/20/2017;12:00:00 UTC+1.0;10.587606;3.787606";
> 
> $dat << EOD
> 8/20/2017;12:00:00 UTC+1.0;10h 35m 50.60s;+03° 47' 15.38";
> 8/21/2017;12:00:00 UTC+1.0;10h 33m 13.63s;+04° 02' 23.70";
> 8/22/2017;12:00:00 UTC+1.0;10h 30m 23.99s;+04° 20' 34.64";
> EOD
> 
> set dataf sep "; \"
> set table
> 
> plot $dat us \
> 	(timecolumn(1,"%d/%m/%Y")):\
> 	(timecolumn(2,"%H:%M:%S")):\
> 	(timecolumn(4,"%Hh %Mm %Ss")/3600):\
> 	(($7+$8/60+$9/3600)) w table
> 
> gives
> 
>  1.53369e+009 43200 10.5974 3.78761
>  1.53636e+009 43200 10.5538 4.03992
>  1.53896e+009 43200 10.5067 4.34296
> 
> 	
> This ignores the timezone specifier, I'm not sure how to catch that.

Wasn't hard:

set dataf sep "; " # the backslash was nonsense
set table

plot $dat us \
(timecolumn(1,"%d/%m/%Y")):\
(timecolumn(2,"%H:%M:%S")+strcol(3)[4:8]*3600):\
(timecolumn(4,"%Hh %Mm %Ss")/3600):\
(($7+$8/60+$9/3600)) w table

Expects the format "UTC+x.y" or "UTC-x.y", anything else fails of
course.

[toc] | [prev] | [next] | [standalone]

#3717

On Monday, August 21, 2017 at 4:59:16 AM UTC+2, Karl Ratzsch wrote:
> Am 20.08.2017 um 22:56 schrieb Karl Ratzsch:
> > 
> >> Hi,
> >>
> >> I have difficulties to plot ephemerides data.
> >> They are generated by SkytechX in the following format:
> >>
> >> Date;Time;Local R.A.;Local Dec.;
> >> 8/20/2017;12:00:00 UTC+1.0;10h 35m 50.60s;+03° 47' 15.38";
> >> 8/21/2017;12:00:00 UTC+1.0;10h 33m 13.63s;+04° 02' 23.70";
> >> 8/22/2017;12:00:00 UTC+1.0;10h 30m 23.99s;+04° 20' 34.64";
> >> ...
> >>
> >> The column "Local R.A." is in time format: Hours Minits Seconds,
> >> should be converted to: Hours + Minits/60 + Seconds/3600.
> >>
> >> Column "Local Dec." is in angle format: Degrees Arcminits Arcseconds,
> >> should be converted to Degrees + Arcminits/60 + Arcseconds/3600.
> >>
> >> So the first record should become:
> >> 8/20/2017;12:00:00 UTC+1.0;10.587606;3.787606";
> > 
> > $dat << EOD
> > 8/20/2017;12:00:00 UTC+1.0;10h 35m 50.60s;+03° 47' 15.38";
> > 8/21/2017;12:00:00 UTC+1.0;10h 33m 13.63s;+04° 02' 23.70";
> > 8/22/2017;12:00:00 UTC+1.0;10h 30m 23.99s;+04° 20' 34.64";
> > EOD
> > 
> > set dataf sep "; \"
> > set table
> > 
> > plot $dat us \
> > 	(timecolumn(1,"%d/%m/%Y")):\
> > 	(timecolumn(2,"%H:%M:%S")):\
> > 	(timecolumn(4,"%Hh %Mm %Ss")/3600):\
> > 	(($7+$8/60+$9/3600)) w table
> > 
> > gives
> > 
> >  1.53369e+009 43200 10.5974 3.78761
> >  1.53636e+009 43200 10.5538 4.03992
> >  1.53896e+009 43200 10.5067 4.34296
> > 
> > 	
> > This ignores the timezone specifier, I'm not sure how to catch that.
> 
> Wasn't hard:
> 
> set dataf sep "; " # the backslash was nonsense
> set table
> 
> plot $dat us \
> (timecolumn(1,"%d/%m/%Y")):\
> (timecolumn(2,"%H:%M:%S")+strcol(3)[4:8]*3600):\
> (timecolumn(4,"%Hh %Mm %Ss")/3600):\
> (($7+$8/60+$9/3600)) w table
> 
> Expects the format "UTC+x.y" or "UTC-x.y", anything else fails of
> course.



Hi,

Thank you very much, I'm getting closer !

However I still have an issue:

When I add the code for the timezone specifier ( +strcol(3)[4:8]*3600) )
I get the errormessage "Non-numeric string found where a numeric expression was expected"

How to convert to numeric ?

Thanks

[toc] | [prev] | [next] | [standalone]

#3719

Am 22.08.2017 um 00:01 schrieb robertvanpass@gmail.com:
> On Monday, August 21, 2017 at 4:59:16 AM UTC+2, Karl Ratzsch wrote:

>>> $dat << EOD
>>> 8/20/2017;12:00:00 UTC+1.0;10h 35m 50.60s;+03° 47' 15.38";
>>> 8/21/2017;12:00:00 UTC+1.0;10h 33m 13.63s;+04° 02' 23.70";
>>> 8/22/2017;12:00:00 UTC+1.0;10h 30m 23.99s;+04° 20' 34.64";
>>> EOD
>>>
>> set dataf sep "; " # the backslash was nonsense
>> set table
>>
>> plot $dat us \
>> (timecolumn(1,"%d/%m/%Y")):\
>> (timecolumn(2,"%H:%M:%S")+strcol(3)[4:8]*3600):\
>> (timecolumn(4,"%Hh %Mm %Ss")/3600):\
>> (($7+$8/60+$9/3600)) w table
>>


> However I still have an issue:
> 
> When I add the code for the timezone specifier ( +strcol(3)[4:8]*3600) )
> I get the errormessage "Non-numeric string found where a numeric expression was expected"
> 

gnuplot automatically converts strings to numberic variables in
expressions (i.e. if you do math like *1.0) if possible.

Can you make an example like the one above that I can directly copy
into gnuplot to reproduce your problem?

[toc] | [prev] | [next] | [standalone]

#3722

On Tuesday, August 22, 2017 at 3:59:32 AM UTC+2, Karl Ratzsch wrote:

> > On Monday, August 21, 2017 at 4:59:16 AM UTC+2, Karl Ratzsch wrote:
> 
> >>> $dat << EOD
> >>> 8/20/2017;12:00:00 UTC+1.0;10h 35m 50.60s;+03° 47' 15.38";
> >>> 8/21/2017;12:00:00 UTC+1.0;10h 33m 13.63s;+04° 02' 23.70";
> >>> 8/22/2017;12:00:00 UTC+1.0;10h 30m 23.99s;+04° 20' 34.64";
> >>> EOD
> >>>
> >> set dataf sep "; " # the backslash was nonsense
> >> set table
> >>
> >> plot $dat us \
> >> (timecolumn(1,"%d/%m/%Y")):\
> >> (timecolumn(2,"%H:%M:%S")+strcol(3)[4:8]*3600):\
> >> (timecolumn(4,"%Hh %Mm %Ss")/3600):\
> >> (($7+$8/60+$9/3600)) w table
> >>
> 
> 
> > However I still have an issue:
> > 
> > When I add the code for the timezone specifier ( +strcol(3)[4:8]*3600) )
> > I get the errormessage "Non-numeric string found where a numeric expression was expected"
> > 
> 
> gnuplot automatically converts strings to numberic variables in
> expressions (i.e. if you do math like *1.0) if possible.
> 
> Can you make an example like the one above that I can directly copy
> into gnuplot to reproduce your problem?

No problem:

1. Putting these lines in a CSV file will give the errormessage:

Date;Time;Local R.A.;Local Dec.;
8/20/2017;12:00:00 UTC+1.0;10h 35m 50.60s;+03° 47' 15.38";
8/21/2017;12:00:00 UTC+1.0;10h 33m 13.63s;+04° 02' 23.70";
8/22/2017;12:00:00 UTC+1.0;10h 30m 23.99s;+04° 20' 34.64";


2. I saw another strange behaviour:
when values go negative they show strange jumps.
sample data below:

Date;Time;Local R.A.;Local Dec.;
9/15/2016;12:00:00 UTC+1.0;12h 06m 27.95s;+00° 29' 21.10";
9/16/2016;12:00:00 UTC+1.0;12h 07m 15.28s;+00° 24' 12.97";
9/17/2016;12:00:00 UTC+1.0;12h 08m 02.66s;+00° 19' 04.67";
9/18/2016;12:00:00 UTC+1.0;12h 08m 50.08s;+00° 13' 56.25";
9/19/2016;12:00:00 UTC+1.0;12h 09m 37.55s;+00° 08' 47.73";
9/20/2016;12:00:00 UTC+1.0;12h 10m 25.06s;+00° 03' 39.15";
9/21/2016;12:00:00 UTC+1.0;12h 11m 12.60s;-00° 01' 29.46";
9/22/2016;12:00:00 UTC+1.0;12h 12m 00.17s;-00° 06' 38.07";
9/23/2016;12:00:00 UTC+1.0;12h 12m 47.76s;-00° 11' 46.64";
9/24/2016;12:00:00 UTC+1.0;12h 13m 35.38s;-00° 16' 55.14";
9/25/2016;12:00:00 UTC+1.0;12h 14m 23.01s;-00° 22' 03.55";
9/26/2016;12:00:00 UTC+1.0;12h 15m 10.65s;-00° 27' 11.81";
9/27/2016;12:00:00 UTC+1.0;12h 15m 58.31s;-00° 32' 19.91";
9/28/2016;12:00:00 UTC+1.0;12h 16m 45.96s;-00° 37' 27.81";
9/29/2016;12:00:00 UTC+1.0;12h 17m 33.62s;-00° 42' 35.48";
9/30/2016;12:00:00 UTC+1.0;12h 18m 21.27s;-00° 47' 42.88";
10/1/2016;12:00:00 UTC+1.0;12h 19m 08.92s;-00° 52' 49.97";
10/2/2016;12:00:00 UTC+1.0;12h 19m 56.55s;-00° 57' 56.74";
10/3/2016;12:00:00 UTC+1.0;12h 20m 44.16s;-01° 03' 03.13";
10/4/2016;12:00:00 UTC+1.0;12h 21m 31.75s;-01° 08' 09.13";
10/5/2016;12:00:00 UTC+1.0;12h 22m 19.31s;-01° 13' 14.69";
10/6/2016;12:00:00 UTC+1.0;12h 23m 06.84s;-01° 18' 19.78";
10/7/2016;12:00:00 UTC+1.0;12h 23m 54.34s;-01° 23' 24.38";
10/8/2016;12:00:00 UTC+1.0;12h 24m 41.80s;-01° 28' 28.43";
10/9/2016;12:00:00 UTC+1.0;12h 25m 29.21s;-01° 33' 31.92";
10/10/2016;12:00:00 UTC+1.0;12h 26m 16.58s;-01° 38' 34.80";
10/11/2016;12:00:00 UTC+1.0;12h 27m 03.89s;-01° 43' 37.05";
10/12/2016;12:00:00 UTC+1.0;12h 27m 51.14s;-01° 48' 38.62";
10/13/2016;12:00:00 UTC+1.0;12h 28m 38.33s;-01° 53' 39.49";
10/14/2016;12:00:00 UTC+1.0;12h 29m 25.45s;-01° 58' 39.62";
10/15/2016;12:00:00 UTC+1.0;12h 30m 12.50s;-02° 03' 38.98";
10/16/2016;12:00:00 UTC+1.0;12h 30m 59.48s;-02° 08' 37.53";
10/17/2016;12:00:00 UTC+1.0;12h 31m 46.37s;-02° 13' 35.23";
10/18/2016;12:00:00 UTC+1.0;12h 32m 33.18s;-02° 18' 32.06";
10/19/2016;12:00:00 UTC+1.0;12h 33m 19.90s;-02° 23' 27.98";
10/20/2016;12:00:00 UTC+1.0;12h 34m 06.52s;-02° 28' 22.95";
10/21/2016;12:00:00 UTC+1.0;12h 34m 53.04s;-02° 33' 16.94";
10/22/2016;12:00:00 UTC+1.0;12h 35m 39.45s;-02° 38' 09.91";
10/23/2016;12:00:00 UTC+1.0;12h 36m 25.76s;-02° 43' 01.84";

Thanks for your help.

[toc] | [prev] | [next] | [standalone]

#3725

Am 22.08.2017 um 16:57 schrieb robertvanpass@gmail.com:
>>
>>> However I still have an issue:
>>>
>>> When I add the code for the timezone specifier ( +strcol(3)[4:8]*3600) )
>>> I get the errormessage "Non-numeric string found where a numeric expression was expected"
>>>
>>
>> gnuplot automatically converts strings to numberic variables in
>> expressions (i.e. if you do math like *1.0) if possible.
>>
>> Can you make an example like the one above that I can directly copy
>> into gnuplot to reproduce your problem?
> 
> No problem:
> 
> 1. Putting these lines in a CSV file will give the errormessage:
> 

I meant *exactly* ;-) like this:

---- gnuplot input ----
$dat << EOD
8/20/2017;12:00:00 UTC+1.0;10h 35m 50.60s;+03° 47' 15.38";
8/21/2017;12:00:00 UTC+1.0;10h 33m 13.63s;+04° 02' 23.70";
8/22/2017;12:00:00 UTC+1.0;10h 30m 23.99s;-04° 20' 34.64";
EOD

set dataf sep "; "
set table

plot $dat us \
(timecolumn(1,"%d/%m/%Y")):\
(timecolumn(2,"%H:%M:%S")+strcol(3)[4:8]*3600):\
(timecolumn(4,"%Hh %Mm %Ss")/3600):\
(($7 + ($7<0?-$8:$8)/60 + ($7<0?-$9:$9)/3600)) w table
---- end ----

, only with the plot command that fails for you.

Then I can just c&p the whole thing into the gnuplot console, without
creating files and guessing at your exact plot setup.

The negative declinations come out wrong because the minutes and seconds
get the wrong sign then. I fixed the above plot command to take that
into account.

  K

[toc] | [prev] | [next] | [standalone]

#3727

On Tuesday, August 22, 2017 at 7:09:32 PM UTC+2, Karl Ratzsch wrote:
> Am 22.08.2017 um 16:57 schrieb robertvanpass@gmail.com:
> >>
> >>> However I still have an issue:
> >>>
> >>> When I add the code for the timezone specifier ( +strcol(3)[4:8]*3600) )
> >>> I get the errormessage "Non-numeric string found where a numeric expression was expected"
> >>>
> >>
> >> gnuplot automatically converts strings to numberic variables in
> >> expressions (i.e. if you do math like *1.0) if possible.
> >>
> >> Can you make an example like the one above that I can directly copy
> >> into gnuplot to reproduce your problem?
> > 
> > No problem:
> > 
> > 1. Putting these lines in a CSV file will give the errormessage:
> > 
> 
> I meant *exactly* ;-) like this:
> 
> ---- gnuplot input ----
> $dat << EOD
> 8/20/2017;12:00:00 UTC+1.0;10h 35m 50.60s;+03° 47' 15.38";
> 8/21/2017;12:00:00 UTC+1.0;10h 33m 13.63s;+04° 02' 23.70";
> 8/22/2017;12:00:00 UTC+1.0;10h 30m 23.99s;-04° 20' 34.64";
> EOD
> 
> set dataf sep "; "
> set table
> 
> plot $dat us \
> (timecolumn(1,"%d/%m/%Y")):\
> (timecolumn(2,"%H:%M:%S")+strcol(3)[4:8]*3600):\
> (timecolumn(4,"%Hh %Mm %Ss")/3600):\
> (($7 + ($7<0?-$8:$8)/60 + ($7<0?-$9:$9)/3600)) w table
> ---- end ----
> 
> , only with the plot command that fails for you.
> 
> Then I can just c&p the whole thing into the gnuplot console, without
> creating files and guessing at your exact plot setup.
> 
> The negative declinations come out wrong because the minutes and seconds
> get the wrong sign then. I fixed the above plot command to take that
> into account.
> 
>   K



Got it !

I use a slightly different plot command to correct for declination values between -00° and -01°, but basically it's like you say.

Now the last hurdle for me is that the X-axis of the graph is not in a readable Date-Time format:

$dat << EOD
8/20/2017;12:00:00 UTC+1.0;10h 35m 50.60s;+03° 47' 15.38";
8/21/2017;12:00:00 UTC+1.0;10h 33m 13.63s;+04° 02' 23.70";
8/22/2017;12:00:00 UTC+1.0;10h 30m 23.99s;-04° 20' 34.64";
EOD

set dataf sep "; "
set table 'ephem'
set term qt 1
plot $dat us \
(timecolumn(1,"%d/%m/%Y")):\
(timecolumn(2,"%H:%M:%S")+strcol(3)[4:8]*3600):\
(timecolumn(4,"%Hh %Mm %Ss")/3600):\
(($7 + ($7<0?-$8:$8)/60 + ($7<0?-$9:$9)/3600)) w table 
unset table

plot 'ephem' u 2:4 w lp, \
                  '' u 2:5 w lp

Do you have an idea how to get date and time on the X-axis of the graph ?

Thank you very much !

[toc] | [prev] | [next] | [standalone]

#3728

Am 22.08.2017 um 23:09 schrieb robertvanpass@gmail.com:
> 
> I use a slightly different plot command to correct for declination values between -00° and -01°, but basically it's like you say.
> 
> Now the last hurdle for me is that the X-axis of the graph is not in a readable Date-Time format:
> 

> Do you have an idea how to get date and time on the X-axis of the graph ?

simple: "set xtics timedate" . Check the help ("help
time_specifiers") if you don't like the default format.

You can also plot the table to a named datablock instead of a file:

  set table $ephem

and do the second plot from there. Doesn't clutter up your disk with
redundant files.

Or just do it directly.

   Karl

[toc] | [prev] | [next] | [standalone]

#3730

On Wednesday, August 23, 2017 at 1:45:27 AM UTC+2, Karl Ratzsch wrote:
> Am 22.08.2017 um 23:09 schrieb robertvanpass@gmail.com:
> > 
> > I use a slightly different plot command to correct for declination values between -00° and -01°, but basically it's like you say.
> > 
> > Now the last hurdle for me is that the X-axis of the graph is not in a readable Date-Time format:
> > 
> 
> > Do you have an idea how to get date and time on the X-axis of the graph ?
> 
> simple: "set xtics timedate" . Check the help ("help
> time_specifiers") if you don't like the default format.
> 
> You can also plot the table to a named datablock instead of a file:
> 
>   set table $ephem
> 
> and do the second plot from there. Doesn't clutter up your disk with
> redundant files.
> 
> Or just do it directly.
> 
>    Karl


Great, it works !
I couldn't have done this by myself - thanks a million !

[toc] | [prev] | [standalone]

Back to top | Article view | comp.graphics.apps.gnuplot

csiph-web