Path: csiph.com!weretis.net!feeder4.news.weretis.net!feeder5.news.weretis.net!news.solani.org!.POSTED!not-for-mail From: Karl Ratzsch Newsgroups: comp.graphics.apps.gnuplot Subject: Re: Ephemerides data Date: Tue, 22 Aug 2017 19:09:30 +0200 Organization: solani.org Lines: 48 Message-ID: References: <09265b97-7ca8-4f2d-baa2-7e9ff9abf37e@googlegroups.com> Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 8bit X-Trace: solani.org 1503421770 14916 eJwNyckRwDAIBLCWDIZdUw7H0H8Jib7yC0HT4DBfXxVkKtZiqiXZ4pW0qBdMzp+4OsJ8etb7fCD2ESU= (22 Aug 2017 17:09:30 GMT) X-Complaints-To: abuse@news.solani.org NNTP-Posting-Date: Tue, 22 Aug 2017 17:09:30 +0000 (UTC) User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:52.0) Gecko/20100101 Thunderbird/52.3.0 Cancel-Lock: sha1:esVk2t4lnesHNO5BEmBzHE9E5x4= X-NNTP-Posting-Host: eJwNyskBwCAIBMCWwrVgOa5C/yWYeU8YBCcdAY+Jsa4C2zU7UzbXLj3ufwBvbF2cFqFemn1sPhnoEV0= Content-Language: en-GB X-User-ID: eJwNx8kBwDAIA7CZSDDHOBTw/iOk+gnXxNrVYAqCN//MHEWxMndk6RVfm6xLx1WeaG5ZehbkAS6iEas= In-Reply-To: <09265b97-7ca8-4f2d-baa2-7e9ff9abf37e@googlegroups.com> Xref: csiph.com comp.graphics.apps.gnuplot: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