Path: csiph.com!eternal-september.org!feeder.eternal-september.org!reader01.eternal-september.org!.POSTED!not-for-mail
From: =?UTF-8?Q?J=c3=b6rg_Buchholz?= <bookwood4new@freenet.de>
Newsgroups: comp.graphics.apps.gnuplot
Subject: Re: plot datafile with a blank between +- and some data
Date: Thu, 4 Apr 2019 10:10:38 +0200
Organization: A noiseless patient Spider
Lines: 65
Message-ID: <q84e5s$i2f$1@dont-email.me>
References: <q81kck$ifh$1@dont-email.me> <q82uah$dg1$1@dont-email.me> <q833op$5u4$1@solani.org>
Mime-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 8bit
Injection-Date: Thu, 4 Apr 2019 08:10:36 -0000 (UTC)
Injection-Info: reader02.eternal-september.org; posting-host="2346efe3b5da99042ff6e1cff6b3f9ab"; logging-data="18511"; mail-complaints-to="abuse@eternal-september.org";	posting-account="U2FsdGVkX1/DfFhjt4IgqzBH8eAE3w5V2EWgC13ASok="
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:60.0) Gecko/20100101 Thunderbird/60.6.1
Cancel-Lock: sha1:Z9zt+ZTDqnefeNw0E/tnuZAnYYg=
In-Reply-To: <q833op$5u4$1@solani.org>
Content-Language: en-GB
Xref: csiph.com comp.graphics.apps.gnuplot:4175

On 03.04.2019 22:06, Karl Ratzsch wrote:
> Am 03.04.2019 um 20:33 schrieb Jörg Buchholz:
>> On 03.04.19 08:38, Jörg Buchholz wrote:
>>> Hello,
>>>
>>> how can I plot the following data structure out of a file?
>>>
>>> 33,2019/03/21 10:00:06,  0,+ 6.786,+ 2.507
>>> 34,2019/03/21 10:00:07,  0,+10.089,+ 5.564
>>> 35,2019/03/21 10:00:08,  0,- 0.584,+ 3.324
>>> 36,2019/03/21 10:00:09,  0,+10.156,+ 0.024
>>> 37,2019/03/21 10:00:10,  0,+ 4.278,+ 5.439
>>> 38,2019/03/21 10:00:11,  0,+ 0.431,+ 0.262
>>>
>>>
>>> set datafile separator comma
>>> plot 'datafile' u 1:4 w p pt 7
>>>
>>> gives only the values higher than 10 to me, cause there are no space
>>> between the sign and the value.
>>>
>>> How can I set the format correct to get all points?
>>
>> At the moment my solution is the following:
>>
>> system "sed -i 's/+ / +/g;s/- / -/g' datafile"
>>
>> Is there a solution to do it inside gnuplot?
> 
> Possibly, but it will be very cumbersome. Putting a space between
> the sign and the decimal number is really a mean trick.

It is a .csv file that comes from a "Graphtec GL240 midi Data Logger".

> "plot" can use a scanf() format specifier, but it cannot handle the
> date/time column, and only seven values in total, plus it would
> still stumble over the space in column 4.
> 
> Be that as it may:
> 
> $dat << EOD
> 33,2019/03/21 10:00:06,  0,+ 6.786,+ 2.507
> 34,2019/03/21 10:00:07,  0,+10.089,+ 5.564
> 35,2019/03/21 10:00:08,  0,- 0.584,+ 3.324
> 36,2019/03/21 10:00:09,  0,+10.156,+ 0.024
> 37,2019/03/21 10:00:10,  0,+ 4.278,+ 5.439
> 38,2019/03/21 10:00:11,  0,+ 0.431,+ 0.262
> EOD
> 
> set dataf sep "X"
> 
> col1(st) = st[1:2] * 1.0
> col4(st) = (st[28:28] eq "-" ? -1 : 1) * st[29:34]
> stcol(n) = stringcolumn(n)
> 
> set table
> plot $dat us (col1(stcol(1))):(col4(stcol(1))) with table
> unset table


Thanks for your solution. There are some more columns. So I think, for
me, it is easer do handle the datafile with sed. From inside gnuplot or
direct after the files where produced.

Jörg