Path: csiph.com!v102.xanadu-bbs.net!xanadu-bbs.net!feeder.erje.net!us.feeder.erje.net!newsfeed.fsmpi.rwth-aachen.de!eternal-september.org!feeder.eternal-september.org!news.eternal-september.org!.POSTED!not-for-mail
From: Ethan A Merritt <EAMerritt@gmail.com>
Newsgroups: comp.graphics.apps.gnuplot
Subject: Re: difficult for loop synax :-|
Date: Wed, 20 Aug 2014 17:50:52 -0700
Organization: made entirely of Lego
Lines: 66
Message-ID: <lt3fpe$550$1@dont-email.me>
References: <9bc28e9e-f776-4407-8acc-f0bc7a964bcf@googlegroups.com> <lt0s51$jbh$1@dont-email.me> <7d92eab8-bceb-49bb-80aa-ff7834bc9a73@googlegroups.com> <lt2l49$hgn$1@dont-email.me> <9066cc28-d4c2-4bb6-a081-4ce966222a7e@googlegroups.com>
Mime-Version: 1.0
Content-Type: text/plain; charset="ISO-8859-1"
Content-Transfer-Encoding: 7Bit
Injection-Date: Thu, 21 Aug 2014 00:50:54 +0000 (UTC)
Injection-Info: mx05.eternal-september.org; posting-host="7bbfccb08e3cb0f92ebf95516f5f5a81"; logging-data="5280"; mail-complaints-to="abuse@eternal-september.org";	posting-account="U2FsdGVkX1/CI3G/LcM7UXDuh5oC2Apg"
User-Agent: KNode/4.11.4
Cancel-Lock: sha1:KOmYHvm73TRe4+5ZvlReH20Q3EU=
Xref: csiph.com comp.graphics.apps.gnuplot:2558

Kalin Kozhuharov wrote:

> Hello Ethan,
> 
> Thank you for your time! I was trying to make self-contained example,
> but because it was simple enough you "optimized" it ;-) "Premature
> optimization is the root of all evil", said D. Knuth :-D
> 
> On Thursday, August 21, 2014 2:15:51 AM UTC+9, Ethan A Merritt wrote:
>>> Well, may be I oversimplified it, but the real thing is too complex
>>> anyway. What about this:
>>> 
>>> reset; f1(x) = (C=1.134324*x,0); f2(x) = sin(C*x); set xrange
>>> [0:3.14]; plot a=f1(1), f2(x) title "n=1", a=f1(2), f2(x) title
>>> "n=2" , a=f1(3), f2(x) title "n=3";
>> 
>> It is still not clear what you are aiming for.
>> The sequence of commands above does not make sense.
>> 
>> The function f1(x) is never plotted, so it plays no role except for
>> its side effect of setting C to a constant.
>>
> Well, yes I'd have to admit I am using side-effects here.
> And my code is turning to a small mess because I am trying to have
> gnuplot do statistics in-line, instead of preprocessing the data.
> 
> My current code is here:
> https://github.com/thinrope/fixed_sensor_visualization/blob/master/timeplot_all.gpl#L26
> 
> plot for [ID in IDs] "tmp/" . ID . ".data" u 1:2 axes x1y2 w l lw 2
> title graph_title(ID);
> 
> What I wanted to do is add another SMA filter on each datafile (or 2
> with different window) and still have it plot similar graph.
> 
> so having defined SMA as in
> https://github.com/thinrope/fixed_sensor_visualization/blob/master/timeplot.gpl#L46
> I was planning to write something along:
> 
> plot for [ID in IDs] a=sma_24_init(0), "tmp/" . ID . ".data" u
> 1:(sma_24($2)) axes x1y2 w l lw 2 title graph_title(ID);

I think that statement would work if you just remove the unnecessary comma
after the definition a=<something>.
Doesn't it?

	Ethan




> IDs are list of numbers coming form the environment/Makefile, but can
> be strings.
> 
> May be possible with `set table` and `do for [ID in IDs]{init; plot;}`
> but I'd rather avoid it.
> 
> 
> Alternatively, something along
> 
> init_OK=0; f3(x,n) = (init_OK ? a=f1(n): 0, f2(x));
> plot for [n=1:3] a=f3(x,n);
>          undefined variable: x
> :-|
> 
> Kalin.