Re: [Gnuplot-info] How to realize a single plot in a double loop?

[theozh <th...@gm...>]

Hi Iva,

thank you very much for your multi-OS, all-inclusive code...
Well, the codes get's pretty lengthy and complicated in order to check
all sorts of things. This probably can't be avoided.
I'm sure with some testing and maybe some modifications your code will
work and will give the desired result :-).

>> It looks like a threefold nested "for loop" is also possible. But how to
>> move on to the next multiplot and place "label 1 Data_<Index1>"
>> in every plot?
>
> I'm not sure what would like to code. Could you give an example?
>

Well, as I said: I have data files with a clear file name structure, namely
"Data_<Index1>_<Index2>_<Index3>.dat"
The goal is to decide depending on the filename where to plot the data
in a multiplot and how to color it. Sounds like a rather simple task...

I currently see basically two options:

A) put all exisiting files into a list and depending on the filename
distribute it into the corresponding plot. However, deciding based on
the filename in which graph to plot the data, would probably require
some kind of "Regular Expressions" within gnuplot which I do not see how
to realize.

B) loop all possible filenames in a threefold plot loop within a certain
maximum range. Depending on how large the <Index> ranges are you might
get lengthy list of error messages (but, ok, who cares?). However, with
a threefold plot loop I don't (yet) see a way to distribute the data
into the corresponding multiplots.

Since I do not see a solution for A) and for B) is probably has to be a
mix of both. Or as your code is... three loops with a few if {} cases
and helper variables...

I just thought, I overlooked a very simple, short and efficient method.
Seems not to be the case. Thank you for your help!