[Kst] Prune data to visible range

Nicolas Brisset nicolas.brisset at free.fr
Wed Sep 17 20:16:06 UTC 2014


Hi Barth,

Yes, your description fits the idea quite well. 

I imagined selecting a vector (or more) from the change data sample ranges tool, but a RMB entry in a plot would also be fine. I think we can/should even offer both...

One thing I'm unsure about is whether to consider the x and y or only x range of the "reference" vector, it's not 100 % clear in your description. I'd consider the x vector only for a start but maybe I'm missing some use cases.

Another thing I'm wondering about is how to handle different sample rates and/or datasources. If they support time, then we can use time information. But otherwise I can only image setting the start sample and number of samples equal, which in many cases (ASCII files for one) will work. Maybe using the samples per frame information we can even handle different sampling rates gracefully. But there will always be corner cases we can't cover - which I think is OK.

Thanks for your prompt answer, and if you are working on something else, you can give me some hints and I can try to implement it.

Nicolas

----- Mail original -----
> De: "Barth Netterfield" <barth.netterfield at utoronto.ca>
> À: kst at kde.org
> Cc: "Nicolas Brisset" <nicolas.brisset at airbus.com>
> Envoyé: Mercredi 17 Septembre 2014 16:52:57
> Objet: Re: [Kst] Prune data to visible range
> 
> Let me make sure I understand this:
> 
> (2) is a plot tool which, for example could be:
> plotRBM->prune to visible range (or some better name)
> 
> which would set the vector range of all tied plots to the range
> displayed in
> the plot.
> 
> The logic could be:
> go through each vector in each displayed curve in a tied plot
>   -find the index of the first XY pair which is in the plot: f0
>   -find the index of the last XY pair which is in the plot: f1
> Set the new vector range accordingly.
> 
> Notes:
>   -we need to keep different indices for each data source.
>   -If the curves are not monotonic, we might still read samples that
>   are not
> displayed, but we will for sure read all samples that are displayed.
>   -using the tied zoom system lets us select which vectors we trim at
>   the cost
> of some decreased convenience.
> 
> If this is what you mean, then I think it is a very good idea.  I
> would be
> happy to put it on my short list....


More information about the Kst mailing list