[Kst] Fwd: [Google Code-in 2011] New comment on Kst: date/time support for the ASCII data source by kharvd

[nicolas.brisset at free.fr]

Hi Ben,

thanks for the feedback. I agree that reading NI files would be nice, but I would not like to complicate the ASCII source too much, and it seems to depart a bit too much from the usual stuff. To do it properly, we'd need to extract scalars (or even date/time info) from header lines...

I'd rather do a specific datasource for NI. Can you provide more info on the possible extensions, headers, etc? That would possibly also be a nice Code-In task, by the way...
Do others agree that we should rather do a specific source for that format?

Nicolas 

----- Mail original -----
> De: "Ben Lewis" <egretengineering at gmail.com>
> À: kst at kde.org
> Envoyé: Dimanche 4 Décembre 2011 06:48:55
> Objet: Re: [Kst] Fwd: [Google Code-in 2011] New comment on Kst: date/time support for the ASCII data source by kharvd
> 
> Hi Nicolas,
> 
> My requirements for ASCII date/time support are slightly different
> from what you have described here.
> 
> What I need may be outside the scope for this project but I thought I
> would mention it anyway, just
> in case this functionality can be included.
> 
> Below are the first 15 lines of a typical ACSII file I use.
> 
> --------------------------------------------------------------------------------------------------------------------------
> channel names:
> 02/12/2011 04:36:24 PM - Force (Bridge) - ch1    02/12/2011 04:36:24
> PM - Force (Bridge) - ch2
> start times:
> 2/12/2011 16:36:24.609407    2/12/2011 16:36:24.609407
> dt:
> 0.000620
> data:
> -3.400914E+1    -1.141607E+1
> 2.413543E+0    -7.194291E+0
> -2.100850E+1    -5.889673E-2
> -9.260725E-1    2.200645E+0
> -8.395482E+0    -2.604362E+1
> -1.209291E+1    -8.294330E+0
> -1.063183E+1    -1.064306E+1
> 8.108781E+0    -8.318977E-1
> --------------------------------------------------------------------------------------------------------------------------
> 
> The header gives the start date and time of the test with a delta t
> (dt) value.  The first row of
> data corresponds to the start date and time.  Each subsequent row is
> offset, in time, by the value
> of delta t
> 
> Would it be possible to generate the time vector needed for kst from
> this data?
> 
> This file format is used by National Instruments.  If kst could
> support this, I think it would be
> useful for a lot of people.
> 
> Cheers, Ben