How do we get Igor more widely adopted

Hi,

Man I love Igor Pro. I have been struggling with some scripting in JMP to support a client with some analysis. It is pure pain. I really appreciate when I can back to Igor.

We need to get it more widely adopted.
It would be good to increase the power of the scripting language.

I'm working with 2D arrays, where each column is a separate dataset. I've often found myself wanting to do functionOnASingleSet(dataArray[][1]), and I can't even though I've given the function what it wants. I need to write additional code to get out a slice of that array, put it into a 1d array, and do that, then if I alter it, I need to then copy it back into the array.
masheroz wrote:
It would be good to increase the power of the scripting language.

I'm working with 2D arrays, where each column is a separate dataset. I've often found myself wanting to do functionOnASingleSet(dataArray[][1]), and I can't even though I've given the function what it wants. I need to write additional code to get out a slice of that array, put it into a 1d array, and do that, then if I alter it, I need to then copy it back into the array.


It is true that many commands cannot handle mutlidimensional arrays, however, i think writing a generic function to extract a column or row (using matrixOp and potentially a free wave) is a fairly easy workaround.

Having said that, I have asked wavemetrics to have wavestats handle ranges from multidimensional waves :)


But the command isn't handling a multidimensional array, I'm giving it a 1D array that just happens to be part of a larger array.
Okapi wrote:
Having said that, I have asked wavemetrics to have wavestats handle ranges from multidimensional waves :)


If 2-D range is sufficient, often ImageStats is your friend:
ImageStats [/M=val /P=planeNumber ]/G={startP, endP, startQ, endQ } imageWave


P is rows, Q is columns.

A column's stats would be

ImageStats /G={0, DimSize(wave,0)-1, column, column} wave

--Jim Prouty
Software Engineer, WaveMetrics, Inc.
Quote:
If 2-D range is sufficient, often ImageStats is your friend:


Thanks Jim. I just noticed that your tip was also in the wavestats help file. I don't know how I managed to overlook it for so long.