Loadwave: p in column detection

Hi,

One option is to load the whole file as text string and then trimstring() to remove the outside whitespaces and then parse the resulting string into the various waves.

Andy

Your text file may be in the fixed field format. In that format each column is a fixed number of characters. You need to use LoadWave with the /F flag if that is the case.

/F={numColumns, defaultFieldWidth, flags }
    Indicates that the file uses the fixed field file format. Most FORTRAN programs generate files in this format.

Thanks. In fact I did not find a way to load the file as a single column.

The unit change mid column from eV to keV is also not nice. I think hegedus's suggestion may be the best option, but I would probably load the file as one big singe-column text wave and then use SScanF on each line to extract the numerical values and units

Could you suggest me a command line to load file as a single-column text wave?

OK I got it, using /L

Thanks for your help.

If you post the file or a representative section of it, I may be able to recommend how to handle it.

Hi hrodstein,

Here is the file. I have fixed the issue with the following code, but it might be improved.

Thanks for your help.

Erik

Function LoadSRIM_files(element, name) 
	string element, name
	variable i, N, v1,v2,v3,v4, v5,v6
	string str, s1, s2, s3, s4, nom_e, nom_n, pr,lon_strag, lat_strag, unit_s, unit_lat_strag, unit_lon_strag, energy, unit_energy, unit_pr
	
	string path, nom_fic_loaded
	path=" DD:Users:quiricoe:DOSSIERS-ERIC:RECHERCHE:PROJETS_EN_COURS:TNOs:NEWHORIZONS:ARROKOTH:TNOs_SURFACE_IRRADIATION:SRIM_Files:Se_Sn_CH3OH_300eV_30_MeV_MeV_per_mm_d0.8:"+name
	print path

	loadwave /N=filetodealwith /K=2 /V={"","",0,0}  /J /L={0,25,131,0,1}  path
	nom_fic_loaded="filetodealwith0"
	wave /T wave20=$nom_fic_loaded
	
	N=numpnts(wave20)
	print N


	nom_n="Sn_"+element+"_SW"
	nom_e="Se_"+element+"_SW"
	
	pr="SW_"+element+"_PR"
	lon_strag="SW_"+element+"_long_strag"
	lat_strag="SW_"+element+"_lat_strag"
	unit_s="SW_"+element+"_unit_S"
	unit_lat_strag="SW_"+element+"_unit_lat_strag"
	unit_lon_strag="SW_"+element+"_unit_long_strag"
	energy="SW_"+element+"_energy"
	unit_energy="SW_"+element+"_unit_energy"
	unit_pr="SW_"+element+"_unit_pr"
	
	Make /T /O /N=(N) wave21
	wave21=wave20
	 
	Make /O /N=(N) $energy, $nom_e, $nom_n, $pr, $lon_strag, $lat_strag
	Make /O /T /N=(N) $unit_lat_strag, $unit_lon_strag, $unit_energy, $unit_pr
	
	wave wr0=$energy
	wave wr1=$nom_e
	wave wr2=$nom_n
	wave wr3=$pr
	wave wr4=$lon_strag
	wave wr5=$lat_strag
	wave /T wr6=$unit_lat_strag
	wave/T wr7=$unit_lon_strag
	wave /T wr8=$unit_energy
	wave /T wr9=$unit_pr

	i=0
	do
	str=replacestring(",", wave20[i], ".")
	wave21[i]=str
	sscanf str, "%e%s%e%e%e%s%e%s%e%s", v1,s1, v2, v3, v4, s2, v5,s3, v6, s4
	wr0[i]=v1
	wr1[i]=v2
	wr2[i]=v3
	wr3[i]=v4
	wr4[i]=v5
	wr5[i]=v6
	wr6[i]=s4
	wr7[i]=s3
	wr8[i]=s1	
	wr9[i]=s2	
	
	if (cmpstr(wr8[i],"eV")==0)
	wr0[i]=wr0[i]/1E6
	endif
	
	if (cmpstr(wr8[i],"keV")==0)
	wr0[i]=wr0[i]/1E3
	endif
	
	if (cmpstr(wr9[i],"A")==0)
	wr3[i]=wr3[i]/1E4
	endif
	
	if (cmpstr(wr9[i],"mm")==0)
	wr3[i]=wr3[i]*1E3
	endif
	
	if (cmpstr(wr6[i],"A")==0)
	wr5[i]=wr5[i]/1E4
	endif
	
	if (cmpstr(wr7[i],"A")==0)
	wr4[i]=wr4[i]/1E4
	endif
	
	//Killwaves wr8,  wr9, wr6, wr7
	
	i=i+1
	while( i<N)

end

I am attaching a zip file containing the data file that Erik provided and an Igor experiment containing procedures that load the file. It also contains a notebook explaining the procedures.

In brief, I used "Load Fixed Field Text" (LoadWave/F) to treat the file as fixed field. I treated the units as separate columns. I loaded the numeric and units columns, and then scaled the numeric waves by the units to produce waves in keV and um units.

The procedure require Igor Pro 7 at least because I used ints.

A BIG thank you! It works very well.

Erik