Text-to-iraf conversion
Lic. Mónica G. Grosso wrote on Mar 16, 2000
Hi!
I was tryng to convert an ASCII file in IRAF image and I couldn�t.
When I use the rtextimage task in dataio package I obtain an file.imh but I
can not display with splot the image.
I obtain a follow message:
error: MWCS: dimension mismatch (mw-open)
I do not understand what is the problem. My ASCII file have 2 columns
(wavelength and flux).
Please, may you help me or said something do you think that is happening??
Regards,
Monica
Lic. M�nica G. Grosso
CASLEO
Casilla de Correo 467
5400 San Juan- Argentina
phone: 0054 264 4213653
fax: 0054 264 4213693
Phil Hodge wrote on Mar 16, 2000
M���nica, You can convert an ascii table to a multispec image in the following way. Load the following packages: stsdas (an external package), hst_calib, and I presume you already have loaded noao and onedspec. Suppose your input ASCII file is called file.txt. You can do the following: cl> titable file.txt file.tab row=1 cl> tomultispec file.tab file.imh flux_col=c2 wave_col=c1 cl> delete file.tab # this was a temporary file Then file.imh will be in multispec format, and you can work on it with splot. The above assumes that you have wavelengths in the first column and fluxes in the second. If they are reversed, use flux_col=c1 and wave_col=c2 when you run tomultispec. Phil > Hi! > I was tryng to convert an ASCII file in IRAF image and I couldn���t. > When I use the rtextimage task in dataio package I obtain an file.imh but I > can not display with splot the image. > I obtain a follow message: > > error: MWCS: dimension mismatch (mw-open) > > I do not understand what is the problem. My ASCII file have 2 columns > (wavelength and flux). > Please, may you help me or said something do you think that is happening?? > > Regards, > Monica
Olga Kuhn wrote on Mar 16, 2000
Hi, I was wondering about this when the question on ascii to *.imh came up --- If the wavelengths read in from the ascii table are not equally spaced, will splot then be able to correctly compute line equivalent widths or fluxes? (assuming no header information to give the dispersion) Thanks, Olga Kuhn > > You can convert an ascii table to a multispec image in the following way. > Load the following packages: stsdas (an external package), hst_calib, > and I presume you already have loaded noao and onedspec. Suppose your > input ASCII file is called file.txt. You can do the following: > > cl> titable file.txt file.tab row=1 > cl> tomultispec file.tab file.imh flux_col=c2 wave_col=c1 > cl> delete file.tab # this was a temporary file > > Then file.imh will be in multispec format, and you can work on it with splot. > The above assumes that you have wavelengths in the first column and fluxes > in the second. If they are reversed, use flux_col=c1 and wave_col=c2 when > you run tomultispec. >
Rica S. French wrote on Mar 16, 2000
titable requires input already in stsdas tables format. you can create a table from ascii using tcreate, but it might be simpler to use rspectext to go straight from an ascii spectrum to an iraf image (if this works for you). [<>][<>][<>][<>][<>][<>][<>][<>][<>][<>][<>][<>][<>][<>][<>][<>][<>] A bus station is where a bus stops. A train station is where a train stops. On my desk I have a work station............. <Steven Wright> Rica Sirbaugh French rfrench@astro.as.utexas.edu Department of Astronomy http://physics.mtsu.edu/~sirbaugh/ University of Texas at Austin OFC: 512.471.8443 RLM 16.212 Austin, TX 78712 USA FAX: 512.471.6016 On Thu, 16 Mar 2000, Phil Hodge wrote: > M�nica, > > You can convert an ascii table to a multispec image in the following way. > Load the following packages: stsdas (an external package), hst_calib, > and I presume you already have loaded noao and onedspec. Suppose your > input ASCII file is called file.txt. You can do the following: > > cl> titable file.txt file.tab row=1 > cl> tomultispec file.tab file.imh flux_col=c2 wave_col=c1 > cl> delete file.tab # this was a temporary file > > Then file.imh will be in multispec format, and you can work on it with splot. > The above assumes that you have wavelengths in the first column and fluxes > in the second. If they are reversed, use flux_col=c1 and wave_col=c2 when > you run tomultispec. > > Phil > > > Hi! > > I was tryng to convert an ASCII file in IRAF image and I couldn�t. > > When I use the rtextimage task in dataio package I obtain an file.imh but I > > can not display with splot the image. > > I obtain a follow message: > > > > error: MWCS: dimension mismatch (mw-open) > > > > I do not understand what is the problem. My ASCII file have 2 columns > > (wavelength and flux). > > Please, may you help me or said something do you think that is happening?? > > > > Regards, > > Monica >
Ludovic Murat wrote on Mar 16, 2000
Actually im not to sure ,,so I wont give you an answer that is incorrect ,,,so hows your day??/ -----Original Message----- From: irafmail-gateway@noao.edu [SMTP:irafmail-gateway@noao.edu] Sent: Thursday, March 16, 2000 2:33 PM To: lmurat@ubisoft.qc.ca Subject: Re: Text-to-iraf conversion Hi, I was wondering about this when the question on ascii to *.imh came up --- If the wavelengths read in from the ascii table are not equally spaced, will splot then be able to correctly compute line equivalent widths or fluxes? (assuming no header information to give the dispersion) Thanks, Olga Kuhn > > You can convert an ascii table to a multispec image in the following way. > Load the following packages: stsdas (an external package), hst_calib, > and I presume you already have loaded noao and onedspec. Suppose your > input ASCII file is called file.txt. You can do the following: > > cl> titable file.txt file.tab row=1 > cl> tomultispec file.tab file.imh flux_col=c2 wave_col=c1 > cl> delete file.tab # this was a temporary file > > Then file.imh will be in multispec format, and you can work on it with splot. > The above assumes that you have wavelengths in the first column and fluxes > in the second. If they are reversed, use flux_col=c1 and wave_col=c2 when > you run tomultispec. >
Phil Hodge wrote on Mar 16, 2000
> titable requires input already in stsdas tables format. No, the ttools tasks can read ascii text files. The default column names are c1, c2, c3, etc. That's why I specified the column names for flux and wavelength as c2 and c1 respectively. > but it might be simpler to use > rspectext to go straight from an ascii spectrum to an iraf image (if > this works for you). If there are not too many lines, that would indeed be simpler. Using tomultispec assumes that the wavelengths vary smoothly so they can be fit using a polynomial. Phil
Frank Valdes wrote on Mar 16, 2000
In addition to the STSDAS MKMULTISPEC method described by Phil there is also the task RSPECTEXT in the ONEDSPEC package. This works on just a simple wavelength/pixel value ASCII file with an optional header. You can think of it as a version of RTEXTIMAGE for spectra. You can select whether to define the linear wavelength scale, take the values from FITS keyword type of header, or have it fit a dispersion function to the lines. The latter case does not require evenly spaced data and functionally is pretty much like MKMULTISPEC. It is easy to use and requires no intermediate files. There is also the reverse task called WSPECTEXT that creates a two column text file of a spectrum in an image. Frank Valdes
Frank Valdes wrote on Mar 16, 2000
> If the wavelengths read in from the ascii table are not equally spaced, > will splot then be able to correctly compute line equivalent widths > or fluxes? (assuming no header information to give the dispersion) I forgot to add that once a spectrum is stored in an image, either with mkmultispec or rspectext, and whether the pixels are spaced evenly in wavelength with a linear dispersion function or unevenly with a non-linear dispersion function, SPLOT will correctly measure equivalent widths, etc. Frank Valdes
Last post on Mar 16, 2000