Institut für Astronomie und AstrophysikAbteilung AstronomieSand 1, D-72076 Tübingen, Germany |
data_dat
Read in data set from ascii file.
cafe
data
data, filename["["xcol,ycol[,errcol][,xbincol][,headskip][,commentchar]"]"]
Read the file as an ascii file containing the data in columns (error column need not be given): x | y | error. Number separators are non-number characters, recommended are spaces or tabs. Comment lines which start with "#" will be skipped.
(optional) parameters are ordered as follows: xcol - The column number for the X axis. Default is column 0. ycol - The column number for the Y axis. Default is column 1. errcol - The column number for the ERROR axis. Default is column 2. Empty string skips the error column. xbincol - The column number for the XBIN axis. If not given the XBIN axis will be omitted. headskip - The number of header lines to skip. Default is 0. commentchar - The comment character. Default is "#". All parameters are separated with ",". If one parameter is not defined (empty string) the default value is used.
All keywords/options may be set with the "set" command. The command prefix is "data_dat", the setup keyword is as used for parameters above. Additionally following setup keywords are defined: nostrict - do not check for proper number on each column. This allows to read files which contain some columns with irregular numbers. Default is off. sepexpr - Use given expression as separator between the numbers. Must be string. Default is ""[^0-9eE.+-]+" numexpr - Use given expression to check for correct number representation. Might be used to restrict/enhance representation. Default is: > "(^(\+|-)?[0-9]+(\.[0-9]*)?((e|E)(\+|-)?[0-9]+)?)$"
Loads ascii file data to group defined and free subgroup.
> data, test.dat:5, -> loads in data from ascii file test.dat (marked via extension dat) to first free subgroup in group 5.
$Id: cafe_data_dat.pro,v 1.22 2005/02/17 07:52:25 goehler Exp $
[Home Page] [Software, Documentation] [IDL Documentation] [Quick Reference] [Feedback]