barak.io.readtxt¶

barak.io.readtxt(fh, sep=None, usecols=None, comment='#', skip=0, arrays=True, names=None, readnames=False, converters=None, mintype=<type 'int'>)[source]¶

Reads columns from a text file into arrays, converting to int, float or str where appropriate.

By default the column separator is whitespace. rows can be either an input filename or an iterable (e.g. a file object, list or iterator).

Parameters :

rows : filename or iterable object Input data. sep : str (default None) A string used to separate items on a row (also known as a delimiter). Default is None, which means whitespace. usecols : int or tuple of ints, optional Indices of columns to be read. By default all columns are read. comment : str (default #) Character marking the start of a comment. skip : int (default 0) Number of rows to skip (not counting commented or blank lines) before reading data. arrays : bool (True) If True, all columns are converted to Numpy arrays. If False, columns are returned as lists. names : str or sequence of str (default None) If names is given and arrays is True, the data are placed in a Numpy record array with field names given by names. Can also be a single string of comma-separated values. readnames : bool (False) If readnames is True the first line of the file is read to find the field names. This overrides the names keyword. converters : dict (None) Functions to apply to each entry of a column. Each (key,value) pair gives the column index (key) and the function to be applied to each entry in that column (value). Returns either structured array or lists. :

Examples

>>> list_of_all_cols = readtxt('filename')
>>> ninthcol, fifthcol = readtxt('filename', sep=',', usecols=(8,4)])
>>> firstcol = readtxt('filename', comment='%', usecols=[0])
>>> recarray = readtxt('filename', sep=',', usecols=(1,3), names='x,y'])