latqcdtools.base.readWrite

_lab(num) -> str:
'''
Create a short string label for each column of a data table. Needed for writeTable. 
'''

readTable(filename, unpack=True, col=None, minVal=-inf, maxVal=inf, excludeAtVal=None, **kwargs) -> numpy.ndarray:
'''
Wrapper for np.loadtxt. It unpacks by default to prevent transposition errors, and also optionally
allows the user to restrict the table based on the range of one of the columns.

Args:
    filename (str):
        Name of the file to open. 
    unpack (bool, optional): 
        If False, reads the table in a confusing, useless way. Defaults to True.
    col (int, optional): 
        Use this column to restrict the range of the rest of the table. Defaults to None.
    minVal (float, optional): 
        Minimum value for above restriction. Defaults to -np.inf.
    maxVal (float, optional):
        Maximum value for above restriction. Defaults to np.inf.
    excludeAtVal (float, optional):
        Throw out all rows for which column col has value excludeAtVal. Exclusion is
        carried out after range restriction. 
        
Returns:
    np.array: Data table. 
'''

writeTable(filename, *args, **kwargs):
'''
Wrapper for np.savetxt. The idea is that you can use it like this:

writeTable('file.txt',col1,col2,header=['header1','header2])

This works for an arbitrary number of 1-d columns col. It seems much more intuitive to me 
that you pass columns as arguments than whatever np.savetxt is doing.

Args:
    filename (str): output file name
'''