DataFile 0.1.6

Easy way to read data files

This package provides an easy way to open a datafile by returning a “modified” dictionnary (see Guessing) of numpy arrays that corresponds to the differents columns.If the column contains strings it will return an array of strings.

By default it will try to guess the indexation from the commentaries, and if failed will index it from 0 to n:

#!/usr/bin/env python

from datafile.extractfile import dataFile

f = dataFile("./datafile/test/testdocfile.txt")
# the index will be guessed and if not found it will
# be indexed from 0 to n

f = dataFile("./datafile/test/testdocfile.txt",index="auto")
#f will be indexed from 0 to n

#f["0"] will return a numpy array of the first column.
#f["0",1] will return a numpy array of the index number 1 of the first column,
# if the column is divided in index (ensemble of points separated by one or more space)

f = dataFile("./datafile/test/testdocfile.txt" , "A B C D")
#f will be indexed on "A" "B" "C" (if three columns)
f = dataFile("./datafile/test/testdocfile.txt" , ["A", "B","C","D"]
#f will be indexed on "A" "B" "C"

Guessing

if the datafile contains in the commentaries a line like: #Lx(nm) Ly(nm) D(A) and the file at three column, it will index the three columns with “Lx(nm)” “Ly(nm)” , “D(A)”:

#!/usr/bin/env python
f = dataFile("./file")
f["Lx"]
f["x"]
#both are going to return the first column
f["nm"]
#will return a key error

Main usage

I use it with ipython to plot datas and explore them. Lauching ipython -pylab will launch a ipython with matplotlib loaded. After loading the file with dataFile you can plot the data with only typping plot(f[“A”],f[“B”])

email : [e-mail]