读取.dat文件的日期字符串

Question

In Python:

Got a .dat file one column is a datestr 'yyyy-mm-dd'. Column years range from 2000, to 2010 I only want to use 2005.

How can I successfully read using np.loadtxt, keeping in the same format.

I am then going to use:

time_string = yyyy-mm-dd
doy = int (time.strftime ("%j", time.strptime ( time_string, "%Y, %m, %d")))

to convert yyyy-mm-dd to day of year (1-365)

Answer 1

The question doesn't point any reason to the use of loadtxt from numpy , so you actually don't care about how your data is loaded. Said that, in this case you simply use dtype=object for loading it.

Suppose this is your .dat file, let us call it d1.dat:

1 2000-01-01 blah
2 2005-01-01 bleh
3 2006-02-03 blih
4 2008-03-04 bloh
5 2010-04-05 bluh
6 2005-03-12 blahr

Then (for example) to load it using numpy :

import numpy
data = numpy.loadtxt('d1.dat', usecols=[1,2], dtype=object)

Now you can apply your function to extract the day of the year from the first column in data:

for date, _ in data:
    print time.strftime("%j", time.strptime(date, "%Y-%m-%d"))

Answer 2

from datetime import datetime
time_string = '2012-12-31'
dt = datetime.strptime(time_string, '%Y-%m-%d')
print dt.timetuple().tm_yday


366