CODEWITHSUNDEEP
pythondatetime
Shantanu Gupta
Shantanu Gupta

Reputation: 21

Converting Object to datetime in Python

I have a string of series in this format: 2017-12-04T08:30:00+11:00. I am trying to convert this into a datetime object.Highlighted with yellow colour.See Image:

dataframe-String

How to fetch date from the start, end & updated columns and convert these object to date?

I have tried:

def ISOtstr(iso):
    dcomponents = [1,1,1]
    dcomponents[0] = iso[:4]
    dcomponents[1] = iso[5:7]
    dcomponents[2] = iso[8:10]
    tcomponents = [1,1,1]
    tcomponents[0] = iso[11:13]
    tcomponents[1] = iso[14:16]
    tcomponents[2] = iso[17:19]
    d = dcomponents
    t = tcomponents
    string = "{}-{}-{} {}:{}:{}".format(d[0],d[1],d[2],t[0],t[1],t[2])
    return string

import datetime

string = a.iloc[1]['start']
date_string = ISOtstr(string)
date_obj = datetime.datetime.strptime(date_string, '%Y-%m-%d %H:%M:%S')
print(date_obj)
print(type(date_obj)) 

for item in df['start'].iteritems():
    datetime.datetime.strptime(df['start'], "%a-%b-%d-%H-%M-%S-%Z-%Y")

import datetime

date_time_str = a['start']
date_time_obj = datetime.datetime.strptime(date_time_str, '%Y-%m-%d %H:%M:%S.%f')
print('Date:', date_time_obj.date())
print('Time:', date_time_obj.time())
print('Date-time:', date_time_obj)

TypeError: strptime() argument 1 must be str, not Series

IPython Notebook: https://drive.google.com/file/d/1YbQZOCxtLLUiB4YyivRhM5W6n6CVVh3y/view?usp=sharing

Error Message

Upvotes: 2

Views: 510

Answers (3)

adrtam
adrtam

Reputation: 7241

There is a dateutil module in Python that does all magical date parsing job:

>>> import dateutil.parser
>>> dateutil.parser.parse('2017-12-04T08:30:00+11:00')
datetime.datetime(2017, 12, 4, 8, 30, tzinfo=tzoffset(None, 39600))

Upvotes: 0

Boris the Barb
Boris the Barb

Reputation: 51

You need to pass datetime.datetime.strptime() a string, not series. df['start'] is indexing your 'start' column. If you want to stick with specifically the datetime module, you could look into using df.apply, or iterate over your data.

However, others mentioned before I could, pandas has a built-in datetime method that will do exactly as you wish!

To start you off (sorry haven't much time right now!), you'd essentially pass a series of your df as an argument to the pandas class like so

df['start']=pd.to_datetime(df['start'],format='%foo%bar)

The returned object will be a series you can assign to your 'start' series.

Upvotes: 1

Igor Rivin
Igor Rivin

Reputation: 4864

pandas has a very useful pd.to_datetime function, which does what you want.

Upvotes: 1

Related Questions