I have three date formats: YYYY-MM-DD, DD.MM.YYYY, DD/MM/YYYY.
Is it possible to validate and parse strings such as 2014-05-18 or 18.5.2014 or 18/05/2019?
5 Answers
Try each format and see if it works:
from datetime import datetime
def try_parsing_date(text):
for fmt in ('%Y-%m-%d', '%d.%m.%Y', '%d/%m/%Y'):
try:
return datetime.strptime(text, fmt)
except ValueError:
pass
raise ValueError('no valid date format found')
1 Comment
Taylor
what if we have huge data and multiple date formats, then looping through a loop is feasible in terms of time complexity? will it not be slow?
>>> import dateutil.parser
>>> dateutil.parser.parse(date_string)
This should take care of most of the standard date formats in Python 2.7+. If you really have super custom date formats, you can always fall back to the one mentioned by Jon Clements.
1 Comment
nedned
This works fine, but if the date formats are all known in advance, the solution by Jon Clements of trying each format and catching exceptions will likely be much faster.
This actually a problem i was facing and this how i approached it, my main purpose was to the date seperators
class InputRequest:
"This class contain all inputs function that will be required in this program. "
def __init__(self, stockTickerName = 'YHOO', stockSite='yahoo', startDate = None,
endDate = datetime.date.today()):
def requestInput(self, requestType =''):
"Fro requesting input from user"
self.__requestInput = input(requestType)
return self.__requestInput
def dateFormat(self, dateType=''):
'''
this function handles user date input
this repeats until the correct format is supplied
dataType: this is the type of date, eg: DOF, Date of Arriveal, etc
'''
while True:
try:
dateString = InputRequest.requestInput(self,dateType)
dtFormat = ('%Y/%m/%d','%Y-%m-%d','%Y.%m.%d','%Y,%m,%d','%Y\%m\%d') #you can add extra formats needed
for i in dtFormat:
try:
return datetime.datetime.strptime(dateString, i).strftime(i)
except ValueError:
pass
except ValueError:
pass
print('\nNo valid date format found. Try again:')
print("Date must be seperated by either [/ - , . \] (eg: 2012/12/31 --> ): ")
Comments
In order to add a bit to @jon-clements already great answer,
added some verbosity on the exception raised for traceability purposes
from datetime import datetime
def try_parsing_date(possible_date, field):
"""
Try to parse a date using several formats, warn about
problematic value if the possible_date does not match
any of the formats tried
"""
for fmt in ('%Y-%m-%d', '%d/%m/%Y', '%m/%d/%Y'):
try:
return datetime.strptime(possible_date, fmt)
except ValueError:
pass
raise ValueError(f"Non-valid date format for field {field}: '{possible_date}'")
Here, possible_date is the text that may contain a date in one of the formats, and field would be the variable or column that you are testing. This makes it easier to find the problematic value
Comments
Pure python:
from datetime import datetime
my_datetime = datetime.strptime('2014-05-18', '%Y-%m-%d')
repr(my_datetime)
>>> 'datetime.datetime(2014,5,18,0,0)'
Check datetime.strptime() format documentation for more format strings.
3 Comments
Hans Then
The solution should be expanded to use all date formats and explain how to choose which one fits.
Gabriel M
@HansThen you're right, but answer above already answers the question correctly and there's nothing to add to it.
Hans Then
Yes, it is hard to improve upon @Jon Clements.
strptimeing each 3 in turn to see if they work, otherwise fail...