Validating a Date and Time |
In Iguana or Chameleon, dates and times are normally in one of the formats shown below:
YYYYMMDDHHmm YYYYMMDDHHmmSS YYYYMMDDHHmmSS.SSSS |
In these date-time formats:
- YYYY is a four-digit year;
- MM is a two-digit month between 1 and 12;
- DD is a two-digit day of the month;
- HH is the hour portion of the time, and is between 0 and 23;
- mm is the minute portion of the time, and is between 0 and 59;
- SS.SSSS is the seconds and (optional) tenths of milliseconds portion of the time, with the seconds portion between 0 and 59.
In addition, any of these time formats can have a time zone correction appended. This consists of + or - and a four-digit number representing the hours and minutes to add or subtract from the time. (In this manual page, this time zone correction is represented as +ZZZZ or -ZZZZ.)
Occasionally, you may need to check whether a date and time are complete and valid. For example:
datetime = '200702151532' # this is valid datetime = '20070215153201.7890+0200' # so is this datetime = '200802292532' # this is invalid - the hour field is '29' datetime = '20080229204901.7890+2730' # invalid - the hour portion of the time zone correction is too large |
The following Python function, parse_time(), validates dates and times provided in any of the date-time formats shown above. It calls the days_in_month() function to obtain the number of days in the specified month. days_in_month() is defined in the calendar.py module, which is included with Iguana and Chameleon.
import re
from ifware.calendar import days_in_month
def get_integers(alist):
result = []
for x in alist:
try: result.append(int(x))
except ValueError: result.append(0)
return result
# A regular expression that matches timestamps.
ts_regex = re.compile(
r'^'
r'(\d{4})' # Year (4 digits)
r'(\d{2})' # Month
r'(\d{2})' # Day
r'(\d{2})' # Hour
r'(\d{2})' # Minute
r'(?:'
r'(\d{2})' # Seconds (optional)
r'(?:\.(\d{4}))?' # Tenths of milliseconds (may follow seconds when present)
r')?'
r'([+-]\d{4})?' # Timezone (optional)
r'$'
)
def parse_time(timestamp):
matches = ts_regex.findall(timestamp)
if len(matches) < 1:
raise ValueError, 'Not a timestamp: ' + timestamp
# findall() returns the matching strings, but we want integers.
(Y,M,D,h,m,s,ms,tz) = get_integers(matches[0])
# Here we validate each part of the timestamp.
if Y < 1900 or M < 1 or M > 12 or D < 1 or D > days_in_month(Y,M):
raise ValueError, 'Invalid date: ' + timestamp
if h > 23 or m > 59 or s > 59:
raise ValueError, 'Invalid time: ' + timestamp
if tz < -2359 or tz > 2359 or abs(tz) % 100 > 59:
raise ValueError, 'Invalid timezone: ' + timestamp
# All values are valid, so we return them.
return (Y,M,D,h,m,s,ms,tz)
|
parse_time() expects one parameter, which is the timestamp that needs to be validated. If it detects an invalid timestamp component, it raises a ValueError exception; otherwise, it returns the components of the timestamp.
Here is some sample code that calls parse_time():
def valid_time(timestamp):
try:
return parse_time(timestamp)
except ValueError:
return None
|
This function returns the components of a valid timestamp, which can also be interpreted as true. It returns None (false) if the timestamp is invalid.
 | ||
| For details on functions that are provided with the calendar.py module, see Calendar and Age Calculation Functions. | ||
 |