datetime_z.py

"""Functions to parse datetime objects."""

# We're using regular expressions rather than time.strptime because:
# - They provide both validation and parsing.
# - They're more flexible for datetimes.
# - The date/datetime/time constructors produce friendlier error messages.

import datetime
import re

# UTC and local time zones

ZERO = datetime.timedelta(0)


class FixedOffset(datetime.tzinfo):
    """
    Fixed offset in minutes east from UTC. Taken from Python's docs.
    Kept as close as possible to the reference version. __init__ was changed
    to make its arguments optional, according to Python's requirement that
    tzinfo subclasses can be instantiated without arguments.
    """

    def __init__(self, offset=None, name=None):
        if offset is not None:
            self.__offset = datetime.timedelta(minutes=offset)
        if name is not None:
            self.__name = name

    def utcoffset(self, dt):
        return self.__offset

    def tzname(self, dt):
        return self.__name

    def dst(self, dt):
        return ZERO

    def __repr__(self):
        return self.__name


class UTC(datetime.tzinfo):
    zone = "UTC"

    def utcoffset(self, dt):
        return ZERO

    def tzname(self, dt):
        return "UTC"

    def dst(self, dt):
        return ZERO

    def __repr__(self):
        return "UTC"

# UTC time zone as a tzinfo instance.
utc = UTC()


def get_fixed_timezone(offset):
    """Return a tzinfo instance with a fixed offset from UTC."""
    if isinstance(offset, datetime.timedelta):
        offset = offset.total_seconds() // 60
    sign = '-' if offset < 0 else '+'
    hhmm = '%02d%02d' % divmod(abs(offset), 60)
    name = sign + hhmm
    return FixedOffset(offset, name)


date_re = re.compile(
    r'(?P<year>\d{4})-(?P<month>\d{1,2})-(?P<day>\d{1,2})$'
)

time_re = re.compile(
    r'(?P<hour>\d{1,2}):(?P<minute>\d{1,2})'
    r'(?::(?P<second>\d{1,2})(?:\.(?P<microsecond>\d{1,6})\d{0,6})?)?'
)

datetime_re = re.compile(
    r'(?P<year>\d{4})-(?P<month>\d{1,2})-(?P<day>\d{1,2})'
    r'[T ](?P<hour>\d{1,2}):(?P<minute>\d{1,2})'
    r'(?::(?P<second>\d{1,2})(?:\.(?P<microsecond>\d{1,6})\d{0,6})?)?'
    r'(?P<tzinfo>Z|[+-]\d{2}(?::?\d{2})?)?$'
)

standard_duration_re = re.compile(
    r'^'
    r'(?:(?P<days>-?\d+) (days?, )?)?'
    r'((?:(?P<hours>-?\d+):)(?=\d+:\d+))?'
    r'(?:(?P<minutes>-?\d+):)?'
    r'(?P<seconds>-?\d+)'
    r'(?:\.(?P<microseconds>\d{1,6})\d{0,6})?'
    r'$'
)

# Support the sections of ISO 8601 date representation that are accepted by
# timedelta
iso8601_duration_re = re.compile(
    r'^(?P<sign>[-+]?)'
    r'P'
    r'(?:(?P<days>\d+(.\d+)?)D)?'
    r'(?:T'
    r'(?:(?P<hours>\d+(.\d+)?)H)?'
    r'(?:(?P<minutes>\d+(.\d+)?)M)?'
    r'(?:(?P<seconds>\d+(.\d+)?)S)?'
    r')?'
    r'$'
)

# Support PostgreSQL's day-time interval format, e.g. "3 days 04:05:06". The
# year-month and mixed intervals cannot be converted to a timedelta and thus
# aren't accepted.
postgres_interval_re = re.compile(
    r'^'
    r'(?:(?P<days>-?\d+) (days? ?))?'
    r'(?:(?P<sign>[-+])?'
    r'(?P<hours>\d+):'
    r'(?P<minutes>\d\d):'
    r'(?P<seconds>\d\d)'
    r'(?:\.(?P<microseconds>\d{1,6}))?'
    r')?$'
)


def parse_date(value):
    """Parse a string and return a datetime.date.

    Raise ValueError if the input is well formatted but not a valid date.
    Return None if the input isn't well formatted.
    """
    match = date_re.match(value)
    if match:
        kw = {k: int(v) for k, v in match.groupdict().items()}
        return datetime.date(**kw)


def parse_time(value):
    """Parse a string and return a datetime.time.

    This function doesn't support time zone offsets.

    Raise ValueError if the input is well formatted but not a valid time.
    Return None if the input isn't well formatted, in particular if it
    contains an offset.
    """
    match = time_re.match(value)
    if match:
        kw = match.groupdict()
        if kw['microsecond']:
            kw['microsecond'] = kw['microsecond'].ljust(6, '0')
        kw = {k: int(v) for k, v in kw.items() if v is not None}
        return datetime.time(**kw)


def parse_datetime(value):
    """Parse a string and return a datetime.datetime.

    This function supports time zone offsets. When the input contains one,
    the output uses a timezone with a fixed offset from UTC.

    Raise ValueError if the input is well formatted but not a valid datetime.
    Return None if the input isn't well formatted.

    >>> from datetime_z import parse_datetime
    >>> parse_datetime('2013-07-23T15:10:59.342107+01:00')
    datetime.datetime(2013, 7, 23, 15, 10, 59, 342107, tzinfo=+0100)
    >>> parse_datetime('2013-07-23T15:10:59.34210Z')
    datetime.datetime(2013, 7, 23, 15, 10, 59, 342100, tzinfo=UTC)
    """
    match = datetime_re.match(value)
    if match:
        kw = match.groupdict()
        if kw['microsecond']:
            kw['microsecond'] = kw['microsecond'].ljust(6, '0')
        tzinfo = kw.pop('tzinfo')
        if tzinfo == 'Z':
            tzinfo = utc
        elif tzinfo is not None:
            offset_mins = int(tzinfo[-2:]) if len(tzinfo) > 3 else 0
            offset = 60 * int(tzinfo[1:3]) + offset_mins
            if tzinfo[0] == '-':
                offset = -offset
            tzinfo = get_fixed_timezone(offset)
        kw = {k: int(v) for k, v in kw.items() if v is not None}
        kw['tzinfo'] = tzinfo
        return datetime.datetime(**kw)


def parse_duration(value):
    """Parse a duration string and return a datetime.timedelta.

    The preferred format for durations in Django is '%d %H:%M:%S.%f'.

    Also supports ISO 8601 representation and PostgreSQL's day-time interval
    format.
    """
    match = (
        standard_duration_re.match(value) or
        iso8601_duration_re.match(value) or
        postgres_interval_re.match(value)
    )
    if match:
        kw = match.groupdict()
        days = datetime.timedelta(float(kw.pop('days', 0) or 0))
        sign = -1 if kw.pop('sign', '+') == '-' else 1
        if kw.get('microseconds'):
            kw['microseconds'] = kw['microseconds'].ljust(6, '0')
        if kw.get('seconds') and kw.get('microseconds') and kw['seconds'].startswith('-'):
            kw['microseconds'] = '-' + kw['microseconds']
        kw = {k: float(v) for k, v in kw.items() if v is not None}
        return days + sign * datetime.timedelta(**kw)


if __name__ == "__main__":
    import doctest
    doctest.testmod()