dateparse.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

from django.utils.regex_helper import _lazy_re_compile
from django.utils.timezone import get_fixed_timezone, utc

date_re = _lazy_re_compile(r"(?P<year>\d{4})-(?P<month>\d{1,2})-(?P<day>\d{1,2})$")

time_re = _lazy_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 = _lazy_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"\s*(?P<tzinfo>Z|[+-]\d{2}(?::?\d{2})?)?$"
)

standard_duration_re = _lazy_re_compile(
    r"^"
    r"(?:(?P<days>-?\d+) (days?, )?)?"
    r"(?P<sign>-?)"
    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 = _lazy_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 = _lazy_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.
    """
    try:
        return datetime.date.fromisoformat(value)
    except ValueError:
        if match := date_re.match(value):
            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.
    """
    try:
        # The fromisoformat() method takes time zone info into account and
        # returns a time with a tzinfo component, if possible. However, there
        # are no circumstances where aware datetime.time objects make sense, so
        # remove the time zone offset.
        return datetime.time.fromisoformat(value).replace(tzinfo=None)
    except ValueError:
        if match := time_re.match(value):
            kw = match.groupdict()
            kw["microsecond"] = kw["microsecond"] and 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.
    """
    try:
        return datetime.datetime.fromisoformat(value)
    except ValueError:
        if match := datetime_re.match(value):
            kw = match.groupdict()
            kw["microsecond"] = kw["microsecond"] and 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}
            return datetime.datetime(**kw, tzinfo=tzinfo)


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()
        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.replace(",", ".")) for k, v in kw.items() if v is not None}
        days = datetime.timedelta(kw.pop("days", 0.0) or 0.0)
        if match.re == iso8601_duration_re:
            days *= sign
        return days + sign * datetime.timedelta(**kw)