TODO/OPEN
=========
- The Python implementation is missing docstrings in many places.

- LaTeXize the docs.


CLOSED
======
- Pickle incompatibility?  It occurs to me that, because the
  implementations are so different, a pickle created by the Python
  implementation won't be unpicklable by the C implementation, or vice
  versa.  If this is important, the Python implementation will have to
  change, as the C implementation of pickling took days to get working
  and has no wiggle room (TOOWTDI indeed <wink>).

  Resolution:  Jim Fulton suggested adding various __reduce__ methods
  to the Python implementation, and that did lead to pickles that the
  C implementation groks.  I'm not sure why, and tzinfo subclasses must
  have an __init__ that can be called without arguments now (if
  they want to play with pickles).  It appears they must also set
  class attr __safe_for_unpickling__ = True, but only if they want a
  pickle written by the Python implementation to be readable by the C
  implementation.

- The test suite doesn't pass under 2.2.2, due to what Guido tracked
  to a bug in 2.2.2's implementation of __cmp__ for new-style classes.
  Later:  the test suite grew a version check to avoid provoking this
  bug under 2.2.2.

- What should str() do?  It generally acts like a synonym for isoformat()
  now.  But

    >>> print time(2)
    02:00:00.000000
    >>>

  is arguably better as '2:00:00' or even '2:00'.  The Python
  implementation has (overridden) "pretty __str__" according to one
  person's idea of "pretty", for a couple types.  Rat hole.
  Guido sez:  chop ".000000" when microseconds are 0, and that's it.
  Tim sez:  and having a fixed-size string when they are will make
  life easier for people implementing their own ideas of "pretty".
  Later:  isoformat() should also elide trailing microsecond when it's
  0.
  Done.

- pickles still "feel too big".  But this is a general issue w/ new-style
  classes.

- The type objects aren't exposed at the C level, due to headaches with
  DLL/.so.  We're going to ignore this; if you need to construct, e.g.,
  a datetime object from C, the constructors can still be gotten via
  importing the datetime module and calling the appropriate method.

- Every function involving a timestamp inherits platform C limitations
  on year range, on resolution, and on precision.  If we were to invent
  our own timestamp format, an IEEE double doesn't have enough bits to
  cover 9999 years to microsecond resolution.  Tough.  The only methods
  in this module that *produce* timestamps (as opposed to consuming them)
  are in datetimetz, and people will have to live with that limitation.
  Other methods are timestamp consumers, and suffer platform limitations
  no matter what we do here.


Overview
========
The datetime module supplies classes for manipulating dates and times,
in both simple and complex ways.  While date and time arithmetic is
supported, the focus of the implementation is on efficient field
extraction, for output formatting and manipulation.

There are two kinds of date and time objects:  "naive" and "aware".
These refer to whether they have any notion of time zone, daylight
savings time, or other kind of algorithmic or political time adjustment.
Whether a naive datetime object represents UTC, local time, or time in
some other timezone is purely up to the program, just like it's up to
the program whether a particular number represents meters, miles, or
mass.  Naive datetime objects are easy to understand and to work with,
at the cost of ignoring some aspects of reality.

For applications requiring more, so-called aware datetime subclasses add
an optional time zone information object to the basic naive classes.
These tzinfo objects capture information about offset from UTC time,
time zone name, and whether DST is in effect.  Note that no concrete
tzinfo classes are supplied by the datetime module.  Instead, they
provide a framework for incorporating the level of detail an app may
require.  The rules for time adjustment across the world are more
political than rational, and there is no standard suitable for every
app.

class date
    year, month, and day
    An idealized naive date, assuming the current Gregorian calendar
    always was, and always will be, in effect.

class time
    hour, minute, second, and microsecond
    An idealized naive time, independent of any particular day, assuming
    that every day has exactly 24*60*60 seconds (there is no notion
    of "leap seconds" here).

class datetime
    year, month, day, hour, minute, second, and microsecond
    A combination of a naive date and a naive time.

class timedelta
    A duration, the difference between two dates, times, or datetimes, to
    microsecond resolution.

class tzinfo
    An abstract base class for time zone information objects.  These are
    used by the datetimetz and timetz classes to provided a customizable
    notion of time adjustment (for example, to account for time zone
    and/or daylight savings time).

class timetz
    An aware subclass of time, supporting a customizable notion of
    time adjustment.

class datetimetz
    An aware subclass of datetime, supporting a customizable notion of
    time adjustment.

Objects of these types are immutable.


Subclass relationships
======================
    object
        timedelta
        tzinfo
        time
            timetz
        date
            datetime
                datetimetz


Module constants
================
MINYEAR
    The smallest year number allowed in a date, datetime, or datetimetz
    object.  MINYEAR is 1.

MAXYEAR
    The largest year number allowed in a date, datetime, or datetimetz
    object.  MAXYEAR is 9999.


strftime() Behavior
===================
date, datetime, datetimetz, time, and timetz objects all support
a strftime(format) method, to create a string representing the time
under the control of an explicit format string.  Broadly speaking,
    d.strftime(fmt)
acts like the time module's
    time.strftime(fmt, d.timetuple())
although not all objects support a timetuple() method.

For time and timetz objects, format codes for year, month, and day
should not be used, as time objects have no such values.  1900 is
used for the year, and 0 for the month and day.

For date objects, format codes for hours, minutes, and seconds should
not be used, as date objects have no such values.  0 is used insted.

For a naive object, the %z and %Z format codes are replaced by
empty strings.

For an aware object:

- %z:  self.utcoffset() is transformed into a 5-character
  string of the form +HHMM or -HHMM, where HH is a 2-digit string
  giving the number of UTC offset hours, and MM is a 2-digit string
  giving the number of UTC offset minutes.  For example, if
  utcoffset() returns -180, %z is replaced with string "-0300".

- %Z:  If self.tzname() returns None, %Z is replaced by an empty string.
  Else %Z is replaced by the returned value, which must be a string.

The full set of format codes supported varies across platforms,
because Python calls the platform C library's strftime() function,
and platform variations are common.  The documentation for Python's
time module lists the format codes that the C standard (1989 version)
requires, and those will work on all platforms with a standard C
implementation.  Note that the 1999 version of the C standard added
additional format codes.

The exact range of years for which strftime() works also varies
across platforms.  Regardless of platform, years before 1900 cannot
be used.


class timedelta
===============
A timedelta object represents a duration, the difference between two
dates or times.

Constructor:

    timedelta(days=0, seconds=0, microseconds=0,
              # The following should only be used as keyword args:
              milliseconds=0, minutes=0, hours=0, weeks=0)

    All arguments are optional.  Arguments may be ints, longs, or floats,
    and may be positive or negative.

    Only days, seconds and microseconds are stored internally.  Arguments
    are converted to those units:

        A millisecond is converted 1000 microseconds.
        A minute is converted to 60 seconds.
        An hour is converted to 3600 seconds.
        A week is converted to 7 days.

    and days, seconds and microseconds are then normalized so that the
    representation is unique, with

        0 <= microseconds < 1000000
        0 <= seconds < 3600*24 (the number of seconds in one day)
        -999999999 <= days <= 999999999

    If any argument is a float, and there are fractional microseconds,
    the fractional microseconds left over from all arguments are combined
    and their sum is rounded to the nearest microsecond.  If no
    argument is a flost, the conversion and normalization processes
    are exact (no information is lost).

    If the normalized value of days lies outside the indicated range,
    OverflowError is raised.

    Note that normalization of negative values may be surprising at first.
    For example,

    >>> d = timedelta(microseconds=-1)
    >>> (d.days, d.seconds, d.microseconds)
    (-1, 86399, 999999)
    >>>


Class attributes:

    .min
        The most negative timedelta object, timedelta(-999999999).

    .max
        The most positive timedelta object,
        timedelta(days=999999999, hours=23, minutes=59, seconds=59,
                  microseconds=999999)

    .resolution
        The smallest possible difference between non-equal timedelta
        objects, timedelta(microseconds=1).

    Note that, because of normalization, timedelta.max > -timedelta.min.
    -timedelta.max is not representable as a timedelta object.

Instance attributes (read-only):

    .days           between -999999999 and 999999999 inclusive
    .seconds        between 0 and 86399 inclusive
    .microseconds   between 0 and 999999 inclusive

Supported operations:

    - timedelta + timedelta -> timedelta
      This is exact, but may overflow.  After
          t1 = t2 + t3
      t1-t2 == t3 and t1-t3 == t2 are true.

    - timedelta - timedelta -> timedelta
      This is exact, but may overflow.  After
          t1 = t2 - t3
       t2 == t1 + t3 is true.

    - timedelta * (int or long) -> timedelta
      (int or long) * timedelta -> timedelta
      This is exact, but may overflow.  After
          t1 = t2 * i
      t1 // i == t2 is true, provided i != 0.  In general,
          t * i == t * (i-1) + t
      is true.

    - timedelta // (int or long) -> timedelta
      The floor is computed and the remainder (if any) is thrown away.
      Division by 0 raises ZeroDivisionError.

    - certain additions and subtractions with date, datetime, and datimetz
      objects (see below)

    - +timedelta -> timedelta
      Returns a timedelta object with the same value.

    - -timedelta -> timedelta
      -t is equivalent to timedelta(-t.days, -t.seconds, -t.microseconds),
      and to t*-1.  This is exact, but may overflow (for example,
      -timedelta.max is not representable as a timedelta object).

    - abs(timedelta) -> timedelta
      abs(t) is equivalent to +t when t.days >= 0, and to -t when
      t.days < 0.  This is exact, and cannot overflow.

    - comparison of timedelta to timedelta; the timedelta representing
      the smaller duration is considered to be the smaller timedelta

    - hash, use as dict key

    - efficient pickling

    - in Boolean contexts, a timedelta object is considred to be true
      if and only if it isn't equal to timedelta(0)


class date
==========
A date object represents a date (year, month and day) in an idealized
calendar, the current Gregorian calendar indefinitely extended in both
directions.  January 1 of year 1 is called day number 1, January 2 of year
1 is called day number 2, and so on.  This matches the definition of the
"proleptic Gregorian" calendar in Dershowitz and Reingold's book
"Calendrical Calculations", where it's the base calendar for all
computations.  See the book for algorithms for converting between
proleptic Gregorian ordinals and many other calendar systems.

Constructor:

    date(year, month, day)

    All arguments are required.  Arguments may be ints or longs, in the
    following ranges:

        MINYEAR <= year <= MAXYEAR
        1 <= month <= 12
        1 <= day <= number of days in the given month and year

    If an argument outside those ranges is given, ValueError is raised.

Other constructors (class methods):

  - today()
    Return the current local date.  This is equivalent to
    date.fromtimestamp(time.time()).

  - fromtimestamp(timestamp)
    Return the local date corresponding to the POSIX timestamp, such as
    is returned by time.time().  This may raise ValueError, if the
    timestamp is out of the range of values supported by the platform C
    localtime() function.  It's common for this to be restricted to
    years in 1970 through 2038.

  - fromordinal(ordinal)
    Return the date corresponding to the proleptic Gregorian ordinal,
    where January 1 of year 1 has ordinal 1.  ValueError is raised
    unless 1 <= ordinal <= date.max.toordinal().  For any date d,
    date.fromordinal(d.toordinal()) == d.

Class attributes:

    .min
        The earliest representable date, date(MINYEAR, 1, 1).

    .max
        The latest representable date, date(MAXYEAR, 12, 31).

    .resolution
        The smallest possible difference between non-equal date
        objects, timedelta(days=1).

Instance attributes (read-only):

    .year           between MINYEAR and MAXYEAR inclusive
    .month          between 1 and 12 inclusive
    .day            between 1 and the number of days in the given month
                    of the given year

Supported operations:

    - date1 + timedelta -> date2
      timedelta + date1 -> date2
      date2 is timedelta.days days removed from the date1, moving forward
      in time if timedelta.days > 0, or backward if timedetla.days < 0.
      date2 - date1 == timedelta.days after.  timedelta.seconds and
      timedelta.microseconds are ignored.  OverflowError is raised if
      date2.year would be smaller than MINYEAR or larger than MAXYEAR.

    - date1 - timedelta -> date2
      Computes the date2 such that date2 + timedelta == date1.  This
      isn't quite equivalent to date1 + (-timedelta), because -timedelta
      in isolation can overflow in cases where date1 - timedelta does
      not.  timedelta.seconds and timedelta.microseconds are ignored.

    - date1 - date2 -> timedelta
      This is exact, and cannot overflow.  timedelta.seconds and
      timedelta.microseconds are 0, and date2 + timedelta == date1
      after.

    - comparison of date to date, where date1 is considered less than
      date2 when date1 precedes date2 in time.  In other words,
      date1 < date2 if and only if date1.toordinal() < date2.toordinal().

    - hash, use as dict key

    - efficient pickling

    - in Boolean contexts, all date objects are considered to be true

Instance methods:

  - timetuple()
    Return a 9-element tuple of the form returned by time.localtime().
    The hours, minutes and seconds are 0, and the DST flag is -1.
    d.timetuple() is equivalent to
        (d.year, d.month, d.day,
         0, 0, 0,  # h, m, s
         d.weekday(),  # 0 is Monday
         d.toordinal() - date(d.year, 1, 1).toordinal() + 1, # day of year
         -1)

  - toordinal()
    Return the proleptic Gregorian ordinal of the date, where January 1
    of year 1 has ordinal 1.  For any date object d,
    date.fromordinal(d.toordinal()) == d.

  - weekday()
    Return the day of the week as an integer, where Monday is 0 and
    Sunday is 6.  For example, date(2002, 12, 4).weekday() == 2, a
    Wednesday.
    See also isoweekday().

  - isoweekday()
    Return the day of the week as an integer, where Monday is 1 and
    Sunday is 7.  For example, date(2002, 12, 4).isoweekday() == 3, a
    Wednesday.
    See also weekday() and isocalendar().

  - isocalendar()
    Return a 3-tuple, (ISO year, ISO week number, ISO weekday).

    The ISO calendar is a widely used variant of the Gregorian calendar.
    See <http://www.phys.uu.nl/~vgent/calendar/isocalendar.htm>
    for a good explanation.

    The ISO year consists of 52 or 53 full weeks, and where a week starts
    on a Monday and ends on a Sunday.  The first week of an ISO year is
    the first (Gregorian) calendar week of a year containing a Thursday.
    This is called week number 1, and the ISO year of that Thursday is
    the same as its Gregorian year.

    For example, 2004 begins on a Thursday, so the first week of ISO
    year 2004 begins on Monday, 29 Dec 2003 and ends on Sunday, 4 Jan
    2004, so that

    date(2003, 12, 29).isocalendar() == (2004, 1, 1)
    date(2004, 1, 4).isocalendar() == (2004, 1, 7)

  - isoformat()
    Return a string representing the date in ISO 8601 format,
    'YYYY-MM-DD'.  For example,
    date(2002, 12, 4).isoformat() == '2002-12-04'.

  - __str__()
    For a date d, str(d) is equivalent to d.isoformat().

  - ctime()
    Return a string representing the date, for example
    date(2002, 12, 4).ctime() == 'Wed Dec  4 00:00:00 2002'.
    d.ctime() is equivalent to time.ctime(time.mktime(d.timetuple()))
    on platforms where the native C ctime() function (which time.ctime()
    invokes, but which date.ctime() does not invoke) conforms to the
    C standard.

  - strftime(format)
    Return a string representing the date, controlled by an explicit
    format string.  Format codes referring to hours, minutes or seconds
    will see 0 values.  See the section on strftime() behavior.


class datetime
==============
A datetime object is a single object containing all the information from
a date object and a time object.  Like a date object, datetime assumes
the current Gregorian calendar extended in both directions; like a time
object, datetime assumes there are exactly 3600*24 seconds in every day.

Constructor:

    datetime(year, month, day,
             hour=0, minute=0, second=0, microsecond=0)

    The year, month and day arguments are required.  Arguments may be ints
    or longs, in the following ranges:

        MINYEAR <= year <= MAXYEAR
        1 <= month <= 12
        1 <= day <= number of days in the given month and year
        0 <= hour < 24
        0 <= minute < 60
        0 <= second < 60
        0 <= microsecond < 1000000

    If an argument outside those ranges is given, ValueError is raised.

Other constructors (class methods):

  - today()
    Return the current local datetime.  This is equivalent to
    datetime.fromtimestamp(time.time()).
    See also now(), fromtimestamp().

  - now()
    Return the current local datetime.  This is like today(), but, if
    possible, supplies more precision than can be gotten from going
    through a time.time() timestamp (for example, this may be possible
    on platforms that supply the C gettimeofday() function).
    See also today(), utcnow().

  - utcnow()
    Return the current UTC datetime.  This is like now(), but returns
    the current UTC date and time.
    See also now().

  - fromtimestamp(timestamp)
    Return the local datetime corresponding to the POSIX timestamp, such
    as is returned by time.time().  This may raise ValueError, if the
    timestamp is out of the range of values supported by the platform C
    localtime() function.  It's common for this to be restricted to
    years in 1970 through 2038.
    See also utcfromtimestamp().

  - utcfromtimestamp(timestamp)
    Return the UTC datetime corresponding to the POSIX timestamp.
    This may raise ValueError, if the timestamp is out of the range of
    values supported by the platform C gmtime() function.  It's common
    for this to be restricted to years in 1970 through 2038.
    See also fromtimestamp().

  - fromordinal(ordinal)
    Return the datetime corresponding to the proleptic Gregorian ordinal,
    where January 1 of year 1 has ordinal 1.  ValueError is raised
    unless 1 <= ordinal <= datetime.max.toordinal().  The hour, minute,
    second and microsecond of the result are all 0.

  - combine(date, time)
    Return a new datetime object whose date components are equal to the
    given date object's, and whose time components are equal to the given
    time object's.  For any datetime object d,
    d == datetime.combine(d.date(), d.time()).
    If date is a datetime or datetimetz object, its time components are
    ignored.  If date is datetimetz object, its tzinfo component is also
    ignored.  If time is a timetz object, its tzinfo component is ignored.

Class attributes:

    .min
        The earliest representable datetime,
        datetime(MINYEAR, 1, 1).

    .max
        The latest representable datetime,
        datetime(MAXYEAR, 12, 31, 23, 59, 59, 999999).

    .resolution
        The smallest possible difference between non-equal datetime
        objects, timedelta(microseconds=1).

Instance attributes (read-only):

    .year           between MINYEAR and MAXYEAR inclusive
    .month          between 1 and 12 inclusive
    .day            between 1 and the number of days in the given month
                    of the given year
    .hour           in range(24)
    .minute         in range(60)
    .second         in range(60)
    .microsecond    in range(1000000)

Supported operations:

    - datetime1 + timedelta -> datetime2
      timedelta + datetime1 -> datetime2
      datetime2 is a duration of timedelta removed from datetime1, moving
      forward in time if timedelta.days > 0, or backward if
      timedelta.days < 0.  datetime2 - datetime1 == timedelta after.
      OverflowError is raised if datetime2.year would be smaller than
      MINYEAR or larger than MAXYEAR.

    - datetime1 - timedelta -> datetime2
      Computes the datetime2 such that datetime2 + timedelta == datetime1.
      This isn't quite equivalent to datetime1 + (-timedelta), because
      -timedelta in isolation can overflow in cases where
      datetime1 - timedelta does not.

    - datetime1 - datetime2 -> timedelta
      This is exact, and cannot overflow.
      datetime2 + timedelta == datetime1 after.

    - comparison of datetime to datetime, where datetime1 is considered
      less than datetime2 when datetime1 precedes datetime2 in time.

    - hash, use as dict key

    - efficient pickling

    - in Boolean contexts, all datetime objects are considered to be true

Instance methods:

  - date()
    Return date object with same year, month and day.

  - time()
    Return time object with same hour, minute, second and microsecond.

  - timetuple()
    Return a 9-element tuple of the form returned by time.localtime().
    The DST flag is -1.   d.timetuple() is equivalent to
        (d.year, d.month, d.day,
         d.hour, d.minute, d.second,
         d.weekday(),  # 0 is Monday
         d.toordinal() - date(d.year, 1, 1).toordinal() + 1, # day of year
         -1)

  - toordinal()
    Return the proleptic Gregorian ordinal of the date.  The same as
    date.toordinal().

  - weekday()
    Return the day of the week as an integer, where Monday is 0 and
    Sunday is 6.  The same as date.weekday().
    See also isoweekday().

  - isoweekday()
    Return the day of the week as an integer, where Monday is 1 and
    Sunday is 7.  The same as date.isoweekday().
    See also weekday() and isocalendar().

  - isocalendar()
    Return a 3-tuple, (ISO year, ISO week number, ISO weekday).  The
    same as date.isocalendar().

  - isoformat(sep='T')
    Return a string representing the date and time in ISO 8601 format,
        YYYY-MM-DDTHH:MM:SS.mmmmmm
    or, if self.microsecond is 0,
        YYYY-MM-DDTHH:MM:SS
    Optional argument sep (default 'T') is a one-character separator,
    placed between the date and time portions of the result.  For example,
        datetime(2002, 12, 4, 1, 2, 3, 4).isoformat(' ') ==
        '2002-12-04 01:02:03.000004'

  - __str__()
    For a datetime d, str(d) is equivalent to d.isoformat(' ').

  - ctime()
    Return a string representing the date, for example
    datetime(2002, 12, 4, 20, 30, 40).ctime() == 'Wed Dec  4 20:30:40 2002'.
    d.ctime() is equivalent to time.ctime(time.mktime(d.timetuple()))
    on platforms where the native C ctime() function (which time.ctime()
    invokes, but which datetime.ctime() does not invoke) conforms to the
    C standard.

  - strftime(format)
    Return a string representing the date and time, controlled by an
    explicit format string.  See the section on strftime() behavior.


class time
==========
A time object represents an idealized time of day, independent of day
and timezone.

Constructor:

    time(hour=0, minute=0, second=0, microsecond=0)

    All arguments are optional.  They may be ints or longs, in the
    following ranges:

        0 <= hour < 24
        0 <= minute < 60
        0 <= second < 60
        0 <= microsecond < 1000000

    If an argument outside those ranges is given, ValueError is raised.

Other constructors (class methods):

    None

Class attributes:

    .min
        The earliest representable time, time(0, 0, 0, 0).

    .max
        The latest representable time, time(23, 59, 59, 999999).

    .resolution
        The smallest possible difference between non-equal time
        objects, timedelta(microseconds=1), although note that
        arithmetic on time objects is not supported.

Instance attributes (read-only):

    .hour           in range(24)
    .minute         in range(60)
    .second         in range(60)
    .microsecond    in range(1000000)

Supported operations:

    - comparison of time to time, where time1 is considered
      less than time2 when time1 precedes time2 in time.

    - hash, use as dict key

    - efficient pickling

    - in Boolean contexts, a time object is considered to be true
      if and only if it isn't equal to time(0)

Instance methods:

  - isoformat()
    Return a string representing the time in ISO 8601 format,
        HH:MM:SS.mmmmmm
    or, if self.microsecond is 0
        HH:MM:SS

  - __str__()
    For a time t, str(t) is equivalent to t.isoformat().

  - strftime(format)
    Return a string representing the time, controlled by an explicit
    format string.  See the section on strftime() behavior.


class tzinfo
============
tzinfo is an abstract base clase, meaning that objects directly of this
class should not be instantiated.  You need to derive a concrete
subclass, and (at least) supply implementations of the standard tzinfo
methods needed by the datetime methods you use. The datetime module does
not supply any concrete subclasses of tzinfo.

An instance of (a concrete subclass of) tzinfo can be passed to the
constructors for datetimetz and timetz objects.  The latter objects
view their fields as being in local time, and the tzinfo object supports
methods revealing offset of local time from UTC, the name of the time
zone, and DST offset, all relative to a date or time object passed
to them.

Special requirement for pickling:  A tzinfo subclass must have an
__init__ method that can be called with no arguments, else it can
be pickled but possibly not unpickled again.  This is a technical
requirement that may be relaxed in the future.

A concrete subclass of tzinfo may need to implement the following
methods.  Exactly which methods are needed depends on the uses made
of aware datetime objects; if in doubt, simply implement all of them.
The methods are called by a datetimetz or timetz object, passing itself
as the argument.  A tzinfo subclass's methods should be prepared to
accept a dt argument of type None, timetz, or datetimetz.  If the dt
argument is not None, and dt.tzinfo is not None and not equal to self,
an exception should be raised.

  - utcoffset(dt)
    Return offset of local time from UTC, in minutes east of UTC.  If
    local time is west of UTC, this should be negative.  Note that this
    is intended to be the total offset from UTC; for example, if a
    tzinfo object represents both time zone and DST adjustments,
    utcoffset() should return their sum.  If the UTC offset isn't known,
    return None.  Else the value returned must be an int, long, or
    timedelta object, in the range -1439 to 1439 inclusive (1440 = 24*60;
    the magnitude of the offset must be less than one day, and must be
    a whole number of minutes).

  - tzname(dt)
    Return the timezone name corresponding to the datetime represented
    by dt, as a string.  Nothing about string names is defined by the
    datetime module, and there's no requirement that it mean anything
    in particular.  For example, "GMT", "UTC", "-500", "-5:00", "EDT",
    "US/Eastern", "America/New York" are all valid replies.  Return
    None if a string name isn't known.  Note that this is a method
    rather than a fixed string primarily because some tzinfo objects
    will wish to return different names depending on the specific value
    of dt passed, especially if the tzinfo class is accounting for DST.

  - dst(dt)
    Return the DST offset, in minutes east of UTC, or None if DST
    information isn't known.  Return 0 if DST is not in effect.
    If DST is in effect, return an int, long, or timedelta object, in
    the range -1439 to 1439 inclusive.  This must be a whole number of
    minutes.  Note that DST offset, if applicable, has already been added
    to the UTC offset returned by utcoffset(), so there's no need to
    consult dst() unless you're interested in displaying DST info
    separately.  For example, datetimetz.timetuple() calls its
    tzinfo object's dst() method to determine how the tm_isdst flag
    should be set.

Example tzinfo classes:

   class UTC(tzinfo):
        "UTC"
        def utcoffset(self, dt):
            return 0
        def tzname(self, dt):
            return "UTC"
        def dst(self, dt):
            return 0

    class FixedOffset(tzinfo):
        "Fixed offset in minutes east from UTC"
        def __init__(self, offset, name):
            self.__offset = offset
            self.__name = name
        def utcoffset(self, dt):
            return self.__offset
        def tzname(self, dt):
            return self.__name
        def dst(self, dt):
            # It depends on more than we know in an example.
            return None # Indicate we don't know

    import time
    class LocalTime(tzinfo):
        "Local time as defined by the operating system"
        def _isdst(self, dt):
            t = (dt.year, dt.month, dt.day, dt.hour, dt.minute, dt.second,
                 -1, -1, -1)
            # XXX This may fail for years < 1970 or >= 2038
            t = time.localtime(time.mktime(t))
            return t.tm_isdst > 0
        def utcoffset(self, dt):
            if self._isdst(dt):
                return -time.timezone/60
            else:
                return -time.altzone/60
        def tzname(self, dt):
            return time.tzname[self._isdst(dt)]


Naive and Aware Objects
=======================
Objects directly of these types are always naive:  date, datetime, and
time.

An object d of type timetz or datetimetz may be naive or aware.  d
is aware if d.tzinfo is not None, and d.tzinfo.utcoffset(d) does
not return None.  If d.tzinfo is None, or if d.tzinfo is not None
but d.tzinfo.utcoffset(d) returns None, d is naive.

The distinction between naive and aware doesn't apply to timedelta
objects.


class timetz
============
A time object represents a (local) time of day, independent of any
particular day, and subject to adjustment via a tzinfo object.

Constructor:

    time(hour=0, minute=0, second=0, microsecond=0, tzinfo=None)

    All arguments are optional.  tzinfo may be None, or an instance of
    a tzinfo subclass.  The remaining arguments may be ints or longs, in
    the following ranges:

        0 <= hour < 24
        0 <= minute < 60
        0 <= second < 60
        0 <= microsecond < 1000000

    If an argument outside those ranges is given, ValueError is raised.

Other constructors (class methods):

    None

Class attributes:

    .min
        The earliest representable time, timetz(0, 0, 0, 0).

    .max
        The latest representable time, timetz(23, 59, 59, 999999).

    .resolution
        The smallest possible difference between non-equal timetz
        objects, timedelta(microseconds=1), although note that
        arithmetic on timetz objects is not supported.

Instance attributes (read-only):

    .hour           in range(24)
    .minute         in range(60)
    .second         in range(60)
    .microsecond    in range(1000000)
    .tzinfo         the object passed as the tzinfo argument to the
                    timetz constructor, or None if none was passed.

Supported operations:

    - comparison of timetz to timetz, where timetz1 is considered
      less than timetz2 when timetz1 precedes timetz2 in time, and
      where the timetz objects are first adjusted by subtracting
      their UTC offsets (obtained from self.utcoffset()).

    - hash, use as dict key

    - pickling

    - in Boolean contexts, a timetz object is considered to be true
      if and only if, after converting it to minutes and subtracting
      self.utcoffset() (or 0 if that's None), the result is non-zero.

Instance methods:

  - isoformat()
    Return a string representing the time in ISO 8601 format,
        HH:MM:SS.mmmmmm
    or, if self.microsecond is 0
        HH:MM:SS
    If self.utcoffset() does not return None, a 6-character string is
    appended, giving the UTC offset in (signed) hours and minutes:
        HH:MM:SS.mmmmmm+HH:MM
    or, if self.microsecond is 0
        HH:MM:SS+HH:MM

  - __str__()
    For a timetz t, str(t) is equivalent to t.isoformat().

  - strftime(format)
    Return a string representing the time, controlled by an explicit
    format string.  See the section on strftime() behavior.

  - utcoffset()
    If self.tzinfo is None, returns None, else self.tzinfo.utcoffset(self),
    converted a timedelta object.

  - tzname():
    If self.tzinfo is None, returns None, else self.tzinfo.tzname(self).

  - dst()
    If self.tzinfo is None, returns None, else self.tzinfo.dst(self),
    converted to a timedelta object.


class datetimetz
================
A datetimetz object is a single object containing all the information
from a date object and a timetz object.

Constructor:

    datetimetz(year, month, day,
               hour=0, minute=0, second=0, microsecond=0, tzinfo=None)

    The year, month and day arguments are required.  tzinfo may be None,
    or an instance of a tzinfo subclass.  The remaining arguments may be
    ints or longs, in the following ranges:

        MINYEAR <= year <= MAXYEAR
        1 <= month <= 12
        1 <= day <= number of days in the given month and year
        0 <= hour < 24
        0 <= minute < 60
        0 <= second < 60
        0 <= microsecond < 1000000

    If an argument outside those ranges is given, ValueError is raised.

Other constructors (class methods):

  - today()
    utcnow()
    utcfromtimestamp(timestamp)
    fromordinal(ordinal)

    These are the same as the datetime class methods of the same names,
    except that they construct a datetimetz object, with tzinfo None.

  - now([tzinfo=None])
    fromtimestamp(timestamp[, tzinfo=None])

    These are the same as the datetime class methods of the same names,
    except that they accept an additional, optional tzinfo argument, and
    construct a datetimetz object with that tzinfo object attached.

  - combine(date, time)
    This is the same as datetime.combine(), except that it constructs
    a datetimetz object, and, if the time object is of type timetz,
    the datetimetz object has the same tzinfo object as the time object.

Class attributes:

    .min
        The earliest representable datetimetz,
        datetimetz(MINYEAR, 1, 1).

    .max
        The latest representable datetime,
        datetimetz(MAXYEAR, 12, 31, 23, 59, 59, 999999).

    .resolution
        The smallest possible difference between non-equal datetimetz
        objects, timedelta(microseconds=1).

Instance attributes (read-only):

    .year           between MINYEAR and MAXYEAR inclusive
    .month          between 1 and 12 inclusive
    .day            between 1 and the number of days in the given month
                    of the given year
    .hour           in range(24)
    .minute         in range(60)
    .second         in range(60)
    .microsecond    in range(1000000)
    .tzinfo         the object passed as the tzinfo argument to the
                    datetimetz constructor, or None if none was passed.

Supported operations:

    - datetimetz1 + timedelta -> datetimetz2
      timedelta + datetimetz1 -> datetimetz2
      The same as addition of datetime objects, except that
      datetimetz2.tzinfo is set to datetimetz1.tzinfo.

    - datetimetz1 - timedelta -> datetimetz2
      The same as addition of datetime objects, except that
      datetimetz2.tzinfo is set to datetimetz1.tzinfo.

    - aware_datetimetz1 - aware_datetimetz2 -> timedelta
      naive_datetimetz1 - naive_datetimetz2 -> timedelta
      naive_datetimetz1 - datetime2 -> timedelta
      datetime1 - naive_datetimetz2 -> timedelta

      Subtraction of a datetime or datetimetz, from a datetime or
      datetimetz, is defined only if both operands are naive, or if
      both are aware.  If one is aware and the other is naive, TypeError
      is raised.

      If both are naive, subtraction acts as for datetime subtraction.

      If both are aware datetimetz objects, a-b acts as if a and b were
      first converted to UTC datetimes (by subtracting a.utcoffset()
      minutes from a, and b.utcoffset() minutes from b), and then doing
      datetime subtraction, except that the implementation never
      overflows.

    - Comparison of datetimetz to datetime or datetimetz.  As for
      subtraction, comparison is defined only if both operands are
      naive or both are aware.  If both are naive, comparison is as
      for datetime objects with the same date and time components.
      If both are aware, comparison acts as if both were converted to
      UTC datetimes first, except the the implementation never
      overflows.  If one comparand is naive and the other aware,
      TypeError is raised.

    - hash, use as dict key

    - efficient pickling

    - in Boolean contexts, all datetimetz objects are considered to be
      true

Instance methods:

  - date()
    time()
    toordinal()
    weekday()
    isoweekday()
    isocalendar()
    ctime()
    __str__()
    strftime(format)

    These are the same as the datetime methods of the same names.

  - timetz()
    Return timetz object with same hour, minute, second, microsecond,
    and tzinfo.

  - utcoffset()
    If self.tzinfo is None, returns None, else self.tzinfo.utcoffset(self).
    converted to a timedelta object.

  - tzname():
    If self.tzinfo is None, returns None, else self.tzinfo.tzname(self).

  - dst()
    If self.tzinfo is None, returns None, else self.tzinfo.dst(self),
    converted to a timedelta object.

  - timetuple()
    Like datetime.timetuple(), but sets the tm_isdst flag according to
    the dst() method:  if self.dst() returns None, tm_isdst is set to -1;
    else if self.dst() returns a non-zero value, tm_isdst is set to 1;
    else tm_isdst is set to 0.

  - utctimetuple()
    If datetimetz d is naive, this is the same as d.timetuple() except
    that tm_isdst is forced to 0 regardless of what d.dst() returns.
    DST is never in effect for a UTC time.

    If d is aware, d is normalized to UTC time, by subtracting
    d.utcoffset() minutes, and a timetuple for the normalized time is
    returned.  tm_isdst is forced to 0.  Note that the result's
    tm_year field may be MINYEAR-1 or MAXYEAR+1, if d.year was MINYEAR
    or MAXYEAR and UTC adjustment spills over a year boundary.

  - isoformat(sep='T')
    Return a string representing the date and time in ISO 8601 format,
        YYYY-MM-DDTHH:MM:SS.mmmmmm
    or, if self.microsecond is 0,
        YYYY-MM-DDTHH:MM:SS

    If self.utcoffset() does not return None, a 6-character string is
    appended, giving the UTC offset in (signed) hours and minutes:
        YYYY-MM-DDTHH:MM:SS.mmmmmm+HH:MM
    or, if self.microsecond is 0
        YYYY-MM-DDTHH:MM:SS+HH:MM

    Optional argument sep (default 'T') is a one-character separator,
    placed between the date and time portions of the result.  For example,

        >>> from datetime import *
        >>> class TZ(tzinfo):
        ...     def utcoffset(self, dt): return -399
        ...
        >>> datetimetz(2002, 12, 25, tzinfo=TZ()).isoformat(' ')
        '2002-12-25 00:00:00-06:39'
        >>>

    str(d) is equivalent to d.isoformat(' ').


C API
=====
Struct typedefs:

    PyDateTime_Date
    PyDateTime_DateTime
    PyDateTime_DateTimeTZ
    PyDateTime_Time
    PyDateTime_TimeTZ
    PyDateTime_Delta
    PyDateTime_TZInfo

Type-check macros:

    PyDate_Check(op)
    PyDate_CheckExact(op)

    PyDateTime_Check(op)
    PyDateTime_CheckExact(op)

    PyDateTimeTZ_Check(op)
    PyDateTimeTZ_CheckExact(op)

    PyTime_Check(op)
    PyTime_CheckExact(op)

    PyTimeTZ_Check(op)
    PyTimeTZ_CheckExact(op)

    PyDelta_Check(op)
    PyDelta_CheckExact(op)

    PyTZInfo_Check(op)
    PyTZInfo_CheckExact(op

Accessor macros:

All objects are immutable, so accessors are read-only.  All macros
return ints:

    For date, datetime, and datetimetz instances:
        PyDateTime_GET_YEAR(o)
        PyDateTime_GET_MONTH(o)
        PyDateTime_GET_DAY(o)

    For datetime and datetimetz instances:
        PyDateTime_DATE_GET_HOUR(o)
        PyDateTime_DATE_GET_MINUTE(o)
        PyDateTime_DATE_GET_SECOND(o)
        PyDateTime_DATE_GET_MICROSECOND(o)

    For time and timetz instances:
        PyDateTime_TIME_GET_HOUR(o)
        PyDateTime_TIME_GET_MINUTE(o)
        PyDateTime_TIME_GET_SECOND(o)
        PyDateTime_TIME_GET_MICROSECOND(o)