Changing to a different date parsing module.

This commit is contained in:
Jeremy Fincher 2004-09-30 07:40:45 +00:00
parent 6878510d7e
commit 27fe6554c7
5 changed files with 2190 additions and 229 deletions

View File

@ -0,0 +1,8 @@
"""
Copyright (c) 2003 Gustavo Niemeyer <niemeyer@conectiva.com>
This module offers extensions to the standard python 2.3+
datetime module.
"""
__author__ = "Gustavo Niemeyer <niemeyer@conectiva.com>"
__license__ = "PSF License"

867
others/dateutil/parser.py Normal file
View File

@ -0,0 +1,867 @@
# -*- coding:iso-8859-1 -*-
"""
Copyright (c) 2003 Gustavo Niemeyer <niemeyer@conectiva.com>
This module offers extensions to the standard python 2.3+
datetime module.
"""
__author__ = "Gustavo Niemeyer <niemeyer@conectiva.com>"
__license__ = "PSF License"
import os.path
import string
import sys
import time
import datetime
import relativedelta
import tz
__all__ = ["parse", "parserinfo"]
# Some pointers:
#
# http://www.cl.cam.ac.uk/~mgk25/iso-time.html
# http://www.iso.ch/iso/en/prods-services/popstds/datesandtime.html
# http://www.w3.org/TR/NOTE-datetime
# http://ringmaster.arc.nasa.gov/tools/time_formats.html
# http://search.cpan.org/author/MUIR/Time-modules-2003.0211/lib/Time/ParseDate.pm
# http://stein.cshl.org/jade/distrib/docs/java.text.SimpleDateFormat.html
try:
from cStringIO import StringIO
except ImportError:
from StringIO import StringIO
class _timelex:
def __init__(self, instream):
if isinstance(instream, basestring):
instream = StringIO(instream)
self.instream = instream
self.wordchars = ('abcdfeghijklmnopqrstuvwxyz'
'ABCDEFGHIJKLMNOPQRSTUVWXYZ_'
'ßàáâãäåæçèéêëìíîïðñòóôõöøùúûüýþÿ'
'ÀÁÂÃÄÅÆÇÈÉÊËÌÍÎÏÐÑÒÓÔÕÖØÙÚÛÜÝÞ')
self.numchars = '0123456789'
self.whitespace = ' \t\r\n'
self.charstack = []
self.tokenstack = []
self.eof = False
def get_token(self):
if self.tokenstack:
return self.tokenstack.pop(0)
seenletters = False
token = None
state = None
wordchars = self.wordchars
numchars = self.numchars
whitespace = self.whitespace
while not self.eof:
if self.charstack:
nextchar = self.charstack.pop(0)
else:
nextchar = self.instream.read(1)
if not nextchar:
self.eof = True
break
elif not state:
token = nextchar
if nextchar in wordchars:
state = 'a'
elif nextchar in numchars:
state = '0'
elif nextchar in whitespace:
token = ' '
break # emit token
else:
break # emit token
elif state == 'a':
seenletters = True
if nextchar in wordchars:
token += nextchar
elif nextchar == '.':
token += nextchar
state = 'a.'
else:
self.charstack.append(nextchar)
break # emit token
elif state == '0':
if nextchar in numchars:
token += nextchar
elif nextchar == '.':
token += nextchar
state = '0.'
else:
self.charstack.append(nextchar)
break # emit token
elif state == 'a.':
seenletters = True
if nextchar == '.' or nextchar in wordchars:
token += nextchar
elif nextchar in numchars and token[-1] == '.':
token += nextchar
state = '0.'
else:
self.charstack.append(nextchar)
break # emit token
elif state == '0.':
if nextchar == '.' or nextchar in numchars:
token += nextchar
elif nextchar in wordchars and token[-1] == '.':
token += nextchar
state = 'a.'
else:
self.charstack.append(nextchar)
break # emit token
if (state in ('a.', '0.') and
(seenletters or token.count('.') > 1 or token[-1] == '.')):
l = token.split('.')
token = l[0]
for tok in l[1:]:
self.tokenstack.append('.')
if tok:
self.tokenstack.append(tok)
return token
def __iter__(self):
return self
def next(self):
token = self.get_token()
if token is None:
raise StopIteration
return token
def split(cls, s):
return list(cls(s))
split = classmethod(split)
class _resultbase(object):
def __init__(self):
for attr in self.__slots__:
setattr(self, attr, None)
def _repr(self, classname):
l = []
for attr in self.__slots__:
value = getattr(self, attr)
if value is not None:
l.append("%s=%s" % (attr, `value`))
return "%s(%s)" % (classname, ", ".join(l))
def __repr__(self):
return self._repr(self.__class__.__name__)
class parserinfo:
# m from a.m/p.m, t from ISO T separator
JUMP = [" ", ".", ",", ";", "-", "/", "'",
"at", "on", "and", "ad", "m", "t", "of",
"st", "nd", "rd", "th"]
WEEKDAYS = [("Mon", "Monday"),
("Tue", "Tuesday"),
("Wed", "Wednesday"),
("Thu", "Thursday"),
("Fri", "Friday"),
("Sat", "Saturday"),
("Sun", "Sunday")]
MONTHS = [("Jan", "January"),
("Feb", "February"),
("Mar", "March"),
("Apr", "April"),
("May", "May"),
("Jun", "June"),
("Jul", "July"),
("Aug", "August"),
("Sep", "September"),
("Oct", "October"),
("Nov", "November"),
("Dec", "December")]
HMS = [("h", "hour", "hours"),
("m", "minute", "minutes"),
("s", "second", "seconds")]
AMPM = [("am", "a"),
("pm", "p")]
UTCZONE = ["UTC", "GMT", "Z"]
PERTAIN = ["of"]
TZOFFSET = {}
def __init__(self, dayfirst=False, yearfirst=False):
self._jump = self._convert(self.JUMP)
self._weekdays = self._convert(self.WEEKDAYS)
self._months = self._convert(self.MONTHS)
self._hms = self._convert(self.HMS)
self._ampm = self._convert(self.AMPM)
self._utczone = self._convert(self.UTCZONE)
self._pertain = self._convert(self.PERTAIN)
self.dayfirst = dayfirst
self.yearfirst = yearfirst
self._year = time.localtime().tm_year
self._century = self._year/100*100
def _convert(self, lst):
dct = {}
for i in range(len(lst)):
v = lst[i]
if isinstance(v, tuple):
for v in v:
dct[v.lower()] = i
else:
dct[v.lower()] = i
return dct
def jump(self, name):
return name.lower() in self._jump
def weekday(self, name):
if len(name) >= 3:
try:
return self._weekdays[name.lower()]
except KeyError:
pass
return None
def month(self, name):
if len(name) >= 3:
try:
return self._months[name.lower()]+1
except KeyError:
pass
return None
def hms(self, name):
try:
return self._hms[name.lower()]
except KeyError:
return None
def ampm(self, name):
try:
return self._ampm[name.lower()]
except KeyError:
return None
def pertain(self, name):
return name.lower() in self._pertain
def utczone(self, name):
return name.lower() in self._utczone
def tzoffset(self, name):
if name in self._utczone:
return 0
return self.TZOFFSET.get(name)
def convertyear(self, year):
if year < 100:
year += self._century
if abs(year-self._year) >= 50:
if year < self._year:
year += 100
else:
year -= 100
return year
def validate(self, res):
# move to info
if res.year is not None:
res.year = self.convertyear(res.year)
if res.tzoffset == 0 and not res.tzname or res.tzname == 'Z':
res.tzname = "UTC"
res.tzoffset = 0
elif res.tzoffset != 0 and res.tzname and self.utczone(res.tzname):
res.tzoffset = 0
return True
class parser:
def __init__(self, parserinfo=parserinfo):
self.info = parserinfo()
def parse(self, timestr, default=None,
ignoretz=False, tzinfos=None,
**kwargs):
if not default:
default = datetime.datetime.now().replace(hour=0, minute=0,
second=0, microsecond=0)
res = self._parse(timestr, **kwargs)
if res is None:
raise ValueError, "unknown string format"
repl = {}
for attr in ["year", "month", "day", "hour",
"minute", "second", "microsecond"]:
value = getattr(res, attr)
if value is not None:
repl[attr] = value
ret = default.replace(**repl)
if res.weekday is not None and not res.day:
ret = ret+relativedelta.relativedelta(weekday=res.weekday)
if not ignoretz:
if callable(tzinfos) or tzinfos and res.tzname in tzinfos:
if callable(tzinfos):
tzdata = tzinfos(res.tzname, res.tzoffset)
else:
tzdata = tzinfos.get(res.tzname)
if isinstance(tzdata, datetime.tzinfo):
tzinfo = tzdata
elif isinstance(tzdata, basestring):
tzinfo = tz.tzstr(tzdata)
elif isinstance(tzdata, int):
tzinfo = tz.tzoffset(res.tzname, tzdata)
else:
raise ValueError, "offset must be tzinfo subclass, " \
"tz string, or int offset"
ret = ret.replace(tzinfo=tzinfo)
elif res.tzname and res.tzname in time.tzname:
ret = ret.replace(tzinfo=tz.tzlocal())
elif res.tzoffset == 0:
ret = ret.replace(tzinfo=tz.tzutc())
elif res.tzoffset:
ret = ret.replace(tzinfo=tz.tzoffset(res.tzname, res.tzoffset))
return ret
class _result(_resultbase):
__slots__ = ["year", "month", "day", "weekday",
"hour", "minute", "second", "microsecond",
"tzname", "tzoffset"]
def _parse(self, timestr, dayfirst=None, yearfirst=None, fuzzy=False):
info = self.info
if dayfirst is None:
dayfirst = info.dayfirst
if yearfirst is None:
yearfirst = info.yearfirst
res = self._result()
l = _timelex.split(timestr)
try:
# year/month/day list
ymd = []
# Index of the month string in ymd
mstridx = -1
len_l = len(l)
i = 0
while i < len_l:
# Check if it's a number
try:
value = float(l[i])
except ValueError:
value = None
if value is not None:
# Token is a number
len_li = len(l[i])
i += 1
if (len(ymd) == 3 and len_li in (2, 4)
and (i >= len_l or l[i] != ':')):
# 19990101T23[59]
s = l[i-1]
res.hour = int(s[:2])
if len_li == 4:
res.minute = int(s[2:])
elif len_li == 6 or (len_li > 6 and l[i-1].find('.') == 6):
# YYMMDD or HHMMSS[.ss]
s = l[i-1]
if not ymd and l[i-1].find('.') == -1:
ymd.append(info.convertyear(int(s[:2])))
ymd.append(int(s[2:4]))
ymd.append(int(s[4:]))
else:
# 19990101T235959[.59]
res.hour = int(s[:2])
res.minute = int(s[2:4])
value = float(s[4:])
res.second = int(value)
if value%1:
res.microsecond = int(1000000*(value%1))
elif len_li == 8:
# YYYYMMDD
s = l[i-1]
ymd.append(int(s[:4]))
ymd.append(int(s[4:6]))
ymd.append(int(s[6:]))
elif len_li in (12, 14):
# YYYYMMDDhhmm[ss]
s = l[i-1]
ymd.append(int(s[:4]))
ymd.append(int(s[4:6]))
ymd.append(int(s[6:8]))
res.hour = int(s[8:10])
res.minute = int(s[10:12])
if len_li == 14:
res.second = int(s[12:])
elif ((i < len_l and info.hms(l[i]) is not None) or
(i+1 < len_l and l[i] == ' ' and
info.hms(l[i+1]) is not None)):
# HH[ ]h or MM[ ]m or SS[.ss][ ]s
if l[i] == ' ':
i += 1
idx = info.hms(l[i])
while True:
if idx == 0:
res.hour = int(value)
if value%1:
res.minute = int(60*(value%1))
elif idx == 1:
res.minute = int(value)
if value%1:
res.second = int(60*(value%1))
elif idx == 2:
res.second = int(value)
if value%1:
res.microsecond = int(1000000*(value%1))
i += 1
if i >= len_l or idx == 2:
break
# 12h00
try:
value = float(l[i])
except ValueError:
break
else:
i += 1
idx += 1
if i < len_l:
newidx = info.hms(l[i])
if newidx is not None:
idx = newidx
elif i+1 < len_l and l[i] == ':':
# HH:MM[:SS[.ss]]
res.hour = int(value)
i += 1
value = float(l[i])
res.minute = int(value)
if value%1:
res.second = int(60*(value%1))
i += 1
if i < len_l and l[i] == ':':
value = float(l[i+1])
res.second = int(value)
if value%1:
res.microsecond = int(1000000*(value%1))
i += 2
elif i < len_l and l[i] in ('-', '/', '.'):
sep = l[i]
ymd.append(int(value))
i += 1
if i < len_l and not info.jump(l[i]):
try:
# 01-01[-01]
ymd.append(int(l[i]))
except ValueError:
# 01-Jan[-01]
value = info.month(l[i])
if value is not None:
ymd.append(value)
assert mstridx == -1
mstridx = len(ymd)-1
else:
return None
i += 1
if i < len_l and l[i] == sep:
# We have three members
i += 1
value = info.month(l[i])
if value is not None:
ymd.append(value)
mstridx = len(ymd)-1
assert mstridx == -1
else:
ymd.append(int(l[i]))
i += 1
elif i >= len_l or info.jump(l[i]):
if i+1 < len_l and info.ampm(l[i+1]) is not None:
# 12 am
res.hour = int(value)
if res.hour < 12 and info.ampm(l[i+1]) == 1:
res.hour += 12
elif res.hour == 12 and info.ampm(l[i+1]) == 0:
res.hour = 0
i += 1
else:
# Year, month or day
ymd.append(int(value))
i += 1
elif info.ampm(l[i]) is not None:
# 12am
res.hour = int(value)
if res.hour < 12 and info.ampm(l[i]) == 1:
res.hour += 12
elif res.hour == 12 and info.ampm(l[i]) == 0:
res.hour = 0
i += 1
elif not fuzzy:
return None
else:
i += 1
continue
# Check weekday
value = info.weekday(l[i])
if value is not None:
res.weekday = value
i += 1
continue
# Check month name
value = info.month(l[i])
if value is not None:
ymd.append(value)
assert mstridx == -1
mstridx = len(ymd)-1
i += 1
if i < len_l:
if l[i] in ('-', '/'):
# Jan-01[-99]
sep = l[i]
i += 1
ymd.append(int(l[i]))
i += 1
if i < len_l and l[i] == sep:
# Jan-01-99
i += 1
ymd.append(int(l[i]))
i += 1
elif (i+3 < len_l and l[i] == l[i+2] == ' '
and info.pertain(l[i+1])):
# Jan of 01
# In this case, 01 is clearly year
try:
value = int(l[i+3])
except ValueError:
# Wrong guess
pass
else:
# Convert it here to become unambiguous
ymd.append(info.convertyear(value))
i += 4
continue
# Check am/pm
value = info.ampm(l[i])
if value is not None:
if value == 1 and res.hour < 12:
res.hour += 12
elif value == 0 and res.hour == 12:
res.hour = 0
i += 1
continue
# Check for a timezone name
if (res.hour is not None and len(l[i]) <= 5 and
res.tzname is None and res.tzoffset is None and
not [x for x in l[i] if x not in string.ascii_uppercase]):
res.tzname = l[i]
res.tzoffset = info.tzoffset(res.tzname)
i += 1
# Check for something like GMT+3, or BRST+3. Notice
# that it doesn't mean "I am 3 hours after GMT", but
# "my time +3 is GMT". If found, we reverse the
# logic so that timezone parsing code will get it
# right.
if i < len_l and l[i] in ('+', '-'):
l[i] = ('+', '-')[l[i] == '+']
res.tzoffset = None
if info.utczone(res.tzname):
# With something like GMT+3, the timezone
# is *not* GMT.
res.tzname = None
continue
# Check for a numbered timezone
if res.hour is not None and l[i] in ('+', '-'):
signal = (-1,1)[l[i] == '+']
i += 1
len_li = len(l[i])
if len_li == 4:
# -0300
res.tzoffset = int(l[i][:2])*3600+int(l[i][2:])*60
elif i+1 < len_l and l[i+1] == ':':
# -03:00
res.tzoffset = int(l[i])*3600+int(l[i+2])*60
i += 2
elif len_li <= 2:
# -[0]3
res.tzoffset = int(l[i][:2])*3600
else:
return None
i += 1
res.tzoffset *= signal
# Look for a timezone name between parenthesis
if (i+3 < len_l and
info.jump(l[i]) and l[i+1] == '(' and l[i+3] == ')' and
3 <= len(l[i+2]) <= 5 and
not [x for x in l[i+2]
if x not in string.ascii_uppercase]):
# -0300 (BRST)
res.tzname = l[i+2]
i += 4
continue
# Check jumps
if not (info.jump(l[i]) or fuzzy):
return None
i += 1
# Process year/month/day
len_ymd = len(ymd)
if len_ymd > 3:
# More than three members!?
return None
elif len_ymd == 1 or (mstridx != -1 and len_ymd == 2):
# One member, or two members with a month string
if mstridx != -1:
res.month = ymd[mstridx]
del ymd[mstridx]
if len_ymd > 1 or mstridx == -1:
if ymd[0] > 31:
res.year = ymd[0]
else:
res.day = ymd[0]
elif len_ymd == 2:
# Two members with numbers
if ymd[0] > 31:
# 99-01
res.year, res.month = ymd
elif ymd[1] > 31:
# 01-99
res.month, res.year = ymd
elif dayfirst and ymd[1] <= 12:
# 13-01
res.day, res.month = ymd
else:
# 01-13
res.month, res.day = ymd
if len_ymd == 3:
# Three members
if mstridx == 0:
res.month, res.day, res.year = ymd
elif mstridx == 1:
if ymd[0] > 31 or (yearfirst and ymd[2] <= 31):
# 99-Jan-01
res.year, res.month, res.day = ymd
else:
# 01-Jan-01
# Give precendence to day-first, since
# two-digit years is usually hand-written.
res.day, res.month, res.year = ymd
elif mstridx == 2:
# WTF!?
if ymd[1] > 31:
# 01-99-Jan
res.day, res.year, res.month = ymd
else:
# 99-01-Jan
res.year, res.day, res.month = ymd
else:
if ymd[0] > 31 or \
(yearfirst and ymd[1] <= 12 and ymd[2] <= 31):
# 99-01-01
res.year, res.month, res.day = ymd
elif ymd[0] > 12 or (dayfirst and ymd[1] <= 12):
# 13-01-01
res.day, res.month, res.year = ymd
else:
# 01-13-01
res.month, res.day, res.year = ymd
except (IndexError, ValueError, AssertionError):
return None
if not info.validate(res):
return None
return res
DEFAULTPARSER = parser()
def parse(timestr, parserinfo=None, **kwargs):
if parserinfo:
return parser(parserinfo).parse(timestr, **kwargs)
else:
return DEFAULTPARSER.parse(timestr, **kwargs)
class _tzparser:
class _result(_resultbase):
__slots__ = ["stdabbr", "stdoffset", "dstabbr", "dstoffset",
"start", "end"]
class _attr(_resultbase):
__slots__ = ["month", "week", "weekday",
"yday", "jyday", "day", "time"]
def __repr__(self):
return self._repr("")
def __init__(self):
_resultbase.__init__(self)
self.start = self._attr()
self.end = self._attr()
def parse(self, tzstr):
res = self._result()
l = _timelex.split(tzstr)
try:
len_l = len(l)
i = 0
while i < len_l:
# BRST+3[BRDT[+2]]
j = i
while j < len_l and not [x for x in l[j]
if x in "0123456789:,-+"]:
j += 1
if j != i:
if not res.stdabbr:
offattr = "stdoffset"
res.stdabbr = "".join(l[i:j])
else:
offattr = "dstoffset"
res.dstabbr = "".join(l[i:j])
i = j
if (i < len_l and
(l[i] in ('+', '-') or l[i][0] in "0123456789")):
if l[i] in ('+', '-'):
signal = (1,-1)[l[i] == '+']
i += 1
else:
signal = -1
len_li = len(l[i])
if len_li == 4:
# -0300
setattr(res, offattr,
(int(l[i][:2])*3600+int(l[i][2:])*60)*signal)
elif i+1 < len_l and l[i+1] == ':':
# -03:00
setattr(res, offattr,
(int(l[i])*3600+int(l[i+2])*60)*signal)
i += 2
elif len_li <= 2:
# -[0]3
setattr(res, offattr,
int(l[i][:2])*3600*signal)
else:
return None
i += 1
if res.dstabbr:
break
else:
break
if i < len_l:
for j in range(i, len_l):
if l[j] == ';': l[j] = ','
assert l[i] == ','
i += 1
if i >= len_l:
pass
elif (8 <= l.count(',') <= 9 and
not [y for x in l[i:] if x != ','
for y in x if y not in "0123456789"]):
# GMT0BST,3,0,30,3600,10,0,26,7200[,3600]
for x in (res.start, res.end):
x.month = int(l[i])
i += 2
if l[i] == '-':
value = int(l[i+1])*-1
i += 1
else:
value = int(l[i])
i += 2
if value:
x.week = value
x.weekday = (int(l[i])-1)%7
else:
x.day = int(l[i])
i += 2
x.time = int(l[i])
i += 2
if i < len_l:
if l[i] in ('-','+'):
signal = (-1,1)[l[i] == "+"]
i += 1
else:
signal = 1
res.dstoffset = (res.stdoffset+int(l[i]))*signal
elif (l.count(',') == 2 and l[i:].count('/') <= 2 and
not [y for x in l[i:] if x not in (',','/','J','M',
'.','-',':')
for y in x if y not in "0123456789"]):
for x in (res.start, res.end):
if l[i] == 'J':
# non-leap year day (1 based)
i += 1
x.jyday = int(l[i])
elif l[i] == 'M':
# month[-.]week[-.]weekday
i += 1
x.month = int(l[i])
i += 1
assert l[i] in ('-', '.')
i += 1
x.week = int(l[i])
if x.week == 5:
x.week = -1
i += 1
assert l[i] in ('-', '.')
i += 1
x.weekday = (int(l[i])-1)%7
else:
# year day (zero based)
x.yday = int(l[i])+1
i += 1
if i < len_l and l[i] == '/':
i += 1
# start time
len_li = len(l[i])
if len_li == 4:
# -0300
x.time = (int(l[i][:2])*3600+int(l[i][2:])*60)
elif i+1 < len_l and l[i+1] == ':':
# -03:00
x.time = int(l[i])*3600+int(l[i+2])*60
i += 2
if i+1 < len_l and l[i+1] == ':':
i += 2
x.time += int(l[i])
elif len_li <= 2:
# -[0]3
x.time = (int(l[i][:2])*3600)
else:
return None
i += 1
assert i == len_l or l[i] == ','
i += 1
assert i >= len_l
except (IndexError, ValueError, AssertionError):
return None
return res
DEFAULTTZPARSER = _tzparser()
def _parsetz(tzstr):
return DEFAULTTZPARSER.parse(tzstr)
# vim:ts=4:sw=4:et

View File

@ -0,0 +1,432 @@
"""
Copyright (c) 2003 Gustavo Niemeyer <niemeyer@conectiva.com>
This module offers extensions to the standard python 2.3+
datetime module.
"""
__author__ = "Gustavo Niemeyer <niemeyer@conectiva.com>"
__license__ = "PSF License"
import datetime
import calendar
__all__ = ["relativedelta", "MO", "TU", "WE", "TH", "FR", "SA", "SU"]
class weekday(object):
__slots__ = ["weekday", "n"]
def __init__(self, weekday, n=0):
self.weekday = weekday
self.n = n
def __call__(self, n):
if n == self.n:
return self
else:
return self.__class__(self.weekday, n)
def __eq__(self, other):
try:
if self.weekday != other.weekday or self.n != other.n:
return False
except AttributeError:
return False
return True
def __repr__(self):
s = ("MO", "TU", "WE", "TH", "FR", "SA", "SU")[self.weekday]
if not self.n:
return s
else:
return "%s(%+d)" % (s, self.n)
MO, TU, WE, TH, FR, SA, SU = weekdays = tuple([weekday(x) for x in range(7)])
class relativedelta:
"""
The relativedelta type is based on the specification of the excelent
work done by M.-A. Lemburg in his mx.DateTime extension. However,
notice that this type does *NOT* implement the same algorithm as
his work. Do *NOT* expect it to behave like mx.DateTime's counterpart.
There's two different ways to build a relativedelta instance. The
first one is passing it two date/datetime classes:
relativedelta(datetime1, datetime2)
And the other way is to use the following keyword arguments:
year, month, day, hour, minute, second, microsecond:
Absolute information.
years, months, weeks, days, hours, minutes, seconds, microseconds:
Relative information, may be negative.
weekday:
One of the weekday instances (MO, TU, etc). These instances may
receive a parameter N, specifying the Nth weekday, which could
be positive or negative (like MO(+1) or MO(-2). Not specifying
it is the same as specifying +1. You can also use an integer,
where 0=MO.
leapdays:
Will add given days to the date found, if year is a leap
year, and the date found is post 28 of february.
yearday, nlyearday:
Set the yearday or the non-leap year day (jump leap days).
These are converted to day/month/leapdays information.
Here is the behavior of operations with relativedelta:
1) Calculate the absolute year, using the 'year' argument, or the
original datetime year, if the argument is not present.
2) Add the relative 'years' argument to the absolute year.
3) Do steps 1 and 2 for month/months.
4) Calculate the absolute day, using the 'day' argument, or the
original datetime day, if the argument is not present. Then,
subtract from the day until it fits in the year and month
found after their operations.
5) Add the relative 'days' argument to the absolute day. Notice
that the 'weeks' argument is multiplied by 7 and added to
'days'.
6) Do steps 1 and 2 for hour/hours, minute/minutes, second/seconds,
microsecond/microseconds.
7) If the 'weekday' argument is present, calculate the weekday,
with the given (wday, nth) tuple. wday is the index of the
weekday (0-6, 0=Mon), and nth is the number of weeks to add
forward or backward, depending on its signal. Notice that if
the calculated date is already Monday, for example, using
(0, 1) or (0, -1) won't change the day.
"""
def __init__(self, dt1=None, dt2=None,
years=0, months=0, days=0, leapdays=0, weeks=0,
hours=0, minutes=0, seconds=0, microseconds=0,
year=None, month=None, day=None, weekday=None,
yearday=None, nlyearday=None,
hour=None, minute=None, second=None, microsecond=None):
if dt1 and dt2:
if not isinstance(dt1, datetime.date) or \
not isinstance(dt2, datetime.date):
raise TypeError, "relativedelta only diffs datetime/date"
if type(dt1) is not type(dt2):
if not isinstance(dt1, datetime.datetime):
dt1 = datetime.datetime.fromordinal(dt1.toordinal())
elif not isinstance(dt2, datetime.datetime):
dt2 = datetime.datetime.fromordinal(dt2.toordinal())
self.years = 0
self.months = 0
self.days = 0
self.leapdays = 0
self.hours = 0
self.minutes = 0
self.seconds = 0
self.microseconds = 0
self.year = None
self.month = None
self.day = None
self.weekday = None
self.hour = None
self.minute = None
self.second = None
self.microsecond = None
self._has_time = 0
months = (dt1.year*12+dt1.month)-(dt2.year*12+dt2.month)
self._set_months(months)
dtm = self.__radd__(dt2)
if dt1 < dt2:
while dt1 > dtm:
months += 1
self._set_months(months)
dtm = self.__radd__(dt2)
else:
while dt1 < dtm:
months -= 1
self._set_months(months)
dtm = self.__radd__(dt2)
delta = dt1 - dtm
self.seconds = delta.seconds+delta.days*86400
self.microseconds = delta.microseconds
else:
self.years = years
self.months = months
self.days = days+weeks*7
self.leapdays = leapdays
self.hours = hours
self.minutes = minutes
self.seconds = seconds
self.microseconds = microseconds
self.year = year
self.month = month
self.day = day
self.hour = hour
self.minute = minute
self.second = second
self.microsecond = microsecond
if type(weekday) is int:
self.weekday = weekdays[weekday]
else:
self.weekday = weekday
yday = 0
if nlyearday:
yday = nlyearday
elif yearday:
yday = yearday
if yearday > 59:
self.leapdays = -1
if yday:
ydayidx = [31,59,90,120,151,181,212,243,273,304,334,366]
for idx, ydays in enumerate(ydayidx):
if yday <= ydays:
self.month = idx+1
if idx == 0:
self.day = ydays
else:
self.day = yday-ydayidx[idx-1]
break
else:
raise ValueError, "invalid year day (%d)" % yday
self._fix()
def _fix(self):
if abs(self.microseconds) > 999999:
s = self.microseconds/abs(self.microseconds)
div, mod = divmod(self.microseconds*s, 1000000)
self.microseconds = mod*s
self.seconds += div*s
if abs(self.seconds) > 59:
s = self.seconds/abs(self.seconds)
div, mod = divmod(self.seconds*s, 60)
self.seconds = mod*s
self.minutes += div*s
if abs(self.minutes) > 59:
s = self.minutes/abs(self.minutes)
div, mod = divmod(self.minutes*s, 60)
self.minutes = mod*s
self.hours += div*s
if abs(self.hours) > 23:
s = self.hours/abs(self.hours)
div, mod = divmod(self.hours*s, 24)
self.hours = mod*s
self.days += div*s
if abs(self.months) > 11:
s = self.months/abs(self.months)
div, mod = divmod(self.months*s, 12)
self.months = mod*s
self.years += div*s
if (self.hours or self.minutes or self.seconds or self.microseconds or
self.hour is not None or self.minute is not None or
self.second is not None or self.microsecond is not None):
self._has_time = 1
else:
self._has_time = 0
def _set_months(self, months):
self.months = months
if abs(self.months) > 11:
s = self.months/abs(self.months)
div, mod = divmod(self.months*s, 12)
self.months = mod*s
self.years = div*s
else:
self.years = 0
def __radd__(self, other):
if not isinstance(other, datetime.date):
raise TypeError, "unsupported type for add operation"
elif self._has_time and not isinstance(other, datetime.datetime):
other = datetime.datetime.fromordinal(other.toordinal())
year = (self.year or other.year)+self.years
month = self.month or other.month
if self.months:
assert 1 <= abs(self.months) <= 12
month += self.months
if month > 12:
year += 1
month -= 12
elif month < 1:
year -= 1
month += 12
day = min(calendar.monthrange(year, month)[1],
self.day or other.day)
repl = {"year": year, "month": month, "day": day}
for attr in ["hour", "minute", "second", "microsecond"]:
value = getattr(self, attr)
if value is not None:
repl[attr] = value
days = self.days
if self.leapdays and month > 2 and calendar.isleap(year):
days += self.leapdays
ret = (other.replace(**repl)
+ datetime.timedelta(days=days,
hours=self.hours,
minutes=self.minutes,
seconds=self.seconds,
microseconds=self.microseconds))
if self.weekday:
weekday, nth = self.weekday.weekday, self.weekday.n or 1
jumpdays = (abs(nth)-1)*7
if nth > 0:
jumpdays += (7-ret.weekday()+weekday)%7
else:
jumpdays += (ret.weekday()-weekday)%7
jumpdays *= -1
ret += datetime.timedelta(days=jumpdays)
return ret
def __rsub__(self, other):
return self.__neg__().__radd__(other)
def __add__(self, other):
if not isinstance(other, relativedelta):
raise TypeError, "unsupported type for add operation"
return relativedelta(years=other.years+self.years,
months=other.months+self.months,
days=other.days+self.days,
hours=other.hours+self.hours,
minutes=other.minutes+self.minutes,
seconds=other.seconds+self.seconds,
microseconds=other.microseconds+self.microseconds,
leapdays=other.leapdays or self.leapdays,
year=other.year or self.year,
month=other.month or self.month,
day=other.day or self.day,
weekday=other.weekday or self.weekday,
hour=other.hour or self.hour,
minute=other.minute or self.minute,
second=other.second or self.second,
microsecond=other.second or self.microsecond)
def __sub__(self, other):
if not isinstance(other, relativedelta):
raise TypeError, "unsupported type for sub operation"
return relativedelta(years=other.years-self.years,
months=other.months-self.months,
days=other.days-self.days,
hours=other.hours-self.hours,
minutes=other.minutes-self.minutes,
seconds=other.seconds-self.seconds,
microseconds=other.microseconds-self.microseconds,
leapdays=other.leapdays or self.leapdays,
year=other.year or self.year,
month=other.month or self.month,
day=other.day or self.day,
weekday=other.weekday or self.weekday,
hour=other.hour or self.hour,
minute=other.minute or self.minute,
second=other.second or self.second,
microsecond=other.second or self.microsecond)
def __neg__(self):
return relativedelta(years=-self.years,
months=-self.months,
days=-self.days,
hours=-self.hours,
minutes=-self.minutes,
seconds=-self.seconds,
microseconds=-self.microseconds,
leapdays=self.leapdays,
year=self.year,
month=self.month,
day=self.day,
weekday=self.weekday,
hour=self.hour,
minute=self.minute,
second=self.second,
microsecond=self.microsecond)
def __nonzero__(self):
return not (not self.years and
not self.months and
not self.days and
not self.hours and
not self.minutes and
not self.seconds and
not self.microseconds and
not self.leapdays and
self.year is None and
self.month is None and
self.day is None and
self.weekday is None and
self.hour is None and
self.minute is None and
self.second is None and
self.microsecond is None)
def __mul__(self, other):
f = float(other)
return relativedelta(years=self.years*f,
months=self.months*f,
days=self.days*f,
hours=self.hours*f,
minutes=self.minutes*f,
seconds=self.seconds*f,
microseconds=self.microseconds*f,
leapdays=self.leapdays,
year=self.year,
month=self.month,
day=self.day,
weekday=self.weekday,
hour=self.hour,
minute=self.minute,
second=self.second,
microsecond=self.microsecond)
def __eq__(self, other):
if not isinstance(other, relativedelta):
return False
if self.weekday or other.weekday:
if not self.weekday or not other.weekday:
return False
if self.weekday.weekday != other.weekday.weekday:
return False
n1, n2 = self.weekday.n, other.weekday.n
if n1 != n2 and not (n1 in (0, 1) and n2 in (0, 1)):
return False
return (self.years == other.years and
self.months == other.months and
self.days == other.days and
self.hours == other.hours and
self.minutes == other.minutes and
self.seconds == other.seconds and
self.leapdays == other.leapdays and
self.year == other.year and
self.month == other.month and
self.day == other.day and
self.hour == other.hour and
self.minute == other.minute and
self.second == other.second and
self.microsecond == other.microsecond)
def __ne__(self, other):
return not self.__eq__(other)
def __div__(self, other):
return self.__mul__(1/float(other))
def __repr__(self):
l = []
for attr in ["years", "months", "days", "leapdays",
"hours", "minutes", "seconds", "microseconds"]:
value = getattr(self, attr)
if value:
l.append("%s=%+d" % (attr, value))
for attr in ["year", "month", "day", "weekday",
"hour", "minute", "second", "microsecond"]:
value = getattr(self, attr)
if value is not None:
l.append("%s=%s" % (attr, `value`))
return "%s(%s)" % (self.__class__.__name__, ", ".join(l))
# vim:ts=4:sw=4:et

883
others/dateutil/tz.py Normal file
View File

@ -0,0 +1,883 @@
"""
Copyright (c) 2003 Gustavo Niemeyer <niemeyer@conectiva.com>
This module offers extensions to the standard python 2.3+
datetime module.
"""
__author__ = "Gustavo Niemeyer <niemeyer@conectiva.com>"
__license__ = "PSF License"
import datetime
import struct
import time
relativedelta = None
parser = None
rrule = None
__all__ = ["tzutc", "tzoffset", "tzlocal", "tzfile",
"tzrange", "tzstr", "tzical", "gettz"]
ZERO = datetime.timedelta(0)
EPOCHORDINAL = datetime.datetime.utcfromtimestamp(0).toordinal()
class tzutc(datetime.tzinfo):
def utcoffset(self, dt):
return ZERO
def dst(self, dt):
return ZERO
def tzname(self, dt):
return "UTC"
def __eq__(self, other):
return (isinstance(other, tzutc) or
(isinstance(other, tzoffset) and other._offset == ZERO))
def __ne__(self, other):
return not self.__eq__(other)
def __repr__(self):
return "%s()" % self.__class__.__name__
class tzoffset(datetime.tzinfo):
def __init__(self, name, offset):
self._name = name
self._offset = datetime.timedelta(seconds=offset)
def utcoffset(self, dt):
return self._offset
def dst(self, dt):
return ZERO
def tzname(self, dt):
return self._name
def __eq__(self, other):
return (isinstance(other, tzoffset) and
self._offset == other._offset)
def __ne__(self, other):
return not self.__eq__(other)
def __repr__(self):
return "%s(%s, %s)" % (self.__class__.__name__,
`self._name`,
self._offset.days*86400+self._offset.seconds)
class tzlocal(datetime.tzinfo):
_std_offset = datetime.timedelta(seconds=-time.timezone)
if time.daylight:
_dst_offset = datetime.timedelta(seconds=-time.altzone)
else:
_dst_offset = _std_offset
def utcoffset(self, dt):
if self._isdst(dt):
return self._dst_offset
else:
return self._std_offset
def dst(self, dt):
if self._isdst(dt):
return self._dst_offset-self._std_offset
else:
return ZERO
def tzname(self, dt):
return time.tzname[self._isdst(dt)]
def _isdst(self, dt):
# We can't use mktime here. It is unstable when deciding if
# the hour near to a change is DST or not.
#
# timestamp = time.mktime((dt.year, dt.month, dt.day, dt.hour,
# dt.minute, dt.second, dt.weekday(), 0, -1))
# return time.localtime(timestamp).tm_isdst
#
# The code above yields the following result:
#
#>>> import tz, datetime
#>>> t = tz.tzlocal()
#>>> datetime.datetime(2003,2,15,23,tzinfo=t).tzname()
#'BRDT'
#>>> datetime.datetime(2003,2,16,0,tzinfo=t).tzname()
#'BRST'
#>>> datetime.datetime(2003,2,15,23,tzinfo=t).tzname()
#'BRST'
#>>> datetime.datetime(2003,2,15,22,tzinfo=t).tzname()
#'BRDT'
#>>> datetime.datetime(2003,2,15,23,tzinfo=t).tzname()
#'BRDT'
#
# Here is a more stable implementation:
#
timestamp = ((dt.toordinal() - EPOCHORDINAL) * 86400
+ dt.hour * 3600
+ dt.minute * 60
+ dt.second)
return time.localtime(timestamp+time.timezone).tm_isdst
def __eq__(self, other):
if not isinstance(other, tzlocal):
return False
return (self._std_offset == other._std_offset and
self._dst_offset == other._dst_offset)
return True
def __ne__(self, other):
return not self.__eq__(other)
def __repr__(self):
return "%s()" % self.__class__.__name__
class _ttinfo(object):
__slots__ = ["offset", "delta", "isdst", "abbr", "isstd", "isgmt"]
def __init__(self):
for attr in self.__slots__:
setattr(self, attr, None)
def __repr__(self):
l = []
for attr in self.__slots__:
value = getattr(self, attr)
if value is not None:
l.append("%s=%s" % (attr, `value`))
return "%s(%s)" % (self.__class__.__name__, ", ".join(l))
def __eq__(self, other):
if not isinstance(other, _ttinfo):
return False
return (self.offset == other.offset and
self.delta == other.delta and
self.isdst == other.isdst and
self.abbr == other.abbr and
self.isstd == other.isstd and
self.isgmt == other.isgmt)
def __ne__(self, other):
return not self.__eq__(other)
class tzfile(datetime.tzinfo):
# http://www.twinsun.com/tz/tz-link.htm
# ftp://elsie.nci.nih.gov/pub/tz*.tar.gz
def __init__(self, fileobj):
if isinstance(fileobj, basestring):
self._s = fileobj
fileobj = open(fileobj)
elif hasattr(fileobj, "name"):
self._s = fileobj.name
else:
self._s = `fileobj`
# From tzfile(5):
#
# The time zone information files used by tzset(3)
# begin with the magic characters "TZif" to identify
# them as time zone information files, followed by
# sixteen bytes reserved for future use, followed by
# six four-byte values of type long, written in a
# ``standard'' byte order (the high-order byte
# of the value is written first).
if fileobj.read(4) != "TZif":
raise ValueError, "magic not found"
fileobj.read(16)
(
# The number of UTC/local indicators stored in the file.
ttisgmtcnt,
# The number of standard/wall indicators stored in the file.
ttisstdcnt,
# The number of leap seconds for which data is
# stored in the file.
leapcnt,
# The number of "transition times" for which data
# is stored in the file.
timecnt,
# The number of "local time types" for which data
# is stored in the file (must not be zero).
typecnt,
# The number of characters of "time zone
# abbreviation strings" stored in the file.
charcnt,
) = struct.unpack(">6l", fileobj.read(24))
# The above header is followed by tzh_timecnt four-byte
# values of type long, sorted in ascending order.
# These values are written in ``standard'' byte order.
# Each is used as a transition time (as returned by
# time(2)) at which the rules for computing local time
# change.
if timecnt:
self._trans_list = struct.unpack(">%dl" % timecnt,
fileobj.read(timecnt*4))
else:
self._trans_list = []
# Next come tzh_timecnt one-byte values of type unsigned
# char; each one tells which of the different types of
# ``local time'' types described in the file is associated
# with the same-indexed transition time. These values
# serve as indices into an array of ttinfo structures that
# appears next in the file.
if timecnt:
self._trans_idx = struct.unpack(">%dB" % timecnt,
fileobj.read(timecnt))
else:
self._trans_idx = []
# Each ttinfo structure is written as a four-byte value
# for tt_gmtoff of type long, in a standard byte
# order, followed by a one-byte value for tt_isdst
# and a one-byte value for tt_abbrind. In each
# structure, tt_gmtoff gives the number of
# seconds to be added to UTC, tt_isdst tells whether
# tm_isdst should be set by localtime(3), and
# tt_abbrind serves as an index into the array of
# time zone abbreviation characters that follow the
# ttinfo structure(s) in the file.
ttinfo = []
for i in range(typecnt):
ttinfo.append(struct.unpack(">lbb", fileobj.read(6)))
abbr = fileobj.read(charcnt)
# Then there are tzh_leapcnt pairs of four-byte
# values, written in standard byte order; the
# first value of each pair gives the time (as
# returned by time(2)) at which a leap second
# occurs; the second gives the total number of
# leap seconds to be applied after the given time.
# The pairs of values are sorted in ascending order
# by time.
# Not used, for now
if leapcnt:
leap = struct.unpack(">%dl" % leapcnt*2,
fileobj.read(leapcnt*8))
# Then there are tzh_ttisstdcnt standard/wall
# indicators, each stored as a one-byte value;
# they tell whether the transition times associated
# with local time types were specified as standard
# time or wall clock time, and are used when
# a time zone file is used in handling POSIX-style
# time zone environment variables.
if ttisstdcnt:
isstd = struct.unpack(">%db" % ttisstdcnt,
fileobj.read(ttisstdcnt))
# Finally, there are tzh_ttisgmtcnt UTC/local
# indicators, each stored as a one-byte value;
# they tell whether the transition times associated
# with local time types were specified as UTC or
# local time, and are used when a time zone file
# is used in handling POSIX-style time zone envi-
# ronment variables.
if ttisgmtcnt:
isgmt = struct.unpack(">%db" % ttisgmtcnt,
fileobj.read(ttisgmtcnt))
# ** Everything has been read **
# Build ttinfo list
self._ttinfo_list = []
for i in range(typecnt):
tti = _ttinfo()
tti.offset = ttinfo[i][0]
tti.delta = datetime.timedelta(seconds=ttinfo[i][0])
tti.isdst = ttinfo[i][1]
tti.abbr = abbr[ttinfo[i][2]:abbr.find('\x00', ttinfo[i][2])]
tti.isstd = (ttisstdcnt > i and isstd[i] != 0)
tti.isgmt = (ttisgmtcnt > i and isgmt[i] != 0)
self._ttinfo_list.append(tti)
# Replace ttinfo indexes for ttinfo objects.
trans_idx = []
for idx in self._trans_idx:
trans_idx.append(self._ttinfo_list[idx])
self._trans_idx = tuple(trans_idx)
# Set standard, dst, and before ttinfos. before will be
# used when a given time is before any transitions,
# and will be set to the first non-dst ttinfo, or to
# the first dst, if all of them are dst.
self._ttinfo_std = None
self._ttinfo_dst = None
self._ttinfo_before = None
if self._ttinfo_list:
if not self._trans_list:
self._ttinfo_std = self._ttinfo_first = self._ttinfo_list[0]
else:
for i in range(timecnt-1,-1,-1):
tti = self._trans_idx[i]
if not self._ttinfo_std and not tti.isdst:
self._ttinfo_std = tti
elif not self._ttinfo_dst and tti.isdst:
self._ttinfo_dst = tti
if self._ttinfo_std and self._ttinfo_dst:
break
else:
if self._ttinfo_dst and not self._ttinfo_std:
self._ttinfo_std = self._ttinfo_dst
for tti in self._ttinfo_list:
if not tti.isdst:
self._ttinfo_before = tti
break
else:
self._ttinfo_before = self._ttinfo_list[0]
# Now fix transition times to become relative to wall time.
#
# I'm not sure about this. In my tests, the tz source file
# is setup to wall time, and in the binary file isstd and
# isgmt are off, so it should be in wall time. OTOH, it's
# always in gmt time. Let me know if you have comments
# about this.
laststdoffset = 0
self._trans_list = list(self._trans_list)
for i in range(len(self._trans_list)):
tti = self._trans_idx[i]
if not tti.isdst:
# This is std time.
self._trans_list[i] += tti.offset
laststdoffset = tti.offset
else:
# This is dst time. Convert to std.
self._trans_list[i] += laststdoffset
self._trans_list = tuple(self._trans_list)
def _find_ttinfo(self, dt, laststd=0):
timestamp = ((dt.toordinal() - EPOCHORDINAL) * 86400
+ dt.hour * 3600
+ dt.minute * 60
+ dt.second)
idx = 0
for trans in self._trans_list:
if timestamp < trans:
break
idx += 1
else:
return self._ttinfo_std
if idx == 0:
return self._ttinfo_before
if laststd:
while idx > 0:
tti = self._trans_idx[idx-1]
if not tti.isdst:
return tti
idx -= 1
else:
return self._ttinfo_std
else:
return self._trans_idx[idx-1]
def utcoffset(self, dt):
if not self._ttinfo_std:
return ZERO
return self._find_ttinfo(dt).delta
def dst(self, dt):
if not self._ttinfo_dst:
return ZERO
tti = self._find_ttinfo(dt)
if not tti.isdst:
return ZERO
# The documentation says that utcoffset()-dst() must
# be constant for every dt.
return self._find_ttinfo(dt, laststd=1).delta-tti.delta
# An alternative for that would be:
#
# return self._ttinfo_dst.offset-self._ttinfo_std.offset
#
# However, this class stores historical changes in the
# dst offset, so I belive that this wouldn't be the right
# way to implement this.
def tzname(self, dt):
if not self._ttinfo_std:
return None
return self._find_ttinfo(dt).abbr
def __eq__(self, other):
if not isinstance(other, tzfile):
return False
return (self._trans_list == other._trans_list and
self._trans_idx == other._trans_idx and
self._ttinfo_list == other._ttinfo_list)
def __ne__(self, other):
return not self.__eq__(other)
def __repr__(self):
return "%s(%s)" % (self.__class__.__name__, `self._s`)
class tzrange(datetime.tzinfo):
def __init__(self, stdabbr, stdoffset=None,
dstabbr=None, dstoffset=None,
start=None, end=None):
global relativedelta
if not relativedelta:
from dateutil import relativedelta
self._std_abbr = stdabbr
self._dst_abbr = dstabbr
if stdoffset is not None:
self._std_offset = datetime.timedelta(seconds=stdoffset)
else:
self._std_offset = ZERO
if dstoffset is not None:
self._dst_offset = datetime.timedelta(seconds=dstoffset)
elif dstabbr and stdoffset is not None:
self._dst_offset = self._std_offset+datetime.timedelta(hours=+1)
else:
self._dst_offset = ZERO
if start is None:
self._start_delta = relativedelta.relativedelta(
hours=+2, month=4, day=1, weekday=relativedelta.SU(+1))
else:
self._start_delta = start
if end is None:
self._end_delta = relativedelta.relativedelta(
hours=+1, month=10, day=31, weekday=relativedelta.SU(-1))
else:
self._end_delta = end
def utcoffset(self, dt):
if self._isdst(dt):
return self._dst_offset
else:
return self._std_offset
def dst(self, dt):
if self._isdst(dt):
return self._dst_offset-self._std_offset
else:
return ZERO
def tzname(self, dt):
if self._isdst(dt):
return self._dst_abbr
else:
return self._std_abbr
def _isdst(self, dt):
if not self._start_delta:
return False
year = datetime.date(dt.year,1,1)
start = year+self._start_delta
end = year+self._end_delta
dt = dt.replace(tzinfo=None)
if start < end:
return dt >= start and dt < end
else:
return dt >= start or dt < end
def __eq__(self, other):
if not isinstance(other, tzrange):
return False
return (self._std_abbr == other._std_abbr and
self._dst_abbr == other._dst_abbr and
self._std_offset == other._std_offset and
self._dst_offset == other._dst_offset and
self._start_delta == other._start_delta and
self._end_delta == other._end_delta)
def __ne__(self, other):
return not self.__eq__(other)
def __repr__(self):
return "%s(...)" % self.__class__.__name__
class tzstr(tzrange):
def __init__(self, s):
global parser
if not parser:
from dateutil import parser
self._s = s
res = parser._parsetz(s)
if res is None:
raise ValueError, "unknown string format"
# We must initialize it first, since _delta() needs
# _std_offset and _dst_offset set. Use False in start/end
# to avoid building it two times.
tzrange.__init__(self, res.stdabbr, res.stdoffset,
res.dstabbr, res.dstoffset,
start=False, end=False)
self._start_delta = self._delta(res.start)
if self._start_delta:
self._end_delta = self._delta(res.end, isend=1)
def _delta(self, x, isend=0):
kwargs = {}
if x.month is not None:
kwargs["month"] = x.month
if x.weekday is not None:
kwargs["weekday"] = relativedelta.weekday(x.weekday, x.week)
if x.week > 0:
kwargs["day"] = 1
else:
kwargs["day"] = 31
elif x.day:
kwargs["day"] = x.day
elif x.yday is not None:
kwargs["yearday"] = x.yday
elif x.jyday is not None:
kwargs["nlyearday"] = x.jyday
if not kwargs:
# Default is to start on first sunday of april, and end
# on last sunday of october.
if not isend:
kwargs["month"] = 4
kwargs["day"] = 1
kwargs["weekday"] = relativedelta.SU(+1)
else:
kwargs["month"] = 10
kwargs["day"] = 31
kwargs["weekday"] = relativedelta.SU(-1)
if x.time is not None:
kwargs["seconds"] = x.time
else:
# Default is 2AM.
kwargs["seconds"] = 7200
if isend:
# Convert to standard time, to follow the documented way
# of working with the extra hour. See the documentation
# of the tzinfo class.
delta = self._dst_offset-self._std_offset
kwargs["seconds"] -= delta.seconds+delta.days*86400
return relativedelta.relativedelta(**kwargs)
def __repr__(self):
return "%s(%s)" % (self.__class__.__name__, `self._s`)
class _tzicalvtzcomp:
def __init__(self, tzoffsetfrom, tzoffsetto, isdst,
tzname=None, rrule=None):
self.tzoffsetfrom = datetime.timedelta(seconds=tzoffsetfrom)
self.tzoffsetto = datetime.timedelta(seconds=tzoffsetto)
self.tzoffsetdiff = self.tzoffsetto-self.tzoffsetfrom
self.isdst = isdst
self.tzname = tzname
self.rrule = rrule
class _tzicalvtz(datetime.tzinfo):
def __init__(self, tzid, comps=[]):
self._tzid = tzid
self._comps = comps
self._cachedate = []
self._cachecomp = []
def _find_comp(self, dt):
if len(self._comps) == 1:
return self._comps[0]
dt = dt.replace(tzinfo=None)
try:
return self._cachecomp[self._cachedate.index(dt)]
except ValueError:
pass
lastcomp = None
lastcompdt = None
for comp in self._comps:
if not comp.isdst:
# Handle the extra hour in DST -> STD
compdt = comp.rrule.before(dt-comp.tzoffsetdiff, inc=True)
else:
compdt = comp.rrule.before(dt, inc=True)
if compdt and (not lastcompdt or lastcompdt < compdt):
lastcompdt = compdt
lastcomp = comp
if not lastcomp:
# RFC says nothing about what to do when a given
# time is before the first onset date. We'll look for the
# first standard component, or the first component, if
# none is found.
for comp in self._comps:
if not comp.isdst:
lastcomp = comp
break
else:
lastcomp = comp[0]
self._cachedate.insert(0, dt)
self._cachecomp.insert(0, lastcomp)
if len(self._cachedate) > 10:
self._cachedate.pop()
self._cachecomp.pop()
return lastcomp
def utcoffset(self, dt):
return self._find_comp(dt).tzoffsetto
def dst(self, dt):
comp = self._find_comp(dt)
if comp.isdst:
return comp.tzoffsetdiff
else:
return ZERO
def tzname(self, dt):
return self._find_comp(dt).tzname
def __repr__(self):
return "<tzicalvtz %s>" % `self._tzid`
class tzical:
def __init__(self, fileobj):
global rrule
if not rrule:
from dateutil import rrule
if isinstance(fileobj, basestring):
self._s = fileobj
fileobj = open(fileobj)
elif hasattr(fileobj, "name"):
self._s = fileobj.name
else:
self._s = `fileobj`
self._vtz = {}
self._parse_rfc(fileobj.read())
def keys(self):
return self._vtz.keys()
def get(self, tzid=None):
if tzid is None:
keys = self._vtz.keys()
if len(keys) == 0:
raise "no timezones defined"
elif len(keys) > 1:
raise "more than one timezone available"
tzid = keys[0]
return self._vtz.get(tzid)
def _parse_offset(self, s):
s = s.strip()
if not s:
raise ValueError, "empty offset"
if s[0] in ('+', '-'):
signal = (-1,+1)[s[0]=='+']
s = s[1:]
else:
signal = +1
if len(s) == 4:
return (int(s[:2])*3600+int(s[2:])*60)*signal
elif len(s) == 6:
return (int(s[:2])*3600+int(s[2:4])*60+int(s[4:]))*signal
else:
raise ValueError, "invalid offset: "+s
def _parse_rfc(self, s):
lines = s.splitlines()
if not lines:
raise ValueError, "empty string"
# Unfold
i = 0
while i < len(lines):
line = lines[i].rstrip()
if not line:
del lines[i]
elif i > 0 and line[0] == " ":
lines[i-1] += line[1:]
del lines[i]
else:
i += 1
invtz = False
comptype = None
for line in lines:
if not line:
continue
name, value = line.split(':', 1)
parms = name.split(';')
if not parms:
raise ValueError, "empty property name"
name = parms[0].upper()
parms = parms[1:]
if invtz:
if name == "BEGIN":
if value in ("STANDARD", "DAYLIGHT"):
# Process component
pass
else:
raise ValueError, "unknown component: "+value
comptype = value
founddtstart = False
tzoffsetfrom = None
tzoffsetto = None
rrulelines = []
tzname = None
elif name == "END":
if value == "VTIMEZONE":
if comptype:
raise ValueError, \
"component not closed: "+comptype
if not tzid:
raise ValueError, \
"mandatory TZID not found"
if not comps:
raise ValueError, \
"at least one component is needed"
# Process vtimezone
self._vtz[tzid] = _tzicalvtz(tzid, comps)
invtz = False
elif value == comptype:
if not founddtstart:
raise ValueError, \
"mandatory DTSTART not found"
if not tzoffsetfrom:
raise ValueError, \
"mandatory TZOFFSETFROM not found"
if not tzoffsetto:
raise ValueError, \
"mandatory TZOFFSETFROM not found"
# Process component
rr = None
if rrulelines:
rr = rrule.rrulestr("\n".join(rrulelines),
compatible=True,
ignoretz=True,
cache=True)
comp = _tzicalvtzcomp(tzoffsetfrom, tzoffsetto,
(comptype == "DAYLIGHT"),
tzname, rr)
comps.append(comp)
comptype = None
else:
raise ValueError, \
"invalid component end: "+value
elif comptype:
if name == "DTSTART":
rrulelines.append(line)
founddtstart = True
elif name in ("RRULE", "RDATE", "EXRULE", "EXDATE"):
rrulelines.append(line)
elif name == "TZOFFSETFROM":
if parms:
raise ValueError, \
"unsupported %s parm: %s "%(name, parms[0])
tzoffsetfrom = self._parse_offset(value)
elif name == "TZOFFSETTO":
if parms:
raise ValueError, \
"unsupported TZOFFSETTO parm: "+parms[0]
tzoffsetto = self._parse_offset(value)
elif name == "TZNAME":
if parms:
raise ValueError, \
"unsupported TZNAME parm: "+parms[0]
tzname = value
elif name == "COMMENT":
pass
else:
raise ValueError, "unsupported property: "+name
else:
if name == "TZID":
if parms:
raise ValueError, \
"unsupported TZID parm: "+parms[0]
tzid = value
elif name in ("TZURL", "LAST-MODIFIED", "COMMENT"):
pass
else:
raise ValueError, "unsupported property: "+name
elif name == "BEGIN" and value == "VTIMEZONE":
tzid = None
comps = []
invtz = True
def __repr__(self):
return "%s(%s)" % (self.__class__.__name__, `self._s`)
TZFILES = ["/etc/localtime", "localtime"]
TZPATHS = ["/usr/share/zoneinfo", "/usr/lib/zoneinfo", "/etc/zoneinfo"]
import os
def gettz(name=None):
tz = None
if not name:
try:
name = os.environ["TZ"]
except KeyError:
pass
if name is None:
for filepath in TZFILES:
if not os.path.isabs(filepath):
filename = filepath
for path in TZPATHS:
filepath = os.path.join(path, filename)
if os.path.isfile(filepath):
break
else:
continue
if os.path.isfile(filepath):
try:
tz = tzfile(filepath)
break
except (IOError, OSError, ValueError):
pass
else:
if name and name[0] == ":":
name = name[:-1]
for path in TZPATHS:
filepath = os.path.join(path, name)
if not os.path.isfile(filepath):
filepath = filepath.replace(' ','_')
if not os.path.isfile(filepath):
continue
try:
tz = tzfile(filepath)
break
except (IOError, OSError, ValueError):
pass
else:
for c in name:
# name must have at least one offset to be a tzstr
if c in "0123456789":
try:
tz = tzstr(name)
except ValueError:
pass
break
else:
if name in ("GMT", "UTC"):
tz = tzutc()
elif name in time.tzname:
tz = tzlocal()
return tz
# vim:ts=4:sw=4:et

View File

@ -1,229 +0,0 @@
#
# Copyright (c) 2004, Mike Taylor
# All rights reserved.
#
# Redistribution and use in source and binary forms, with or without
# modification, are permitted provided that the following conditions are met:
#
# * Redistributions of source code must retain the above copyright notice,
# this list of conditions, and the following disclaimer.
# * Redistributions in binary form must reproduce the above copyright notice,
# this list of conditions, and the following disclaimer in the
# documentation and/or other materials provided with the distribution.
# * Neither the name of the author of this software nor the name of
# contributors to this software may be used to endorse or promote products
# derived from this software without specific prior written consent.
#
# THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
# AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
# IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
# ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
# LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
# CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
# SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
# INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
# CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
# ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
# POSSIBILITY OF SUCH DAMAGE.
#
__author__ = "Mike Taylor <bear@code-bear.com>"
__copyright__ = "Copyright (c) 2004 Mike Taylor"
__license__ = "BSD"
__revision__ = "$Id$"
import os, string, re, time
RE_SPECIAL = r'(?P<special>^[in|last|next]+)\s+'
RE_UNITS = r'\s+(?P<units>[hour|minute|second|day|week|month|year]+)'
RE_QUNITS = r'(?P<qunits>[0-9]+[hmsdwmy])'
RE_MODIFIER = r'(?P<modifier>[from|before|after|ago|prior]+)\s+'
CRE_SPECIAL = re.compile(RE_SPECIAL, re.IGNORECASE)
CRE_UNITS = re.compile(RE_UNITS, re.IGNORECASE)
CRE_QUNITS = re.compile(RE_QUNITS, re.IGNORECASE)
CRE_MODIFIER = re.compile(RE_MODIFIER, re.IGNORECASE)
# Used to adjust the returned date before/after the source
_Modifiers = {'from': 1,
'before': -1,
'after': 1,
'ago': 1,
'prior': -1}
_Minute = 60
_Hour = 60 * _Minute
_Day = 24 * _Hour
_Week = 7 * _Day
_Month = 30 * _Day
_Year = 365 * _Day
# This looks hokey - but it is a nice simple way to get
# the proper unit value and it has the advantage that
# later I can morph it into something localized.
# Any trailing s will be removed before lookup.
_Units = {'second': 1,
'sec': 1,
's': 1,
'minute': _Minute,
'min': _Minute,
'm': _Minute,
'hour': _Hour,
'hr': _Hour,
'h': _Hour,
'day': _Day,
'dy': _Day,
'd': _Day,
'week': _Week,
'wk': _Week,
'w': _Week,
'month': _Month,
'mth': _Month,
'm': _Month,
'year': _Year,
'yr': _Year,
'y': _Year}
def _buildTime(sourceTime, quantity, modifier, units):
"""Take quantity, modifier and units strings and convert them
into values, calcuate the time and return the adjusted
sourceTime
"""
# print '[%s][%s][%s]' % (quantity, modifier, units)
q = int(quantity)
if _Modifiers.has_key(modifier):
q = q * _Modifiers[modifier]
if units[-1] == 's':
units = units[:-1]
if _Units.has_key(units):
u = _Units[units]
else:
u = 1
# print 'sourceTime [%d]' % sourceTime
# print 'quantity [%d]' % q
# print 'units [%d]' % u
return sourceTime + (q * u)
def parse(timeString, sourceTime=None):
"""Parse timeString and return the number of seconds from sourceTime
that the timeString expression represents.
This version of parse understands only the more basic of expressions
in this form:
<quantity> <units> <modifier> <target>
Example:
5 minutes from now
last week
2 hours before noon
Valid units - hour, minute, second, month, week, day and year
(including their plural forms)
Valid modifiers - from, before, after, ago, prior
"""
if sourceTime == None:
sourceTime = int(time.time())
else:
sourceTime = int(sourceTime)
quantity = ''
units = ''
modifier = ''
target = ''
s = string.strip(string.lower(timeString))
m = CRE_SPECIAL.search(s)
if m <> None:
target = 'now'
if m.group('special') == 'last':
modifier = 'before'
else:
modifier = 'from'
s = s[m.end('special'):]
m = CRE_UNITS.search(s)
if m <> None:
units = m.group('units')
quantity = s[:m.start('units')]
s = s[m.end('units'):]
else:
m = CRE_MODIFIER.search(s)
if m <> None:
modifier = m.group('modifier')
target = s[m.end('modifier'):]
s = s[:m.start('modifier'):]
m = CRE_UNITS.search(s)
if m <> None:
units = m.group('units')
quantity = s[:m.start('units')]
target = s[m.end('units'):]
return _buildTime(sourceTime, quantity, modifier, units)
def _test(text, value):
print text
v = parse(text)
print '\t%s\t%d\t%d' % ((v == value), value, v)
#
# TODO
#
# - make month unit adjustment aware of the actual number of days in each
# month between source and target
# - handle edge case where quantity and unit are merged: 5s for 5 sec
# - handle compound/nested quantites and modifiers
# - bring unit test over from prototype into this file
# - convert 'five' to 5 and also 'twenty five' to 25
if __name__ == '__main__':
start = int(time.time())
tests = { '5 minutes from now': start + 5 * _Minute,
'5 min from now': start + 5 * _Minute,
'in 5 minutes': start + 5 * _Minute,
'5 days from today': start + 5 * _Day,
'5 days before today': start - 5 * _Day,
'5 minutes': start + 5 * _Minute,
'5 min': start + 5 * _Minute,
'30 seconds from now': start + 30,
'30 sec from now': start + 30,
'30 seconds': start + 30,
'30 sec': start + 30,
'1 week': start + _Week,
'1 wk': start + _Week,
'1 week': start + _Week,
'5 days': start + 5 * _Day,
'1 year': start + _Year,
'2 years': start + 2 * _Year}
for test in tests.keys():
_test(test, tests[test])