from datetime import date, datetime, timedelta, timezone
import re
#: The names of the months in English
MONTH_FULL_NAMES = [
'January',
'February',
'March',
'April',
'May',
'June',
'July',
'August',
'September',
'October',
'November',
'December',
]
#: The abbreviated names of the months in English
MONTH_SNAMES = [
'Jan', 'Feb', 'Mar', 'Apr', 'May', 'Jun',
'Jul', 'Aug', 'Sep', 'Oct', 'Nov', 'Dec',
]
#: Compiled regex for an Apache timestamp
APACHE_TS_RGX = re.compile(r'''
^\[?
(?P<day>\d\d) / (?P<month>\w\w\w) / (?P<year>\d{4,})
:(?P<hour>\d\d) : (?P<minute>\d\d) : (?P<second>\d\d)
\s* (?P<tzoffset_sign>[-+]) (?P<tzoffset_hour>\d\d) (?P<tzoffset_min>\d\d)
\]?$
''', flags=re.X)
[docs]def parse_apache_timestamp(s):
"""
Parse an Apache timestamp into a `datetime.datetime` object. The month
name in the timestamp is expected to be an abbreviated English name
regardless of the current locale.
>>> parse_apache_timestamp('[01/Nov/2017:07:28:29 +0000]')
datetime.datetime(2017, 11, 1, 7, 28, 29, tzinfo=datetime.timezone.utc)
:param str s: a string of the form ``DD/Mon/YYYY:HH:MM:SS +HHMM``
(optionally enclosed in square brackets)
:return: an aware `datetime.datetime`
:raises ValueError: if ``s`` is not in the expected format
"""
# Apache timestamps always use English month abbreviations. Thus, parsing
# with strptime like the below will fail when in a locale with different
# month snames:
#return datetime.strptime(s.strip('[]'), '%d/%b/%Y:%H:%M:%S %z')
if s is None:
return None
m = APACHE_TS_RGX.match(s)
if not m:
raise ValueError(s)
data = m.groupdict()
for k in 'year day hour minute second'.split():
data[k] = int(data[k])
try:
data['month'] = MONTH_SNAMES.index(data['month']) + 1
except ValueError:
raise ValueError(s)
tzoffset = timedelta(
hours = int(data.pop('tzoffset_hour')),
minutes = int(data.pop('tzoffset_min')),
)
if data.pop('tzoffset_sign') == '-':
tzoffset *= -1
data['tzinfo'] = timezone(tzoffset)
return datetime(**data)
def unescape(s):
"""
Unescape the escape sequences in the string ``s``, returning a `bytes`
string
"""
# Escape sequences used by Apache: \b \n \r \t \v \\ \" \xHH
# cf. ap_escape_logitem() in server/util.c
return re.sub(r'\\(x[0-9A-Fa-f]{2}|.)', _unesc, s).encode('iso-8859-1')
_unescapes = {
't': '\t',
'n': '\n',
'r': '\r',
'b': '\b',
'v': '\v',
# Not emitted by Apache (as of v2.4), but other servers might use it:
'f': '\f',
}
def _unesc(m):
esc = m.group(1)
if esc[0] == 'x':
return chr(int(esc[1:], 16))
else:
return _unescapes.get(esc, esc)
def assemble_datetime(fields):
"""
Given a `dict` of time fields, return a `datetime.datetime` object if there
is enough information to create one, `None` otherwise.
"""
if fields.get("timestamp") is not None:
return fields["timestamp"]
elif fields.get("microepoch") is not None:
return datetime.fromtimestamp(
fields["microepoch"] / 1000000,
fields.get("timezone") or timezone.utc,
# fields["timezone"] may be None, in which case we still want the
# timezone to be UTC
)
elif fields.get("milliepoch") is not None:
return datetime.fromtimestamp(
fields["milliepoch"] / 1000,
fields.get("timezone") or timezone.utc,
# fields["timezone"] may be None, in which case we still want the
# timezone to be UTC
)
elif fields.get("epoch") is not None:
return datetime.fromtimestamp(
fields["epoch"],
fields.get("timezone") or timezone.utc,
# fields["timezone"] may be None, in which case we still want the
# timezone to be UTC
)
else:
if fields.get("year") is not None:
year = fields["year"]
elif fields.get("date") is not None:
year = fields["date"].year
elif fields.get("century") is not None \
and fields.get("abbrev_year") is not None:
year = fields["century"] * 100 + fields["abbrev_year"]
else:
return None
if fields.get("mon") is not None:
month = fields["mon"]
elif fields.get("date") is not None:
month = fields["date"].month
elif fields.get("yday") is not None:
month = (date(year, 1, 1) + timedelta(days=fields["yday"]-1)).month
elif fields.get("full_mon") in MONTH_FULL_NAMES:
month = MONTH_FULL_NAMES.index(fields["full_mon"]) + 1
elif fields.get("abbrev_mon") in MONTH_SNAMES:
month = MONTH_SNAMES.index(fields["abbrev_mon"]) + 1
else:
return None
if fields.get("mday") is not None:
day = fields["mday"]
elif fields.get("date") is not None:
day = fields["date"].day
elif fields.get("yday") is not None:
day = (date(year, 1, 1) + timedelta(days=fields["yday"]-1)).day
else:
return None
if fields.get("hour") is not None:
hour = fields["hour"]
elif fields.get("time") is not None:
hour = fields["time"].hour
elif fields.get("hour_min") is not None:
hour = fields["hour_min"].hour
elif fields.get("hour12") is not None \
and fields.get("am_pm") is not None \
and fields["am_pm"].upper() in ('AM', 'PM'):
hour = fields["hour12"] % 12
if fields["am_pm"].upper() == "PM":
hour += 12
else:
return None
if fields.get("min") is not None:
minute = fields["min"]
elif fields.get("time") is not None:
minute = fields["time"].minute
elif fields.get("hour_min") is not None:
minute = fields["hour_min"].minute
else:
return None
if fields.get("sec") is not None:
second = fields["sec"]
elif fields.get("time") is not None:
second = fields["time"].second
else:
return None
if fields.get("usec_frac") is not None:
microsecond = fields["usec_frac"]
elif fields.get("msec_frac") is not None:
microsecond = fields["msec_frac"] * 1000
else:
microsecond = 0
return datetime(
year = year,
month = month,
day = day,
hour = hour,
minute = minute,
second = second,
microsecond = microsecond,
tzinfo = fields.get("timezone"),
)