def normalize_lay_can(date_item, rpt_date):
"""Transform non isoformat dates to isoformat
Examples:
>>> normalize_lay_can('8/14 PM', '14 Aug 2019')
'2019-08-14T00:00:00'
>>> normalize_lay_can('2018-02-04T00:00:00', '14 Aug 2019')
'2018-02-04T00:00:00'
>>> normalize_lay_can('12/31 PM', '29 Dec 2019')
'2019-12-31T00:00:00'
>>> normalize_lay_can('12/31 PM', '01 Jan 2020')
'2019-12-31T00:00:00'
>>> normalize_lay_can('01/01 PM', '31 Dec 2019')
'2020-01-01T00:00:00'
>>> normalize_lay_can('7/06/2015 1750', '31 Dec 2019')
'2015-07-06T17:50:00'
Args:
date_item (str):
Returns:
str:
"""
if is_isoformat(date_item):
return date_item
if not is_isoformat(date_item):
year = parse_date(rpt_date, dayfirst=True).year
_date = date_item.partition(' ')[0]
if len(_date.split('/')) == 2:
_month, _day = _date.split('/')
if 'Dec' in rpt_date and (str(_month) == '1'
or str(_month) == '01'):
year += 1
if 'Jan' in rpt_date and str(_month) == '12':
year -= 1
return to_isoformat(f'{_day} {_month} {year}', dayfirst=True)
if len(_date.split('/')) == 3:
_time = date_item.partition(' ')[2]
_hour = _time.replace(' ', '')[:2] if _time.replace(
' ', '')[:2].isdigit() else '00'
_min = _time.replace(' ', '')[2:] if _time.replace(
' ', '')[2:].isdigit() else '00'
try:
return to_isoformat(f'{_date} {_hour}:{_min}', dayfirst=False)
except Exception:
logger.error('Skipping date row: %s', date_item)
return date_item
logger.error('Skipping date row: %s', date_item)
return date_item
def normalize_laycan(raw_laycan):
"""normalize laycans
Args:
raw_laycan (str):
Returns:
str:
Examples:
>>> normalize_laycan('2020-01-01T00:00:00')
'2020-01-01T00:00:00'
>>> normalize_laycan('20Feb20')
'2020-02-20T00:00:00'
"""
if is_isoformat(raw_laycan):
return raw_laycan
else:
try:
_match = re.match(r'(\d+)([A-z]+)(\d+)', raw_laycan)
if _match:
day, month, year = _match.groups()
return parse_date(f'20{year} {month} {day}').isoformat()
except Exception:
return None
def normalize_pc_date(date_str):
"""Cleanup portcall-related date.
Args:
date_str (str):
Returns:
str | None: date string without "am" or 'pm' features
"""
if not may_strip(date_str) or any(sub in date_str
for sub in STRING_BLACKLIST):
return None
# remove 'am' and 'pm' abbreviations in portcall-dates
if not is_isoformat(date_str):
for abbreviation in ABBREVIATIONS:
if re.compile(abbreviation).search(date_str):
date_str = date_str.replace(abbreviation, '')
# assume hours of arrival for 'am' and 'pm' equal to 06:00 and 18:00
if abbreviation == 'a.m' or abbreviation == 'am':
date_str = date_str + '06:00'
else:
date_str = date_str + '18:00'
# parse raw date and format it
date_str = may_remove_substring(date_str, ["'", '.', ' '])
date_str = datetime.strptime(date_str, '%d-%b%H:%M')
return date_str
def field_mapping():
return {
'vessel': ('vessel_name', None),
'arr': ('arrival', lambda x: x if is_isoformat(x) else None),
'eta': ('eta', lambda x: x if is_isoformat(x) else None),
'agency': ignore_key('irrelevant'),
'shipping line': ignore_key('irrelevant'),
'load port': ignore_key('irrelevant'),
'origin': ignore_key('irrelevant'),
'charterer/trader': ignore_key('irrelevant'),
'exporter/shipper': ignore_key('irrelevant'),
'exporter': ignore_key('irrelevant'),
'importer': ('cargo_buyer', may_strip),
'cargo type': ('cargo_product', None),
'import tonnes': ('cargo_volume_dis', lambda x: x if is_number(x) else None),
'export tonnes': ('cargo_volume_load', lambda x: x if is_number(x) else None),
'remarks': ignore_key('irrelevant'),
'port_name': ('port_name', None),
'provider_name': ('provider_name', None),
'reported_date': ('reported_date', None),
}
def field_mapping():
return {
'vessel': ('vessel_name', None),
'arrival': ('arrival', lambda x: x if is_isoformat(x) else None),
'berthing': ('berthed', lambda x: x if is_isoformat(x) else None),
'departure': ('departure', lambda x: x if is_isoformat(x) else None),
'second atb': ('second_port_date', lambda x: x
if is_isoformat(x) else None),
'tip(days)':
ignore_key('irrelevant'),
'load/last port': ('departure_zone', may_strip),
'shipper': ('cargo_seller', None),
'main receiver': ('cargo_buyer', None),
'cargo type': ('cargo_product', None),
'import tonnes': ('cargo_volume_disc', lambda x: x
if is_number(x) else None),
'load tonnes': ('cargo_volume_load', lambda x: x
if is_number(x) else None),
'next port':
ignore_key('irrelevant'),
'berth': ('raw_port_name', None),
'provider_name': ('provider_name', None),
'reported_date': ('reported_date', None),
}
def normalize_lay_can(date_item):
"""Transform non isoformat dates to isoformat
Examples:
>>> normalize_lay_can('04-Feb-2018')
'2018-02-04T00:00:00'
>>> normalize_lay_can('2018-02-04T00:00:00')
'2018-02-04T00:00:00'
Args:
date_item (str):
Returns:
str:
"""
return date_item if is_isoformat(date_item) else to_isoformat(
date_item, dayfirst=True)
def normalize_date(raw_date):
"""Normalize laycan
Examples:
- '02.02.19'
Examples:
>>> normalize_date('02.02.19')
'2019-02-02T00:00:00'
Args:
raw_laycan (str):
Returns:
str:
"""
return raw_date if is_isoformat(raw_date) else to_isoformat(raw_date, dayfirst=True)
def normalize_date(raw_date, reported_date):
"""Normalize ETA, ETB, ETS date.
Args:
raw_eta (str):
reported_date (str): date is in ISO-8601 format
Returns:
str | None: date in ISO-8601 format
Examples:
>>> normalize_date('6-Oct', '2018-09-21T00:00:00')
'2018-10-06T00:00:00'
>>> normalize_date('6-Jan', '2018-12-21T00:00:00')
'2019-01-06T00:00:00'
"""
if not raw_date:
return
if is_isoformat(raw_date):
return raw_date
if len(raw_date.split('-')) != 2:
logger.warning('Not a valid date: {}'.format(raw_date))
return
day, month = raw_date.split('-')
# in case month is in spanish
month = SPANISH_MONTH_MAPPING.get(month, month)
_reported_date = parse_date(reported_date, dayfirst=False)
try:
date = parse_date(f'{day} {month} {_reported_date.year}', dayfirst=True)
except ValueError:
logger.warning('Not a valid date: {}'.format(raw_date))
return None
# sanity check for cases where there is year rollover
if date - _reported_date < dt.timedelta(days=-180):
date += relativedelta(years=1)
return date.isoformat()
def normalize_dates(raw_date, raw_year):
"""Normalize raw laycan date.
Args:
raw_date (str):
raw_year (str):
Returns:
str:
Examples:
>>> normalize_dates('2/8', '2019')
'2019-02-08T00:00:00'
>>> normalize_dates('2/8 11:00', '2019')
'2019-02-08T11:00:00'
"""
if not is_isoformat(raw_date):
datetime_array = raw_date.split(' ')
if len(datetime_array) == 1:
try:
return to_isoformat(f'{datetime_array[0]}/{raw_year}',
dayfirst=False)
except Exception:
return raw_date
if len(datetime_array) == 2:
if datetime_array[1] in ['2400', '24:00']:
datetime_array[1] = '0000'
if datetime_array[1].replace('.', '').lower() == 'am':
datetime_array[1] = '0900'
if datetime_array[1].replace('.', '').lower() == 'pm':
datetime_array[1] = '1500'
try:
return to_isoformat(
f'{datetime_array[0]}/{raw_year} {datetime_array[1]}',
dayfirst=False)
except Exception:
return raw_date
return raw_date
def normalize_dates(raw_date):
""" Normalize dates
Args:
raw_date (str):
Returns:
str:
Examples:
>>> normalize_dates('2019-10-10T00:00:00')
'2019-10-10T00:00:00'
>>> normalize_dates('11/01/2019')
'2019-01-11T00:00:00'
>>> normalize_dates('TBA')
"""
if is_isoformat(raw_date):
return raw_date
try:
return to_isoformat(raw_date, dayfirst=True)
except Exception:
return None
def normalize_dates(raw_date, rpt_date):
"""Normalize dates
Args:
vessel_name (str):
Examples:
>>> normalize_dates('1400 hrs 01.01.2020', '2020-01-01T00:00:00')
'2020-01-01T14:00:00'
>>> normalize_dates('2106 hrs/02.01.2020', '2020-01-01T00:00:00')
'2020-01-02T21:06:00'
>>> normalize_dates('2106 hrs /02.01.2020', '2020-01-01T00:00:00')
'2020-01-02T21:06:00'
>>> normalize_dates('02.01.2020(NOR)', '2020-01-01T00:00:00')
'2020-01-02T00:00:00'
>>> normalize_dates('02.01.2020', '2020-01-01T00:00:00')
'2020-01-02T00:00:00'
>>> normalize_dates('02.01.20', '2020-01-01T00:00:00')
'2020-01-02T00:00:00'
>>> normalize_dates('22.03.2020-1800', '2020-01-01T00:00:00')
'2020-03-22T18:00:00'
>>> normalize_dates('22.03.2020-AM', '2020-01-01T00:00:00')
'2020-03-22T06:00:00'
>>> normalize_dates(' am 02.01.2020', '2020-01-01T00:00:00')
'2020-01-02T06:00:00'
>>> normalize_dates('02.01.2020/0700 hrs', '2020-01-01T00:00:00')
'2020-01-02T07:00:00'
>>> normalize_dates('pm hrs 06.01.2020', '2020-01-01T00:00:00')
'2020-01-06T15:00:00'
>>> normalize_dates('26/01', '2020-01-01T00:00:00')
'2020-01-26T00:00:00'
>>> normalize_dates('26/01', '2019-12-30T00:00:00')
'2020-01-26T00:00:00'
Returns:
str:
"""
# normalize dates strings
raw_date = raw_date.lower().replace('am', '0600/').replace('pm', '1500/')
if is_isoformat(raw_date.upper()):
return raw_date.upper()
# if dd/mm data is provided, guess the year and return date
if '/' in raw_date and len(raw_date.split('/')) == 2:
if all(is_number(rd) for rd in raw_date.split('/')):
potential_date, _ = get_date_range(raw_date, '/', '-', rpt_date)
return potential_date
# detect date and time fields
date_hour = [
may_strip(_d)
for _d in re.split(r'(hrs /|hrs/|/ hrs|hrs|/|\()|\-', raw_date) if _d
]
_date, _time = None, ''
for dh in date_hour:
if is_number(dh):
_time = dh
continue
if len(dh.split('.')) == 3:
_date = dh
continue
if not _date:
return None
try:
return to_isoformat(may_strip(f'{_date} {_time}'), dayfirst=True)
except Exception:
return None
