def archive_aux_data(self):
'''
Description:
Defines the main processing method for the class.
'''
# Figure out the names of the files to retrieve
names = list(self.get_name_list())
# Establish a logged in session
session = Web.Session(
block_size=Config.get('http_transfer_block_size'))
# Log in
session.login(Config.get('ucar.login_credentials.login_url'),
Config.get('ucar.login_credentials.login_data'))
for name in names:
filename = '{0}.tar'.format(name)
self.logger.info('Retrieving {0}'.format(filename))
year = name[7:11]
url = Config.get('ucar.url_format').format(year, filename)
session.http_transfer_file(url, filename)
# Extract the tar'd data
cmd = ['tar', '-xvf', filename]
cmd = ' '.join(cmd)
grib_files = System.execute_cmd(cmd)
if grib_files is not None and len(grib_files) > 0:
self.logger.info(grib_files)
# For each parameter we need
for variable in Config.get('narr_variables'):
self.logger.info('Processing Variable [{0}]'.format(variable))
for grib_file in grib_files.split():
self.process_grib_for_variable(variable, grib_file)
# Cleanup - Extracted grib files
for grib_file in grib_files.split():
if os.path.exists(grib_file):
os.unlink(grib_file)
# Cleanup - The Tar ball
if os.path.exists(filename):
os.unlink(filename)
def archive_aux_data(self):
'''
Description:
Defines the main processing method for the class.
'''
# Figure out the names of the files to retrieve
names = list(self.get_name_list())
# Establish a logged in session
session = Web.Session(
block_size=Config.get('http_transfer_block_size'))
# Log in
session.login(Config.get('ucar.login_credentials.login_url'),
Config.get('ucar.login_credentials.login_data'))
for name in names:
filename = '{0}.tar'.format(name)
self.logger.info('Retrieving {0}'.format(filename))
year = name[7:11]
url = Config.get('ucar.url_format').format(year, filename)
session.http_transfer_file(url, filename)
# Extract the tar'd data
cmd = ['tar', '-xvf', filename]
cmd = ' '.join(cmd)
grib_files = System.execute_cmd(cmd)
if grib_files is not None and len(grib_files) > 0:
self.logger.info(grib_files)
# For each parameter we need
for variable in Config.get('narr_variables'):
self.logger.info('Processing Variable [{0}]'.format(variable))
for grib_file in grib_files.split():
self.process_grib_for_variable(variable, grib_file)
# Cleanup - Extracted grib files
for grib_file in grib_files.split():
if os.path.exists(grib_file):
os.unlink(grib_file)
# Cleanup - The Tar ball
if os.path.exists(filename):
os.unlink(filename)
def get_name_list(self):
'''
Description:
Determines all of the base filenames to process based on the dates
provided.
Notes:
Files typically contain 3 days.
Special assumptions are coded for the end of the month; which
may have 1, 2, 3, or 4 days ... depending on the month and year.
'''
name_format = Config.get('ucar.remote_name_format')
days_3 = timedelta(days=3)
c_date = self.s_date
while c_date <= self.e_date:
if c_date.day == 28:
days = calendar.monthrange(c_date.year, c_date.month)[1]
yield(name_format.format(c_date.year, c_date.month,
c_date.day, days))
delta = timedelta(days=(days - 28 + 1))
c_date += delta
else:
yield(name_format.format(c_date.year, c_date.month,
c_date.day, c_date.day + 2))
c_date += days_3
def get_name_list(self):
'''
Description:
Determines all of the base filenames to process based on the dates
provided.
Notes:
Files typically contain 3 days.
Special assumptions are coded for the end of the month; which
may have 1, 2, 3, or 4 days ... depending on the month and year.
'''
name_format = Config.get('ucar.remote_name_format')
days_3 = timedelta(days=3)
c_date = self.s_date
while c_date <= self.e_date:
if c_date.day == 28:
days = calendar.monthrange(c_date.year, c_date.month)[1]
yield (name_format.format(c_date.year, c_date.month,
c_date.day, days))
delta = timedelta(days=(days - 28 + 1))
c_date += delta
else:
yield (name_format.format(c_date.year, c_date.month,
c_date.day, c_date.day + 2))
c_date += days_3
def main():
'''
Description:
Provides the setup and executaion of the processor for the application.
'''
# The config file is located in the same place as this script
Config.read_config(os.path.dirname(__file__))
# Setup the default logger format and level. log to STDOUT
logging.basicConfig(format=('%(asctime)s.%(msecs)03d %(process)d'
' %(levelname)-8s'
' %(filename)s:%(lineno)d:'
'%(funcName)s -- %(message)s'),
datefmt='%Y-%m-%d %H:%M:%S',
level=logging.INFO,
stream=sys.stdout)
# Get the logger
logger = logging.getLogger(__name__)
# Turn down the requests and urllib3 logging otherwise they fill the log
# file with mostly useless information
requests_logger = logging.getLogger("requests")
requests_logger.setLevel(logging.WARNING)
urllib3_logger = logging.getLogger("urllib3")
urllib3_logger.setLevel(logging.WARNING)
# Process the command line
(s_date, e_date) = parse_commandline()
try:
# Create the processor object
processor = NARR_AuxProcessor(s_date, e_date)
# Call the main processing routine
processor.archive_aux_data()
except Exception:
logger.exception('Processing Failed')
sys.exit(1) # EXIT FAILURE
sys.exit(0) # EXIT SUCCESS
def main():
'''
Description:
Provides the setup and executaion of the processor for the application.
'''
# The config file is located in the same place as this script
Config.read_config(os.path.dirname(__file__))
# Setup the default logger format and level. log to STDOUT
logging.basicConfig(format=('%(asctime)s.%(msecs)03d %(process)d'
' %(levelname)-8s'
' %(filename)s:%(lineno)d:'
'%(funcName)s -- %(message)s'),
datefmt='%Y-%m-%d %H:%M:%S',
level=logging.INFO,
stream=sys.stdout)
# Get the logger
logger = logging.getLogger(__name__)
# Turn down the requests and urllib3 logging otherwise they fill the log
# file with mostly useless information
requests_logger = logging.getLogger("requests")
requests_logger.setLevel(logging.WARNING)
urllib3_logger = logging.getLogger("urllib3")
urllib3_logger.setLevel(logging.WARNING)
# Process the command line
(s_date, e_date) = parse_commandline()
try:
# Create the processor object
processor = NARR_AuxProcessor(s_date, e_date)
# Call the main processing routine
processor.archive_aux_data()
except Exception:
logger.exception('Processing Failed')
sys.exit(1) # EXIT FAILURE
sys.exit(0) # EXIT SUCCESS
def main():
'''
Description:
Ensures all data between start_date and end_date are up to date.
Precondition:
start_date and end_date are of type datetime.datetime
start_date and end_date can also be of type datetime.date
'''
# The config file is located in the same place as this script
Config.read_config(os.path.dirname(__file__))
cmd_args = parse_arguments()
setup_logging(cmd_args.debug, cmd_args.verbose)
logger = logging.getLogger(__name__)
try:
# Determine the data that exists within the date range
data = NarrData.get_next_narr_data_gen(cmd_args.start_date,
cmd_args.end_date)
# Determine which files are stale or missing internally.
data_to_be_updated = filter(lambda x: x.need_to_update(), data)
if len(data_to_be_updated) == 0:
logger.info('No data found for updating archive')
else:
logger.info('Will download {0} files'.format(
len(data_to_be_updated)))
if cmd_args.report:
report(list(data_to_be_updated))
else:
update(data_to_be_updated)
except Exception:
logger.exception('Processing Failed')
sys.exit(1) # EXIT FAILURE
sys.exit(0) # EXIT SUCCESS
def main():
'''
Description:
Ensures all data between start_date and end_date are up to date.
Precondition:
start_date and end_date are of type datetime.datetime
start_date and end_date can also be of type datetime.date
'''
# The config file is located in the same place as this script
Config.read_config(os.path.dirname(__file__))
cmd_args = parse_arguments()
setup_logging(cmd_args.debug, cmd_args.verbose)
logger = logging.getLogger(__name__)
try:
# Determine the data that exists within the date range
data = NarrData.get_next_narr_data_gen(cmd_args.start_date,
cmd_args.end_date)
# Determine which files are stale or missing internally.
data_to_be_updated = filter(lambda x: x.need_to_update(), data)
if len(data_to_be_updated) == 0:
logger.info('No data found for updating archive')
else:
logger.info('Will download {0} files'.
format(len(data_to_be_updated)))
if cmd_args.report:
report(list(data_to_be_updated))
else:
update(data_to_be_updated)
except Exception:
logger.exception('Processing Failed')
sys.exit(1) # EXIT FAILURE
sys.exit(0) # EXIT SUCCESS
def get_url(filename):
'''Return the URL for external retrieval of the file'''
return Config.get('ncep.url_format').format(filename)
def get_arch_dir(cls, year, month, day):
'''TODO TODO TODO'''
return (Config.get('archive_directory_format')
.format(cls.get_base_aux_dir(), year, month, day))
def parse_arguments():
'''
Description:
Parses arguments from the command line.
'''
version_number = Version.version_number()
default_date_range = int(Config.get('default_date_range'))
# Create a command line arugment parser
description = ('Downloads LST auxillary inputs, then archives them for'
' future use. Dates must be the in the format: "YYYYMMDD"')
parser = ArgumentParser(description=description,
formatter_class=ArgumentDefaultsHelpFormatter)
# ---- Add parameters ----
parser.add_argument('--start-date',
action='store', dest='start_date',
metavar='YYYYMMDD', type=input_date_validation,
required=False,
default=(date.today() -
timedelta(days=default_date_range)),
help='The start date of the date range of auxiliary'
' data to download.')
parser.add_argument('--end-date',
action='store', dest='end_date',
metavar='YYYYMMDD', type=input_date_validation,
required=False,
default=date.today(),
help='The end date of the date range of auxiliary'
' data to download.')
parser.add_argument('--date',
action='store', dest='date',
metavar='YYYYMMDD', type=input_date_validation,
required=False,
help='Sets both start and end date to this date.'
' Overrides start-date and end-date arguments.')
parser.add_argument('--report',
action='store_true', dest='report',
default=False,
help='Only report what will happen.')
parser.add_argument('--verbose',
action='store_true', dest='verbose',
default=False,
help='Turn verbose logging on.')
parser.add_argument('--debug',
action='store_true', dest='debug',
default=False,
help='Turn debug logging on.')
parser.add_argument('--version',
action='version',
version='%(prog)s {0}'.format(version_number),
help='Displays the version of the software.')
# Parse the command line parameters
args = parser.parse_args()
# Check if date was specified. If so then override start and end.
if args.date is not None:
args.start_date = args.date
args.end_date = args.date
return args
def parse_arguments():
'''
Description:
Parses arguments from the command line.
'''
version_number = Version.version_number()
default_date_range = int(Config.get('default_date_range'))
# Create a command line arugment parser
description = ('Downloads LST auxillary inputs, then archives them for'
' future use. Dates must be the in the format: "YYYYMMDD"')
parser = ArgumentParser(description=description,
formatter_class=ArgumentDefaultsHelpFormatter)
# ---- Add parameters ----
parser.add_argument('--start-date',
action='store',
dest='start_date',
metavar='YYYYMMDD',
type=input_date_validation,
required=False,
default=(date.today() -
timedelta(days=default_date_range)),
help='The start date of the date range of auxiliary'
' data to download.')
parser.add_argument('--end-date',
action='store',
dest='end_date',
metavar='YYYYMMDD',
type=input_date_validation,
required=False,
default=date.today(),
help='The end date of the date range of auxiliary'
' data to download.')
parser.add_argument('--date',
action='store',
dest='date',
metavar='YYYYMMDD',
type=input_date_validation,
required=False,
help='Sets both start and end date to this date.'
' Overrides start-date and end-date arguments.')
parser.add_argument('--report',
action='store_true',
dest='report',
default=False,
help='Only report what will happen.')
parser.add_argument('--verbose',
action='store_true',
dest='verbose',
default=False,
help='Turn verbose logging on.')
parser.add_argument('--debug',
action='store_true',
dest='debug',
default=False,
help='Turn debug logging on.')
parser.add_argument('--version',
action='version',
version='%(prog)s {0}'.format(version_number),
help='Displays the version of the software.')
# Parse the command line parameters
args = parser.parse_args()
# Check if date was specified. If so then override start and end.
if args.date is not None:
args.start_date = args.date
args.end_date = args.date
return args
def get_arch_filename(cls, variable, year, month, day, hour, ext):
'''TODO TODO TODO'''
return (Config.get('archive_name_format')
.format(variable, year, month, day, hour*100, ext))
def get_url(filename):
'''Return the URL for external retrieval of the file'''
return Config.get('ncep.url_format').format(filename)
def get_filename(year, month, day, hour):
'''Return the filename to grab on the external system'''
fmt = Config.get('ncep.name_format')
return fmt.format(year, month, day, hour)
def get_arch_dir(cls, year, month, day):
'''TODO TODO TODO'''
return (Config.get('archive_directory_format').format(
cls.get_base_aux_dir(), year, month, day))
def get_filename(year, month, day, hour):
'''Return the filename to grab on the external system'''
fmt = Config.get('ncep.name_format')
return fmt.format(year, month, day, hour)
def process_grib_for_variable(self, variable, grib_file):
'''
Description:
Extract the specified variable from the grib file and archive it.
'''
self.logger.info("Processing [{0}]".format(grib_file))
# Get the date information from the grib file
parts = grib_file.split('.')
year = int(parts[1][:4])
month = int(parts[1][4:6])
day = int(parts[1][6:8])
hour = int(parts[1][8:])
# Figure out the filenames to create
hdr_name = (Config.get('archive_name_format').format(
variable, year, month, day, hour * 100, 'hdr'))
grb_name = (Config.get('archive_name_format').format(
variable, year, month, day, hour * 100, 'grb'))
# Create inventory/header file to extract the variable data
cmd = ['wgrib', grib_file, '|', 'grep', variable, '>', hdr_name]
cmd = ' '.join(cmd)
self.logger.info('Executing [{0}]'.format(cmd))
output = System.execute_cmd(cmd)
if output is not None and len(output) > 0:
self.logger.info(output)
# Create grib files for each variable
cmd = [
'cat', hdr_name, '|', 'wgrib', grib_file, '-i', '-grib', '-o',
grb_name
]
cmd = ' '.join(cmd)
output = ''
self.logger.info('Executing [{0}]'.format(cmd))
output = System.execute_cmd(cmd)
if output is not None and len(output) > 0:
self.logger.info(output)
# Create new inventory/header file for the variable
cmd = ['wgrib', grb_name, '|', 'grep', variable, '>', hdr_name]
cmd = ' '.join(cmd)
self.logger.info('Executing [{0}]'.format(cmd))
output = System.execute_cmd(cmd)
if output is not None and len(output) > 0:
self.logger.info(output)
# Determine the directory to place the data and create it if it does
# not exist
dest_path = (Config.get('archive_directory_format').format(
self.base_aux_dir, year, month, day))
System.create_directory(dest_path)
# Archive the files
self.logger.info('Archiving into [{0}]'.format(dest_path))
# GRIB
dest_file = os.path.join(dest_path, grb_name)
shutil.copyfile(grb_name, dest_file)
# HEADER
dest_file = os.path.join(dest_path, hdr_name)
shutil.copyfile(hdr_name, dest_file)
# Cleanup the working directory
if os.path.exists(grb_name):
os.unlink(grb_name)
if os.path.exists(hdr_name):
os.unlink(hdr_name)
def process_grib_for_variable(self, variable, grib_file):
'''
Description:
Extract the specified variable from the grib file and archive it.
'''
self.logger.info("Processing [{0}]".format(grib_file))
# Get the date information from the grib file
parts = grib_file.split('.')
year = int(parts[1][:4])
month = int(parts[1][4:6])
day = int(parts[1][6:8])
hour = int(parts[1][8:])
# Figure out the filenames to create
hdr_name = (Config.get('archive_name_format')
.format(variable, year, month, day, hour*100, 'hdr'))
grb_name = (Config.get('archive_name_format')
.format(variable, year, month, day, hour*100, 'grb'))
# Create inventory/header file to extract the variable data
cmd = ['wgrib', grib_file, '|', 'grep', variable, '>', hdr_name]
cmd = ' '.join(cmd)
self.logger.info('Executing [{0}]'.format(cmd))
output = System.execute_cmd(cmd)
if output is not None and len(output) > 0:
self.logger.info(output)
# Create grib files for each variable
cmd = ['cat', hdr_name, '|',
'wgrib', grib_file, '-i', '-grib', '-o', grb_name]
cmd = ' '.join(cmd)
output = ''
self.logger.info('Executing [{0}]'.format(cmd))
output = System.execute_cmd(cmd)
if output is not None and len(output) > 0:
self.logger.info(output)
# Create new inventory/header file for the variable
cmd = ['wgrib', grb_name, '|', 'grep', variable, '>', hdr_name]
cmd = ' '.join(cmd)
self.logger.info('Executing [{0}]'.format(cmd))
output = System.execute_cmd(cmd)
if output is not None and len(output) > 0:
self.logger.info(output)
# Determine the directory to place the data and create it if it does
# not exist
dest_path = (Config.get('archive_directory_format')
.format(self.base_aux_dir, year, month, day))
System.create_directory(dest_path)
# Archive the files
self.logger.info('Archiving into [{0}]'.format(dest_path))
# GRIB
dest_file = os.path.join(dest_path, grb_name)
shutil.copyfile(grb_name, dest_file)
# HEADER
dest_file = os.path.join(dest_path, hdr_name)
shutil.copyfile(hdr_name, dest_file)
# Cleanup the working directory
if os.path.exists(grb_name):
os.unlink(grb_name)
if os.path.exists(hdr_name):
os.unlink(hdr_name)
def get_arch_filename(cls, variable, year, month, day, hour, ext):
'''TODO TODO TODO'''
return (Config.get('archive_name_format').format(
variable, year, month, day, hour * 100, ext))
