def parse(self, url, resp_text):
sel = Selector(resp_text)
directories = []
files = []
for anchor_sel in sel.css('a'):
link = anchor_sel.css('::attr(href)').extract_first()
if link.startswith('.'):
continue
link_url = f'{url}{link}'
if link.endswith('/'):
directories.append(link_url)
else:
fingerprint = anchor_sel.xpath(
'./following-sibling::text()[1]').extract_first()
match = re.match(r'\s*(\d+-\w+-\d+ \d+:\d+)\s+(\d+)',
fingerprint)
last_modified = dateutil.parser.parse(
match.group(1)).replace(tzinfo=tzutc())
file_size = int(match.group(2))
parser_cls = get_parser(link)
if parser_cls and not parser_cls(url=link_url).should_skip():
files.append({
'url': link_url,
'parser': parser_cls.__name__,
'last_modified': last_modified,
'file_size': file_size,
})
self.logger.debug("Found %d directories and %d files at %s",
len(directories), len(files), url)
yield from files
for dir_url in directories:
yield from self.poll_url(dir_url)
def poll(enqueue=False):
updated_files = DWDPoller().poll()
if enqueue:
from brightsky.worker import huey, process
if (expired_locks := huey.expire_locks(1800)):
logger.warning('Removed expired locks: %s',
', '.join(expired_locks))
pending_urls = [
t.args[0] for t in huey.pending() if t.name == 'process'
]
enqueued = 0
for updated_file in updated_files:
url = updated_file['url']
if url in pending_urls:
logger.debug('Skipping "%s": already queued', url)
continue
elif huey.is_locked(url):
logger.debug('Skipping "%s": already running', url)
continue
logger.debug('Enqueueing "%s"', url)
parser_cls = get_parser(os.path.basename(url))
process(url, priority=parser_cls.PRIORITY)
enqueued += 1
logger.info('Enqueued %d updated files for processing. Queue size: %d',
enqueued, enqueued + len(pending_urls))
def test_get_parser():
synop_with_timestamp = (
'Z__C_EDZW_20200617114802_bda01,synop_bufr_GER_999999_999999__MW_617'
'.json.bz2')
synop_latest = (
'Z__C_EDZW_latest_bda01,synop_bufr_GER_999999_999999__MW_XXX.json.bz2')
expected = {
'10minutenwerte_extrema_wind_00427_akt.zip':
(WindGustsObservationsParser),
'stundenwerte_FF_00011_akt.zip': WindObservationsParser,
'stundenwerte_FF_00090_akt.zip': WindObservationsParser,
'stundenwerte_N_01766_akt.zip': CloudCoverObservationsParser,
'stundenwerte_P0_00096_akt.zip': PressureObservationsParser,
'stundenwerte_RR_00102_akt.zip': PrecipitationObservationsParser,
'stundenwerte_SD_00125_akt.zip': SunshineObservationsParser,
'stundenwerte_TD_01766.zip': DewPointObservationsParser,
'stundenwerte_TU_00161_akt.zip': TemperatureObservationsParser,
'stundenwerte_VV_00161_akt.zip': VisibilityObservationsParser,
'MOSMIX_S_LATEST_240.kmz': MOSMIXParser,
'K611_-BEOB.csv': CurrentObservationsParser,
synop_with_timestamp: SYNOPParser,
synop_latest: None,
}
for filename, expected_parser in expected.items():
assert get_parser(filename) is expected_parser
def parse(path=None, url=None, export=False):
if not path and not url:
raise ValueError('Please provide either path or url')
parser_cls = get_parser(os.path.basename(path or url))
parser = parser_cls(path=path, url=url)
if url:
parser.download()
fingerprint = {
'url': url,
**dwd_fingerprint(parser.path),
}
else:
fingerprint = None
records = list(parser.parse())
parser.cleanup()
if export:
exporter = parser.exporter()
exporter.export(records, fingerprint=fingerprint)
return records
