Become more aggressive in parsing
This commit is contained in:
XANTRONIX Industrial
parent
51009d942c
commit
a9bdab632d
3 changed files with 103 additions and 36 deletions
|
|
@ -56,7 +56,7 @@ create table nexrad_afos_message (
|
||||||
phenom TEXT NOT NULL,
|
phenom TEXT NOT NULL,
|
||||||
sig TEXT NOT NULL,
|
sig TEXT NOT NULL,
|
||||||
etn INTEGER NOT NULL,
|
etn INTEGER NOT NULL,
|
||||||
body TEXT NOT NULL,
|
text_raw TEXT NOT NULL,
|
||||||
azimuth FLOAT NOT NULL,
|
azimuth FLOAT NOT NULL,
|
||||||
speed FLOAT NOT NULL,
|
speed FLOAT NOT NULL,
|
||||||
forecaster TEXT NOT NULL
|
forecaster TEXT NOT NULL
|
||||||
|
|
|
||||||
|
|
@ -12,11 +12,20 @@ from nexrad.vtec import VTECEvent
|
||||||
RE_ID = re.compile(r'^(\d+)$')
|
RE_ID = re.compile(r'^(\d+)$')
|
||||||
|
|
||||||
RE_ISSUANCE = re.compile(r'''
|
RE_ISSUANCE = re.compile(r'''
|
||||||
^ (W[A-Z]{3}\d{2})
|
^ ([A-Z]{4}\d+)
|
||||||
[ ]{1} (?P<wfo>[A-Z]{4})
|
\s+ (?P<wfo>[A-Z]{4})
|
||||||
[ ]{1} (?P<day>\d{2})
|
\s+ (?P<day>\d{2}) (?P<hour>\d{2}) (?P<minute>\d{2})
|
||||||
(?P<hour>\d{2}) (?P<minute>\d{2})
|
''', re.X)
|
||||||
$
|
|
||||||
|
RE_DATE = re.compile(r'''
|
||||||
|
^ (?P<hour>\d{1,2})
|
||||||
|
(?P<minute>\d{2})
|
||||||
|
\s+ (AM|PM)
|
||||||
|
\s+ (?P<tz>[A-Z]{3})
|
||||||
|
\s+ (?P<weekday>[A-Za-z]+)
|
||||||
|
\s+ (?P<month>[A-Za-z]+)
|
||||||
|
\s+ (?P<day>\d{1,2})
|
||||||
|
\s+ (?P<year>\d{4})
|
||||||
''', re.X)
|
''', re.X)
|
||||||
|
|
||||||
RE_PRODUCT = re.compile(r'^(?P<product>[A-Z]{3})(?P<wfo>[A-Z]{3})$')
|
RE_PRODUCT = re.compile(r'^(?P<product>[A-Z]{3})(?P<wfo>[A-Z]{3})$')
|
||||||
|
|
@ -27,14 +36,28 @@ RE_MOTION = re.compile(r'''
|
||||||
^ TIME
|
^ TIME
|
||||||
\.\.\. MOT
|
\.\.\. MOT
|
||||||
\.\.\. LOC
|
\.\.\. LOC
|
||||||
[ ]{1} (?P<hour>\d{2})(?P<minute>\d{2})Z
|
\s+ (?P<hour>\d{2})(?P<minute>\d{2})Z
|
||||||
[ ]{1} (?P<azimuth>\d+)DEG
|
\s+ (?P<azimuth>\d+)DEG
|
||||||
[ ]{1} (?P<speed>\d+)KT
|
\s+ (?P<speed>\d+)KT
|
||||||
[ ]{1} (?P<lat>\d+)
|
\s+ (?P<lat>\d+)
|
||||||
[ ]{1} (?P<lon>\d+)
|
\s+ (?P<lon>\d+)
|
||||||
$
|
$
|
||||||
''', re.X)
|
''', re.X)
|
||||||
|
|
||||||
|
MONTHS = {
|
||||||
|
'JAN': 1, 'FEB': 2, 'MAR': 3, 'APR': 4, 'MAY': 5, 'JUN': 6,
|
||||||
|
'JUL': 7, 'AUG': 8, 'SEP': 9, 'OCT': 10, 'NOV': 11, 'DEC': 12,
|
||||||
|
|
||||||
|
'JANUARY': 1, 'FEBRUARY': 2, 'MARCH': 3, 'APRIL': 4,
|
||||||
|
'MAY': 5, 'JUNE': 6, 'JULY': 7, 'AUGUST': 8,
|
||||||
|
'SEPTEMBER': 9, 'OCTOBER': 10, 'NOVEMBER': 11, 'DECEMBER': 12
|
||||||
|
}
|
||||||
|
|
||||||
|
TIMEZONES = {
|
||||||
|
'HST': -10, 'PST': -8, 'PDT': -7, 'MST': -7, 'MDT': -6, 'CST': -6,
|
||||||
|
'CDT': -5, ' EST': -5, 'EDT': -4, 'GMT': 0
|
||||||
|
}
|
||||||
|
|
||||||
def parse_lon(text: str):
|
def parse_lon(text: str):
|
||||||
size = len(text)
|
size = len(text)
|
||||||
return 0 - float(text[0:size-2]) + (float(text[size-2:size]) / 100)
|
return 0 - float(text[0:size-2]) + (float(text[size-2:size]) / 100)
|
||||||
|
|
@ -76,24 +99,28 @@ class AFOSMessage(DatabaseTable):
|
||||||
__columns__ = (
|
__columns__ = (
|
||||||
'id', 'timestamp_issued', 'timestamp_start', 'timestamp_end',
|
'id', 'timestamp_issued', 'timestamp_start', 'timestamp_end',
|
||||||
'serial', 'product', 'vtec_type', 'etn', 'actions', 'wfo',
|
'serial', 'product', 'vtec_type', 'etn', 'actions', 'wfo',
|
||||||
'phenom', 'sig', 'text', 'azimuth', 'speed', 'location',
|
'phenom', 'sig', 'text_raw', 'azimuth', 'speed', 'location',
|
||||||
'forecaster', 'poly',
|
'forecaster', 'poly',
|
||||||
)
|
)
|
||||||
|
|
||||||
__columns_read__ = {
|
__columns_read__ = {
|
||||||
'poly': 'ST_AsText(poly) as poly'
|
'poly': 'ST_AsText(poly) as poly',
|
||||||
|
'location': 'ST_AsText(location) as location'
|
||||||
}
|
}
|
||||||
|
|
||||||
__values_write__ = {
|
__values_write__ = {
|
||||||
'poly': shapely.from_wkt
|
'poly': shapely.from_wkt,
|
||||||
|
'location': shapely.from_wkt
|
||||||
}
|
}
|
||||||
|
|
||||||
__columns_write__ = {
|
__columns_write__ = {
|
||||||
'poly': 'ST_GeomFromText(:poly, {crs})'.format(crs=COORD_SYSTEM)
|
'poly': 'ST_GeomFromText(:poly, {crs})'.format(crs=COORD_SYSTEM),
|
||||||
|
'location': 'ST_GeomFromText(:location, {crs})'.format(crs=COORD_SYSTEM)
|
||||||
}
|
}
|
||||||
|
|
||||||
__values_write__ = {
|
__values_write__ = {
|
||||||
'poly': lambda v: {'poly': shapely.to_wkt(v)}
|
'poly': lambda v: {'poly': shapely.to_wkt(v)},
|
||||||
|
'location': lambda v: {'location': shapely.to_wkt(v)}
|
||||||
}
|
}
|
||||||
|
|
||||||
id: int
|
id: int
|
||||||
|
|
@ -110,7 +137,7 @@ class AFOSMessage(DatabaseTable):
|
||||||
phenom: str
|
phenom: str
|
||||||
sig: str
|
sig: str
|
||||||
etn: int
|
etn: int
|
||||||
body: str
|
text_raw: str
|
||||||
azimuth: int
|
azimuth: int
|
||||||
speed: int
|
speed: int
|
||||||
location: shapely.Point
|
location: shapely.Point
|
||||||
|
|
@ -133,7 +160,7 @@ class AFOSMessage(DatabaseTable):
|
||||||
self.phenom = None
|
self.phenom = None
|
||||||
self.sig = None
|
self.sig = None
|
||||||
self.etn = None
|
self.etn = None
|
||||||
self.body = None
|
self.text_raw = None
|
||||||
self.azimuth = None
|
self.azimuth = None
|
||||||
self.speed = None
|
self.speed = None
|
||||||
self.location = None
|
self.location = None
|
||||||
|
|
@ -143,35 +170,24 @@ class AFOSMessage(DatabaseTable):
|
||||||
@staticmethod
|
@staticmethod
|
||||||
def parse(text: str) -> Self:
|
def parse(text: str) -> Self:
|
||||||
message = AFOSMessage()
|
message = AFOSMessage()
|
||||||
message.text = text
|
message.text_raw = text
|
||||||
|
|
||||||
state = AFOSMessageParserState.SERIAL
|
state = AFOSMessageParserState.SERIAL
|
||||||
|
|
||||||
|
issuance = None
|
||||||
|
timestamp_inline = None
|
||||||
|
|
||||||
for line in text.split('\n'):
|
for line in text.split('\n'):
|
||||||
line = line.rstrip()
|
line = line.rstrip()
|
||||||
|
|
||||||
if state == AFOSMessageParserState.SERIAL:
|
|
||||||
match = RE_ID.match(line)
|
|
||||||
|
|
||||||
if match is not None:
|
|
||||||
message.serial = int(match[1])
|
|
||||||
state = AFOSMessageParserState.ISSUANCE
|
|
||||||
elif state == AFOSMessageParserState.ISSUANCE:
|
|
||||||
match = RE_ISSUANCE.match(line)
|
|
||||||
|
|
||||||
if match is not None:
|
|
||||||
state = AFOSMessageParserState.PRODUCT
|
|
||||||
elif state == AFOSMessageParserState.PRODUCT:
|
|
||||||
match = RE_PRODUCT.match(line)
|
|
||||||
|
|
||||||
if match is not None:
|
|
||||||
message.product = match['product']
|
|
||||||
|
|
||||||
state = AFOSMessageParserState.BODY
|
|
||||||
elif state == AFOSMessageParserState.BODY:
|
|
||||||
if line == '':
|
if line == '':
|
||||||
continue
|
continue
|
||||||
elif line[0] == '/':
|
elif line[0] == '/' and line[-1] == '/':
|
||||||
|
#
|
||||||
|
# The VTEC line can appear anywhere in the message
|
||||||
|
# text, therefore, parsing must be able to occur in
|
||||||
|
# all states.
|
||||||
|
#
|
||||||
vtec = VTECEvent.parse(line)
|
vtec = VTECEvent.parse(line)
|
||||||
|
|
||||||
if vtec is not None:
|
if vtec is not None:
|
||||||
|
|
@ -184,7 +200,43 @@ class AFOSMessage(DatabaseTable):
|
||||||
message.phenom = vtec.phenom
|
message.phenom = vtec.phenom
|
||||||
message.sig = vtec.sig
|
message.sig = vtec.sig
|
||||||
message.etn = vtec.etn
|
message.etn = vtec.etn
|
||||||
elif line == '&&':
|
|
||||||
|
if state == AFOSMessageParserState.SERIAL:
|
||||||
|
match = RE_ID.match(line)
|
||||||
|
|
||||||
|
if match is not None:
|
||||||
|
message.serial = int(match[1])
|
||||||
|
state = AFOSMessageParserState.ISSUANCE
|
||||||
|
elif state == AFOSMessageParserState.ISSUANCE:
|
||||||
|
match = RE_ISSUANCE.match(line)
|
||||||
|
|
||||||
|
if match is not None:
|
||||||
|
state = AFOSMessageParserState.PRODUCT
|
||||||
|
issuance = match
|
||||||
|
elif state == AFOSMessageParserState.PRODUCT:
|
||||||
|
match = RE_PRODUCT.match(line)
|
||||||
|
|
||||||
|
if match is not None:
|
||||||
|
message.product = match['product']
|
||||||
|
|
||||||
|
state = AFOSMessageParserState.BODY
|
||||||
|
elif state == AFOSMessageParserState.BODY:
|
||||||
|
if timestamp_inline is None:
|
||||||
|
match = RE_DATE.match(line)
|
||||||
|
|
||||||
|
if match is not None:
|
||||||
|
offset = TIMEZONES[match['tz'].upper()]
|
||||||
|
timestamp_inline = datetime.datetime(
|
||||||
|
year = int(match['year']),
|
||||||
|
month = MONTHS[match['month'].upper()],
|
||||||
|
day = int(match['day']),
|
||||||
|
hour = int(match['hour']),
|
||||||
|
minute = int(match['minute']),
|
||||||
|
second = 0,
|
||||||
|
tzinfo = datetime.timezone(datetime.timedelta(hours=offset))
|
||||||
|
).astimezone(datetime.UTC)
|
||||||
|
|
||||||
|
if line == '&&':
|
||||||
state = AFOSMessageParserState.TAGS
|
state = AFOSMessageParserState.TAGS
|
||||||
elif state == AFOSMessageParserState.TAGS:
|
elif state == AFOSMessageParserState.TAGS:
|
||||||
if line == '$$':
|
if line == '$$':
|
||||||
|
|
@ -205,6 +257,22 @@ class AFOSMessage(DatabaseTable):
|
||||||
if line != '':
|
if line != '':
|
||||||
message.forecaster = line
|
message.forecaster = line
|
||||||
|
|
||||||
|
if message.timestamp_issued is None:
|
||||||
|
if timestamp_inline is not None:
|
||||||
|
message.timestamp_issued = timestamp_inline
|
||||||
|
message.timestamp_start = timestamp_inline
|
||||||
|
message.timestamp_end = timestamp_inline + datetime.timedelta(hours=1)
|
||||||
|
else:
|
||||||
|
message.timestamp_issued = datetime.datetime(
|
||||||
|
year = message.timestamp_start.year,
|
||||||
|
month = message.timestamp_start.month,
|
||||||
|
day = int(issuance['day']),
|
||||||
|
hour = int(issuance['hour']),
|
||||||
|
minute = int(issuance['minute']),
|
||||||
|
second = 0,
|
||||||
|
tzinfo = datetime.UTC
|
||||||
|
)
|
||||||
|
|
||||||
return message
|
return message
|
||||||
|
|
||||||
def is_watch(self):
|
def is_watch(self):
|
||||||
|
|
|
||||||
|
|
@ -91,8 +91,7 @@ class VTECEvent():
|
||||||
|
|
||||||
class VTECHydroEvent():
|
class VTECHydroEvent():
|
||||||
__slots__ = (
|
__slots__ = (
|
||||||
'severity', 'cause', 'record',
|
'severity', 'cause', 'record', 'timestamp_start', 'timestamp_end'
|
||||||
'timestamp_start', 'timestamp_end'
|
|
||||||
)
|
)
|
||||||
|
|
||||||
severity: str
|
severity: str
|
||||||
|
|
|
||||||
Loading…
Add table
Reference in a new issue