Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Change in behaviour, WSI no longer estimated by default, arg now requ… #26

Merged
merged 3 commits into from
Feb 6, 2024
Merged

Change in behaviour, WSI no longer estimated by default, arg now requ… #26

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
0dbfce1
Change in behaviour, WSI no longer estimated by default, arg now requ…
david-i-berry Nov 22, 2023
7f7bb1b
Revert of JSON schema.
david-i-berry Feb 6, 2024
88f9e78
Update test_bufr2geojson.py
tomkralidis Feb 6, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
105 changes: 65 additions & 40 deletions bufr2geojson/__init__.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -418,14 +418,14 @@ def get_time(self) -> str:

return time_

def get_wsi(self) -> str:
def get_wsi(self, guess_wsi: bool = False) -> str:
"""
Function returns WIGOS station ID as string

:returns: WIGOS station ID.
"""

return self.get_identification()["wsi"]
return self.get_identification(guess_wsi)["wsi"]

def get_tsi(self) -> str:
"""
Expand All @@ -436,13 +436,16 @@ def get_tsi(self) -> str:

return self.get_identification()["tsi"]

def get_identification(self) -> dict:
def get_identification(self, guess_wsi: bool = False) -> dict:
"""
Function extracts identification information from qualifiers.

:returns: dictionary containing any class 01 qualifiers and WSI as dict. # noqa
"""

# default WSI value
wsi = None

# see https://library.wmo.int/doc_num.php?explnum_id=11021
# page 19 for allocation of WSI if not set
# check to see what identification we have
Expand All @@ -468,27 +471,35 @@ def get_identification(self) -> dict:
if all(x in self.qualifiers["01"] for x in _types): # noqa
block = self.get_qualifier("01", "block_number")
station = self.get_qualifier("01", "station_number")
wsi_series = 0
wsi_issuer = 20000
wsi_number = 0
wsi_local = strip2(f"{block:02d}{station:03d}")
tsi = strip2(f"{block:02d}{station:03d}")
if guess_wsi:
wsi_series = 0
wsi_issuer = 20000
wsi_number = 0
wsi_local = tsi
wsi = f"{wsi_series}-{wsi_issuer}-{wsi_number}-{wsi_local}"

return {
"wsi": f"{wsi_series}-{wsi_issuer}-{wsi_number}-{wsi_local}",
"tsi": wsi_local,
"wsi": wsi,
"tsi": tsi,
"type": "{}_and_{}".format(*_types)
}

# ship or mobile land station identifier (001011)
_type = "ship_or_mobile_land_station_identifier"
if _type in self.qualifiers["01"]:
callsign = self.get_qualifier("01", _type)
wsi_series = 0
wsi_issuer = 20004
wsi_number = 0
wsi_local = strip2(callsign)
tsi = strip2(callsign)
if guess_wsi:
wsi_series = 0
wsi_issuer = 20004
wsi_number = 0
wsi_local = tsi
wsi = f"{wsi_series}-{wsi_issuer}-{wsi_number}-{wsi_local}"

return {
"wsi": f"{wsi_series}-{wsi_issuer}-{wsi_number}-{wsi_local}",
"tsi": wsi_local,
"wsi": wsi,
"tsi": tsi,
"type": _type
}

Expand All @@ -500,13 +511,17 @@ def get_identification(self) -> dict:
wmo_region = self.get_qualifier("region_number")
wmo_subregion = self.get_qualifier("wmo_region_sub_area")
wmo_number = self.get_qualifier("buoy_or_platform_identifier")
wsi_series = 0
wsi_issuer = 20002
wsi_number = 0
wsi_local = strip2(f"{wmo_region:01d}{wmo_subregion:01d}{wmo_number:05d}") # noqa
tsi = strip2(f"{wmo_region:01d}{wmo_subregion:01d}{wmo_number:05d}") # noqa
if guess_wsi:
wsi_series = 0
wsi_issuer = 20002
wsi_number = 0
wsi_local = tsi # noqa
wsi = f"{wsi_series}-{wsi_issuer}-{wsi_number}-{wsi_local}"

return {
"wsi": f"{wsi_series}-{wsi_issuer}-{wsi_number}-{wsi_local}",
"tsi": wsi_local,
"wsi": wsi,
"tsi": tsi,
"type": "5_digit_marine_observing_platform_identifier"
}

Expand All @@ -515,13 +530,17 @@ def get_identification(self) -> dict:
_type = "stationary_buoy_platform_identifier_e_g_c_man_buoys"
if _type in self.qualifiers["01"]:
id_ = self.get_qualifier("01", _type)
wsi_series = 0
wsi_issuer = 20002
wsi_number = 0
wsi_local = strip2(id_)
tsi = strip2(id_)
if guess_wsi:
wsi_series = 0
wsi_issuer = 20002
wsi_number = 0
wsi_local = tsi
wsi = f"{wsi_series}-{wsi_issuer}-{wsi_number}-{wsi_local}"

return {
"wsi": f"{wsi_series}-{wsi_issuer}-{wsi_number}-{wsi_local}",
"tsi": wsi_local,
"wsi": wsi,
"tsi": tsi,
"type": _type
}

Expand All @@ -530,13 +549,17 @@ def get_identification(self) -> dict:
_type = "marine_observing_platform_identifier"
if _type in self.qualifiers["01"]:
id_ = self.get_qualifier("01", _type)
wsi_series = 0
wsi_issuer = 20002
wsi_number = 0
wsi_local = strip2(id_)
tsi = strip2(id_)
if guess_wsi:
wsi_series = 0
wsi_issuer = 20002
wsi_number = 0
wsi_local = tsi
wsi = f"{wsi_series}-{wsi_issuer}-{wsi_number}-{wsi_local}"

return {
"wsi": f"{wsi_series}-{wsi_issuer}-{wsi_number}-{wsi_local}",
"tsi": wsi_local,
"wsi": wsi,
"tsi": tsi,
"type": "7_digit_marine_observing_platform_identifier"
}

Expand Down Expand Up @@ -577,7 +600,7 @@ def get_code_value(self, fxxyyy: str, code: int) -> str:
return decoded

def as_geojson(self, bufr_handle: int, id: str,
serialize: bool = False) -> dict:
serialize: bool = False, guess_wsi: bool = False) -> dict:
"""
Function to return GeoJSON representation of BUFR message

Expand Down Expand Up @@ -730,7 +753,7 @@ def as_geojson(self, bufr_handle: int, id: str,
attributes, append)
continue
if value is not None:
self.get_identification()
# self.get_identification()
metadata = self.get_qualifiers()
metadata_hash = hashlib.md5(json.dumps(metadata).encode("utf-8")).hexdigest() # noqa
md = {
Expand All @@ -739,7 +762,7 @@ def as_geojson(self, bufr_handle: int, id: str,
}
for idx in range(len(metadata)):
md["metadata"].append(metadata[idx])
wsi = self.get_wsi()
wsi = self.get_wsi(guess_wsi)
feature_id = f"WIGOS_{wsi}_{characteristic_date}T{characteristic_time}" # noqa
feature_id = f"{feature_id}{id}-{index}"
phenomenon_time = self.get_time()
Expand Down Expand Up @@ -787,12 +810,14 @@ def as_geojson(self, bufr_handle: int, id: str,
return data


def transform(data: bytes, serialize: bool = False) -> Iterator[dict]:
def transform(data: bytes, serialize: bool = False,
guess_wsi: bool = False) -> Iterator[dict]:
"""
Main transformation

:param data: byte string of BUFR data
:param serialize: whether to return as JSON string (default is False)
:param guess_wsi: whether to 'guess' WSI based on TSI and allocaiotn rules

:returns: `generator` of GeoJSON features
"""
Expand All @@ -804,8 +829,7 @@ def transform(data: bytes, serialize: bool = False) -> Iterator[dict]:
with open(tmp.name, 'wb') as f:
f.write(data)

# check data type, only in situ supported
# not yet implemented
# check data type, only in situ supported (not yet implemented)
# split subsets into individual messages and process
imsg = 0
messages_remaining = True
Expand Down Expand Up @@ -847,7 +871,8 @@ def transform(data: bytes, serialize: bool = False) -> Iterator[dict]:
tag = f"-{idx}"
try:
data = parser.as_geojson(single_subset, id=tag,
serialize=serialize)
serialize=serialize,
guess_wsi=guess_wsi) # noqa

except Exception as e:
LOGGER.error("Error parsing BUFR to GeoJSON, no data written") # noqa
Expand Down
2 changes: 1 addition & 1 deletion bufr2geojson/resources/schemas/wmo-om-profile-geojson.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -99,4 +99,4 @@ examples:
type: application/geo+json
hreflang: en
title: 'Trierer Strasse 70, 53115 Bonn'
length: 0
length: 0
5 changes: 3 additions & 2 deletions tests/test_bufr2geojson.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -128,7 +128,7 @@ def geojson_output():


def test_multi(multimsg_bufr):
results = transform(multimsg_bufr)
results = transform(multimsg_bufr, guess_wsi=True)
# count number of geojsons
icount = 0
for res in results:
Expand All @@ -140,7 +140,8 @@ def test_multi(multimsg_bufr):
def test_transform(geojson_schema, geojson_output):
test_bufr_file = 'A_ISIA21EIDB202100_C_EDZW_20220320210902_11839953.bin'
with open(test_bufr_file, 'rb') as fh:
messages1, messages2 = itertools.tee(transform(fh.read()))
messages1, messages2 = itertools.tee(transform(fh.read(),
guess_wsi=True))

# validate against JSON Schema
for message in messages1:
Expand Down
Loading