def test_parse_metar_df_positional_datetime_failure():
"""Test that positional year, month arguments fail for parse_metar_to_dataframe."""
# pylint: disable=too-many-function-args
with pytest.raises(TypeError,
match='takes 1 positional argument but 3 were given'):
parse_metar_to_dataframe(
'K6B0 261200Z AUTO 00000KT 10SM CLR 20/M17'
'A3002 RMK AO2 T01990165=', 2019, 6)
def test_missing_temp_dewp():
"""Test when missing both temperature and dewpoint."""
df = parse_metar_to_dataframe(
'KIOW 011152Z AUTO A3006 RMK AO2 SLPNO 70020 51013 PWINO=')
assert_almost_equal(df.air_temperature.values, np.nan)
assert_almost_equal(df.dew_point_temperature.values, np.nan)
assert_almost_equal(df.cloud_coverage.values, 10)
def test_broken_clouds():
"""Test for skycover when there are broken clouds."""
df = parse_metar_to_dataframe(
'METAR KLOT 261155Z AUTO 00000KT 10SM BKN100 05/00 A3001 '
'RMK AO2=')
assert df.low_cloud_type.values == 'BKN'
assert df.cloud_coverage.values == 6
def test_few_clouds_():
"""Test for skycover when there are few clouds."""
df = parse_metar_to_dataframe(
'METAR KMKE 266155Z AUTO /////KT 10SM FEW100 05/00 A3001 '
'RMK AO2=')
assert df.low_cloud_type.values == 'FEW'
assert df.cloud_coverage.values == 2
assert_almost_equal(df.wind_direction.values, np.nan)
assert_almost_equal(df.wind_speed.values, np.nan)
assert_almost_equal(df.date_time.values, np.nan)
def test_named_tuple_test1():
"""Test the named tuple parsing function."""
df = parse_metar_to_dataframe(
'KDEN 012153Z 09010KT 10SM FEW060 BKN110 BKN220 27/13 '
'A3010 RMK AO2 LTG DSNT SW AND W SLP114 OCNL LTGICCG '
'DSNT SW CB DSNT SW MOV E T02670128')
assert df.wind_direction.values == 90
assert df.wind_speed.values == 10
assert df.air_temperature.values == 27
assert df.dew_point_temperature.values == 13
def test_date_time_given():
"""Test for when date_time is given."""
df = parse_metar_to_dataframe(
'K6B0 261200Z AUTO 00000KT 10SM CLR 20/M17 A3002 RMK AO2 '
'T01990165=',
year=2019,
month=6)
assert_equal(df['date_time'][0], datetime(2019, 6, 26, 12))
assert df.eastward_wind.values == 0
assert df.northward_wind.values == 0
def test_parse_no_pint_objects_in_df():
"""Test that there are no Pint quantities in dataframes created by parser."""
input_file = get_test_data('metar_20190701_1200.txt', mode='rt')
metar_str = (
'KSLK 011151Z AUTO 21005KT 1/4SM FG VV002 14/13 A1013 RMK AO2 SLP151 70043 '
'T01390133 10139 20094 53002=')
for df in (parse_metar_file(input_file),
parse_metar_to_dataframe(metar_str)):
for column in df:
assert not isinstance(df[column][0], units.Quantity)
def test_station_id_not_in_dictionary():
"""Test for when the METAR does not correspond to a station in the dictionary."""
df = parse_metar_to_dataframe(
'METAR KLBG 261155Z AUTO 00000KT 10SM CLR 05/00 A3001 RMK '
'AO2=')
assert df.station_id.values == 'KLBG'
assert_almost_equal(df.latitude.values, np.nan)
assert_almost_equal(df.longitude.values, np.nan)
assert_almost_equal(df.elevation.values, np.nan)
assert_almost_equal(df.wind_direction.values, 0)
assert_almost_equal(df.wind_speed.values, 0)
def test_date_time_given():
"""Test for when date_time is given."""
df = parse_metar_to_dataframe(
'K6B0 261200Z AUTO 00000KT 10SM CLR 20/M17 A3002 RMK AO2 '
'T01990165=',
year=2019,
month=6)
assert df.date_time[0] == datetime(2019, 6, 26, 12)
assert df.eastward_wind[0] == 0
assert df.northward_wind[0] == 0
assert_almost_equal(df.air_pressure_at_sea_level[0], 1016.56)
assert_almost_equal(df.visibility.values, 16093.44)
def test_parse_metar_to_dataframe():
"""Test parsing a single METAR to a DataFrame."""
df = parse_metar_to_dataframe(
'KDEN 012153Z 09010KT 10SM FEW060 BKN110 BKN220 27/13 '
'A3010 RMK AO2 LTG DSNT SW AND W SLP114 OCNL LTGICCG '
'DSNT SW CB DSNT SW MOV E T02670128')
assert df.wind_direction.values == 90
assert df.wind_speed.values == 10
assert_almost_equal(df.eastward_wind.values, -10)
assert_almost_equal(df.northward_wind.values, 0)
assert_almost_equal(df.visibility.values, 16093.44)
assert df.air_temperature.values == 27
assert df.dew_point_temperature.values == 13
def test_missing_values():
"""Test for missing values from nearly every field."""
df = parse_metar_to_dataframe(
'METAR KBOU 011152Z AUTO 02006KT //// // ////// 42/02 '
'Q1004=')
assert_almost_equal(df.current_wx1.values, np.nan)
assert_almost_equal(df.current_wx2.values, np.nan)
assert_almost_equal(df.current_wx3.values, np.nan)
assert_almost_equal(df.current_wx1_symbol.values, 0)
assert_almost_equal(df.current_wx2_symbol.values, 0)
assert_almost_equal(df.current_wx3_symbol.values, 0)
assert_almost_equal(df.low_cloud_type.values, np.nan)
assert_almost_equal(df.medium_cloud_type.values, np.nan)
assert_almost_equal(df.high_cloud_type.values, np.nan)
def test_all_weather_given():
"""Test when all possible weather slots are given."""
df = parse_metar_to_dataframe(
'METAR RJOI 261155Z 00000KT 4000 -SHRA BR VCSH BKN009 '
'BKN015 OVC030 OVC040 22/21 A2987 RMK SHRAB35E44 SLP114 '
'VCSH S-NW P0000 60021 70021 T02220206 10256 20211 55000=')
assert df.station_id.values == 'RJOI'
assert_almost_equal(df.latitude.values, 34.14, decimal=2)
assert_almost_equal(df.longitude.values, 132.22, decimal=2)
assert df.current_wx1.values == '-SHRA'
assert df.current_wx2.values == 'BR'
assert df.current_wx3.values == 'VCSH'
assert df.low_cloud_type.values == 'BKN'
assert df.low_cloud_level.values == 900
assert df.high_cloud_type.values == 'OVC'
assert df.high_cloud_level.values == 3000
def parse_metar_file(file, wx_subset=True):
"""
Parses METAR file from NCDC
Input:
--------
file = Text file downloaded from NCDC
wx_subset = Flag to determine whether or not to drop non-current weather obs (if True, only returns obs with observed weather)
Output:
--------
df = Pandas dataframe filtered for times where current weather is not 'nan'
"""
# Read in the file using Pandas
df = pd.read_csv(file, header=None)
# Pull the timestamp from the filename
timestamp = datetime.strptime(file[-10:], '%Y%m.dat')
# Iterrate over rows to parse METARS
df_list = []
for index, row in df.iterrows():
try:
df_list.append(
parse_metar_to_dataframe(row.values[0][52:],
year=timestamp.year,
month=timestamp.month))
except:
print('Error with METAR: ', row.values[0][52:])
#
merged_df = pd.concat(df_list)
# Drop datasets that do not include current weather
merged_df = merged_df.dropna(subset=['current_wx1'])
# Change the index to datetime
merged_df.index = merged_df.date_time
# Return the merged dataset sorted by datetime
return merged_df.sort_index()
def test_vertical_vis():
"""Test for when vertical visibility is given."""
df = parse_metar_to_dataframe(
'KSLK 011151Z AUTO 21005KT 1/4SM FG VV002 14/13 A1013 RMK '
'AO2 SLP151 70043 T01390133 10139 20094 53002=')
assert df.low_cloud_type.values == 'VV'
