def test_lookup_asdataframe():
"""lookup - asdataframe"""
r = xy.lookup(test_points, asdataframe=True)
assert r.shape[0] == len(test_points)
assert r.shape[1] >= 4
assert 'DataFrame' in str(type(r))
assert isinstance(r, pd.DataFrame)
r = xy.lookup(test_points,
shoredistance=True,
grids=False,
areas=False,
asdataframe=True)
assert r.shape[0] == len(test_points)
assert r.shape[1] == 1
r = xy.lookup(test_points,
shoredistance=False,
grids=True,
areas=False,
asdataframe=True)
assert r.shape[0] == len(test_points)
assert r.shape[1] >= 3
r = xy.lookup(test_points,
shoredistance=False,
grids=False,
areas=True,
asdataframe=True)
assert r.shape[0] == len(test_points)
assert r.shape[1] == 1
r = xy.lookup(test_points,
shoredistance=True,
grids=True,
areas=True,
asdataframe=True)
assert r.shape[0] == len(test_points)
assert r.shape[1] >= 5
def test_lookup_duplicate_points():
"""Lookup duplicate points"""
points = test_points * 10000
r = xy.lookup(points,
shoredistance=True,
grids=False,
areas=False,
asdataframe=False)
assert len(r) == len(points)
r = xy.lookup(points,
shoredistance=True,
grids=False,
areas=False,
asdataframe=True)
assert r.shape[0] == len(points)
assert r.shape[1] >= 1
def test_lookup_na_points():
"""Lookup NA points with duplicates"""
import math
points = [[0, 1], [float('nan'), 3], [4, float('nan')], [5, 6]] * 3
r = xy.lookup(points,
shoredistance=True,
grids=False,
areas=False,
asdataframe=False)
assert len(r) == len(points)
for i in [1, 2, 5, 6, 9, 10]:
assert r[i] == {}
r = xy.lookup(points,
shoredistance=True,
grids=True,
areas=True,
asdataframe=True)
assert r.shape[0] == len(points)
assert r.shape[1] >= 1
for i in [1, 2, 5, 6, 9, 10]:
assert math.isnan(r["shoredistance"][i])
def test_lookup_na_points():
"""Lookup NA points"""
import math
points = [[0, 1], [float('nan'), 3], [4, float('nan')], [5, 6]]
r = xy.lookup(points,
shoredistance=True,
grids=False,
areas=False,
asdataframe=False)
assert len(r) == len(points)
assert r[1] == {}
assert r[2] == {}
r = xy.lookup(points,
shoredistance=True,
grids=True,
areas=True,
asdataframe=True)
assert r.shape[0] == len(points)
assert r.shape[1] >= 1
assert math.isnan(r["shoredistance"][1])
assert math.isnan(r["shoredistance"][2])
def test_lookup_many_points():
"""Lookup many points"""
import random
random.seed(42)
points = [[random.uniform(-180, 180),
random.uniform(-90, 90)] for _ in range(50000)]
r = xy.lookup(points,
shoredistance=True,
grids=True,
areas=True,
asdataframe=True)
assert r.shape[0] == len(points)
assert r.shape[1] >= 5
def environmental(points, duplicate_indices, mad_coef, iqr_coef, qcstats=None, return_values=False):
if mad_coef is None:
mad_coef = MAD_COEF
if iqr_coef is None:
iqr_coef = IQR_COEF
env = xy.lookup(points, shoredistance=False, grids=True, areas=False, asdataframe=True)
qc = {}
for grid in ['bathymetry', 'sssalinity', 'sstemperature']:
values = env.get(grid, np.full(len(points), np.nan))
if qcstats is None:
median, mad, q1, q3 = stats.get_values_stats(values)
else:
median, mad, q1, q3 = qcstats[grid]
qc[grid] = _values_qc(values, median, mad, q1, q3, mad_coef, iqr_coef, duplicate_indices, return_values)
return qc
def do_xylookup(results):
output = [None] * len(results)
indices = []
coordinates = []
for i in range(len(results)):
result = results[i]
if "decimalLongitude" in result[
"annotations"] and "decimalLatitude" in result["annotations"]:
indices.append(i)
coordinates.append([
result["annotations"]["decimalLongitude"],
result["annotations"]["decimalLatitude"]
])
if len(coordinates) > 0:
xy = pyxylookup.lookup(coordinates,
shoredistance=True,
grids=True,
areas=True)
for i in range(len(indices)):
output[indices[i]] = xy[i]
return output
def do_xylookup(records):
""" derived from equivalent in obis-qc - takes a list of records already QCd
for LAT - LON presence and validity - QC field must
be present
:param - records IMPORTANT - with VALID LAT, LON
:returns - a list of result records as retrieved from pyxylookup for each point """
import pyxylookup as pxy # and then use pyxylookup.lookup
output = [None] * len(records)
indices = []
coordinates = []
performLookup = False
for i in range(len(records)):
record = records[i]
# The record has been already checked LAT LON validity, but verify anyway...
if "decimalLongitude" in record and \
"decimalLatitude" in record and \
"qc" in record and \
record["qc"] & qc_flags.QCFlag.GEO_LAT_LON_VALID.bitmask and \
record["qc"] & qc_flags.QCFlag.GEO_LAT_LON_PRESENT.bitmask:
indices.append(i)
lon = check_float(record["decimalLongitude"])["float"]
lat = check_float(record["decimalLatitude"])["float"]
coordinates.append([lon, lat])
if lat is not None and lon is not None:
performLookup = True
if performLookup:
xy = pxy.lookup(coordinates,
shoredistance=True,
grids=True,
areas=True)
for i in range(len(indices)):
output[indices[i]] = xy[i]
return output
def test_lookup_wrong_points_error(points):
xy.lookup(points)
def check_lookup_all_false_error(asdataframe):
xy.lookup(test_points,
shoredistance=False,
grids=False,
areas=False,
asdataframe=asdataframe)
def test_lookup_points_tuples_list():
"""lookup - points tuples list"""
test_points_tuples = [(4, 5), (6, 7)]
r = xy.lookup(test_points_tuples)
assert len(r) == 2
def test_lookup_pandas_points():
"""lookup - numpy points"""
points = np.asarray(test_points)
r = xy.lookup(points)
assert len(r) == len(test_points)
def test_lookup_points_array():
"""lookup - points array"""
r = xy.lookup(test_points)
assert len(r) == len(test_points)
