-
Notifications
You must be signed in to change notification settings - Fork 1
/
Copy pathdl_data.py
194 lines (159 loc) · 6.81 KB
/
dl_data.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
"""
Author: Ryan Grout
"""
import pandas as pd
import numpy as np
import requests
import argparse
from itertools import groupby, zip_longest
import pathlib
import random
import asyncio
import concurrent.futures
BASE_URL = "https://nwis.waterservices.usgs.gov/nwis/iv/"
HEADERS = {"Accept-Encoding": "gzip, compress"}
PARAMS = {"format": "json", "startDT": "1970-01-01", "parameterCd": [], "siteStatus": "all"}
#COLUMN_ORDER = ['siteCode', 'streamflow (ft^3/s)', 'sf qualifiers', 'gage height (ft)', 'gh qualifiers', 'longitude', 'latitude']
#COLUMN_ORDER = ['siteCode', 'Precip Total (in)', 'pt qualifiers', 'Physical Precip Total (in/wk)', 'ppr qualifiers', 'longitude', 'latitude']
#COLUMN_ORDER = ['siteCode', 'Elevation ocean/est (ft NAVD88)', 'elv qualifiers', 'longitude', 'latitude']
VAR_CODES = {'00060': 'streamflow (ft^3/s)', '00065': 'gage height (ft)',
'00045': 'Precip Total (in)',
'00046': 'Physical Precip Total (in/wk)',
'62620': "Elevation ocean/est (ft NAVD88)",
'62615': "Water Level (ft)",
'62614': "Water surface elevation above NGVD1929 (ft)"}
ABBREV = {'00060': 'sf', '00065': 'gh',
'00045': 'pt', '00046': 'ppr', '62620': 'elv', '62615': 'wl', '62614': 'wse'}
GROUP_SIZE = 1
def grouper(iterable, n, fillvalue=None):
"Collect data into fixed-length chunks or blocks"
# grouper('ABCDEFG', 3, 'x') --> ABC DEF Gxx"
args = [iter(iterable)] * n
chunks = zip_longest(*args, fillvalue=fillvalue)
for chunk in chunks:
yield filter(None, chunk)
def join_params(params, sep=','):
joined_params = {}
for k, v in params.items():
if isinstance(v, (list, tuple)):
joined_params[k] = ','.join(v)
else:
joined_params[k] = v
return joined_params
def make_df(variable):
# Columns
# dateTime (UTC), streamflow (ft^3/s), sf qualifiers, gage height (ft), gh qualifiers, siteCode, longitude, latitude
# Create the dataframe and cast to proper types.
df = pd.DataFrame.from_records(variable['values'][0]['value'])
#df['dateTime'] = pd.to_datetime(df['dateTime'], utc=True, format="%Y-%m-%dT%H:%M:%S.%f%z")
df['dateTime'] = pd.to_datetime(df['dateTime'], utc=True, infer_datetime_format=True)
duptimes = df['dateTime'].duplicated()
if not duptimes.empty:
df = df.loc[~duptimes]
df = df.set_index('dateTime', verify_integrity=True)
df.index = df.index.set_names("Date (utc)")
df['value'] = pd.to_numeric(df['value'])
df['qualifiers'] = df['qualifiers'].astype('string')
# set nodata values to nan
nodata = variable['variable']['noDataValue']
df[df['value'] == nodata] = np.nan
vcd = variable['variable']['variableCode'][0]['value']
renames = {'value': VAR_CODES[vcd], 'qualifiers': f"{ABBREV[vcd]} qualifiers"}
df = df.rename(columns=renames)
coords = variable['sourceInfo']['geoLocation']['geogLocation']
keys = {'siteCode': variable['sourceInfo']['siteCode'][0]['value'],
'longitude': float(coords['longitude']),
'latitude': float(coords['latitude'])}
return keys, df
def get_values(resp):
dfs = []
for v in resp['value']['timeSeries']:
if v and v['values'][0]['value']:
dfs.append(make_df(v))
return dfs
def groupby_sitecode_join(dfs):
keyfunc = lambda x: x[0]['siteCode']
dfs = sorted(dfs, key=keyfunc)
_dfs = []
for code, g in groupby(dfs, key=keyfunc):
frames = tuple(g)
try:
df = pd.concat((x[1] for x in frames), axis='columns', copy=False)
except:
breakpoint()
lon = frames[0][0]['longitude']
lat = frames[0][0]['latitude']
df[['longitude', 'latitude', 'siteCode']] = lon, lat, code
_dfs.append(df)
return pd.concat(_dfs, axis='index', copy=False)
def process_request(result, output):
if result.ok:
print(result.url)
dfs = get_values(result.json())
if dfs:
DF = groupby_sitecode_join(dfs)
print("Processing data...", len(DF), "records")
DF = DF.sort_index()
for code, df in DF.groupby('siteCode'):
fname = output / f"US{code}.csv.xz"
print("Writing to", fname)
try:
df.to_csv(fname)
except KeyboardInterrupt:
print("Removing partial file", fname.absolute())
fname.unlink()
break
else:
print("Finished processing", len(DF), "records")
def make_request2(params):
print("Requesting", params['sites'].count(',') + 1, "sites")
return requests.get(BASE_URL, params=params, headers=HEADERS)
def request_task(params, output):
response = make_request2(params)
process_request(response, output)
async def download_data(sites, output):
tasks = []
completed = 0
with concurrent.futures.ProcessPoolExecutor(max_workers=10) as pool:
for s in grouper(sites, GROUP_SIZE):
sitecodes = list(s)
params = PARAMS.copy()
params['sites'] = sitecodes
tasks.append(pool.submit(request_task, join_params(params), output))
#request_task(join_params(params), output)
for t in concurrent.futures.as_completed(tasks):
if exc := t.exception():
raise exc
else:
completed += 1
print("Completed", completed, "of", len(tasks), f"{round(100*(completed/len(tasks)), 2)}%")
#waited_tasks = concurrent.futures.wait(tasks, return_when=concurrent.futures.FIRST_EXCEPTION)
print("Done awaiting futures")
print("Shut down pool")
def get_options():
parser = argparse.ArgumentParser()
parser.add_argument('sites', type=pathlib.Path, help='List of sites to download')
parser.add_argument('-p', '--param', dest='params', action='append', help="Parameters to request")
parser.add_argument('-o', '--output', type=pathlib.Path, default=pathlib.Path(), help="output directory")
args = parser.parse_args()
return args
if __name__ == "__main__":
args = get_options()
PARAMS['parameterCd'] = args.params
sites = []
# Check if args.output exists. Create if not
args.output = args.output.resolve()
if not args.output.exists():
args.output.mkdir(parents=True)
print("Writing to", args.output)
print(PARAMS)
existing = set(x.name for x in args.output.glob('*.csv.xz'))
with open(args.sites, 'r') as fi:
for L in map(str.strip, fi):
if L.startswith('#'):
continue
if f"US{L}.csv.xz" not in existing:
sites.append(L)
print("Read", len(sites), "sites from", args.sites)
random.shuffle(sites)
asyncio.run(download_data(sites, args.output))