pthelma 0.99.3.dev0__cp311-cp311-musllinux_1_2_x86_64.whl
Sign up to get free protection for your applications and to get access to all the features.
- enhydris_api_client/__init__.py +252 -0
- enhydris_cache/__init__.py +5 -0
- enhydris_cache/cli.py +150 -0
- enhydris_cache/enhydris_cache.py +69 -0
- evaporation/__init__.py +4 -0
- evaporation/cli.py +729 -0
- evaporation/evaporation.py +437 -0
- haggregate/__init__.py +5 -0
- haggregate/cli.py +91 -0
- haggregate/haggregate.py +155 -0
- haggregate/regularize.c +15105 -0
- haggregate/regularize.cpython-311-x86_64-linux-musl.so +0 -0
- haggregate/regularize.pyx +193 -0
- hspatial/__init__.py +4 -0
- hspatial/cli.py +310 -0
- hspatial/hspatial.py +425 -0
- hspatial/test.py +27 -0
- htimeseries/__init__.py +2 -0
- htimeseries/htimeseries.py +574 -0
- htimeseries/timezone_utils.py +44 -0
- pthelma/__init__.py +0 -0
- pthelma/_version.py +16 -0
- pthelma-0.99.3.dev0.dist-info/LICENSE.rst +34 -0
- pthelma-0.99.3.dev0.dist-info/METADATA +54 -0
- pthelma-0.99.3.dev0.dist-info/RECORD +28 -0
- pthelma-0.99.3.dev0.dist-info/WHEEL +5 -0
- pthelma-0.99.3.dev0.dist-info/entry_points.txt +5 -0
- pthelma-0.99.3.dev0.dist-info/top_level.txt +7 -0
@@ -0,0 +1,252 @@
|
|
1
|
+
from copy import copy
|
2
|
+
from io import StringIO
|
3
|
+
from urllib.parse import urljoin
|
4
|
+
|
5
|
+
try:
|
6
|
+
from zoneinfo import ZoneInfo
|
7
|
+
except ImportError:
|
8
|
+
from backports.zoneinfo import ZoneInfo
|
9
|
+
|
10
|
+
import iso8601
|
11
|
+
import requests
|
12
|
+
|
13
|
+
from htimeseries import HTimeseries
|
14
|
+
|
15
|
+
|
16
|
+
class EnhydrisApiClient:
|
17
|
+
def __init__(self, base_url, token=None):
|
18
|
+
self.base_url = base_url
|
19
|
+
self.token = token
|
20
|
+
self.session = requests.Session()
|
21
|
+
if token is not None:
|
22
|
+
self.session.headers.update({"Authorization": f"token {self.token}"})
|
23
|
+
|
24
|
+
# My understanding from requests' documentation is that when I make a post
|
25
|
+
# request, it shouldn't be necessary to specify Content-Type:
|
26
|
+
# application/x-www-form-urlencoded, and that requests adds the header
|
27
|
+
# automatically. However, when running in Python 3, apparently requests does not
|
28
|
+
# add the header (although it does convert the post data to
|
29
|
+
# x-www-form-urlencoded format). This is why I'm specifying it explicitly.
|
30
|
+
self.session.headers.update(
|
31
|
+
{"Content-Type": "application/x-www-form-urlencoded"}
|
32
|
+
)
|
33
|
+
|
34
|
+
def __enter__(self):
|
35
|
+
self.session.__enter__()
|
36
|
+
return self
|
37
|
+
|
38
|
+
def __exit__(self, *args):
|
39
|
+
self.session.__exit__(*args)
|
40
|
+
|
41
|
+
def check_response(self, expected_status_code=None):
|
42
|
+
try:
|
43
|
+
self._raise_HTTPError_on_error(expected_status_code=expected_status_code)
|
44
|
+
except requests.HTTPError as e:
|
45
|
+
if self.response.text:
|
46
|
+
raise requests.HTTPError(
|
47
|
+
f"{str(e)}. Server response: {self.response.text}"
|
48
|
+
)
|
49
|
+
|
50
|
+
def _raise_HTTPError_on_error(self, expected_status_code):
|
51
|
+
self._check_status_code_is_nonerror()
|
52
|
+
self._check_status_code_is_the_one_expected(expected_status_code)
|
53
|
+
|
54
|
+
def _check_status_code_is_nonerror(self):
|
55
|
+
self.response.raise_for_status()
|
56
|
+
|
57
|
+
def _check_status_code_is_the_one_expected(self, expected_status_code):
|
58
|
+
if expected_status_code and self.response.status_code != expected_status_code:
|
59
|
+
raise requests.HTTPError(
|
60
|
+
f"Expected status code {expected_status_code}; "
|
61
|
+
f"got {self.response.status_code} instead"
|
62
|
+
)
|
63
|
+
|
64
|
+
def get_token(self, username, password):
|
65
|
+
if not username:
|
66
|
+
return
|
67
|
+
|
68
|
+
# Get a csrftoken
|
69
|
+
login_url = urljoin(self.base_url, "api/auth/login/")
|
70
|
+
data = f"username={username}&password={password}"
|
71
|
+
self.response = self.session.post(login_url, data=data, allow_redirects=False)
|
72
|
+
self.check_response()
|
73
|
+
key = self.response.json()["key"]
|
74
|
+
self.session.headers.update({"Authorization": f"token {key}"})
|
75
|
+
return key
|
76
|
+
|
77
|
+
def get_station(self, station_id):
|
78
|
+
url = urljoin(self.base_url, f"api/stations/{station_id}/")
|
79
|
+
self.response = self.session.get(url)
|
80
|
+
self.check_response()
|
81
|
+
return self.response.json()
|
82
|
+
|
83
|
+
def post_station(self, data):
|
84
|
+
self.response = self.session.post(
|
85
|
+
urljoin(self.base_url, "api/stations/"), data=data
|
86
|
+
)
|
87
|
+
self.check_response()
|
88
|
+
return self.response.json()["id"]
|
89
|
+
|
90
|
+
def put_station(self, station_id, data):
|
91
|
+
self.response = self.session.put(
|
92
|
+
urljoin(self.base_url, f"api/stations/{station_id}/"), data=data
|
93
|
+
)
|
94
|
+
self.check_response()
|
95
|
+
|
96
|
+
def patch_station(self, station_id, data):
|
97
|
+
self.response = self.session.patch(
|
98
|
+
urljoin(self.base_url, f"api/stations/{station_id}/"), data=data
|
99
|
+
)
|
100
|
+
self.check_response()
|
101
|
+
|
102
|
+
def delete_station(self, station_id):
|
103
|
+
url = urljoin(self.base_url, f"api/stations/{station_id}/")
|
104
|
+
self.response = self.session.delete(url)
|
105
|
+
self.check_response(expected_status_code=204)
|
106
|
+
|
107
|
+
def get_timeseries_group(self, station_id, timeseries_group_id):
|
108
|
+
url = urljoin(
|
109
|
+
self.base_url,
|
110
|
+
f"api/stations/{station_id}/timeseriesgroups/{timeseries_group_id}/",
|
111
|
+
)
|
112
|
+
self.response = self.session.get(url)
|
113
|
+
self.check_response()
|
114
|
+
return self.response.json()
|
115
|
+
|
116
|
+
def post_timeseries_group(self, station_id, data):
|
117
|
+
url = urljoin(self.base_url, f"api/stations/{station_id}/timeseriesgroups/")
|
118
|
+
self.response = self.session.post(url, data=data)
|
119
|
+
self.check_response()
|
120
|
+
return self.response.json()["id"]
|
121
|
+
|
122
|
+
def put_timeseries_group(self, station_id, timeseries_group_id, data):
|
123
|
+
url = urljoin(
|
124
|
+
self.base_url,
|
125
|
+
f"api/stations/{station_id}/timeseriesgroups/{timeseries_group_id}/",
|
126
|
+
)
|
127
|
+
self.response = self.session.put(url, data=data)
|
128
|
+
self.check_response()
|
129
|
+
return self.response.json()["id"]
|
130
|
+
|
131
|
+
def patch_timeseries_group(self, station_id, timeseries_group_id, data):
|
132
|
+
url = urljoin(
|
133
|
+
self.base_url,
|
134
|
+
f"api/stations/{station_id}/timeseriesgroups/{timeseries_group_id}/",
|
135
|
+
)
|
136
|
+
self.response = self.session.patch(url, data=data)
|
137
|
+
self.check_response()
|
138
|
+
|
139
|
+
def delete_timeseries_group(self, station_id, timeseries_group_id):
|
140
|
+
url = urljoin(
|
141
|
+
self.base_url,
|
142
|
+
f"api/stations/{station_id}/timeseriesgroups/{timeseries_group_id}/",
|
143
|
+
)
|
144
|
+
self.response = self.session.delete(url)
|
145
|
+
self.check_response(expected_status_code=204)
|
146
|
+
|
147
|
+
def list_timeseries(self, station_id, timeseries_group_id):
|
148
|
+
url = urljoin(
|
149
|
+
self.base_url,
|
150
|
+
f"api/stations/{station_id}/timeseriesgroups/{timeseries_group_id}/"
|
151
|
+
"timeseries/",
|
152
|
+
)
|
153
|
+
self.response = self.session.get(url)
|
154
|
+
self.check_response()
|
155
|
+
return self.response.json()["results"]
|
156
|
+
|
157
|
+
def get_timeseries(self, station_id, timeseries_group_id, timeseries_id):
|
158
|
+
url = urljoin(
|
159
|
+
self.base_url,
|
160
|
+
f"api/stations/{station_id}/timeseriesgroups/{timeseries_group_id}/"
|
161
|
+
f"timeseries/{timeseries_id}/",
|
162
|
+
)
|
163
|
+
self.response = self.session.get(url)
|
164
|
+
self.check_response()
|
165
|
+
return self.response.json()
|
166
|
+
|
167
|
+
def post_timeseries(self, station_id, timeseries_group_id, data):
|
168
|
+
self.response = self.session.post(
|
169
|
+
urljoin(
|
170
|
+
self.base_url,
|
171
|
+
f"api/stations/{station_id}/timeseriesgroups/{timeseries_group_id}/"
|
172
|
+
"timeseries/",
|
173
|
+
),
|
174
|
+
data=data,
|
175
|
+
)
|
176
|
+
self.check_response()
|
177
|
+
return self.response.json()["id"]
|
178
|
+
|
179
|
+
def delete_timeseries(self, station_id, timeseries_group_id, timeseries_id):
|
180
|
+
url = urljoin(
|
181
|
+
self.base_url,
|
182
|
+
f"api/stations/{station_id}/timeseriesgroups/{timeseries_group_id}/"
|
183
|
+
f"timeseries/{timeseries_id}/",
|
184
|
+
)
|
185
|
+
self.response = self.session.delete(url)
|
186
|
+
self.check_response(expected_status_code=204)
|
187
|
+
|
188
|
+
def read_tsdata(
|
189
|
+
self,
|
190
|
+
station_id,
|
191
|
+
timeseries_group_id,
|
192
|
+
timeseries_id,
|
193
|
+
start_date=None,
|
194
|
+
end_date=None,
|
195
|
+
timezone=None,
|
196
|
+
):
|
197
|
+
url = urljoin(
|
198
|
+
self.base_url,
|
199
|
+
f"api/stations/{station_id}/timeseriesgroups/{timeseries_group_id}/"
|
200
|
+
f"timeseries/{timeseries_id}/data/",
|
201
|
+
)
|
202
|
+
params = {"fmt": "hts"}
|
203
|
+
tzinfo = ZoneInfo(timezone) if timezone else None
|
204
|
+
dates_are_aware = (start_date is None or start_date.tzinfo is not None) and (
|
205
|
+
end_date is None or end_date.tzinfo is not None
|
206
|
+
)
|
207
|
+
if not dates_are_aware:
|
208
|
+
raise ValueError("start_date and end_date must be aware")
|
209
|
+
params["start_date"] = start_date and start_date.isoformat()
|
210
|
+
params["end_date"] = end_date and end_date.isoformat()
|
211
|
+
params["timezone"] = timezone
|
212
|
+
self.response = self.session.get(url, params=params)
|
213
|
+
self.check_response()
|
214
|
+
if self.response.text:
|
215
|
+
return HTimeseries(StringIO(self.response.text), default_tzinfo=tzinfo)
|
216
|
+
else:
|
217
|
+
return HTimeseries()
|
218
|
+
|
219
|
+
def post_tsdata(self, station_id, timeseries_group_id, timeseries_id, ts):
|
220
|
+
f = StringIO()
|
221
|
+
data = copy(ts.data)
|
222
|
+
try:
|
223
|
+
data.index = data.index.tz_convert("UTC")
|
224
|
+
except AttributeError:
|
225
|
+
assert data.empty
|
226
|
+
data.to_csv(f, header=False)
|
227
|
+
url = urljoin(
|
228
|
+
self.base_url,
|
229
|
+
f"api/stations/{station_id}/timeseriesgroups/{timeseries_group_id}/"
|
230
|
+
f"timeseries/{timeseries_id}/data/",
|
231
|
+
)
|
232
|
+
self.response = self.session.post(
|
233
|
+
url, data={"timeseries_records": f.getvalue(), "timezone": "UTC"}
|
234
|
+
)
|
235
|
+
self.check_response()
|
236
|
+
return self.response.text
|
237
|
+
|
238
|
+
def get_ts_end_date(
|
239
|
+
self, station_id, timeseries_group_id, timeseries_id, timezone=None
|
240
|
+
):
|
241
|
+
url = urljoin(
|
242
|
+
self.base_url,
|
243
|
+
f"api/stations/{station_id}/timeseriesgroups/{timeseries_group_id}/"
|
244
|
+
f"timeseries/{timeseries_id}/bottom/",
|
245
|
+
)
|
246
|
+
self.response = self.session.get(url, params={"timezone": timezone})
|
247
|
+
self.check_response()
|
248
|
+
try:
|
249
|
+
datestring = self.response.text.strip().split(",")[0]
|
250
|
+
return iso8601.parse_date(datestring, default_timezone=None)
|
251
|
+
except (IndexError, iso8601.ParseError):
|
252
|
+
return None
|
enhydris_cache/cli.py
ADDED
@@ -0,0 +1,150 @@
|
|
1
|
+
import configparser
|
2
|
+
import datetime as dt
|
3
|
+
import logging
|
4
|
+
import os
|
5
|
+
import sys
|
6
|
+
import traceback
|
7
|
+
|
8
|
+
import click
|
9
|
+
|
10
|
+
from enhydris_cache import TimeseriesCache
|
11
|
+
from pthelma._version import __version__
|
12
|
+
|
13
|
+
|
14
|
+
class WrongValueError(configparser.Error):
|
15
|
+
pass
|
16
|
+
|
17
|
+
|
18
|
+
class App:
|
19
|
+
def __init__(self, configfilename):
|
20
|
+
self.configfilename = configfilename
|
21
|
+
|
22
|
+
def run(self):
|
23
|
+
self.config = AppConfig(self.configfilename)
|
24
|
+
self.config.read()
|
25
|
+
self._setup_logger()
|
26
|
+
self._execute_with_error_handling()
|
27
|
+
|
28
|
+
def _execute_with_error_handling(self):
|
29
|
+
self.logger.info("Starting enhydris-cache, " + dt.datetime.today().isoformat())
|
30
|
+
try:
|
31
|
+
self._execute()
|
32
|
+
except Exception as e:
|
33
|
+
self.logger.error(str(e))
|
34
|
+
self.logger.debug(traceback.format_exc())
|
35
|
+
self.logger.info(
|
36
|
+
"enhydris-cache terminated with error, "
|
37
|
+
+ dt.datetime.today().isoformat()
|
38
|
+
)
|
39
|
+
raise click.ClickException(str(e))
|
40
|
+
else:
|
41
|
+
self.logger.info(
|
42
|
+
"Finished enhydris-cache, " + dt.datetime.today().isoformat()
|
43
|
+
)
|
44
|
+
|
45
|
+
def _setup_logger(self):
|
46
|
+
self.logger = logging.getLogger("spatialize")
|
47
|
+
self._set_logger_handler()
|
48
|
+
self.logger.setLevel(self.config.loglevel.upper())
|
49
|
+
|
50
|
+
def _set_logger_handler(self):
|
51
|
+
if getattr(self.config, "logfile", None):
|
52
|
+
self.logger.addHandler(logging.FileHandler(self.config.logfile))
|
53
|
+
else:
|
54
|
+
self.logger.addHandler(logging.StreamHandler())
|
55
|
+
|
56
|
+
def _execute(self):
|
57
|
+
os.chdir(self.config.cache_dir)
|
58
|
+
self.cache = TimeseriesCache(self.config.timeseries_group)
|
59
|
+
self.cache.update()
|
60
|
+
|
61
|
+
|
62
|
+
class AppConfig:
|
63
|
+
config_file_general_options = {
|
64
|
+
"logfile": {"fallback": ""},
|
65
|
+
"loglevel": {"fallback": "warning"},
|
66
|
+
"cache_dir": {"fallback": os.getcwd()},
|
67
|
+
}
|
68
|
+
config_file_timeseries_options = {
|
69
|
+
"base_url": {},
|
70
|
+
"station_id": {},
|
71
|
+
"timeseries_id": {},
|
72
|
+
"timeseries_group_id": {},
|
73
|
+
"auth_token": {"fallback": None},
|
74
|
+
"file": {},
|
75
|
+
}
|
76
|
+
|
77
|
+
def __init__(self, configfilename):
|
78
|
+
self.configfilename = configfilename
|
79
|
+
|
80
|
+
def read(self):
|
81
|
+
try:
|
82
|
+
self._parse_config()
|
83
|
+
except (OSError, configparser.Error) as e:
|
84
|
+
sys.stderr.write(str(e))
|
85
|
+
raise click.ClickException(str(e))
|
86
|
+
|
87
|
+
def _parse_config(self):
|
88
|
+
self._read_config_file()
|
89
|
+
self._parse_general_section()
|
90
|
+
self._parse_timeseries_sections()
|
91
|
+
|
92
|
+
def _read_config_file(self):
|
93
|
+
self.config = configparser.ConfigParser(interpolation=None)
|
94
|
+
with open(self.configfilename) as f:
|
95
|
+
self.config.read_file(f)
|
96
|
+
|
97
|
+
def _parse_general_section(self):
|
98
|
+
options = {
|
99
|
+
opt: self.config.get("General", opt, **kwargs)
|
100
|
+
for opt, kwargs in self.config_file_general_options.items()
|
101
|
+
}
|
102
|
+
for key, value in options.items():
|
103
|
+
setattr(self, key, value)
|
104
|
+
self._parse_log_level()
|
105
|
+
|
106
|
+
def _parse_log_level(self):
|
107
|
+
log_levels = ("ERROR", "WARNING", "INFO", "DEBUG")
|
108
|
+
self.loglevel = self.loglevel.upper()
|
109
|
+
if self.loglevel not in log_levels:
|
110
|
+
raise WrongValueError("loglevel must be one of " + ", ".join(log_levels))
|
111
|
+
|
112
|
+
def _parse_timeseries_sections(self):
|
113
|
+
self.timeseries_group = []
|
114
|
+
for section in self.config:
|
115
|
+
if section in ("General", "DEFAULT"):
|
116
|
+
continue
|
117
|
+
item = self._read_section(section)
|
118
|
+
self.timeseries_group.append(item)
|
119
|
+
|
120
|
+
def _read_section(self, section):
|
121
|
+
options = {
|
122
|
+
opt: self.config.get(section, opt, **kwargs)
|
123
|
+
for opt, kwargs in self.config_file_timeseries_options.items()
|
124
|
+
}
|
125
|
+
options["name"] = section
|
126
|
+
try:
|
127
|
+
options["station_id"] = int(options["station_id"])
|
128
|
+
options["timeseries_group_id"] = int(options["timeseries_group_id"])
|
129
|
+
options["timeseries_id"] = int(options["timeseries_id"])
|
130
|
+
except ValueError:
|
131
|
+
raise WrongValueError(
|
132
|
+
'"{}" or "{}" or "{}" is not a valid integer'.format(
|
133
|
+
options["station_id"],
|
134
|
+
options["timeseries_group_id"],
|
135
|
+
options["timeseries_id"],
|
136
|
+
)
|
137
|
+
)
|
138
|
+
return options
|
139
|
+
|
140
|
+
|
141
|
+
@click.command()
|
142
|
+
@click.argument("configfile")
|
143
|
+
@click.version_option(
|
144
|
+
version=__version__, message="%(prog)s from pthelma v.%(version)s"
|
145
|
+
)
|
146
|
+
def main(configfile):
|
147
|
+
"""Spatial integration"""
|
148
|
+
|
149
|
+
app = App(configfile)
|
150
|
+
app.run()
|
@@ -0,0 +1,69 @@
|
|
1
|
+
import datetime as dt
|
2
|
+
|
3
|
+
import pandas as pd
|
4
|
+
|
5
|
+
from enhydris_api_client import EnhydrisApiClient
|
6
|
+
from htimeseries import HTimeseries
|
7
|
+
|
8
|
+
|
9
|
+
class TimeseriesCache(object):
|
10
|
+
def __init__(self, timeseries_group):
|
11
|
+
self.timeseries_group = timeseries_group
|
12
|
+
|
13
|
+
def update(self):
|
14
|
+
for item in self.timeseries_group:
|
15
|
+
self.base_url = item["base_url"]
|
16
|
+
if self.base_url[-1] != "/":
|
17
|
+
self.base_url += "/"
|
18
|
+
self.station_id = item["station_id"]
|
19
|
+
self.timeseries_group_id = item["timeseries_group_id"]
|
20
|
+
self.timeseries_id = item["timeseries_id"]
|
21
|
+
self.auth_token = item["auth_token"]
|
22
|
+
self.filename = item["file"]
|
23
|
+
self._update_for_one_timeseries()
|
24
|
+
|
25
|
+
def _update_for_one_timeseries(self):
|
26
|
+
cached_ts = self._read_timeseries_from_cache_file()
|
27
|
+
end_date = self._get_timeseries_end_date(cached_ts)
|
28
|
+
start_date = end_date + dt.timedelta(minutes=1)
|
29
|
+
new_ts = self._append_newer_timeseries(start_date, cached_ts)
|
30
|
+
with open(self.filename, "w", encoding="utf-8") as f:
|
31
|
+
new_ts.write(f, format=HTimeseries.FILE)
|
32
|
+
|
33
|
+
def _read_timeseries_from_cache_file(self):
|
34
|
+
try:
|
35
|
+
with open(self.filename, newline="\n") as f:
|
36
|
+
return HTimeseries(f)
|
37
|
+
except (FileNotFoundError, ValueError):
|
38
|
+
# If file is corrupted or nonexistent, continue with empty time series
|
39
|
+
return HTimeseries()
|
40
|
+
|
41
|
+
def _get_timeseries_end_date(self, timeseries):
|
42
|
+
try:
|
43
|
+
end_date = timeseries.data.index[-1]
|
44
|
+
except IndexError:
|
45
|
+
# Timeseries is totally empty; no start and end date
|
46
|
+
end_date = dt.datetime(1, 1, 1, 0, 0, tzinfo=dt.timezone.utc)
|
47
|
+
return end_date
|
48
|
+
|
49
|
+
def _append_newer_timeseries(self, start_date, old_ts):
|
50
|
+
with EnhydrisApiClient(self.base_url, token=self.auth_token) as api_client:
|
51
|
+
ts = api_client.read_tsdata(
|
52
|
+
self.station_id,
|
53
|
+
self.timeseries_group_id,
|
54
|
+
self.timeseries_id,
|
55
|
+
start_date=start_date,
|
56
|
+
)
|
57
|
+
new_data = ts.data
|
58
|
+
|
59
|
+
# For appending to work properly, both time series need to have the same
|
60
|
+
# tz.
|
61
|
+
if len(old_ts.data):
|
62
|
+
new_data.index = new_data.index.tz_convert(old_ts.data.index.tz)
|
63
|
+
else:
|
64
|
+
old_ts.data.index = old_ts.data.index.tz_convert(new_data.index.tz)
|
65
|
+
|
66
|
+
ts.data = pd.concat(
|
67
|
+
[old_ts.data, new_data], verify_integrity=True, sort=False
|
68
|
+
)
|
69
|
+
return ts
|
evaporation/__init__.py
ADDED