|
1
|
|
|
#!/usr/bin/env python3 |
|
2
|
|
|
# -*- coding: utf-8 -*- |
|
3
|
|
|
""" |
|
4
|
|
|
Created on Thu Dec 19 16:28:46 2019 |
|
5
|
|
|
|
|
6
|
|
|
@author: Paolo Cozzi <[email protected]> |
|
7
|
|
|
""" |
|
8
|
|
|
|
|
9
|
|
|
import requests |
|
10
|
|
|
import logging |
|
11
|
|
|
|
|
12
|
|
|
from dateutil.parser import parse as parse_date |
|
13
|
|
|
|
|
14
|
|
|
from . import __version__ |
|
15
|
|
|
from .auth import Auth |
|
16
|
|
|
|
|
17
|
|
|
logger = logging.getLogger(__name__) |
|
18
|
|
|
|
|
19
|
|
|
|
|
20
|
|
|
# https://stackoverflow.com/a/25341965/4385116 |
|
21
|
|
|
def is_date(string, fuzzy=False): |
|
22
|
|
|
""" |
|
23
|
|
|
Return whether the string can be interpreted as a date. |
|
24
|
|
|
|
|
25
|
|
|
:param string: str, string to check for date |
|
26
|
|
|
:param fuzzy: bool, ignore unknown tokens in string if True |
|
27
|
|
|
""" |
|
28
|
|
|
try: |
|
29
|
|
|
parse_date(string, fuzzy=fuzzy) |
|
30
|
|
|
return True |
|
31
|
|
|
|
|
32
|
|
|
except ValueError: |
|
33
|
|
|
return False |
|
34
|
|
|
|
|
35
|
|
|
|
|
36
|
|
|
class Client(): |
|
37
|
|
|
"""A class to deal with EBI submission API. It perform request |
|
38
|
|
|
modelling user token in request headers. You need to call this class after |
|
39
|
|
|
instantiating an :py:class:`Auth <pyUSIrest.auth.Auth>` object:: |
|
40
|
|
|
|
|
41
|
|
|
import getpass |
|
42
|
|
|
from pyUSIrest.auth import Auth |
|
43
|
|
|
from pyUSIrest.client import Client |
|
44
|
|
|
auth = Auth(user=<you_aap_user>, password=getpass.getpass()) |
|
45
|
|
|
client = Client(auth) |
|
46
|
|
|
response = client.get("https://submission-test.ebi.ac.uk/api/") |
|
47
|
|
|
|
|
48
|
|
|
Attributes: |
|
49
|
|
|
headers (dict): default headers for requests |
|
50
|
|
|
last_response (requests.Response): last response object read by this |
|
51
|
|
|
class |
|
52
|
|
|
last_satus_code (int): last status code read by this class |
|
53
|
|
|
session (request.Session): a session object |
|
54
|
|
|
auth (Auth): a pyUSIrest Auth object |
|
55
|
|
|
""" |
|
56
|
|
|
|
|
57
|
|
|
headers = { |
|
58
|
|
|
'Accept': 'application/hal+json', |
|
59
|
|
|
'User-Agent': 'pyUSIrest %s' % (__version__) |
|
60
|
|
|
} |
|
61
|
|
|
|
|
62
|
|
|
def __init__(self, auth): |
|
63
|
|
|
"""Instantiate the class |
|
64
|
|
|
|
|
65
|
|
|
Args: |
|
66
|
|
|
auth (Auth): a valid :py:class:`Auth <pyUSIrest.auth.Auth>` object |
|
67
|
|
|
|
|
68
|
|
|
""" |
|
69
|
|
|
|
|
70
|
|
|
# my attributes |
|
71
|
|
|
self._auth = None |
|
72
|
|
|
self.last_response = None |
|
73
|
|
|
self.last_status_code = None |
|
74
|
|
|
self.session = requests.Session() |
|
75
|
|
|
|
|
76
|
|
|
# setting auth object |
|
77
|
|
|
self.auth = auth |
|
78
|
|
|
|
|
79
|
|
|
@property |
|
80
|
|
|
def auth(self): |
|
81
|
|
|
"""Get/Set :py:class:`Auth <pyUSIrest.auth.Auth>` object""" |
|
82
|
|
|
|
|
83
|
|
|
return self._auth |
|
84
|
|
|
|
|
85
|
|
|
@auth.setter |
|
86
|
|
|
def auth(self, auth): |
|
87
|
|
|
logger.debug("Auth type is %s" % (type(auth))) |
|
88
|
|
|
|
|
89
|
|
|
# assign Auth object or create a new one |
|
90
|
|
|
if isinstance(auth, Auth): |
|
91
|
|
|
logger.debug("Assigning an Auth object") |
|
92
|
|
|
self._auth = auth |
|
93
|
|
|
|
|
94
|
|
|
else: |
|
95
|
|
|
logger.debug("Creating an Auth object") |
|
96
|
|
|
self._auth = Auth(token=auth) |
|
97
|
|
|
|
|
98
|
|
|
logger.debug("Updating headers with token") |
|
99
|
|
|
self.headers['Authorization'] = "Bearer {token}".format( |
|
100
|
|
|
token=self._auth.token) |
|
101
|
|
|
|
|
102
|
|
|
def check_headers(self, headers=None): |
|
103
|
|
|
"""Checking headers and token |
|
104
|
|
|
|
|
105
|
|
|
Args: |
|
106
|
|
|
headers (dict): custom header for request |
|
107
|
|
|
|
|
108
|
|
|
Returns: |
|
109
|
|
|
headers (dict): an update headers tocken""" |
|
110
|
|
|
|
|
111
|
|
|
if self.auth.is_expired(): |
|
112
|
|
|
raise RuntimeError("Your token is expired") |
|
113
|
|
|
|
|
114
|
|
|
if not headers: |
|
115
|
|
|
logger.debug("Using default headers") |
|
116
|
|
|
new_headers = self.headers |
|
117
|
|
|
|
|
118
|
|
|
else: |
|
119
|
|
|
new_headers = self.headers.copy() |
|
120
|
|
|
new_headers.update(headers) |
|
121
|
|
|
|
|
122
|
|
|
return new_headers |
|
123
|
|
|
|
|
124
|
|
|
def check_status(self, response, expected_status=200): |
|
125
|
|
|
"""Check response status. See `HTTP status codes <https://submission. |
|
126
|
|
|
ebi.ac.uk/api/docs/ref_overview.html#_http_status_codes>`_ |
|
127
|
|
|
|
|
128
|
|
|
Args: |
|
129
|
|
|
response (requests.Reponse): the reponse returned by requests |
|
130
|
|
|
method |
|
131
|
|
|
""" |
|
132
|
|
|
|
|
133
|
|
|
# check with status code. deal with 50X statuses (internal error) |
|
134
|
|
|
if int(response.status_code / 100) == 5: |
|
135
|
|
|
raise ConnectionError( |
|
136
|
|
|
"Problems with API endpoints: %s" % response.text) |
|
137
|
|
|
|
|
138
|
|
|
# TODO: evaluate a list of expected status? |
|
|
|
|
|
|
139
|
|
|
if response.status_code != expected_status: |
|
140
|
|
|
raise ConnectionError( |
|
141
|
|
|
"%s:%s" % (response.status_code, response.text)) |
|
142
|
|
|
|
|
143
|
|
View Code Duplication |
def get(self, url, headers={}, params={}): |
|
|
|
|
|
|
144
|
|
|
"""Generic GET method |
|
145
|
|
|
|
|
146
|
|
|
Args: |
|
147
|
|
|
url (str): url to request |
|
148
|
|
|
headers (dict): custom headers for get request |
|
149
|
|
|
params (dict): custom params for get request |
|
150
|
|
|
|
|
151
|
|
|
Returns: |
|
152
|
|
|
requests.Response: a response object |
|
153
|
|
|
""" |
|
154
|
|
|
|
|
155
|
|
|
logger.debug("Getting %s" % (url)) |
|
156
|
|
|
headers = self.check_headers(headers) |
|
157
|
|
|
response = self.session.get(url, headers=headers, params=params) |
|
158
|
|
|
|
|
159
|
|
|
# track last response |
|
160
|
|
|
self.last_response = response |
|
161
|
|
|
self.last_status_code = response.status_code |
|
162
|
|
|
|
|
163
|
|
|
# check response status code |
|
164
|
|
|
self.check_status(response) |
|
165
|
|
|
|
|
166
|
|
|
return response |
|
167
|
|
|
|
|
168
|
|
|
def post(self, url, payload={}, headers={}, params={}): |
|
169
|
|
|
"""Generic POST method |
|
170
|
|
|
|
|
171
|
|
|
Args: |
|
172
|
|
|
url (str): url to request |
|
173
|
|
|
payload (dict): data to send |
|
174
|
|
|
headers (dict): custom header for request |
|
175
|
|
|
params (dict): custom params for request |
|
176
|
|
|
|
|
177
|
|
|
Returns: |
|
178
|
|
|
requests.Response: a response object |
|
179
|
|
|
""" |
|
180
|
|
|
|
|
181
|
|
|
logger.debug("Posting data to %s" % (url)) |
|
182
|
|
|
headers = self.check_headers(headers) |
|
183
|
|
|
response = self.session.post( |
|
184
|
|
|
url, json=payload, headers=headers, params=params) |
|
185
|
|
|
|
|
186
|
|
|
# track last response |
|
187
|
|
|
self.last_response = response |
|
188
|
|
|
self.last_status_code = response.status_code |
|
189
|
|
|
|
|
190
|
|
|
# check response status code |
|
191
|
|
|
self.check_status(response, expected_status=201) |
|
192
|
|
|
|
|
193
|
|
|
return response |
|
194
|
|
|
|
|
195
|
|
|
def patch(self, url, payload={}, headers={}, params={}): |
|
196
|
|
|
"""Generic PATCH method |
|
197
|
|
|
|
|
198
|
|
|
Args: |
|
199
|
|
|
url (str): url to request |
|
200
|
|
|
payload (dict): data to send |
|
201
|
|
|
headers (dict): custom header for request |
|
202
|
|
|
params (dict): custom params for request |
|
203
|
|
|
|
|
204
|
|
|
Returns: |
|
205
|
|
|
requests.Response: a response object |
|
206
|
|
|
""" |
|
207
|
|
|
|
|
208
|
|
|
logger.debug("Patching data to %s" % (url)) |
|
209
|
|
|
headers = self.check_headers(headers) |
|
210
|
|
|
response = self.session.patch( |
|
211
|
|
|
url, json=payload, headers=headers, params=params) |
|
212
|
|
|
|
|
213
|
|
|
# track last response |
|
214
|
|
|
self.last_response = response |
|
215
|
|
|
self.last_status_code = response.status_code |
|
216
|
|
|
|
|
217
|
|
|
# check response status code |
|
218
|
|
|
self.check_status(response) |
|
219
|
|
|
|
|
220
|
|
|
return response |
|
221
|
|
|
|
|
222
|
|
|
def delete(self, url, headers={}, params={}): |
|
223
|
|
|
"""Generic DELETE method |
|
224
|
|
|
|
|
225
|
|
|
Args: |
|
226
|
|
|
url (str): url to request |
|
227
|
|
|
headers (dict): custom header for request |
|
228
|
|
|
params (dict): custom params for request |
|
229
|
|
|
|
|
230
|
|
|
Returns: |
|
231
|
|
|
requests.Response: a response object |
|
232
|
|
|
""" |
|
233
|
|
|
|
|
234
|
|
|
logger.debug("Deleting %s" % (url)) |
|
235
|
|
|
|
|
236
|
|
|
headers = self.check_headers(headers) |
|
237
|
|
|
response = self.session.delete(url, headers=headers, params=params) |
|
238
|
|
|
|
|
239
|
|
|
# track last response |
|
240
|
|
|
self.last_response = response |
|
241
|
|
|
self.last_status_code = response.status_code |
|
242
|
|
|
|
|
243
|
|
|
# check response status code |
|
244
|
|
|
self.check_status(response, expected_status=204) |
|
245
|
|
|
|
|
246
|
|
|
return response |
|
247
|
|
|
|
|
248
|
|
View Code Duplication |
def put(self, url, payload={}, headers={}, params={}): |
|
|
|
|
|
|
249
|
|
|
"""Generic PUT method |
|
250
|
|
|
|
|
251
|
|
|
Args: |
|
252
|
|
|
url (str): url to request |
|
253
|
|
|
payload (dict): data to send |
|
254
|
|
|
params (dict): custom params for request |
|
255
|
|
|
headers (dict): custom header for request |
|
256
|
|
|
|
|
257
|
|
|
Returns: |
|
258
|
|
|
requests.Response: a response object |
|
259
|
|
|
""" |
|
260
|
|
|
|
|
261
|
|
|
logger.debug("Putting data to %s" % (url)) |
|
262
|
|
|
headers = self.check_headers(headers) |
|
263
|
|
|
response = self.session.put( |
|
264
|
|
|
url, json=payload, headers=headers, params=params) |
|
265
|
|
|
|
|
266
|
|
|
# track last response |
|
267
|
|
|
self.last_response = response |
|
268
|
|
|
self.last_status_code = response.status_code |
|
269
|
|
|
|
|
270
|
|
|
# check response status code |
|
271
|
|
|
self.check_status(response) |
|
272
|
|
|
|
|
273
|
|
|
return response |
|
274
|
|
|
|
|
275
|
|
|
|
|
276
|
|
|
class Document(Client): |
|
277
|
|
|
"""Base class for pyUSIrest classes. It models common methods and |
|
278
|
|
|
attributes by calling :py:class:`Client` and reading json response from |
|
279
|
|
|
biosample API |
|
280
|
|
|
|
|
281
|
|
|
Attributes: |
|
282
|
|
|
_link (dict): ``_links`` data read from USI response |
|
283
|
|
|
_embeddedd (dict): ``_embedded`` data read from USI response |
|
284
|
|
|
page (dict): ``page`` data read from USI response |
|
285
|
|
|
name (str): name of this object |
|
286
|
|
|
data (dict): data from USI read with |
|
287
|
|
|
:py:meth:`response.json() <requests.Response.json>` |
|
288
|
|
|
|
|
289
|
|
|
""" |
|
290
|
|
|
|
|
291
|
|
|
def __init__(self, auth=None, data=None): |
|
292
|
|
|
# if I get auth, setting appropriate method |
|
293
|
|
|
if auth: |
|
294
|
|
|
Client.__init__(self, auth) |
|
295
|
|
|
|
|
296
|
|
|
# my class attributes |
|
297
|
|
|
self._links = {} |
|
298
|
|
|
self._embedded = {} |
|
299
|
|
|
self.page = {} |
|
300
|
|
|
self.name = None |
|
301
|
|
|
self.data = {} |
|
302
|
|
|
|
|
303
|
|
|
# if I get data, read data into myself |
|
304
|
|
|
if data: |
|
305
|
|
|
self.read_data(data) |
|
306
|
|
|
|
|
307
|
|
|
def get(self, url, force_keys=True): |
|
308
|
|
|
"""Override the Client.get method and read data into object:: |
|
309
|
|
|
|
|
310
|
|
|
document = Document(auth) |
|
311
|
|
|
document.get(settings.ROOT_URL + "/api/") |
|
312
|
|
|
|
|
313
|
|
|
Args: |
|
314
|
|
|
url (str): url to request |
|
315
|
|
|
force_keys (bool): If True, define a new class attribute from data |
|
316
|
|
|
keys |
|
317
|
|
|
|
|
318
|
|
|
Returns: |
|
319
|
|
|
requests.Response: a response object |
|
320
|
|
|
""" |
|
321
|
|
|
|
|
322
|
|
|
# call the base method |
|
323
|
|
|
response = super().get(url) |
|
324
|
|
|
|
|
325
|
|
|
# read data |
|
326
|
|
|
self.read_data(response.json(), force_keys) |
|
327
|
|
|
|
|
328
|
|
|
# act like client object |
|
329
|
|
|
return response |
|
330
|
|
|
|
|
331
|
|
|
def read_data(self, data, force_keys=False): |
|
332
|
|
|
"""Read data from a dictionary object and set class attributes |
|
333
|
|
|
|
|
334
|
|
|
Args: |
|
335
|
|
|
data (dict): a data dictionary object read with |
|
336
|
|
|
:py:meth:`response.json() <requests.Response.json>` |
|
337
|
|
|
force_keys (bool): If True, define a new class attribute from data |
|
338
|
|
|
keys |
|
339
|
|
|
""" |
|
340
|
|
|
|
|
341
|
|
|
# dealing with this type of documents |
|
342
|
|
|
for key in data.keys(): |
|
343
|
|
|
if "date" in key.lower() and is_date(data[key]): |
|
344
|
|
|
self.__update_key(key, parse_date(data[key]), force_keys) |
|
345
|
|
|
|
|
346
|
|
|
else: |
|
347
|
|
|
self.__update_key(key, data[key], force_keys) |
|
348
|
|
|
|
|
349
|
|
|
self.data = data |
|
350
|
|
|
|
|
351
|
|
|
def __update_key(self, key, value, force_keys=False): |
|
352
|
|
|
"""Helper function to update keys""" |
|
353
|
|
|
|
|
354
|
|
|
if hasattr(self, key): |
|
355
|
|
|
if getattr(self, key) and getattr(self, key) != '': |
|
356
|
|
|
# when I reload data, I do a substitution |
|
357
|
|
|
logger.debug("Found %s -> %s" % (key, getattr(self, key))) |
|
358
|
|
|
logger.debug("Updating %s -> %s" % (key, value)) |
|
359
|
|
|
|
|
360
|
|
|
else: |
|
361
|
|
|
# don't have this attribute set |
|
362
|
|
|
logger.debug("Setting %s -> %s" % (key, value)) |
|
363
|
|
|
|
|
364
|
|
|
setattr(self, key, value) |
|
365
|
|
|
|
|
366
|
|
|
else: |
|
367
|
|
|
if force_keys is True: |
|
368
|
|
|
logger.info("Forcing %s -> %s" % (key, value)) |
|
369
|
|
|
setattr(self, key, value) |
|
370
|
|
|
|
|
371
|
|
|
else: |
|
372
|
|
|
logger.warning("key %s not implemented" % (key)) |
|
373
|
|
|
|
|
374
|
|
|
@classmethod |
|
375
|
|
|
def clean_url(cls, url): |
|
376
|
|
|
"""Remove stuff like ``{?projection}`` from url |
|
377
|
|
|
|
|
378
|
|
|
Args: |
|
379
|
|
|
url (str): a string url |
|
380
|
|
|
|
|
381
|
|
|
Returns: |
|
382
|
|
|
str: the cleaned url |
|
383
|
|
|
""" |
|
384
|
|
|
|
|
385
|
|
|
# remove {?projection} from self url. This is unreachable |
|
386
|
|
|
if '{?projection}' in url: |
|
387
|
|
|
logger.debug("removing {?projection} from url") |
|
388
|
|
|
url = url.replace("{?projection}", "") |
|
389
|
|
|
|
|
390
|
|
|
return url |
|
391
|
|
|
|
|
392
|
|
|
@classmethod |
|
393
|
|
|
def read_url(cls, auth, url): |
|
394
|
|
|
"""Read a url and returns a :py:class:`Document` object |
|
395
|
|
|
|
|
396
|
|
|
Args: |
|
397
|
|
|
auth (Auth): an Auth object to pass to result |
|
398
|
|
|
url (str): url to request |
|
399
|
|
|
|
|
400
|
|
|
Returns: |
|
401
|
|
|
Document: a document object |
|
402
|
|
|
""" |
|
403
|
|
|
|
|
404
|
|
|
# clean url |
|
405
|
|
|
url = cls.clean_url(url) |
|
406
|
|
|
|
|
407
|
|
|
# create a new document |
|
408
|
|
|
document = cls(auth=auth) |
|
409
|
|
|
|
|
410
|
|
|
# get url and load data |
|
411
|
|
|
document.get(url) |
|
412
|
|
|
|
|
413
|
|
|
return document |
|
414
|
|
|
|
|
415
|
|
|
def paginate(self): |
|
416
|
|
|
"""Follow all the pages. Return an iterator of document objects |
|
417
|
|
|
|
|
418
|
|
|
Args: |
|
419
|
|
|
response (requests.Response): a response object |
|
420
|
|
|
|
|
421
|
|
|
Yield: |
|
422
|
|
|
Document: a new Document instance |
|
423
|
|
|
""" |
|
424
|
|
|
|
|
425
|
|
|
# return myself |
|
426
|
|
|
yield self |
|
427
|
|
|
|
|
428
|
|
|
# track the current document |
|
429
|
|
|
document = self |
|
430
|
|
|
|
|
431
|
|
|
while 'next' in document._links: |
|
432
|
|
|
url = document._links['next']['href'] |
|
433
|
|
|
document = Document.read_url(self.auth, url) |
|
434
|
|
|
|
|
435
|
|
|
# return the last document |
|
436
|
|
|
yield document |
|
437
|
|
|
|
|
438
|
|
|
def follow_tag(self, tag, force_keys=True): |
|
439
|
|
|
"""Pick a url from data attribute relying on tag, perform a request |
|
440
|
|
|
and returns a document object. For instance:: |
|
441
|
|
|
|
|
442
|
|
|
document.follow_tag('userSubmissions') |
|
443
|
|
|
|
|
444
|
|
|
will return a document instance by requesting with |
|
445
|
|
|
:py:meth:`Client.get` using |
|
446
|
|
|
``document._links['userSubmissions']['href']`` as url |
|
447
|
|
|
|
|
448
|
|
|
Args: |
|
449
|
|
|
tag (str): a key from USI response dictionary |
|
450
|
|
|
force_keys (bool): set a new class attribute if not present |
|
451
|
|
|
|
|
452
|
|
|
Returns: |
|
453
|
|
|
Document: a document object |
|
454
|
|
|
""" |
|
455
|
|
|
|
|
456
|
|
|
logger.debug("Following %s url" % (tag)) |
|
457
|
|
|
|
|
458
|
|
|
url = self._links[tag]['href'] |
|
459
|
|
|
|
|
460
|
|
|
# create a new document |
|
461
|
|
|
document = Document(auth=self.auth) |
|
462
|
|
|
|
|
463
|
|
|
# read data |
|
464
|
|
|
document.get(url, force_keys) |
|
465
|
|
|
|
|
466
|
|
|
return document |
|
467
|
|
|
|
|
468
|
|
|
def follow_self_url(self): |
|
469
|
|
|
"""Follow *self* url and update class attributes. For instance:: |
|
470
|
|
|
|
|
471
|
|
|
document.follow_self_url() |
|
472
|
|
|
|
|
473
|
|
|
will reload document instance by requesting with |
|
474
|
|
|
:py:meth:`Client.get` using |
|
475
|
|
|
``document.data['_links']['self']['href']`` as url""" |
|
476
|
|
|
|
|
477
|
|
|
logger.debug("Following self url") |
|
478
|
|
|
|
|
479
|
|
|
# get a url to follow |
|
480
|
|
|
url = self._links['self']['href'] |
|
481
|
|
|
|
|
482
|
|
|
# clean url |
|
483
|
|
|
url = self.clean_url(url) |
|
484
|
|
|
|
|
485
|
|
|
logger.debug("Updating self") |
|
486
|
|
|
|
|
487
|
|
|
# now follow self url and load data to self |
|
488
|
|
|
self.get(url) |
|
489
|
|
|
|