|
1
|
|
|
# -*- coding: utf-8 -*- |
|
2
|
|
|
import datetime |
|
3
|
|
|
import json |
|
4
|
|
|
|
|
5
|
|
|
import transaction |
|
6
|
|
|
from pyjobs_crawlers.run import Connector |
|
7
|
|
|
from sqlalchemy.orm.exc import NoResultFound |
|
8
|
|
|
|
|
9
|
|
|
from pyjobsweb import model |
|
10
|
|
|
from pyjobsweb.model import DBSession, Log |
|
11
|
|
|
|
|
12
|
|
|
__all__ = ('helpers', 'app_globals') |
|
13
|
|
|
|
|
14
|
|
|
|
|
15
|
|
|
class PyJobsWebConnector(Connector): |
|
16
|
|
|
def add_job(self, job_item): |
|
17
|
|
|
""" |
|
18
|
|
|
|
|
19
|
|
|
Add job to PyJobsWeb database |
|
20
|
|
|
|
|
21
|
|
|
:param job_item: Scrapy pyjobs_crawlers item object |
|
22
|
|
|
:return: |
|
23
|
|
|
""" |
|
24
|
|
|
job_public_id = job_item['url'] |
|
25
|
|
|
|
|
26
|
|
|
if self.job_exist(job_public_id): |
|
27
|
|
|
print 'Skip existing item' |
|
28
|
|
|
return |
|
29
|
|
|
|
|
30
|
|
|
job = model.JobAlchemy() |
|
31
|
|
|
|
|
32
|
|
|
# Populate attributes which do not require special treatments before |
|
33
|
|
|
# population |
|
34
|
|
|
attributes = ['title', 'description', 'company', 'address', |
|
35
|
|
|
'company_url', 'publication_datetime', |
|
36
|
|
|
'publication_datetime_is_fake'] |
|
37
|
|
|
|
|
38
|
|
|
# Populate job attributes if item contain it |
|
39
|
|
|
for attribute in attributes: |
|
40
|
|
|
if attribute in job_item: |
|
41
|
|
|
setattr(job, attribute, job_item[attribute]) |
|
42
|
|
|
|
|
43
|
|
|
job.url = job_item['url'] |
|
44
|
|
|
job.source = job_item['source'] |
|
45
|
|
|
job.crawl_datetime = job_item['initial_crawl_datetime'] |
|
46
|
|
|
|
|
47
|
|
|
# Populate attributes which require special treatments before population |
|
48
|
|
|
if 'tags' in job_item: |
|
49
|
|
|
tags = [{'tag': t.tag, 'weight': t.weight} |
|
50
|
|
|
for t in job_item['tags']] |
|
51
|
|
|
job.tags = json.dumps(tags) |
|
52
|
|
|
|
|
53
|
|
|
# Insert the job offer in the Postgresql database |
|
54
|
|
|
DBSession.add(job) |
|
55
|
|
|
transaction.commit() |
|
56
|
|
|
|
|
57
|
|
|
def job_exist(self, job_url): |
|
58
|
|
|
""" |
|
59
|
|
|
|
|
60
|
|
|
Return count of jobs having this url |
|
61
|
|
|
|
|
62
|
|
|
:param job_url: External identifier of job (url) |
|
63
|
|
|
:return: |
|
64
|
|
|
""" |
|
65
|
|
|
return model.JobAlchemy.job_offer_exists(job_url) |
|
66
|
|
|
|
|
67
|
|
|
def log(self, source, action, more=None): |
|
68
|
|
|
if more is not None: |
|
69
|
|
|
message = '%s (%s)' % (action, more) |
|
70
|
|
|
else: |
|
71
|
|
|
message = action |
|
72
|
|
|
|
|
73
|
|
|
log = Log() |
|
74
|
|
|
log.source = source |
|
75
|
|
|
log.message = message |
|
76
|
|
|
log.datetime = datetime.datetime.now() |
|
77
|
|
|
|
|
78
|
|
|
DBSession.add(log) |
|
79
|
|
|
transaction.commit() |
|
80
|
|
|
|
|
81
|
|
|
def get_most_recent_job_date(self, source): |
|
82
|
|
|
try: |
|
83
|
|
|
return \ |
|
84
|
|
|
model.DBSession.query(model.JobAlchemy.publication_datetime)\ |
|
85
|
|
|
.filter(model.JobAlchemy.source == source)\ |
|
86
|
|
|
.order_by(model.JobAlchemy.publication_datetime.desc())\ |
|
87
|
|
|
.limit(1)\ |
|
88
|
|
|
.one()[0] # First element is publication_datetime value |
|
89
|
|
|
except NoResultFound: |
|
90
|
|
|
return datetime.datetime(1970, 1, 1, 0, 0, 0) |
|
91
|
|
|
|