|
1
|
|
|
import multiprocessing |
|
2
|
|
|
from bs4 import BeautifulSoup |
|
3
|
|
|
|
|
4
|
|
|
from pysmartprice.results import SmartPriceResult |
|
5
|
|
|
from pysmartprice.helpers import scrape, scrape_helper |
|
6
|
|
|
from pysmartprice import constants |
|
7
|
|
|
|
|
8
|
|
|
|
|
9
|
|
|
class BaseParser(object): |
|
10
|
|
|
def __init__(self, mapper, **kwargs): |
|
11
|
|
|
self.mapper = mapper |
|
12
|
|
|
self.params = kwargs |
|
13
|
|
|
self.url = constants.URL_MAPPER[self.mapper] |
|
14
|
|
|
self.response = scrape(self._make_url(self.url), **kwargs) |
|
15
|
|
|
self.soup = BeautifulSoup(self.response, 'lxml') |
|
16
|
|
|
self.result = [ |
|
17
|
|
|
SmartPriceResult(self.get_product_attrs(item)) |
|
18
|
|
|
for item in self.products_html |
|
19
|
|
|
] |
|
20
|
|
|
|
|
21
|
|
|
def _make_url(self, target): |
|
22
|
|
|
return '{}{}'.format(constants.SMARTPRICE_WEB_URL, target) |
|
23
|
|
|
|
|
24
|
|
|
@property |
|
25
|
|
|
def price_results(self): |
|
26
|
|
|
if self.get_page_range: |
|
27
|
|
|
return self.process_multiple_pages() |
|
28
|
|
|
|
|
29
|
|
|
return self.result |
|
30
|
|
|
|
|
31
|
|
|
|
|
32
|
|
|
class ParserMixin(object): |
|
33
|
|
|
def get_product_attrs(self, item): |
|
34
|
|
|
return dict( |
|
35
|
|
|
img=item.find('img').get('src'), |
|
36
|
|
|
title=item.find('a', attrs={'class': 'prdct-item__name'}).text, |
|
37
|
|
|
url=item.find( |
|
38
|
|
|
'a', attrs={'class': 'prdct-item__name'}).get('href'), |
|
39
|
|
|
best_price=item.find( |
|
40
|
|
|
'span', attrs={'class': 'prdct-item__prc-val'}).text, |
|
41
|
|
|
product_id=item.get('data-mspid') |
|
42
|
|
|
) |
|
43
|
|
|
|
|
44
|
|
|
@property |
|
45
|
|
|
def products_html(self): |
|
46
|
|
|
html = self.soup.findAll('div', attrs={'class': 'prdct-item'}) |
|
47
|
|
|
return html |
|
48
|
|
|
|
|
49
|
|
|
def process_multiple_pages(self): |
|
50
|
|
|
results = self.result |
|
51
|
|
|
first_page, last_page = self.get_page_range |
|
52
|
|
|
paged_url = self.get_paged_url |
|
53
|
|
|
page_urls = [] |
|
54
|
|
|
|
|
55
|
|
|
for page in range(first_page+1, last_page+1): |
|
56
|
|
|
url = paged_url.replace('.html', '-{}.html'.format(page)) |
|
57
|
|
|
params = self.params.copy() |
|
58
|
|
|
if self.params.get('page', None): |
|
59
|
|
|
params.update({'page': page}) |
|
60
|
|
|
page_urls.append((self._make_url(url), params)) |
|
61
|
|
|
|
|
62
|
|
|
# Scrape pages in parallel |
|
63
|
|
|
pool = multiprocessing.Pool(processes=multiprocessing.cpu_count()*2) |
|
64
|
|
|
|
|
65
|
|
|
for page in pool.map(scrape_helper, page_urls): |
|
66
|
|
|
self.soup = BeautifulSoup(page, 'lxml') |
|
67
|
|
|
|
|
68
|
|
|
results += [ |
|
69
|
|
|
SmartPriceResult(self.get_product_attrs(item)) |
|
70
|
|
|
for item in self.products_html |
|
71
|
|
|
] |
|
72
|
|
|
return results |
|
73
|
|
|
|
|
74
|
|
|
@property |
|
75
|
|
|
def get_page_range(self): |
|
76
|
|
|
page_range = self.soup.findAll( |
|
77
|
|
|
'span', attrs={'class': 'pgntn__rslt-page'}) |
|
78
|
|
|
|
|
79
|
|
|
if not page_range: |
|
80
|
|
|
return None |
|
81
|
|
|
|
|
82
|
|
|
first_page = int(page_range[0].text) |
|
83
|
|
|
last_page = int(page_range[1].text) |
|
84
|
|
|
return first_page, last_page |
|
85
|
|
|
|
|
86
|
|
|
|
asifpy /
python-smartprice
