update Woblink plugin for website changes

This commit is contained in:
Tomasz Długosz 2013-07-04 00:02:19 +02:00
parent bd4e828668
commit 32fa93d584

View File

@ -56,20 +56,20 @@ class WoblinkStore(BasicStoreConfig, StorePlugin):
counter = max_results
with closing(br.open(url, timeout=timeout)) as f:
doc = html.fromstring(f.read())
for data in doc.xpath('//div[@class="book-item backgroundmix"]'):
for data in doc.xpath('//div[@class="nw_katalog_lista_ksiazka"]'):
if counter <= 0:
break
id = ''.join(data.xpath('.//td[@class="w10 va-t mYHaveItYes"]/a[1]/@href'))
id = ''.join(data.xpath('.//div[@class="nw_katalog_lista_ksiazka_okladka nw_okladka"]/a[1]/@href'))
if not id:
continue
cover_url = ''.join(data.xpath('.//td[@class="w10 va-t mYHaveItYes"]/a[1]/img/@src'))
title = ''.join(data.xpath('.//h2[@class="title"]/a[1]/text()'))
author = ', '.join(data.xpath('.//td[@class="va-t"]/h3/a/text()'))
price = ''.join(data.xpath('.//div[@class="prices"]/span[1]/strong/span/text()'))
cover_url = ''.join(data.xpath('.//div[@class="nw_katalog_lista_ksiazka_okladka nw_okladka"]/a[1]/img/@src'))
title = ''.join(data.xpath('.//h2[@class="nw_katalog_lista_ksiazka_detale_tytul"]/a[1]/text()'))
author = ', '.join(data.xpath('.//h3[@class="nw_katalog_lista_ksiazka_detale_autor"]/a/text()'))
price = ''.join(data.xpath('.//div[@class="nw_katalog_lista_ksiazka_opcjezakupu_cena"]/span/text()'))
price = re.sub('\.', ',', price)
formats = [ form[8:-4].split('.')[0] for form in data.xpath('.//p[3]/img/@src')]
formats = ', '.join(data.xpath('.//p[@class="nw_katalog_lista_ksiazka_detale_formaty"]/span/text()'))
s = SearchResult()
s.cover_url = 'http://woblink.com' + cover_url
@ -77,25 +77,15 @@ class WoblinkStore(BasicStoreConfig, StorePlugin):
s.author = author.strip()
s.price = price + ''
s.detail_item = id.strip()
s.formats = formats
if 'epub_drm' in formats:
if 'EPUB DRM' in formats:
s.drm = SearchResult.DRM_LOCKED
s.formats = 'EPUB'
counter -= 1
yield s
elif 'pdf' in formats:
s.drm = SearchResult.DRM_LOCKED
s.formats = 'PDF'
counter -= 1
yield s
else:
s.drm = SearchResult.DRM_UNLOCKED
if 'MOBI_nieb' in formats:
formats.remove('MOBI_nieb')
formats.append('MOBI')
s.formats = ', '.join(formats).upper()
counter -= 1
yield s