def retrieve_bokelai_detail(self, bokelai_id, log, result_queue, timeout):
detail_url = self.BOKELAI_DETAIL_URL % bokelai_id
log.info(detail_url)
try:
br = self.browser
_raw = br.open_novisit(detail_url, timeout=timeout)
raw = _raw.read()
except Exception as e:
log.exception('Failed to load detail page: %s' % detail_url)
return
root = etree.HTML(raw)
info_json_text = root.xpath(
"//script[@type='application/ld+json']")[0].text
log.info(info_json_text)
info_json = json.loads(info_json_text)
title = info_json['name']
authors = info_json['author'][0]['name'].split(",")
publisher = info_json['publisher'][0]['name']
isbn = info_json['workExample']['workExample']['isbn']
pubdate = info_json['datePublished']
comments = ""
comments_ele = root.xpath("(//div[@class='content'])[1]//text()")
comments = "\n".join(comments_ele)
tags = list()
for ele in root.xpath("//li[contains(text(),'本書分類：')]/a"):
log.info(ele.text)
if "／" in ele.text:
tags.extend(ele.text.split("／"))
if "/" in ele.text:
tags.extend(ele.text.split("/"))
else:
tags.append(ele.text)
cover_url = re.search(r'https[^\?\=\&]*' + bokelai_id + r'[^\?\=\&]*',
info_json['image']).group(0)
if not authors:
authors = [_('Unknown')]
log.info(title, authors, publisher, isbn, pubdate, comments, tags,
cover_url)
mi = Metadata(title, authors)
mi.identifiers = {'bokelai': bokelai_id, 'isbn': isbn}
mi.publisher = publisher
mi.comments = comments
mi.isbn = isbn
mi.tags = tags
if pubdate:
try:
from calibre.utils.date import parse_date, utcnow
default = utcnow().replace(day=15)
mi.pubdate = parse_date(pubdate,
assume_utc=True,
default=default)
except:
log.error('Failed to parse pubdate %r' % pubdate)
if not cover_url is None:
mi.has_bokelai_cover = cover_url
self.cache_identifier_to_cover_url(mi.identifiers['bokelai'],
mi.has_bokelai_cover)
else:
mi.has_bokelai_cover = None
result_queue.put(mi)
