def update(self):
rss = self.source.conf.get('rss')
logger.debug(u'解析 rss %s', rss)
response = self.opener.open(rss, timeout=config.getint('timeout', 30))
tree = etree.parse(response)
songs = []
for item in tree.findall('channel/item'):
song = Song()
for e in item.findall('enclosure'):
t = e.get('type')
if t and t.startswith('audio/'):
song.url = e.get('url')
break
else:
continue
song.title = item.find('title').text
song.id = item.find('guid').text.strip()
song.pubDate = item.find('pubDate').text
if self.last_id and song.id == self.last_id:
break
songs.append(song)
if not self.last_id and len(songs) >= self.init_count:
break
songs.reverse()
return songs
def update(self):
songs = []
home = 'http://soundofhope.org'
label_l1 = util.decode(self.source.conf.get('label_l1'))
label_l2 = util.decode(self.source.conf.get('label_l2'))
response = self.opener.open(home, timeout=config.getint('timeout', 30))
html = etree.parse(response, etree.HTMLParser())
for l1 in html.findall('//div[@id="nav-site"]/ul/li[a]'):
a = l1.find('a')
if a.text.find(label_l1) != -1:
break
else:
logger.warning(u'没有找到一级标签 label_l1 = %s', label_l1)
return songs
for l2 in l1.findall('div/ul/li/a'):
if l2.text.find(label_l2) != -1:
break
else:
logger.warning(u'没有找到二级标签 label_l1 = %s label_l2 = %s', (label_l1, label_l2))
return songs
items = urlparse.urljoin(home, l2.get('href'))
response = self.opener.open(items, timeout=config.getint('timeout', 30))
html = etree.parse(response, etree.HTMLParser())
for item in html.findall('//div[@id="CategoryCol_mid"]/div/ul/li/a'):
url = urlparse.urljoin(items, item.get('href')).strip()
if self.last_id and self.last_id == url:
break
song = rss.Song()
song.id = url
song.title = item.text.strip()
response = self.opener.open(url, timeout=config.getint('timeout', 30))
html = etree.parse(response, etree.HTMLParser())
div = html.find('//div[@id="columnfirstnei2"]')
pubDate = div.find('div[@class="subtitle"]/span[@class="date"]')
song.pubDate = pubDate.text
mp3 = div.find('.//div[@class="mp3_links"]/a')
song.url = urlparse.urljoin(url, mp3.get('href'))
songs.append(song)
if not self.last_id and len(songs) >= self.init_count:
break
songs.reverse()
return songs
def download(self, song):
logger.debug(u'下载歌曲 %s', song.url)
suffix = util.getSuffix(song.url)
if not suffix:
suffix = '.mp3'
fd, path = tempfile.mkstemp(suffix, '', self.source.cachedir)
response = self.opener.open(song.url, timeout=config.getint('timeout', 30))
while True:
data = response.read(4096)
if not data:
break
os.write(fd, data)
response.close()
os.close(fd)
song.file = path
logger.debug(u'下载完成 <%s> %s', path, song.url)
def _open(self, type="n", sid=None, channel=0, pt=None):
params = {}
if type:
params["type"] = type
if sid:
params["sid"] = sid
if channel != None:
params["channel"] = channel
if pt != None:
params["pt"] = "%.1f" % pt
params["from"] = "mainsite"
params["r"] = self.random()
url = self.url
if params:
url = "".join([url, "?", urllib.urlencode(params)])
logger.info(u"请求URL %s", util.decode(url))
response = self.opener.open(url, timeout=config.getint("timeout", 30))
return response