def store(self,feed,entries):
feed_name = feed['name']
feed_refresh_time = None
if 'refresh' in feed.keys():
feed_refresh_time = feed['refresh']
if feed_refresh_time == None:
feed_refresh_time = self.get_feed_refresh_time(feed_name)
c = self.conn.cursor()
for entry in entries:
if feed['type'] == 'ato':
entry_time = datetime.fromtimestamp(mktime(entry.updated_parsed))
elif feed['type'] == 'rss':
entry_time = datetime.fromtimestamp(mktime(entry.published_parsed))
else:
pass
max_time = None
if feed_refresh_time == None or entry_time > feed_refresh_time:
if max_time == None or max_time < entry_time:
max_time = entry_time
logger.info('find new article \'%s\' from feed \'%s\'', entry.title,feed_name)
if feed['type'] == 'ato':
self.save_doc(c,feed_name,entry.id,entry.title,entry.content[0].value,entry.link,entry_time)
elif feed['type'] == 'rss':
self.save_doc(c,feed_name,entry.id,entry.title,entry.description,entry.link,entry_time)
else:
pass
feed['refresh'] = max_time
self.conn.commit()
def download(self,feed,url,filename,extname=None):
if feed.startswith('csdn'):
true_url = url.split('?')[0]
else:
true_url = url
logger.info('downloading image \'%s\'',url)
try:
req = urllib2.Request(true_url)
req.add_header('User-Agent',USER_AGENT)
resp = urllib2.urlopen(req,None,DOWNLOAD_TIMEOUT)
#resp = urlopen(true_url)
#logger.debug(resp.getcode())
data = resp.read(-1)
resp.close()
if extname == None:
#in python3
#content_type = resp.getheader('Content-Type')
#in python2
content_type = resp.info().getheader('Content-Type').lower()
extname = self.get_extname(content_type)
if extname == None:
logger.warning('unsupported content type \'%s\'',content_type)
return (None, STATE_BADURL, 'unsupported content type \''+content_type+'\'')
file_fullname = filename+extname
fn = os.path.join(self.image_dir,os.path.join(file_fullname[:1],os.path.join(file_fullname[1:2],file_fullname)))
f = open(fn,'wb')
f.write(data)
f.close()
return (file_fullname, STATE_SUCCESS, 'success')
except Exception as e:
logger.error('an error accur while downloading %s',url);
logger.exception(e)
return (None, STATE_NETWORK_ERROR, 'network error')
def save_doc(self,cursor,feed_name,doc_id,doc_title,doc_content,doc_url,update_time):
sql = 'select entry_time from contents where entry_url = ?'
cursor.execute(sql,(doc_url,))
rows = cursor.fetchall()
if len(rows) > 0 :
sql = 'delete from contents where entry_url = ?'
cursor.execute(sql,(doc_url,))
sql = 'insert into contents(feed_name,entry_id,entry_title,entry_desc,entry_url,entry_time,state) values(?,?,?,?,?,?,0)'
params = (feed_name,doc_id,doc_title,doc_content,doc_url,update_time)
cursor.execute(sql,params)
logger.info('save new article \'%s\' to database', doc_title)
def check_images(self):
cursor = self.conn.cursor()
# new images
logger.info('start list all new images for download')
sql = 'select id, feed_name, url from images where state = 0'
cursor.execute(sql)
rows = cursor.fetchall()
self.download_images(rows,cursor)
# recent error images
logger.info('start list recent error images for download')
sql = 'select id, feed_name, url from images where state&? = ? and crtime > datetime(\'now\',\'-24 hour\')'
cursor.execute(sql,(STATE_SUCCESS,0))
rows = cursor.fetchall()
self.download_images(rows,cursor)
cursor.close()
self.conn.commit()
def fetch_feed(self,feed):
feed_name = feed['name']
try:
logger.info('start fetch feed \'%s\'', feed_name)
fp = feedparser.parse(feed['url'])
if len(fp.entries) > 0:
feed['type'] = fp.version[0:3]
logger.info('get %d articles from feed \'%s\'',len(fp.entries),feed_name)
self.store(feed,fp.entries)
else:
logger.info('get nothing from feed \'%s\'',feed_name)
except Exception as e:
logger.error('get error while fetching feed \'%s\'',feed_name)
logger.exception(e)
def download_images(self,rows,cursor):
for row in rows:
imgid = row[0]
feed = row[1]
url = row[2]
logger.info('prepare download image#%d \'%s\'',imgid,url)
(filename,state,message) = self.download(feed,url,uuid.uuid1().hex)
if filename != None:
logger.info('download image success, local filename is \'%s\'',filename)
cursor.execute('update images set filename = ?, state = state|? where id = ?',(filename,state,imgid))
else:
logger.info('download image fail: %s',message)
cursor.execute('update images set state = state|? where id = ?',(state,imgid))
self.conn.commit()
def fetch(self):
logger.info('start fetch documents')
for feed in config.feeds:
self.fetch_feed(feed)
logger.info('documents fetch complete')
def extract_url(self):
logger.info('start search images from document')
cursor = self.conn.cursor()
sql = 'select id, feed_name, entry_url, entry_desc from contents where state&1 = 0'
cursor.execute(sql)
rows = cursor.fetchall()
logger.info('get %d new documents', len(rows))
for row in rows:
id = row[0]
feed = row[1]
docurl = row[2]
content = row[3]
logger.info('extract image from doc \'%s\'',docurl)
for img in BeautifulSoup(content).findAll('img'):
if img.has_key('src'):
imgurl = img['src']
logger.info('find image \'%s\'',imgurl)
if not imgurl.startswith('http'):
imgurl = urljoin(docurl,imgurl)
logger.info('image\'s full url is \'%s\'',imgurl)
cursor.execute('select count(*) from images where url = ?',(imgurl,))
img_count = cursor.fetchall()[0][0]
if img_count == 0:
logger.info('it is a new image')
cursor.execute('insert into images (feed_name,url,state) values (?,?,0)',(feed,imgurl))
logger.info('insert this image url to database')
else:
logger.info('this image is already exists')
cursor.execute('update contents set state = state|1 where id = ?',(id,))
self.conn.commit()
def fetch(self):
logger.info('start fetch images')
self.extract_url()
self.check_images()
logger.info('images fetch complete')
