91 lines
3.1 KiB
Python
91 lines
3.1 KiB
Python
from requests import session
|
|
s=session()
|
|
from src import package_directory, download_path,cfg
|
|
from os import path, makedirs
|
|
import os
|
|
import json
|
|
from gevent import spawn
|
|
from src import clogger
|
|
from src.fb import graph
|
|
from hashlib import md5
|
|
from src.database import db_session2
|
|
from models import CrawlUrl
|
|
import errno
|
|
import urlparse
|
|
def announce_articleid(id):
|
|
for u in cfg.announcearticle_url:
|
|
s.get( u % id)
|
|
|
|
def downloadfile(url):
|
|
u=urlparse.urlparse(url)
|
|
relative_name=path.join("downloads",str(md5(url).hexdigest()),u[2].split('/')[-1])
|
|
local_filename = path.join(download_path,relative_name)
|
|
if not os.path.exists(os.path.dirname(local_filename)):
|
|
try:
|
|
os.makedirs(os.path.dirname(local_filename))
|
|
except OSError as exc: # Guard against race condition
|
|
if exc.errno != errno.EEXIST:
|
|
raise
|
|
if not path.exists(local_filename):
|
|
spawn(fetch_load_file, url, local_filename)
|
|
return relative_name
|
|
|
|
from models import CrawlCache
|
|
from datetime import datetime, timedelta
|
|
|
|
|
|
|
|
|
|
def fetch_page(furl,p={}):
|
|
current_time = datetime.utcnow()
|
|
ten_weeks_ago = current_time - timedelta(days=cfg.cache_days)
|
|
u=urlparse.urlparse(furl)
|
|
cu=CrawlUrl.query.filter(CrawlUrl.url==furl).first()
|
|
|
|
if u[0] == '':
|
|
furl=urlparse.urlunsplit(("http",u[1],u[2],u[3],u[4]))
|
|
cc=CrawlCache.query.filter(CrawlCache.url==furl).filter(CrawlCache.fetched>ten_weeks_ago).first()
|
|
if cc is None or u[0]=='fb': # no caching for Facebook
|
|
clogger.debug("fetching url: "+ str(furl))
|
|
if u[0]=='fb':
|
|
fb_time_since = str(int((current_time - timedelta(days=10)-datetime(1970,1,1)).total_seconds()))
|
|
if p.has_key("nofollow") and p["nofollow"]==False:
|
|
furl=u[1]+u[2]+"?fields=story,created_time,id,message,attachments"
|
|
else:
|
|
furl=u[1]+u[2]+"?since="+fb_time_since+"&fields=story,created_time,id,message,attachments"
|
|
cc=CrawlCache.query.filter(CrawlCache.url==furl).filter(CrawlCache.fetched>ten_weeks_ago).first()
|
|
if cc is None:
|
|
tx = json.dumps(graph.get_object(id=furl))
|
|
else:
|
|
tx = cc.raw
|
|
if not cu==None:
|
|
cu.last_fetched = datetime.utcnow()
|
|
db_session2.add(cu)
|
|
db_session2.commit()
|
|
|
|
else:
|
|
tx=s.get(furl).text
|
|
if not cu==None:
|
|
cu.last_fetched = datetime.utcnow()
|
|
db_session2.add(cu)
|
|
db_session2.commit()
|
|
CrawlCache.store(furl,tx)
|
|
else:
|
|
#if furl is not None:
|
|
# clogger.debug("cache hit")
|
|
tx=cc.raw
|
|
return tx
|
|
|
|
def fetch_load_file(furl, path):
|
|
try:
|
|
clogger.info("Downloading "+ str(furl))
|
|
r = s.get(furl, stream=True)
|
|
f = open(path, 'wb')
|
|
for chunk in r.iter_content(chunk_size=1024):
|
|
if chunk: # filter out keep-alive new chunks
|
|
f.write(chunk)
|
|
f.close()
|
|
except Exception, e:
|
|
#clogger.error("Error Occured during fetching:"+str(furl))
|
|
clogger.error(e,exc_info=True)
|