258 lines
8.0 KiB
Python
258 lines
8.0 KiB
Python
from bs4 import BeautifulSoup
|
|
#import crawler.objects.models
|
|
#from crawler.objects.models import Object
|
|
from dateutil.parser import parse
|
|
from datetime import datetime
|
|
import re
|
|
import urlparse
|
|
from src import clogger, cfg
|
|
from fixing import fix_link
|
|
import feedparser
|
|
|
|
#from fetching import downloadfile
|
|
import json
|
|
def do_compile(tpe, cont):
|
|
if type(cont) != dict:
|
|
clogger.error("Type Error for do compile for :"+str(cont["url"]))
|
|
# Starting to compile an generic object
|
|
if "url" not in cont:
|
|
clogger.error("no url can't compile "+tpe)
|
|
else:
|
|
clogger.debug("compile: type:"+str(tpe)+ "| "+ str(cont["url"]))
|
|
if tpe in compiler:
|
|
cont=compiler[tpe](cont["url"], cont["raw"])
|
|
return cont
|
|
|
|
from comp import rssfeed
|
|
from comp import fbfeed
|
|
def dummyarticle(url, raw):
|
|
return {"url": url, "article":{"url": url, "section": "dummysection", "sourcetype": "dummy", "title":"dummytitle", "text": raw, "image": "fff", "author": "me", "published": None}}
|
|
|
|
|
|
|
|
def htufeed(url,raw):
|
|
al=[]
|
|
f=feedparser.parse(raw)
|
|
for e in f['entries']:
|
|
al.append(e['link'])
|
|
return {"url": url, "next_page": None, "article_links": al, "objecttype":"index"}
|
|
|
|
|
|
def htuarticle(url,raw):
|
|
sp=BeautifulSoup(raw)
|
|
d={}
|
|
h=sp.find("meta", {"property": "og:image"})
|
|
if h is not None:
|
|
d["image"]=h.attrs["content"]
|
|
d["image2"]=d["image"]
|
|
h=sp.find("div", {"class": "patternRevInfo"})
|
|
if h is not None:
|
|
# clogger.debug(h.text.strip())
|
|
h1= re.sub(r'.*- (\d+) ([a-zA-Z]+) (\d+) - ([:\d]+)[^\d]*', r'\3/\2/\1 \4',unicode(h.text.strip()))
|
|
# clogger.debug(h1)
|
|
d["published"]=parse(h1)
|
|
# clogger.debug(parse(h1))
|
|
# clogger.debug(d["published"])
|
|
h=h.find("a")
|
|
if h is not None:
|
|
d["author"]=h.text.strip()
|
|
h=sp.find("div", {"class": "foswikiTopic"})
|
|
h1=h.find("h4")
|
|
if h1 is not None:
|
|
d["title"]= h1.text.strip()
|
|
h1.extract() # remove head
|
|
else:
|
|
h1=sp.find("meta", {"name": "WEBTOPIC"})
|
|
d["title"]= h1.attrs["content"]
|
|
d["text"]=(h.encode_contents()).strip()
|
|
d["section"]="HTU"
|
|
d["url"]=url
|
|
# clogger.debug(d)
|
|
return {"article": d}
|
|
|
|
|
|
def fetarticle(url, raw):
|
|
sp=BeautifulSoup(raw)
|
|
d={}
|
|
h=sp.find("h1", {"itemprop": "name"})
|
|
d["title"]=unicode(h.text).strip()
|
|
h=sp.find("div", {"itemprop": "articleBody"})
|
|
if h is not None:
|
|
d["text"]=(h.encode_contents()).strip()
|
|
else:
|
|
d["text"]=""
|
|
d["url"]=url
|
|
h=sp.find("span", {"itemprop": "author"})
|
|
if h is not None:
|
|
d["author"]=h.text.strip()
|
|
h=sp.find("span", {"itemprop": "articleSection"})
|
|
if h is not None:
|
|
d["section"]= "FET - " + h.text.strip()
|
|
|
|
|
|
h=sp.find("span", {"itemprop": "datePublished"})
|
|
if h is not None:
|
|
h1= re.sub(r'[^\d]*(\d+)\.(\d+)\.(\d+)[^\d]*', r'\3/\2/\1',unicode(h.text).strip())
|
|
d["published"]=parse(h1.strip())
|
|
|
|
h=sp.find("meta", {"property": "og:image"})
|
|
if h is not None:
|
|
d["image"]=h.attrs["content"]
|
|
d["image2"]=d["image"]
|
|
# hh=sp.find_all("div", {"class":"media"})
|
|
# for h in hh:
|
|
# if h is not None:
|
|
# h=h.find("div", {"class": "pull-left"})
|
|
# if h is not None:
|
|
# h=h.find("a")
|
|
# if h is not None:
|
|
# d["image2"]=downloadfile(fix_link(url,h.attrs["href"]))
|
|
return {"article": d}
|
|
|
|
|
|
def fsarcharticle(url, raw):
|
|
sp=BeautifulSoup(raw)
|
|
d={}
|
|
h=sp.find("h1", {"class": "title"})
|
|
if h is not None:
|
|
d["title"]=h.text.strip()
|
|
d["url"]=url
|
|
d["published"]=None
|
|
h=sp.find("article")
|
|
if h is not None:
|
|
h=h.find("div", {"class": "content"})
|
|
d["text"]=h.encode_contents().strip()
|
|
h=sp.find("article")
|
|
if h is not None:
|
|
h=h.find("h1", {"class": "title"})
|
|
if h is not None:
|
|
d["title"]=h.text.strip()
|
|
else:
|
|
d["title"]=""
|
|
d["image"]=""
|
|
d["sourcetype"]="fsarcharticle"
|
|
d["section"]="fsarch"
|
|
d["author"]=None
|
|
return {"article": d}
|
|
|
|
def fsbizarticle(url, raw):
|
|
sp=BeautifulSoup(raw)
|
|
d={}
|
|
h=sp.find("h1", {"class": "entry-title"})
|
|
if h is not None:
|
|
d["title"]=h.text.strip()
|
|
d["url"]=url
|
|
h=sp.find("time", {"class": "entry-date"})
|
|
if h is not None:
|
|
d["published"] = parse(h.attrs["datetime"])
|
|
else:
|
|
d["published"]=None
|
|
h=sp.find("div", {"class": "entry-content"})
|
|
if h is not None:
|
|
d["text"]=h.encode_contents().strip()
|
|
d["image"]=""
|
|
d["sourcetype"]="fsbizarticle"
|
|
d["section"]="fsbiz"
|
|
h=sp.find("span", {"class": "author"})
|
|
d["author"]=None
|
|
if h is not None:
|
|
d["author"]=h.find("a").text.strip()
|
|
return {"article": d}
|
|
|
|
def fetindex(url, raw):
|
|
if raw is None:
|
|
raise Error
|
|
# clogger.debug("compile_fetindex: "+str(url))
|
|
html=BeautifulSoup(raw)
|
|
h = html.find("li", {"class": "next_page" })
|
|
if h is not None:
|
|
nl=h.find("a")
|
|
nl=fix_link(url,nl.attrs["href"])
|
|
else:
|
|
nl=None
|
|
h= html.find("ul", {"id": "neuigkeiten"})
|
|
al = []
|
|
if h is not None:
|
|
links=h.find_all("a")
|
|
for t in links:
|
|
al.append(t.attrs["href"])
|
|
return {"url": url, "next_page": nl, "article_links": al, "objecttype": "index" }
|
|
|
|
def fsarchindex(url, raw):
|
|
if raw is None:
|
|
raise Error
|
|
html=BeautifulSoup(raw)
|
|
h= html.find("article")
|
|
print unicode(h)
|
|
links=h.find_all("a")
|
|
al = []
|
|
fl=[]
|
|
for t in links:
|
|
url=t.attrs["href"]
|
|
if re.search("fachschaftarchitektur\.at", url):
|
|
al.append(t.attrs["href"])
|
|
if re.search("facebook\.com/events", url):
|
|
fl.append(t.attrs["href"])
|
|
|
|
return {"url": url, "next_page": None, "article_links": al, "facebook_links": fl,"objecttype":"index"}
|
|
|
|
|
|
def fsbizindex(url, raw):
|
|
if raw is None:
|
|
raise Error
|
|
print "compile_fsbizindex"
|
|
html=BeautifulSoup(raw)
|
|
h= html.find("section", {"id": "primary"})
|
|
links=h.find_all("h1", {"class": "entry-title"})
|
|
al = []
|
|
for t in links:
|
|
|
|
al.append(t.find("a").attrs["href"])
|
|
return {"url": url,"article_links": al,"objecttype": "index"}
|
|
|
|
|
|
|
|
|
|
def fsmbindex(url, raw):
|
|
if raw is None:
|
|
raise Error
|
|
html=BeautifulSoup(raw)
|
|
h= html.find("a",{"class": "next"})
|
|
if h is not None:
|
|
np=h.attrs["href"]
|
|
else:
|
|
np=None
|
|
h=html.find("div", {"id": "main"}).find("div", {"class": "inside"}).find("div", {"class": "mod_newslist"})
|
|
if h is not None:
|
|
ats=h.find_all("div",{"class": "block"})
|
|
articles=[]
|
|
for a in ats:
|
|
aa={}
|
|
h=a.find("h3")
|
|
if h is not None:
|
|
aa["title"] = h.text.strip()
|
|
h=a.find("div", {"class": "ce_text"})
|
|
if h is not None:
|
|
aa["text"] = (h.encode_contents()).strip()
|
|
aa["info"]=[]
|
|
hh=a.find_all("p", {"class": "info"},recursive=False)
|
|
for h in hh:
|
|
aa["info"].append(unicode(h.text))
|
|
if re.search(r'von', str(h)):
|
|
h1= re.sub(r'[^\d]*(\d+)\.(\d+)\.(\d+)[^\d]*', r'\3/\2/\1',unicode(h.text))
|
|
aa["published"] =parse(h1.strip())
|
|
aa["author"]=re.sub(r'^.*von(.*)$', r'\1',unicode(h.text)).strip() #h.text + "--" #+ re.sub(r'[^\d]*(\d+)\.(\d+)\.(\d+)[^\d]*', r'\3/\2/\1',hh)
|
|
aa["section"]="FSMB"
|
|
articles.append(aa)
|
|
return {"url": url, "next_page": np, "articles": articles,"objecttype": "articles"}
|
|
|
|
compiler = {"fetindex": fetindex, "fetarticle": fetarticle, "fsarchindex": fsarchindex, "fsarcharticle": fsarcharticle, "fsmbindex": fsmbindex, "fsbizindex": fsbizindex, "dummyarticle": dummyarticle,"htuarticle": htuarticle, "htufeed": htufeed, "fbfeed": fbfeed, "fschfeed": rssfeed}
|
|
|
|
compiler = cfg.compiler
|
|
for i in compiler:
|
|
compiler[i]=eval(compiler[i])
|
|
|
|
|
|
|
|
article_types={"fetindex" : "fetarticle", "fsarchindex": "fsarcharticle", "fsbizindex": "fsbizarticle", "dummyindex": "dummyarticle", "htufeed": "htuarticle"}
|