dano/websucker-pip
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity

zz

Browse Source
This commit is contained in:
Daniel Hládek 2023-04-17 15:09:49 +02:00
commit 01645b8862
1 changed files with 3 additions and 12 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

15
mongo/mongocrawler.py
View File

@ -66,6 +66,7 @@ def get_bs_links(link,html):
netloc = parsed.netloc
path = os.path.normpath(parsed.path)
scheme = parsed.scheme
query = parsed.query
# internal link
if parsed.netloc == "":
scheme = base.scheme
@ -79,7 +80,7 @@ def get_bs_links(link,html):
if path.endswith(")"):
# javascript
continue
href = urllib.parse.urlunparse((scheme,netloc,path,"","",""))
href = urllib.parse.urlunparse((scheme,netloc,path,"",query,""))
href = courlan.normalize_url(href)
links.add(href)
except ValueError as err:
@ -238,7 +239,6 @@ def index_page(db,original_link,final_link,html,doc,filter_content=True):
state = "good"
link = original_link
if original_link != final_link:
print(original_link,final_link)
linkcol.update_one({"url":original_link},{"$set":{"status":"redirect"}})
link = final_link
if html is None:
@ -277,10 +277,7 @@ def index_page(db,original_link,final_link,html,doc,filter_content=True):
del doc["url"]
contentcol.update_one({"url":link},{"$set":doc},upsert=True)
for chs in doc["paragraph_checksums"]:
try:
checkcol.insert_one({"_id":chs})
except pymongo.errors.DuplicateKeyError as err:
pass
checkcol.update_one({"_id":chs},{"$inc":{"count":1}},upsert=True)
linkdoc = get_link_doc(link,state)
del linkdoc["url"]
@ -308,7 +305,6 @@ def save_batch_info(db,host,states,docs):
"batch_size": batch_size,
}
db["batches"].insert_one(batchdoc)
print(batchdoc)
def extract_links(link_batch:list,responses:list,hostname:str,rules,default_status="frontlink")->list:
@ -319,15 +315,11 @@ def extract_links(link_batch:list,responses:list,hostname:str,rules,default_stat
if html is None or len(html) < 256:
continue
page_links = get_bs_links(final_link,html)
#external_links = courlan.extract_links(html,final_link,external_bool=True,language=LANGUAGE)
#internal_links = courlan.extract_links(html,final_link,external_bool=False,language=LANGUAGE)
#print(extracted_links)
for link in page_links:
if not courlan.is_external(link,final_link) and not is_robot_good(link,rules):
badrobot += 1
continue
status = str(default_status)
#print(link,status)
links[link] = status
outlinks = []
badlink = 0
@ -453,7 +445,6 @@ class LinkClassifier:
goodprob += g
b = math.log(self.badcounter[feature] + self.alpha) - bcc
badprob += b
print(feature,g,b)
pa = math.exp(goodprob + gp)
pb = math.exp(badprob + bp)
return pa - pb #+ random.uniform(-0.001,0.001)