Merge branch 'master' of git.kemt.fei.tuke.sk:dano/websucker-pip

mongo/cli.py

@@ -4,9 +4,9 @@ import rq
 import redis
 import sys
 import os
-
-REDIS_URL= os.getenv("REDIS_URL","redis://localhost:6379/")
-QUEUES=os.getenv("QUEUES","high,default,low")
+import pymongo
+import courlan
+from config import *
 
 @click.group()
 @click.option("--dbname",default=mongocrawler.DBNAME,help="database to use")
@@ -47,6 +47,14 @@ def visit(hostname,filter_content=True):
     """ Hostname to crawl """
     mongocrawler.visit(hostname,filter_content=filter_content)
 
+@cli.command()
+@click.argument("hostname")
+def linksummary(hostname):
+    myclient = pymongo.MongoClient(CONNECTION)
+    db=myclient[DBNAME]
+    mongocrawler.link_summary(db,hostname)
+
+
 @cli.command()
 def summary():
     mongocrawler.crawl_summary()
@@ -55,11 +63,46 @@ def summary():
 def sampledomains():
     mongocrawler.sample_domains()
 
+@cli.command()
+@click.argument("domain")
+def sample(domain):
+    myclient = pymongo.MongoClient(CONNECTION)
+    db=myclient[DBNAME]
+    links = mongocrawler.sample_links(db,domain,"frontlink",BATCH_SIZE)
+    print(links)
+
+@cli.command()
+@click.argument("start_link")
+def fetchlinks(start_link):
+    myclient = pymongo.MongoClient(CONNECTION)
+    db=myclient[DBNAME]
+    start_link,hostname = courlan.check_url(start_link)
+    rules = mongocrawler.fetch_robot(hostname)
+    links = mongocrawler.fetch_front_links(start_link,rules)
+    for link in links:
+        print(link[0])
+    #print(front_links)
+    mongocrawler.index_links(db,links)
+
+
+@cli.command()
+@click.argument(hostname)
+def process_links():
+    rules = mongocrawler.fetch_robot(hostname)
+    outfile = "data.jsonl"
+    links = []
+    for line in sys.stdin:
+        links.append(line.rstrip())
+    extracted_pages, extracted_links = fetch_and_extract(links,rules)
+    for page in extracted_pages:
+        print(page)
+    pass
+
 
 @cli.command(help="Enqueue a list of links into redis queue for crawling")
 def enqueue():
     # TODO: select queues
-    q = rq.Queue(connection=redis.from_url(REDIS_URL))
+    q = rq.Queue(connection=redis.from_url(CONNECTION))
     for l in sys.stdin:
         print(l.strip())
         r = q.enqueue(mongocrawler.visit, l.strip())

mongo/config.py

@@ -0,0 +1,24 @@
+import os
+# database options
+CONNECTION=os.getenv("SUCKER_CONNECTION","mongodb://root:example@localhost:27017/")
+
+QUEUES=os.getenv("QUEUES","high,default,low")
+DBNAME=os.getenv("SUCKER_DBNAME","crawler")
+# retrieving filter
+BATCH_SIZE = int(os.getenv("SUCKER_BATCH_SIZE","10"))
+MIN_FILE_SIZE=int(os.getenv("SUCKER_MIN_FILE_SIZE","300"))
+MAX_FILE_SIZE=int(os.getenv("SUCKER_MAX_FILE_SIZE","10000000"))
+# document originality filter
+MIN_TEXT_SIZE=int(os.getenv("SUCKER_MIN_TEXT_SIZE","200"))
+CHECK_PARAGRAPH_SIZE=int(os.getenv("SUCKER_CHECK_PARAGRAPH_SIZE","150"))
+TEXT_TRASH_RATIO=float(os.getenv("SUCKER_TEXT_TRASH_RATIO","0.6"))
+# link and domain sampling
+DISCOVER_LINK_RATIO = float(os.getenv("SUCKER_DISCOVER_LINK_RATIO","0.3"))
+SAMPLE_SET_SIZE = int(os.getenv("SUCKER_DISCOVER_LINK_RATIO","10000"))
+CLASSIFIER_SET_SIZE = int(os.getenv("SUCKER_DISCOVER_LINK_RATIO","200"))
+# link filter
+LANGUAGE= os.getenv("SUCKER_LANGUAGE","sk")
+DOMAIN = os.getenv("SUCKER_DOMAIN","sk")
+STOP_PATHS=os.getenv("SUCKER_STOP_PATHS","xml,rss,login,admin").split(",")
+
+

mongo/mongocrawler.py

@@ -27,25 +27,7 @@ import os.path
 import binascii
 import json
 
-# database options
-CONNECTION=os.getenv("SUCKER_CONNECTION","mongodb://root:example@localhost:27017/")
-DBNAME=os.getenv("SUCKER_DBNAME","crawler")
-# retrieving filter
-BATCH_SIZE = int(os.getenv("SUCKER_BATCH_SIZE","10"))
-MIN_FILE_SIZE=int(os.getenv("SUCKER_MIN_FILE_SIZE","300"))
-MAX_FILE_SIZE=int(os.getenv("SUCKER_MAX_FILE_SIZE","10000000"))
-# document originality filter
-MIN_TEXT_SIZE=int(os.getenv("SUCKER_MIN_TEXT_SIZE","200"))
-CHECK_PARAGRAPH_SIZE=int(os.getenv("SUCKER_CHECK_PARAGRAPH_SIZE","150"))
-TEXT_TRASH_RATIO=float(os.getenv("SUCKER_TEXT_TRASH_RATIO","0.6"))
-# link and domain sampling
-DISCOVER_LINK_RATIO = float(os.getenv("SUCKER_DISCOVER_LINK_RATIO","0.3"))
-SAMPLE_SET_SIZE = int(os.getenv("SUCKER_DISCOVER_LINK_RATIO","10000"))
-CLASSIFIER_SET_SIZE = int(os.getenv("SUCKER_DISCOVER_LINK_RATIO","200"))
-# link filter
-LANGUAGE= os.getenv("SUCKER_LANGUAGE","sk")
-DOMAIN = os.getenv("SUCKER_DOMAIN","sk")
-STOP_PATHS=os.getenv("SUCKER_STOP_PATHS","xml,rss,login,admin").split(",")
+from config import *
 
 def get_bs_links(link,html):
     # Extrakcia linkov zo stranky
@@ -691,6 +673,20 @@ def index_pages(db,hostname,extracted_pages,filter_content):
         docs.append(doc)
     save_batch_info(db,hostname,final_states,docs)
 
+def fetch_and_extract(links,rules):
+    print("Processing links")
+    responses = []
+    for link in links:
+        responses.append(fetch_page(link))
+    extracted_pages = []
+    for original_link,(final_link,html) in zip(links,responses):
+        doc = None
+        assert original_link is not None
+        doc = extract_page(final_link,html)
+        extracted_pages.append((original_link,html,doc))
+    extracted_links = extract_links(links,responses,rules,"frontlink")
+    return extracted_pages, extracted_links
+
     
 def visit(hostname,filter_content=True):
     myclient = pymongo.MongoClient(CONNECTION)
@@ -710,15 +706,7 @@ def visit(hostname,filter_content=True):
     print(links)
     # index results
     print("Processing links")
-    responses = []
-    for link in links:
-        responses.append(fetch_page(link))
-    extracted_pages = []
-    for original_link,(final_link,html) in zip(links,responses):
-        doc = None
-        assert original_link is not None
-        doc = extract_page(final_link,html)
-        extracted_pages.append((original_link,html,doc))
+    extracted_pages, extracted_links = fetch_and_extract(links,rules)
 
     extracted_links = extract_links(links,responses,hostname,rules,"frontlink")
 
@@ -731,7 +719,6 @@ def visit(hostname,filter_content=True):
         docs.append(doc)
     save_batch_info(db,hostname,final_states,docs)
     index_pages(db,hostname,extracted_pages,filter_content)
-    extracted_links = extract_links(links,responses,rules,"frontlink")
     index_links(db, extracted_links)
     link_summary(db,hostname)
 

mongo/start-docker-devstack.sh

@@ -0,0 +1,5 @@
+#!/usr/bin/sh
+docker pull redis
+docker pull mongo
+docker pull mongo-express
+docker stack deploy -c ./docker-compose.yaml websucker