zz

.dockerignore

@@ -0,0 +1,2 @@
+venv
+websucker.egg-info

Dockerfile

@@ -1,15 +1,16 @@
-FROM python:3.8.0-alpine
+FROM python:3.8
 
-RUN apk add --update --no-cache git curl curl-dev vim py3-lxml gcc make libxml2-dev libxslt-dev libc-dev
+RUN apt-get update && apt-get install -y git curl  libcurl4-openssl-dev build-essential vim libssl-dev python-pip
 
-RUN apk add py2-pip 
 RUN pip2 install cqlsh
 
-RUN addgroup -S appgroup -g 1000 && \
-    adduser -u 1000 -S appuser -G appgroup
+RUN addgroup appgroup && \
+    adduser  appuser  && adduser appuser appgroup
 
 RUN mkdir /app /src
-ADD requirements.txt setup.py ./websucker /src/
-
+ADD requirements.txt  /src/
 RUN CASS_DRIVER_BUILD_CONCURRENCY=4 pip install -r /src/requirements.txt
+WORKDIR /src
+ADD . /src
+RUN python /src/setup.py install
 WORKDIR /app

websucker/agent.py

@@ -283,6 +283,9 @@ class Connection:
         return res
 
 class ParsedDocument:
+    """
+    One document in the database
+    """
     def __init__(self, parser,work_link):
         self.parser = parser
         self.work_link = work_link
@@ -304,6 +307,9 @@ class ParsedDocument:
         self.current_time = datetime.date.today()
 
     def extract(self,content,bs):
+        """
+        Parse content and fill the object
+        """
         self.content = content
         self.bs = bs
 
@@ -336,9 +342,15 @@ class ParsedDocument:
                 pass
 
     def get_links(self):
+        """
+        @return all links
+        """
         return self.link_set
 
     def get_follow_links(self):
+        """
+        @return good normalized links
+        """
         follow_links = set()
         for l in self.link_set:
             if self.parser.is_link_good(l):
@@ -355,24 +367,18 @@ class ParsedDocument:
             if (len(self.body) < 20):
                 r.append(self.body)
             else:
-                r.append(self.body[0:20]) + " ...."
+                r.append(self.body[0:20] + " ....")
         return ">>> ".join(r)
 
 
-def get_domains(arg):
-    domains = []
-    if arg == "-":
-        for l in sys.stdin:
-            domain = l.rstrip()
-            assert(domain is not None)
-            if len(domain) == 0:
-                continue
-            domains.append(domain)
-    else:
-        domains = arg.split(",")
-    return domains
-
 def parse_and_index(work_link,parser,responses,db):
+    """
+    Take all responses from work link, parse and store in db
+    @param work_link - final link from downloader
+    @param parser to use
+    @param responses from the downloader
+    @param db 
+    """
     target_link = work_link
     links = []
     if len(responses) > 0:
@@ -387,6 +393,9 @@ def parse_and_index(work_link,parser,responses,db):
     return target_link,links
 
 def visit_sitemap(domain,connection,parser,db):
+    """
+    get links from sitemap of the domain
+    """
     link = "http://" + domain
     print("Sitemap visit: " + link)
     responses = connection.html_download2(link)
@@ -402,8 +411,10 @@ def visit_sitemap(domain,connection,parser,db):
     return True
 
 
-def visit_links(links,connection,parser,db,is_online):
-    # is is not online, then just check links
+def visit_links(links,connection,parser,db,is_online=True):
+    """
+    if the site is not online, then just check links
+    """
     outlinks = []
     junklinks = []
     badrobotlinks = []
@@ -423,6 +434,14 @@ def visit_links(links,connection,parser,db,is_online):
         db.index_follow_links(parser,outlinks,connection)
 
 def visit_domain(domain,parser,db):
+    """
+    One visit of the domain
+
+    1.Get links from the frontpage,
+    2. visit links and extract new links
+    3. get new links to visit
+    4. repeat visit for parser.crawl_rounds
+    """
     c = Connection()
     p = parser
     # Get links from frontpage
@@ -442,6 +461,10 @@ def visit_domain(domain,parser,db):
     return True
 
 def process_domains(domains,visit,parser,db,queue):
+    """
+    Visit all domains in list.
+    if queue is true, then queue domain instead immediate visit
+    """
     print("Websucker Agenda>>")
     random.shuffle(domains)
     for domain in domains:
@@ -460,6 +483,9 @@ def process_domains(domains,visit,parser,db,queue):
             visit_domain(domain[0],parser,db)
 
 def work_domains(parser,db,queue):
+    """
+    Poll the queue and visit
+    """
     while True:
         print("Waiting for a new job:")
         job = queue.reserve()

websucker/cli.py

@@ -15,7 +15,7 @@ def create_database_from_context(ctx):
     return Data(ctx.obj["cassandra_keyspace"],ctx.obj["cassandra_host"],ctx.obj["cassandra_port"])
 
 def create_queue_from_context(ctx):
-    return greenstalk.Client(ctx.obj["beanstalkd_host"],ctx.obj["beanstalkd_port"],use=ctx.obj["beanstalkd_tube"],watch=ctx.obj["beanstalkd_tube"],encoding="utf8")
+    return greenstalk.Client((ctx.obj["beanstalkd_host"],ctx.obj["beanstalkd_port"]),use=ctx.obj["beanstalkd_tube"],watch=ctx.obj["beanstalkd_tube"],encoding="utf8")
 
 
 @click.group()
@@ -52,7 +52,7 @@ def cli(ctx,cassandra_keyspace,cassandra_host,cassandra_port,beanstalkd_tube,bea
     ctx.obj["queue"] = queue
 
 
-@cli.command(help="All domains")
+@cli.command(help="Get visited domains from db")
 @click.pass_context
 @click.argument("count",type=int,default=20)
 def all(ctx,count):
@@ -63,28 +63,7 @@ def all(ctx,count):
         q = create_queue_from_context(ctx)
     process_domains(res,ctx.obj["visit"],ctx.obj["parser"],db,q)
 
-@cli.command(help="Work queue")
-@click.pass_context
-def work(ctx):
-    db = create_database_from_context(ctx)
-    q = create_queue_from_context(ctx)
-    work_domains(ctx.obj["parser"],db,q)
-
-
-@cli.command(help="find best domains")
-@click.pass_context
-@click.argument("count",type=int,default=20)
-#@click.option("visit",is_flag=True)
-def best(ctx, count):
-    db = create_database_from_context(ctx)
-    p = ctx.obj["parser"]
-    domains = db.get_best_domains(count,p)
-    q = None
-    if ctx.obj["queue"]:
-        q = create_queue_from_context(ctx)
-    process_domains(domains,ctx.obj["visit"],p ,db,q)
-
-@cli.command(help="select random domains")
+@cli.command(help="Get random domains")
 @click.pass_context
 @click.argument("count",type=int,default=20)
 #@click.option("visit",is_flag=True)
@@ -97,7 +76,29 @@ def blind(ctx, count):
         q = create_queue_from_context(ctx)
     process_domains(domains,ctx.obj["visit"],p ,db,q)
 
-@cli.command(help="Find unvisited domains, Visit a site, get links and crawl")
+@cli.command(help="Visit domains from queue")
+@click.pass_context
+def work(ctx):
+    db = create_database_from_context(ctx)
+    q = create_queue_from_context(ctx)
+    work_domains(ctx.obj["parser"],db,q)
+
+
+@cli.command(help="Get best domains from db")
+@click.pass_context
+@click.argument("count",type=int,default=20)
+#@click.option("visit",is_flag=True)
+def best(ctx, count):
+    db = create_database_from_context(ctx)
+    p = ctx.obj["parser"]
+    domains = db.get_best_domains(count,p)
+    q = None
+    if ctx.obj["queue"]:
+        q = create_queue_from_context(ctx)
+    process_domains(domains,ctx.obj["visit"],p ,db,q)
+
+
+@cli.command(help="Get unvisited domains")
 @click.pass_context
 @click.argument("count",type=int,default=20)
 def unvisited(ctx, count):
@@ -124,7 +125,7 @@ def file(ctx, name):
         q = create_queue_from_context(ctx)
     process_domains(domains,ctx.obj["visit"],p,db,q)
 
-@cli.command(help="Visit url and get links. Start here")
+@cli.command(help="Visit one url and get links. Start here")
 @click.pass_context
 @click.argument("link")
 def start(ctx, link):
@@ -132,7 +133,7 @@ def start(ctx, link):
     p = ctx.obj["parser"]
     c = Connection()
     visit_links([link],c,p,db)
-    db.check_domain(domain)
+    #db.check_domain(domain)
 
 @cli.command(help="Continue crawling of seen links from a domain")
 @click.pass_context