Feineigle.com - Craigslist Scnner

Home · Projects · 2014 · Craigslist Scnner

Published: October 11, 2014 (9 years 6 months ago.)
Updated:   September 14, 2016 (7 years 6 months ago.)
Tags:  Python · Software · Sqlite

Threads in python only let you “stack” the network delays.

No interface, simply edit the file to get what you wanted, quick and dirty.

Hasn’t been used in over a year many years.

#!/usr/bin/env python
# Scans Craigslist city/category RSS pages and saves new data to SQLite 
#   (file:dbfile, table:cl)
# 2014/10/21
# 2014/11/01 - added concurrent threads, moved to project folder, added urllib_retry
# 2016-09-14 - updated, formattting, git tracking, getting ready to use in portfolio
"""Logs all posts in a given city and category to SQLite"""
import cookielib
from cookielib import CookieJar
from datetime import datetime
try: import Queue
except: import queue as Queue
import random # to randomly stagger threads for better output
import re
import socket # for timeout/retry
import sqlite3
import string
import thread
import threading
import time
import urllib2
import urllib_retry # local module
# browser setup
cj = CookieJar()
opener = urllib2.build_opener(urllib2.HTTPCookieProcessor(cj)) 
opener.addheaders = [('User-agent','Mozilla/5.0')] # fake header
# table setup
#connection = sqlite3.Connection("dbfile", detect_types=sqlite3.PARSE_COLNAMES)
#c = connection.cursor()
# if the cl table exists, drop it (makes new table each run for testing)
#c.execute("DROP TABLE IF EXISTS cl")
#                            Scanned TEXT,
#                            Posted TEXT,
#                            Links TEXT,
#                            Titles TEXT,
#                            Bodies TEXT)") # make a new cl table

# enter city(s) and category(s)
cities = ['charleston','flagstaff','dallas','boston','miami','lasvegas']
categories = ['rid','act']

q = Queue.Queue()

alive_threads = 0 # track how many threads are alive
lock = thread.allocate_lock() # create a lock object
pages_scanned = 0

class worker_thread(threading.Thread):
  def __init__(self, city, category):
    global alive_threads, pages_scanned
    self.city = city
    self.category = category
    alive_threads+=1 # increment under lock

  def run(self):
    global alive_threads, pages_scanned
    self.connection = sqlite3.Connection("bin/dbfile",
    self.connection.text_factory = str
    self.c = self.connection.cursor()
    self.titles, self.bodies, self.posttimes = [], [], []

    print "Scanning:", self.city, self.category
    scanned_links = self.link_scanner()
    self.titles = self.page_scanner(scanned_links)

    alive_threads-=1 # de-increment under lock
    print "Done scanning %s. %d threads still active." % (self.city, alive_threads)
    pages_scanned = pages_scanned + len(self.titles)# increment # of pages scanned

  @urllib_retry.retry(urllib2.URLError, tries=2, delay=3)
  @urllib_retry.retry(socket.timeout, tries=2, delay=3)
  def link_scanner(self):
    '''scan each starting_URL for links
    url = "http://"+self.city+".craigslist.org/search/"+self.category+"?format=rss"
    html = opener.open(url, timeout=3).read() # open the starting page
    scanned_links = re.findall("<link>(.*?)</link>", html) # find each post link
    for link in scanned_links[:]: #slice new copy, maintains original
      self.c.execute("SELECT Id FROM cl WHERE Links = ?", (link,))
      data = self.c.fetchone()
      if data is not None:         # if there is data... duplicate url found,
	scanned_links.remove(link) # remove it from the unsliced original
    return scanned_links

  @urllib_retry.retry(urllib2.URLError, tries=2, delay=3)
  @urllib_retry.retry(socket.timeout, tries=2, delay=3)
  def page_scanner(self, scanned_links):
    """Scan each link for title, body, etc
    global q
    if len(scanned_links) > 1:
      print ("%d new files in %s's %s." % (len(scanned_links)-1,
      for scanned_link in scanned_links:
	if scanned_link[-3:] == "rss": # skip scanning the rss feed
	  html = opener.open(scanned_link, timeout=3).read()
	  print "-"*80,"\nScanning:", scanned_link
	  # pulls the title from each scanned_URL
	  self.bodies.append(re.findall(r'<section id="postingbody">(.*?)</section>',
                             html ,re.DOTALL)) # DOTALL . matches \n
                                html ,re.DOTALL))
	  # remove the show contact info info, remove the <br>s
          # and strip the whitespace
	  target = '<a href=".*" class="showcontact" \
                     title="click to show contact info" \
                     rel="nofollow">show contact info</a>'
	  source = str(self.bodies[-1][0].replace('<br>','').strip())
          #NOTE bodies/titles[-1] is a list? [-1][0] is the string?? 
          #List of list of strings?
	  self.bodies[-1][0] = re.sub(target, '', source)
	  self.date = datetime.strftime(datetime.now(), "%Y-%m-%d %H:%M:%S")

#          print "SCAN:\t",self.date,
#                "\nPOST:\t", self.posttimes[-1][:],
#                "\nTITLE:\t",self.titles[-1][0],
#                "\nBODY:\t",self.bodies[-1][0],
#                "\n"

#	  insert data into db, cleanup bodies
#          self.c.execute("INSERT INTO cl (Scanned, Posted, Links, Titles, Bodies)\
#                          VALUES (?, ?, ?, ?, ?)",
#                          (self.date,
#	                   str(self.posttimes[-1]),
#                          scanned_link,
#                          str(self.titles[-1][0]),
#                          str(self.bodies[-1][0].replace('<br>', '').strip())))
#	  self.connection.commit()   # move outside loop w/ good internet
	  print scanned_link, "Queued not written." #NOTE, uncomment sql to log
          # write and remove each entry to get around crappy connection
	print ("\n0 new files in %s's %s." % (self.city, self.category))
      return self.titles

if __name__ == '__main__':
  print "%d cities to scan" % len(cities)
  for city in cities:
    for category in categories:
      t = worker_thread(city, category)	
  time.sleep(.1) # need a pause to increment the first alive_threads
  while alive_threads > 0:
  while not q.empty():
    d = q.get()
    print d
  print "\n","-"*88
  print ("Scan of %d items completed at: %s" % (pages_scanned,
                                                          "%Y-%m-%d %H:%M:%S")))
#!/usr/bin/env python
# http://www.saltycrane.com/blog/2009/11/trying-out-retry-decorator-python/
import time
from functools import wraps
def retry(ExceptionToCheck, tries=4, delay=3, backoff=2, logger=None):
    """Retry calling the decorated function using an exponential backoff.

    original from: http://wiki.python.org/moin/PythonDecoratorLibrary#Retry

    :param ExceptionToCheck: the exception to check. may be a tuple of
        exceptions to check
    :type ExceptionToCheck: Exception or tuple
    :param tries: number of times to try (not retry) before giving up
    :type tries: int
    :param delay: initial delay between retries in seconds
    :type delay: int
    :param backoff: backoff multiplier e.g. value of 2 will double the delay
        each retry
    :type backoff: int
    :param logger: logger to use. If None, print
    :type logger: logging.Logger instance
    def deco_retry(f):
        def f_retry(*args, **kwargs):
            mtries, mdelay = tries, delay
            while mtries > 1:
                    return f(*args, **kwargs)
                except ExceptionToCheck, e:
                    msg = "%s, Retrying in %d seconds..." % (str(e), mdelay)
                    if logger:
                        print msg
                    mtries -= 1
                    mdelay *= backoff
            return f(*args, **kwargs)
        return f_retry  # true decorator
    return deco_retry