I’ve implemented a very simple automated comment poster. If you don’t make seemingly useful comments, don’t plan on getting too far, as most people moderate their comments. The code WILL fail if you do not note the following:
At the bottom of the file you’ll see a few variables that you’ll need to set.
blogUrl = “”"HTTP://WWW.YOUR_URL.COM”"”
Your blogUrl should be in the http://DOMAIN.com format.
keyword = “”"YOUR KEYWORD”"”
You need to change that to the keyword that you want. This will find all wp blogs similar to yours.
results = 50
You can turn this up to 100 or more.
{‘author’ : “”"YOUR AUTHOR NAME”"”, ‘email’ : “”"YOUR@EMAIL.COM”"”, ‘comment’ : “”"YOUR COMMENT”"”},
This line can be edited to your liking, and duplicated as many times as you like. I would suggest making at least 20 of these, all a bit different from the last one. Your posts will NOT work if your ‘email’ isn’t in the form of blah@blah.com.
Download the script commentonwordpressblogs.py
If you are trying to do JS in python, you’re going to run into many hurdles. The python javascript project, that I know of, is python-spidermonkey. The project lacks a few important object; document, window, navigator. Fortunately, there are those of us who have added to the project, and made these modules available, at least in some capacity.
You can find the document and window objects on Didier Stevens’ blog (link below). He’s done an excellent job at porting these modules over to spidermonkey.
I’ve put together a very rudimentary navigator object. The values are hard-coded, but the object makes you appear to be running Firefox on a Mac. If you’d like to change the code in anyway, be my guest. I was unable to find a port of this object, so I created on from scratch.
I’ve included all of the configuration files that you will need to work with Didier Stevens’ objects and my own. You need to visit Didier Steven’s blog to download his code.
Go to http://wwwsearch.sourceforge.net/python-spidermonkey/ and download/untar the project
Go to http://blog.didierstevens.com/programs/spidermonkey/ and follow his instructions
Download BlackCodeSeo Navigator.
Run python setup.py install
Python-SpiderMonkey Implemeting Navigator
>>> from spidermonkey import Runtime
>>>
>>> rt = Runtime()
>>> cx = rt.new_context()
>>>
>>> print cx.eval_script('navigator.appCodeName;')
Mozilla
>>> print cx.eval_script('navigator.appName;')
Netscape
>>> print cx.eval_script('navigator.appVersion;')
5.0 (Macintosh; en-US)
>>> print cx.eval_script('navigator.cookieEnabled;')
0
>>> print cx.eval_script('navigator.language;')
en-US
>>> print cx.eval_script('navigator.mimeTypes;')
None
>>> print cx.eval_script('navigator.platform;')
MacIntel
>>> print cx.eval_script('navigator.plugins;')
None
>>> print cx.eval_script('navigator.systemLanguage;')
undefined
>>> print cx.eval_script('navigator.userAgent;')
Mozilla/5.0 (Macintosh; U; Intel Mac OS X 10.5; en-US; rv:1.9.0.3) Gecko/2008092414 Firefox/3.0.3
>>> print cx.eval_script('navigator.userLanguage;')
en-ca
>>> print cx.eval_script('navigator.javaEnabled();')
0
Update: Automatically Installing Wordpress Forum Post
Automating the install of wordpress saves a great deal of time. You can easily write a script to copy wordpress to a directory, append your vhosts, auto-generate a wp-config.php and restart apache, so I won’t be including that part of the code in this post. I will cover the initial install, and also the sql injection to get wordpress up and running.
Automatically Installing Wordpress Part 1
import urllib
# make sure you set WEBLOG_TITLE and ADMIN_EMAIL
params = urllib.urlencode({'weblog_title' : WEBLOG_TITLE, 'admin_email' : ADMIN_EMAIL})
# install wordpress
f = urllib.urlopen('http://MY_DOMAIN/wp-admin/install.php?step=2', params)
data = f.read()
f.close()
At this point, you have installed wordpress, but you’ll want to set some options. This next example shows some examples of how you can manipulate the base installation of wordpress.
Automatically Installing Wordpress Part 2
import MySQLdb
tblPrefix = 'TABLE_PREFIX'
conn = MySQLdb.connect (host='localhost', user='USERNAME', passwd='PASSWORD', db='DATABASE')
cursor = conn.cursor()
# delete hello world post
cursor.execute('delete * from %sposts' % (tblPrefix))
# create a password, wp makes a random default one
cursor.execute("""update %susers set user_pass = md5('PASSWORD') where user_login='admin'""" % (tblPrefix))
# title the blog
cursor.execute('''update %soptions set option_value='TITLE_OF_BLOG' where option_id=3''' % (tblPrefix))
# make the blog public
cursor.execute('''update %soptions set option_value = '1' where option_name = 'blog_public' ''' % (tblPrefix))
# allow pingback
cursor.execute("""update %soptions set option_value = '1' where option_name = 'default_pingback_flag'""" % (tblPrefix))
# open for pings
cursor.execute("""update %soptions set option_value = 'open' where option_name = 'default_ping_status'""" % (tblPrefix))
# open for comments
cursor.execute("""update %soptions set option_value = 'open' where option_name = 'default_comment_status'""" % (tblPrefix))
# disallow comment registration
cursor.execute("""update %soptions set option_value = '0' where option_name = 'comment_registration'""" % (tblPrefix))
# non-static page
cursor.execute("""update %soptions set option_value = '0' where option_name = 'page_on_front'""" % (tblPrefix))
cursor.close()
conn.close()
Programatically posting to wordpress saves a TON of time, no more outsourcing. My prefered method of posting to wordpress is via xml-rpc, NOT sql injection. Michele Ferretti has developed an excellent library for controlling wordpress in such a matter. Wordpresslib allows you to; publish new posts, edit old posts, publish draft posts, delete posts, change post categories, get blog and user informations, upload multimedia files like movies or photos, get last recents post, get last post, get Trackbacks of posts, get Pingbacks of posts.
Automatically Posting To Wordpress
import wordpresslib
url = 'http://www.MYDOMAIN.com/xmlrpc.php'
wp = wordpresslib.WordPressClient(url, 'USERNAME', 'PASSWORD')
wp.selectBlog(0)
post = wordpresslib.WordPressPost()
post.title = 'Sample Post'
post.description = 'Sample Content'
idPost = wp.newPost(post, True)
A quick and easy way to create content is syndication. Rss is among the most popular methods of syndication and the best part about Rss is that you have full control over the content that you are “collecting”. Mark Pilgrim developed a nice rss parser to use with python; Universal Feed Parser.
Let’s “subscribe” to 10 rss feeds from google’s blog search, with our target keyword being “python”.
Parsing Google Blogsearch
import feedparser
d = feedparser.parse('http://blogsearch.google.com/blogsearch_feeds?hl=en&q=python&ie=utf-8&num=10&output=rss')
for feed in d['entries']:
print feed['link']
print feed['title']
print feed['content']
At this point we can grab a part of speech to replace with our keyword. Doing so will build up our likely hood of dominating the SERPS. This would be a good point to point out a few things.
1) Be frugal with the use your keyword(s).
2) Randomizing the order of the content that you’ve scraped makes it more difficult to flag you as a “scraper”.
3) Change the titles! Bloggers are ALWAYS reporting “scrapers”, so mix it up a bit.
4) When you parse URL feeds, use relevant topics, if you are tying to dominate “pay day loans”, don’t search for blogs on “potato salad”. Try to stay on point with your blog searching, I realize that you are going to be collecting lots of feeds. There is a finite number of blogs, seriously!
For the next chunk of code, you’ll need Oliver Steele’s PyWordNet; the python implementation of WordNet. I suggest reading up on this project, I’ve used it for many of my own projects. Let’s take a string, replace all Nouns with the keyword “python”.
Substituting Nouns With A Keyword With WordNet
from wordnet import *
from wntools import *
def substituteKeyword(description, keyword):
words = description.split(' ')
return string.join( map(lambda word: (keyword, word)[morphy(word, 'noun')==None], words), ' ' )
RSS submission sites are a great way to create backlinks quickly. http://www.feedage.com is one of the fastest RSS “reader” sites for giving immediate backlinks (within a few hours).
For this example, you will need mechanize
Submit To Feedage
# mechanize handles cookies, so we can actually log in
import mechanize
USERNAME = ''
PASSWORD = ''
RSS_URL = ''
# Instantiate the mechanize browser
br = mechanize.Browser()
# Add headers to make the "browser" look real
br.addheaders = [("User-agent", "Mozilla/5.0 (Macintosh; U; Intel Mac OS X 10.5; en-US; rv:1.9.0.3) Gecko/2008092414 Firefox/3.0.3"), ("HTTP_CONNECTION", "keep-alive"), ("HTTP_KEEP_ALIVE", "300"), ("HTTP_ACCEPT", "text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8"), ("HTTP_ACCEPT_CHARSET", "ISO-8859-1,utf-8;q=0.7,*;q=0.7"), ("HTTP_ACCEPT_ENCODING", "gzip,deflate"), ("HTTP_ACCEPT_LANGUAGE", "en-us,en;q=0.5"), ("HTTP_USER_AGENT", "Mozilla/5.0 (Macintosh; U; Intel Mac OS X 10.5; en-US; rv:1.9.0.3) Gecko/2008092414 Firefox/3.0.3")]
# We're not a robot
br.set_handle_robots(False)
# Navigate to the login page
br.open('''http://rss2.com/signin/''')
# Count the forms in the HTML, we want the 1st one
br.select_form(nr=0)
# Fill in the form elements by name
br["username"] = USERNAME
br["password"] = PASSWORD
# Click the submit button
login_response = br.submit()
# Read the login buffer, might want to actually do something with this
buffer = login_response.read()
# Navigate to your user's page
br.open('''http://rss2.com/users/%s''' % (USERNAME))
# Count the forms in the HTML, we want the 1st one
br.select_form(nr=0)
# Fill in the form elements by name
br["site_url"] = RSS_URL
# Click the submit button
resp = br.submit()
# View the results after the click
print resp.read()
Newer→
