Run Python Script in Automator

Question

I have a python script (which was written for me), and I would like to make it so that the script executes every x minutes. I know this should be simple to do, but I can't figure it out.
Thus far, I have created a workflow in automator, used the "Run Shell Script" action, and pasted the script into the text field.
"Workflow failed - 1 error
I'm very new to this, so I'm sure it's a simple error. Any help would great.
Here is the script I am trying to execute.
#!/Library/Frameworks/Python.framework/Versions/2.7/bin/python
# you can change the above line to point to the location of python
# on your system (you can check this by typing 'which python' into
# Terminal.app), but this isn't necessary if you execute the script
# using "python ksl.py [URL]"
# change the value of NAME to your desired name
NAME = "Bob Jones"
# change the value of EMAIL to your desired email
EMAIL = "[email protected]"
# your message will be the contact name as mined from the page source,
# followed by whatever message you enter between the following triple quotes
MESSAGE = """Replace this text with your message. Newlines are also OK."""
import mechanize
import re
import sys
def setupBrowser(url):
b = mechanize.Browser()
# b.sethandlerobots(False)
# b.addheaders = [('User-agent', 'Mozilla/4.0 (compatible; MSIE 6.0; Windows XP)')]
try:
b.open(url)
except mechanize._mechanize.BrowserStateError:
print >> sys.stderr, "You have mistyped the URL. It must by in the following format, including the quotes and the preceding 'http://':\n\t %s [--test] 'http://www.blah.com'" % (sys.argv[0])
sys.exit(1)
return b
def grabLinks(b):
"""Takes in a mechanize.Browser() object pointed to a listings URL and returns a listing of classified ad links."""
links = []
for link in b.links():
# change this line if the URL format ever changes
if re.search(r'&ad=', link.url):
links.append(mechanize.urljoin(link.base_url, link.url))
return links
if _name_ == '_main_':
# check for proper command line args
if len(sys.argv) != 2 and len(sys.argv) != 3:
print >> sys.stderr, "Usage: %s [--test] url" % (sys.argv[0])
sys.exit(1)
args = sys.argv[1:]
if len(args) == 1:
# start from listings page
url = args[0]
# set up the mechanize browser object
b = setupBrowser(url)
# grab only the relevant ad links
links = grabLinks(b)
if not links or len(links) == 0:
# the links do not follow the same format as the original page
print >> sys.stderr, "The link format has changed, or you have mistyped the URL."
sys.exit(1)
# open the first link on the listings page
b.open(links[0])
else:
# start from a single listing
if args[0] != "--test":
print >> sys.stderr, "Usage %s [--test] url"
sys.exit(1)
url = args[1]
b = setupBrowser(url)
# grab the HTML so that we can search for the contact name
response = b.response().get_data()
# perform a regex search on the HTML for the contact name
regexSearch = re.search(r'Contact Name:\s*\s(w+)s', response)
contactName = ""
if regexSearch:
# contact name found -- store it
contactName = regexSearch.group(1)
else:
# contact name not found -- use generic greeting
contactName = "Hello"
theOne = ""
# find the "Email Seller" link (stored as "theOne")
for link in b.links():
# again, if the URL changes, change this line
if re.search(r'xmldb=', link.url):
theOne = mechanize.urljoin(link.base_url, link.url)
if theOne == "":
# something went wrong
print >> sys.stderr, "'Email Seller' link has changed formats within the HTML."
sys.exit(1)
b.open(theOne)
# fill out the forms. note that I am grabbing the SECOND form here.
# again, this might change over time, so feel free to change this from
# nr=1 to nr=0 (for the first form), nr=2 (for the third form), etc.
b.select_form(nr=1)
b['form_4'] = NAME
b['form_5'] = EMAIL
# append the contact name to the rest of the message
MESSAGE = contactName + """,\n\n""" + MESSAGE
b['form_6'] = MESSAGE
# submit the form
b.submit()
b.close()

Answer 1

If the script works, and all you need is to execute every x minutes, use launchd, and its simple to write with the Lingon GUI: http://sourceforge.net/projects/lingon/files/