jirabot.py

   1 import zephyr
   2 import feedparser
   3 import pycurl
   4 from BeautifulSoup import BeautifulSoup
   5
   6 """Screen scrapes jira.mit.edu. Supposedly. Doesn't actually work yet.
   7
   8 You need to have a personal certificate named cert.pem in this
   9 directory to run this script. You also need the abovementioned scripts.
  10 The zephyr bindings can be gotten from ebroder's GitHub repository
  11 located: http://github.com/ebroder/python-zephyr/tree/master
  12
  13 SSL server CA authentication is disabled because we don't know how to
  14 make curl stop complaining.
  15
  16 Current difficulty is we are causing a NullPointer exception on
  17 (we think) touchstone's servers."""
  18
  19 class Browser(object):
  20     def __init__(self, **kwargs):
  21         self.opts = kwargs
  22         self.lastResult = None
  23     def _callback(self, buf):
  24         self.lastResult += buf
  25     def get(self, url):
  26         c = pycurl.Curl()
  27         c.setopt(c.URL, url)
  28         c.setopt(c.WRITEFUNCTION, self._callback)
  29         for k,v in self.opts.items():
  30             c.setopt(getattr(c, k), v)
  31         self.lastResult = ''
  32         c.perform()
  33         return self.lastResult
  34     def getPage(self, url):
  35         return Page(url, self.get(url), self)
  36
  37 class Page(object):
  38     def __init__(self, url, contents, browser):
  39         self.url = url
  40         self.soup = BeautifulSoup(contents)
  41         self.browser = browser
  42     def submitForm(self, name, args ={}):
  43         form = self.soup.find(name='form', attrs={"name": name})
  44         if form == None: return False
  45         dest = form.get('action', self.url)
  46         # this has a lot of edge-cases that don't work
  47         inputs = form.findAll(name='input')
  48         vals = {}
  49         for input in inputs:
  50             vals[input[name]] = input[value]
  51         for k,v in args.items():
  52             vals[k] = v
  53         return self.browser.getPage()
  54
  55
  56 b = Browser(
  57     SSLCERT = "cert.pem",
  58     SSLKEY = "cert.pem",
  59     SSL_VERIFYPEER = 0,
  60     COOKIEJAR = "jirabot.cookie",
  61     FOLLOWLOCATION = 1,
  62     AUTOREFERER = 1,
  63     HTTPHEADER = ["Accept-Language: en-us,en;q=0.5"],
  64 )
  65
  66 b.get("https://jira.mit.edu/jira/secure/Dashboard.jspa")
  67 page = b.getPage("https://jira.mit.edu/jira/secure/mit_login.jsp?os_destination=/secure/Dashboard.jspa")
  68
  69 print page.submitForm('wayfForm1')
  70
  71 #c.setopt(c.URL, "https://jira.mit.edu/jira/sr/jira.issueviews:searchrequest-rss/temp/SearchRequest.xml?&pid=10185&updated%3Aprevious=-1w&sorter/field=updated&sorter/order=DESC&tempMax=1000")
  72
  73 #d = feedparser.parse(t.contents)
  74 #print d
  75