import re, sgmllib, sys, time, urllib
from glob import glob
from xml.sax.saxutils import escape
import extractor

PINGABLE   = re.compile('http://www.intertwingly.net/blog/(\d+).html')

class parser(sgmllib.SGMLParser):
  """ extract title and hrefs from a web page"""
  intitle=0
  title = ""
  hrefs = []

  def do_a(self, attrs):
    attrs=dict(attrs)
    if attrs.has_key('href'): self.hrefs.append(attrs['href'])

  def do_title(self, attrs):
    if self.title=="": self.intitle=1

  def unknown_starttag(self, tag, attrs):
    self.intitle=0
  def unknown_endtag(self,tag):
    self.intitle=0

  def handle_charref(self, ref):
    if self.intitle: self.title = self.title + ("&#%s;" % ref)
  def handle_data(self,text):
    if self.intitle: self.title = self.title + text

def excerpt(source):
  baseurl=source.split("#")[0]
  feedurl = extractor.html(baseurl).feed
  feed = minidom.parseString(feedurl)
  entries = feed.getElementsByTagName('entry')
  entries = entries or feed.getElementsByTagName('item')
  for entry in entries:
    (title, alternate, summary, content) = extractor.extract(entry, feedurl)
    if alternate==source:
      summary = summary or content
      summary=re.compile('<.*?>',re.S).sub('',summary)
      summary=re.sub('\s+',' ',summary)
      summary=summary[:summary.rfind(' ',0,250)][:250]
      return (title, summary)
  else:
    return ('', '')

def ping(source, target):
  from post import sanitize, writeComment, existingBacklink

  file = urllib.urlopen(source.split('#')[0])
  page = parser()
  page.feed(file.read())
  file.close()

  if not target in page.hrefs: return "href not found"

  for entry in PINGABLE.findall(target):
    if existingBacklink(entry,source): 
      return "target page has already been pinged"

    if page.title == "": page.title = source

    (title,body)=excerpt(source)
    title = escape(page.title.strip().replace('&amp;','&'))
    if body: body = sanitize(body) + " ...<br /><br />"
    source = escape(source.replace('&amp;','&'))
    body += 'Pingback from <a href="%s">%s</a>' % (source, title)
    writeComment(entry, title, body)

    return "ping successful"

  return "target page not eligable for pings"

def pingback():
  import os, xmlrpclib

  try:
    data = sys.stdin.read(int(os.environ["CONTENT_LENGTH"]))
    if not data: return
    params, method = xmlrpclib.loads(data)
    if method=='pingback.ping':
      message=apply(ping, params)
      response = xmlrpclib.dumps(tuple([message]), methodresponse=1)
    else:
      raise NameError, method
  except:
    import traceback
    tb = ''.join(apply(traceback.format_exception, sys.exc_info()))

    import nonce,time
    nonce.generate(time.time()+86400, tb)

    fault = xmlrpclib.Fault(1, tb)
    response = xmlrpclib.dumps(fault)
 
  print 'Content-type: text/xml'
  print
  print response