+from HTMLParser import HTMLParser
+
+class HTML2Text(HTMLParser):
+ entities = {
+ "amp": "&",
+ "lt": "<",
+ "gt": ">",
+ "pound": "£",
+ "copy": "©",
+ "apos": "'",
+ "quot": "\"",
+ "nbsp": " ",
+ }
+
+ def __init__(self):
+ self.inheadingone = False
+ self.inheadingtwo = False
+ self.inotherheading = False
+ self.inparagraph = True
+ self.inblockquote = False
+ self.inlink = False
+ self.text = u''
+ self.currentparagraph = u''
+ self.headingtext = u''
+ self.blockquote = u''
+ self.inpre = False
+ self.inul = False
+ self.initem = False
+ self.item = u''
+ HTMLParser.__init__(self)
+
+ def handle_starttag(self, tag, attrs):
+ if tag.lower() == "h1":
+ self.inheadingone = True
+ self.inparagraph = False
+ elif tag.lower() == "h2":
+ self.inheadingtwo = True
+ self.inparagraph = False
+ elif tag.lower() in ["h3", "h4", "h5", "h6"]:
+ self.inotherheading = True
+ self.inparagraph = False
+ elif tag.lower() == "a":
+ self.inlink = True
+ elif tag.lower() == "br":
+ self.handle_br()
+ elif tag.lower() == "blockquote":
+ self.inblockquote = True
+ self.text = self.text + u'\n'
+ elif tag.lower() == "p":
+ if self.text != "":
+ self.text = self.text + u'\n\n'
+ if self.inparagraph:
+ self.text = self.text \
+ + u'\n'.join(textwrap.wrap(self.currentparagraph, 70))
+ self.currentparagraph = u''
+ self.inparagraph = True
+ elif tag.lower() == "pre":
+ self.text = self.text + "\n"
+ self.inpre = True
+ self.inparagraph = False
+ self.inblockquote = False
+ elif tag.lower() == "ul":
+ self.item = u''
+ self.inul = True
+ self.text = self.text + "\n"
+ elif tag.lower() == "li":
+ if not self.initem:
+ self.initem = True
+ self.item = u''
+ else:
+ self.text = self.text \
+ + u' * ' \
+ + u'\n '.join([a.strip() for a in \
+ textwrap.wrap(self.item, 67)]) \
+ + u'\n'
+ self.item = u''
+ self.initem = True
+
+ def handle_startendtag(self, tag, attrs):
+ if tag.lower() == "br":
+ self.handle_br()
+
+ def handle_br(self):
+ if self.inparagraph:
+ self.text = self.text \
+ + u'\n'.join( \
+ [a \
+ for a in textwrap.wrap( \
+ self.currentparagraph, 70) \
+ ] \
+ ) \
+ + u'\n'
+ self.currentparagraph = u''
+ elif self.inblockquote:
+ self.text = self.text \
+ + u'\n> ' \
+ + u'\n> '.join( \
+ [a \
+ for a in textwrap.wrap( \
+ self.blockquote.encode("utf-8") \
+ , 68) \
+ ] \
+ ) \
+ + u'\n'
+ self.blockquote = u''
+ else:
+ self.text = self.text + "\n"
+
+ def handle_endtag(self, tag):
+ if tag.lower() == "h1":
+ self.inheadingone = False
+ self.text = self.text \
+ + u'\n\n' \
+ + self.headingtext.encode("utf-8") \
+ + u'\n' \
+ + u'=' * len(self.headingtext.encode("utf-8").strip())
+ self.headingtext = u''
+ elif tag.lower() == "h2":
+ self.inheadingtwo = False
+ self.text = self.text \
+ + u'\n\n' \
+ + self.headingtext.encode("utf-8") \
+ + u'\n' \
+ + u'-' * len(self.headingtext.encode("utf-8").strip())
+ self.headingtext = u''
+ elif tag.lower() in ["h3", "h4", "h5", "h6"]:
+ self.inotherheading = False
+ self.text = self.text \
+ + u'\n\n' \
+ + self.headingtext.encode("utf-8") \
+ + u'\n' \
+ + u'~' * len(self.headingtext.encode("utf-8").strip())
+ self.headingtext = u''
+ elif tag.lower() == "p":
+ self.text = self.text \
+ + u'\n'.join(textwrap.wrap( \
+ self.currentparagraph, 70) \
+ )
+ self.inparagraph = False
+ self.currentparagraph = u''
+ elif tag.lower() == "blockquote":
+ self.text = self.text \
+ + u'\n> ' \
+ + u'\n> '.join( \
+ [a.strip() \
+ for a in textwrap.wrap( \
+ self.blockquote, 68)] \
+ ) \
+ + u'\n'
+ self.inblockquote = False
+ self.blockquote = u''
+ elif tag.lower() == "pre":
+ self.inpre = False
+ elif tag.lower() == "li":
+ self.initem = False
+ if self.item != u'':
+ self.text = self.text \
+ + u' * ' \
+ + u'\n '.join( \
+ [a.strip() for a in textwrap.wrap(self.item, 67)]) \
+ + u'\n'
+ self.item = u''
+ elif tag.lower() == "ul":
+ self.inul = False
+
+ def handle_data(self, data):
+ if self.inheadingone or self.inheadingtwo or self.inotherheading:
+ self.headingtext = self.headingtext \
+ + unicode(data, "utf-8").strip() \
+ + u' '
+ elif self.inblockquote:
+ self.blockquote = self.blockquote \
+ + unicode(data, "utf-8").strip() \
+ + u' '
+ elif self.initem:
+ self.item = self.item + unicode(data, "utf-8")
+ elif self.inparagraph:
+ self.currentparagraph = self.currentparagraph \
+ + unicode(data, "utf-8").strip() \
+ + u' '
+ elif self.inpre:
+ self.text = self.text + unicode(data, "utf-8")
+ else:
+ isallwhitespace = data.strip() == ""
+ if not isallwhitespace:
+ self.text = self.text + unicode(data, "utf-8").strip() + u' '
+
+ def handle_entityref(self, name):
+ entity = name
+ if HTML2Text.entities.has_key(name.lower()):
+ entity = HTML2Text.entities[name.lower()]
+ elif name[0] == "#":
+ entity = unichr(int(name[1:]))
+ else:
+ entity = "&" + name + ";"
+
+ if self.inparagraph:
+ self.currentparagraph = self.currentparagraph \
+ + unicode(entity, "utf-8")
+ elif self.inblockquote:
+ self.blockquote = self.blockquote + unicode(entity, "utf-8")
+ else:
+ self.text = self.text + unicode(entity, "utf-8")
+
+ def gettext(self):
+ data = self.text
+ if self.inparagraph:
+ data = data + "\n".join(textwrap.wrap(self.currentparagraph, 70))
+ if data[-1] != '\n':
+ data = data + '\n'
+ return data
+
+def open_url(method, url):
+ redirectcount = 0
+ while redirectcount < 3:
+ (type, rest) = urllib.splittype(url)
+ (host, path) = urllib.splithost(rest)
+ (host, port) = urllib.splitport(host)
+ if port == None:
+ port = 80
+ try:
+ conn = httplib.HTTPConnection("%s:%s" %(host, port))
+ conn.request(method, path)
+ response = conn.getresponse()
+ if response.status in [301, 302, 303, 307]:
+ headers = response.getheaders()
+ for header in headers:
+ if header[0] == "location":
+ url = header[1]
+ elif response.status == 200:
+ return response
+ except:
+ pass
+ redirectcount = redirectcount + 1
+ return None