+from base64 import b64encode
+import md5
+
+import cgi
+import dbm
+
+from HTMLParser import HTMLParser
+
+entities = {
+ "amp": "&",
+ "lt": "<",
+ "gt": ">",
+ "pound": "£",
+ "copy": "©",
+ "apos": "'",
+ "quote": "\"",
+ "nbsp": " ",
+ }
+
+class HTML2Text(HTMLParser):
+
+ 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
+ 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":
+ if self.inparagraph:
+ self.text = self.text + "\n".join(textwrap.wrap(self.currentparagraph, 70)).encode('utf-8') + "\n"
+ self.currentparagraph = ""
+ elif self.inblockquote:
+ self.text = self.text + "\n> " + "\n> ".join([a.strip() for a in textwrap.wrap(self.blockquote, 68)]).encode("utf-8") + "\n"
+ self.blockquote = u''
+ else:
+ self.text = self.text + "\n"
+ elif tag.lower() == "blockquote":
+ self.inblockquote = True
+ self.text = self.text + "\n"
+ elif tag.lower() == "p":
+ if self.text != "":
+ self.text = self.text + "\n\n"
+ if self.inparagraph:
+ self.text = self.text + "\n".join(textwrap.wrap(self.currentparagraph, 70)).encode("utf-8")
+ self.currentparagraph = u''
+ self.inparagraph = True
+ elif tag.lower() == "pre":
+ self.text = self.text + "\n"
+ self.inpre = True
+ self.inparagraph = False
+ self.inblockquote = False
+
+ def handle_startendtag(self, tag, attrs):
+ if tag.lower() == "br":
+ if self.inparagraph:
+ self.text = self.text + "\n".join(textwrap.wrap(self.currentparagraph, 70)).encode("utf-8") + "\n"
+ self.currentparagraph = u''
+ elif self.inblockquote:
+ self.text = self.text + "\n> " + "\n> ".join([a.strip() for a in textwrap.wrap(self.blockquote, 68)]).encode("utf-8") + "\n"
+ self.blockquote = ""
+ else:
+ self.text = self.text + "\n"
+
+ def handle_endtag(self, tag):
+ if tag.lower() == "h1":
+ self.inheadingone = False
+ self.text = self.text + "\n\n" + self.headingtext + "\n" + "=" * len(self.headingtext.strip())
+ self.headingtext = u''
+ elif tag.lower() == "h2":
+ self.inheadingtwo = False
+ self.text = self.text + "\n\n" + self.headingtext + "\n" + "-" * len(self.headingtext.strip())
+ self.headingtext = u''
+ elif tag.lower() in ["h3", "h4", "h5", "h6"]:
+ self.inotherheading = False
+ self.text = self.text + "\n\n" + self.headingtext + "\n" + "~" * len(self.headingtext.strip())
+ self.headingtext = u''
+ elif tag.lower() == "p":
+ self.text = self.text + "\n".join(textwrap.wrap(self.currentparagraph, 70))
+ self.inparagraph = False
+ elif tag.lower() == "blockquote":
+ self.text = self.text + "\n> " + "\n> ".join([a.strip() for a in textwrap.wrap(self.blockquote, 68)]).encode("utf-8") + "\n"
+ self.inblockquote = False
+ self.blockquote = u''
+ elif tag.lower() == "pre":
+ self.inpre = 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.inparagraph:
+ self.currentparagraph = self.currentparagraph + unicode(data, "utf-8").strip() + u' '
+ elif self.inpre:
+ self.text = self.text + data.encode("utf-8")
+ else:
+ self.text = self.text + unicode(data, "utf-8").strip() + u' '
+
+ def handle_entityref(self, name):
+ entity = name
+ if entities.has_key(name.lower()):
+ entity = entities[name.lower()]
+ elif name[0] == "#":
+ entity = unichr(int(name[1:]))
+ else:
+ entity = "&" + name + ";"
+
+ if self.inparagraph:
+ self.currentparagraph = self.currentparagraph + entity
+ elif self.inblockquote:
+ self.blockquote = self.blockquote + entity
+ else:
+ self.text = self.text + entity
+
+ def gettext(self):
+ data = self.text
+ if self.inparagraph:
+ data = data + "\n".join(textwrap.wrap(self.currentparagraph, 70))
+ return data
+
+def parse_and_deliver(maildir, url, statedir):
+ fp = feedparser.parse(url)
+ db = dbm.open(os.path.join(statedir, "seen"), "c")
+ for item in fp["items"]:
+ # have we seen it before?
+ # need to work out what the content is first...
+
+ if item.has_key("content"):
+ content = item["content"][0]["value"]
+ else:
+ content = item["summary"]
+
+ md5sum = md5.md5(content.encode("utf-8")).hexdigest()
+
+ if db.has_key(url + "|" + item["link"]):
+ data = db[url + "|" + item["link"]]
+ data = cgi.parse_qs(data)
+ if data["contentmd5"][0] == md5sum:
+ continue
+
+ try:
+ author = item["author"]
+ except:
+ author = url
+
+ # create a basic email message
+ msg = MIMEMultipart("alternative")
+ messageid = "<" + datetime.datetime.now().strftime("%Y%m%d%H%M") + "." + "".join([random.choice(string.ascii_letters + string.digits) for a in range(0,6)]) + "@" + socket.gethostname() + ">"
+ msg.add_header("Message-ID", messageid)
+ msg.set_unixfrom("\"%s\" <rss2maildir@localhost>" %(url))
+ msg.add_header("From", "\"%s\" <rss2maildir@localhost>" %(author))
+ msg.add_header("To", "\"%s\" <rss2maildir@localhost>" %(url))
+ createddate = datetime.datetime(*item["updated_parsed"][0:6]).strftime("%a, %e %b %Y %T -0000")
+ msg.add_header("Date", createddate)
+ msg.add_header("Subject", item["title"])
+ msg.set_default_type("text/plain")
+
+ htmlpart = MIMEText(content.encode("utf-8"), "html", "utf-8")
+ textparser = HTML2Text()
+ textparser.feed(content.encode("utf-8"))
+ textcontent = textparser.gettext()
+ textpart = MIMEText(textcontent.encode("utf-8"), "plain", "utf-8")
+ msg.attach(textpart)
+ msg.attach(htmlpart)
+
+ # start by working out the filename we should be writting to, we do
+ # this following the normal maildir style rules
+ fname = str(os.getpid()) + "." + socket.gethostname() + "." + "".join([random.choice(string.ascii_letters + string.digits) for a in range(0,10)]) + "." + datetime.datetime.now().strftime('%s')
+ fn = os.path.join(maildir, "tmp", fname)
+ fh = open(fn, "w")
+ fh.write(msg.as_string())
+ fh.close()
+ # now move it in to the new directory
+ newfn = os.path.join(maildir, "new", fname)
+ os.link(fn, newfn)
+ os.unlink(fn)
+
+ # now add to the database about the item
+ data = urllib.urlencode((("message-id", messageid), ("created", createddate), ("contentmd5", md5sum)))
+ db[url + "|" + item["link"]] = data
+
+ db.close()
+
projects / rss2maildir.git / blobdiff