X-Git-Url: https://git.sommitrealweird.co.uk/rss2maildir.git/blobdiff_plain/79abf12671ca1edc7d95e7f1aaa9c4b45a7de7b9..41170825eb09fcf3b3c69c9812073727ee8a2b4e:/rss2maildir.py

diff --git a/rss2maildir.py b/rss2maildir.py
index 02bd392..25ddb3c 100755
--- a/rss2maildir.py
+++ b/rss2maildir.py
@@ -70,6 +70,9 @@ class HTML2Text(HTMLParser):
         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):
@@ -85,22 +88,16 @@ class HTML2Text(HTMLParser):
         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"
+            self.handle_br()
         elif tag.lower() == "blockquote":
             self.inblockquote = True
-            self.text = self.text + "\n"
+            self.text = self.text + u'\n'
         elif tag.lower() == "p":
             if self.text != "":
-                self.text = self.text + "\n\n"
+                self.text = self.text + u'\n\n'
             if self.inparagraph:
-                self.text = self.text + "\n".join(textwrap.wrap(self.currentparagraph, 70)).encode("utf-8")
+                self.text = self.text \
+                    + u'\n'.join(textwrap.wrap(self.currentparagraph, 70))
             self.currentparagraph = u''
             self.inparagraph = True
         elif tag.lower() == "pre":
@@ -108,50 +105,126 @@ class HTML2Text(HTMLParser):
             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" and self.inul:
+            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''
 
     def handle_startendtag(self, tag, attrs):
         if tag.lower() == "br":
+            self.handle_br()
+
+    def handle_br(self):
             if self.inparagraph:
-                self.text = self.text + "\n".join(textwrap.wrap(self.currentparagraph, 70)).encode("utf-8") + "\n"
+                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 + "\n> " + "\n> ".join([a.strip() for a in textwrap.wrap(self.blockquote, 68)]).encode("utf-8") + "\n"
-                self.blockquote = ""
+                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 + "\n\n" + self.headingtext + "\n" + "=" * len(self.headingtext.strip())
+            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 + "\n\n" + self.headingtext + "\n" + "-" * len(self.headingtext.strip())
+            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 + "\n\n" + self.headingtext + "\n" + "~" * len(self.headingtext.strip())
+            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 + "\n".join(textwrap.wrap(self.currentparagraph, 70))
+            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 + "\n> " + "\n> ".join([a.strip() for a in textwrap.wrap(self.blockquote, 68)]).encode("utf-8") + "\n"
+            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 != "":
+                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' '
+            self.headingtext = self.headingtext \
+                + unicode(data, "utf-8").strip() \
+                + u' '
         elif self.inblockquote:
-            self.blockquote = self.blockquote + unicode(data, "utf-8").strip() + u' '
+            self.blockquote = self.blockquote \
+                + unicode(data, "utf-8").strip() \
+                + u' '
         elif self.inparagraph:
-            self.currentparagraph = self.currentparagraph + unicode(data, "utf-8").strip() + u' '
+            self.currentparagraph = self.currentparagraph \
+                + unicode(data, "utf-8").strip() \
+                + u' '
+        elif self.inul and self.initem:
+            self.item = self.item + unicode(data, "utf-8")
         elif self.inpre:
-            self.text = self.text + data.encode("utf-8")
+            self.text = self.text + unicode(data, "utf-8")
         else:
             self.text = self.text + unicode(data, "utf-8").strip() + u' '
 
@@ -165,11 +238,12 @@ class HTML2Text(HTMLParser):
             entity = "&" + name + ";"
 
         if self.inparagraph:
-            self.currentparagraph = self.currentparagraph + entity
+            self.currentparagraph = self.currentparagraph \
+                + unicode(entity, "utf-8")
         elif self.inblockquote:
-            self.blockquote = self.blockquote + entity
+            self.blockquote = self.blockquote + unicode(entity, "utf-8")
         else:
-            self.text = self.text + entity
+            self.text = self.text + unicode(entity, "utf-8")
 
     def gettext(self):
         data = self.text
@@ -191,9 +265,13 @@ def parse_and_deliver(maildir, url, statedir):
 
         md5sum = md5.md5(content.encode("utf-8")).hexdigest()
 
+        prevmessageid = None
+
         if db.has_key(url + "|" + item["link"]):
             data = db[url + "|" + item["link"]]
             data = cgi.parse_qs(data)
+            if data.has_key("message-id"):
+                prevmessageid = data["message-id"][0]
             if data["contentmd5"][0] == md5sum:
                 continue
 
@@ -204,12 +282,22 @@ def parse_and_deliver(maildir, url, statedir):
 
         # 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() + ">"
+        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")
+        if prevmessageid:
+            msg.add_header("References", prevmessageid)
+        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")
@@ -224,7 +312,14 @@ def parse_and_deliver(maildir, url, statedir):
 
         # 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')
+        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())
@@ -235,7 +330,13 @@ def parse_and_deliver(maildir, url, statedir):
         os.unlink(fn)
 
         # now add to the database about the item
-        data = urllib.urlencode((("message-id", messageid), ("created", createddate), ("contentmd5", md5sum)))
+        if prevmessageid:
+            messageid = prevmessageid + " " + messageid
+        data = urllib.urlencode((
+            ("message-id", messageid), \
+            ("created", createddate), \
+            ("contentmd5", md5sum) \
+            ))
         db[url + "|" + item["link"]] = data
 
     db.close()
@@ -265,7 +366,8 @@ if options.conf != None:
         configfile = options.conf
     except:
         # should exit here as the specified file doesn't exist
-        sys.stderr.write("Config file %s does not exist. Exiting.\n" %(options.conf,))
+        sys.stderr.write( \
+            "Config file %s does not exist. Exiting.\n" %(options.conf,))
         sys.exit(2)
 else:
     # check through the default locations
@@ -294,7 +396,8 @@ if options.statedir != None:
     try:
         mode = os.stat(state_dir)[stat.ST_MODE]
         if not stat.S_ISDIR(mode):
-            sys.stderr.write("State directory (%s) is not a directory\n" %(state_dir))
+            sys.stderr.write( \
+                "State directory (%s) is not a directory\n" %(state_dir))
             sys.exit(1)
     except:
         # try to make the directory
@@ -308,7 +411,8 @@ elif scp.has_option("general", "state_dir"):
     try:
         mode = os.stat(state_dir)[stat.ST_MODE]
         if not stat.S_ISDIR(mode):
-            sys.stderr.write("State directory (%s) is not a directory\n" %(state_dir))
+            sys.stderr.write( \
+                "State directory (%s) is not a directory\n" %(state_dir))
             sys.exit(1)
     except:
         # try to create it
@@ -316,19 +420,22 @@ elif scp.has_option("general", "state_dir"):
             os.mkdir(new_state_dir)
             state_dir = new_state_dir
         except:
-            sys.stderr.write("Couldn't create state directory %s\n" %(new_state_dir))
+            sys.stderr.write( \
+                "Couldn't create state directory %s\n" %(new_state_dir))
             sys.exit(1)
 else:
     try:
         mode = os.stat(state_dir)[stat.ST_MODE]
         if not stat.S_ISDIR(mode):
-            sys.stderr.write("State directory %s is not a directory\n" %(state_dir))
+            sys.stderr.write( \
+                "State directory %s is not a directory\n" %(state_dir))
             sys.exit(1)
     except:
         try:
             os.mkdir(state_dir)
         except:
-            sys.stderr.write("State directory %s could not be created\n" %(state_dir))
+            sys.stderr.write( \
+                "State directory %s could not be created\n" %(state_dir))
             sys.exit(1)
 
 if scp.has_option("general", "maildir_root"):
@@ -337,7 +444,9 @@ if scp.has_option("general", "maildir_root"):
 try:
     mode = os.stat(maildir_root)[stat.ST_MODE]
     if not stat.S_ISDIR(mode):
-        sys.stderr.write("Maildir Root %s is not a directory\n" %(maildir_root))
+        sys.stderr.write( \
+            "Maildir Root %s is not a directory\n" \
+            %(maildir_root))
         sys.exit(1)
 except:
     try:
@@ -398,7 +507,9 @@ for section in feeds:
             os.mkdir(os.path.join(maildir, "cur"))
             os.mkdir(os.path.join(maildir, "tmp"))
         except:
-            sys.stderr.write("Couldn't create required maildir directories for %s\n" %(section,))
+            sys.stderr.write( \
+                "Couldn't create required maildir directories for %s\n" \
+                %(section,))
             sys.exit(1)
 
     # right - we've got the directories, we've got the section, we know the