* Add redirect support

[rss2maildir.git] / rss2maildir.py

diff --git a/rss2maildir.py b/rss2maildir.py
index d721d0e49c19994c84b6e6407766e02c836ce5d4..aa3b0ccac5badb4a6c21f46591470ab3e5fc0531 100755 (executable)
--- a/rss2maildir.py
+++ b/rss2maildir.py
@@ -1,5 +1,5 @@
 #!/usr/bin/python
 #!/usr/bin/python

-# coding=utf8
+# coding=utf-8

 
 # rss2maildir.py - RSS feeds to Maildir 1 email per item
 # Copyright (C) 2007  Brett Parker <iDunno@sommitrealweird.co.uk>
 
 # rss2maildir.py - RSS feeds to Maildir 1 email per item
 # Copyright (C) 2007  Brett Parker <iDunno@sommitrealweird.co.uk>


@@ -17,10 +17,10 @@
 # You should have received a copy of the GNU General Public License
 # along with this program.  If not, see <http://www.gnu.org/licenses/>.
 
 # You should have received a copy of the GNU General Public License
 # along with this program.  If not, see <http://www.gnu.org/licenses/>.
 

-import mailbox

 import sys
 import os
 import stat
 import sys
 import os
 import stat

+import httplib

 import urllib
 
 import feedparser
 import urllib
 
 import feedparser


@@ -53,21 +53,27 @@ entities = {
     "pound": "£",
     "copy": "©",
     "apos": "'",
     "pound": "£",
     "copy": "©",
     "apos": "'",

-    "quote": "\"",
+    "quot": "\"",

     "nbsp": " ",
     }
 
 class HTML2Text(HTMLParser):
     "nbsp": " ",
     }
 
 class HTML2Text(HTMLParser):

-    
+

     def __init__(self):
         self.inheadingone = False
         self.inheadingtwo = False
         self.inotherheading = False
         self.inparagraph = True
     def __init__(self):
         self.inheadingone = False
         self.inheadingtwo = False
         self.inotherheading = False
         self.inparagraph = True

+        self.inblockquote = False

         self.inlink = False
         self.inlink = False

-        self.text = ""
-        self.currentparagraph = ""
-        self.headingtext = ""
+        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):
         HTMLParser.__init__(self)
 
     def handle_starttag(self, tag, attrs):


@@ -83,47 +89,162 @@ class HTML2Text(HTMLParser):
         elif tag.lower() == "a":
             self.inlink = True
         elif tag.lower() == "br":
         elif tag.lower() == "a":
             self.inlink = True
         elif tag.lower() == "br":

-            self.text = self.text + "\n"
+            self.handle_br()
+        elif tag.lower() == "blockquote":
+            self.inblockquote = True
+            self.text = self.text + u'\n'

         elif tag.lower() == "p":
             if self.text != "":
         elif tag.lower() == "p":
             if self.text != "":

-                self.text = self.text + "\n\n"
-                self.currentparagraph = ""
-                self.inparagraph = True
+                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" 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":
 
     def handle_startendtag(self, tag, attrs):
         if tag.lower() == "br":

-            self.text = self.text + "\n"
+            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
 
     def handle_endtag(self, tag):
         if tag.lower() == "h1":
             self.inheadingone = False

-            self.text = self.text + self.headingtext + "\n" + "=" * len(self.headingtext)
-            self.headingtext = ""
+            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
         elif tag.lower() == "h2":
             self.inheadingtwo = False

-            self.text = self.text + self.headingtext + "\n" + "-" * len(self.headingtext)
-            self.headingtext = ""
+            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
         elif tag.lower() in ["h3", "h4", "h5", "h6"]:
             self.inotherheading = False

-            self.text = self.text + self.headingtext + "\n" + "~" * len(self.headingtext)
-            self.headingtext = ""
+            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":
         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.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 != "":
+                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):
 
     def handle_data(self, data):

-        if not self.inheadingone and not self.inheadingtwo and not self.inotherheading and not self.inparagraph:
-            self.text = self.text + data.strip() + " "
+        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:
         elif self.inparagraph:

-            self.currentparagraph = self.currentparagraph + data.strip() + " "
+            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 + unicode(data, "utf-8")

         else:
         else:

-            self.headingtext = self.headingtext + data.strip() + " "
+            self.text = self.text + unicode(data, "utf-8").strip() + u' '

 
     def handle_entityref(self, name):
 
     def handle_entityref(self, name):

+        entity = name

         if entities.has_key(name.lower()):
         if entities.has_key(name.lower()):

-            self.text = self.text + entities[name.lower()]
+            entity = entities[name.lower()]
+        elif name[0] == "#":
+            entity = unichr(int(name[1:]))

         else:
         else:

-            self.text = self.text + "&" + name + ";"
+            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
 
     def gettext(self):
         data = self.text


@@ -131,9 +252,79 @@ class HTML2Text(HTMLParser):
             data = data + "\n".join(textwrap.wrap(self.currentparagraph, 70))
         return data
 
             data = data + "\n".join(textwrap.wrap(self.currentparagraph, 70))
         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
+

 def parse_and_deliver(maildir, url, statedir):
 def parse_and_deliver(maildir, url, statedir):

-    md = mailbox.Maildir(maildir)
-    fp = feedparser.parse(url)
+    feedhandle = None
+    headers = None
+    # first check if we know about this feed already
+    feeddb = dbm.open(os.path.join(statedir, "feeds"), "c")
+    if feeddb.has_key(url):
+        data = feeddb[url]
+        data = cgi.parse_qs(data)
+        response = open_url("HEAD", url)
+        headers = None
+        if response:
+            headers = response.getheaders()
+        ischanged = False
+        try:
+            for header in headers:
+                if header[0] == "content-length":
+                    if header[1] != data["content-length"][0]:
+                        ischanged = True
+                elif header[0] == "etag":
+                    if header[1] != data["etag"][0]:
+                        ischanged = True
+                elif header[0] == "last-modified":
+                    if header[1] != data["last-modified"][0]:
+                        ischanged = True
+                elif header[0] == "content-md5":
+                    if header[1] != data["content-md5"][0]:
+                        ischanged = True
+        except:
+            ischanged = True
+        if ischanged:
+            response = open_url("GET", url)
+            if response != None:
+                headers = response.getheaders()
+                feedhandle = response
+            else:
+                sys.stderr.write("Failed to fetch feed: %s\n" %(url))
+                return
+        else:
+            return # don't need to do anything, nothings changed.
+    else:
+        response = open_url("GET", url)
+        if response != None:
+            headers = response.getheaders()
+            feedhandle = response
+        else:
+            sys.stderr.write("Failed to fetch feed: %s\n" %(url))
+            return
+
+    fp = feedparser.parse(feedhandle)

     db = dbm.open(os.path.join(statedir, "seen"), "c")
     for item in fp["items"]:
         # have we seen it before?
     db = dbm.open(os.path.join(statedir, "seen"), "c")
     for item in fp["items"]:
         # have we seen it before?


@@ -144,11 +335,15 @@ def parse_and_deliver(maildir, url, statedir):
         else:
             content = item["summary"]
 
         else:
             content = item["summary"]
 

-        md5sum = md5.md5(content.encode("utf8")).hexdigest()
+        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 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
 
             if data["contentmd5"][0] == md5sum:
                 continue
 


@@ -159,27 +354,49 @@ def parse_and_deliver(maildir, url, statedir):
 
         # create a basic email message
         msg = MIMEMultipart("alternative")
 
         # 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))
         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.now() \
+            .strftime("%a, %e %b %Y %T -0000")
+        try:
+            createddate = datetime.datetime(*item["updated_parsed"][0:6]) \
+                .strftime("%a, %e %b %Y %T -0000")
+        except:
+            pass

         msg.add_header("Date", createddate)
         msg.add_header("Subject", item["title"])
         msg.set_default_type("text/plain")
 
         msg.add_header("Date", createddate)
         msg.add_header("Subject", item["title"])
         msg.set_default_type("text/plain")
 

-        htmlpart = MIMEText(content.encode("utf8"), "html", "utf8")
+        htmlpart = MIMEText(content.encode("utf-8"), "html", "utf-8")

         textparser = HTML2Text()
         textparser = HTML2Text()

-        textparser.feed(content.encode("utf8"))
+        textparser.feed(content.encode("utf-8"))

         textcontent = textparser.gettext()
         textcontent = textparser.gettext()

-        textpart = MIMEText(textcontent, "plain", "utf8")
+        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
         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')
+        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())
         fn = os.path.join(maildir, "tmp", fname)
         fh = open(fn, "w")
         fh.write(msg.as_string())


@@ -190,10 +407,26 @@ def parse_and_deliver(maildir, url, statedir):
         os.unlink(fn)
 
         # now add to the database about the item
         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[url + "|" + item["link"]] = data
 

+    if headers:
+        data = []
+        for header in headers:
+            if header[0] in ["content-md5", "etag", "last-modified", "content-length"]:
+                data.append((header[0], header[1]))
+        if len(data) > 0:
+            data = urllib.urlencode(data)
+            feeddb[url] = data
+

     db.close()
     db.close()

+    feeddb.close()

 
 # first off, parse the command line arguments
 
 
 # first off, parse the command line arguments
 


@@ -220,7 +453,8 @@ if options.conf != None:
         configfile = options.conf
     except:
         # should exit here as the specified file doesn't exist
         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
         sys.exit(2)
 else:
     # check through the default locations


@@ -249,7 +483,8 @@ if options.statedir != None:
     try:
         mode = os.stat(state_dir)[stat.ST_MODE]
         if not stat.S_ISDIR(mode):
     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
             sys.exit(1)
     except:
         # try to make the directory


@@ -263,7 +498,8 @@ elif scp.has_option("general", "state_dir"):
     try:
         mode = os.stat(state_dir)[stat.ST_MODE]
         if not stat.S_ISDIR(mode):
     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
             sys.exit(1)
     except:
         # try to create it


@@ -271,19 +507,22 @@ elif scp.has_option("general", "state_dir"):
             os.mkdir(new_state_dir)
             state_dir = new_state_dir
         except:
             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.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.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"):
             sys.exit(1)
 
 if scp.has_option("general", "maildir_root"):


@@ -292,7 +531,9 @@ if scp.has_option("general", "maildir_root"):
 try:
     mode = os.stat(maildir_root)[stat.ST_MODE]
     if not stat.S_ISDIR(mode):
 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:
         sys.exit(1)
 except:
     try:


@@ -353,7 +594,9 @@ for section in feeds:
             os.mkdir(os.path.join(maildir, "cur"))
             os.mkdir(os.path.join(maildir, "tmp"))
         except:
             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
             sys.exit(1)
 
     # right - we've got the directories, we've got the section, we know the