* improve handling of unicode data

[rss2maildir.git] / rss2maildir.py

diff --git a/rss2maildir.py b/rss2maildir.py
index 968c647f6190a8daaaf8d39e7e8653141c7acf5d..cacf1ee8eaabe04a6d5fc20d21c3e9e8d664c9c1 100755 (executable)
--- a/rss2maildir.py
+++ b/rss2maildir.py
@@ -1,6 +1,22 @@
 #!/usr/bin/python
 #!/usr/bin/python

+# coding=utf-8
+
+# rss2maildir.py - RSS feeds to Maildir 1 email per item
+# Copyright (C) 2007  Brett Parker <iDunno@sommitrealweird.co.uk>
+# 
+# This program is free software: you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+# the Free Software Foundation, either version 3 of the License, or
+# (at your option) any later version.
+# 
+# This program is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+# GNU General Public License for more details.
+# 
+# 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


@@ -8,11 +24,13 @@ import urllib
 
 import feedparser
 
 
 import feedparser
 

-import email
+from email.MIMEMultipart import MIMEMultipart
+from email.MIMEText import MIMEText

 
 import datetime
 import random
 import string
 
 import datetime
 import random
 import string

+import textwrap

 
 import socket
 
 
 import socket
 


@@ -20,29 +38,293 @@ from optparse import OptionParser
 from ConfigParser import SafeConfigParser
 
 from base64 import b64encode
 from ConfigParser import SafeConfigParser
 
 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
+        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":
+            if self.inparagraph:
+                self.text = self.text \
+                    + u'\n'.join( \
+                        textwrap.wrap(self.currentparagraph, 70)) \
+                    + u'\n'
+                self.currentparagraph = ""
+            elif self.inblockquote:
+                self.text = self.text \
+                    + u'\n> ' \
+                    + u'\n> '.join( \
+                        [a.strip() \
+                            for a in textwrap.wrap(self.blockquote, 68) \
+                        ]) \
+                    + u'\n'
+                self.blockquote = u''
+            else:
+                self.text = self.text + u'\n'
+        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" 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":
+            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 != "":
+                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.inparagraph:
+            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:
+            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 + 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))
+        return data

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

-    md = mailbox.Maildir(maildir)

     fp = feedparser.parse(url)
     fp = feedparser.parse(url)

+    db = dbm.open(os.path.join(statedir, "seen"), "c")

     for item in fp["items"]:
     for item in fp["items"]:

-        # things that we need in the message
-        msg = email.message_from_string("")
+        # 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.set_unixfrom("\"%s\" <rss2maildir@localhost>" %(url))

-        msg.add_header("From", "\"%s\" <rss2maildir@localhost>" %(item["author"]))
+        msg.add_header("From", "\"%s\" <rss2maildir@localhost>" %(author))

         msg.add_header("To", "\"%s\" <rss2maildir@localhost>" %(url))
         msg.add_header("To", "\"%s\" <rss2maildir@localhost>" %(url))

-        msg.add_header("Date", datetime.datetime(*item["created_parsed"][0:6]).strftime("%a, %e %b %Y %T -0000"))
+        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.add_header("Subject", item["title"])

-        msg.set_charset("utf8")
-        msg.set_default_type(item["content"][0]["type"])
-        msg.set_type(item["content"][0]["type"])
-        msg.set_payload(b64encode(item["content"][0]["value"]))
+        msg.set_default_type("text/plain")

 
 

-        msg.add_header("Message-ID", "<" + 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() + ">")
+        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
 
         # 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())


@@ -52,6 +334,16 @@ def parse_and_deliver(maildir, url, statedir):
         os.link(fn, newfn)
         os.unlink(fn)
 
         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()
+

 # first off, parse the command line arguments
 
 oparser = OptionParser()
 # first off, parse the command line arguments
 
 oparser = OptionParser()


@@ -130,6 +422,18 @@ elif scp.has_option("general", "state_dir"):
         except:
             sys.stderr.write("Couldn't create state directory %s\n" %(new_state_dir))
             sys.exit(1)
         except:
             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.exit(1)
+    except:
+        try:
+            os.mkdir(state_dir)
+        except:
+            sys.stderr.write("State directory %s could not be created\n" %(state_dir))
+            sys.exit(1)

 
 if scp.has_option("general", "maildir_root"):
     maildir_root = scp.get("general", "maildir_root")
 
 if scp.has_option("general", "maildir_root"):
     maildir_root = scp.get("general", "maildir_root")