Reformat code ready for adding test suite

[rss2maildir.git] / rss2maildir.py

diff --git a/rss2maildir.py b/rss2maildir.py
index cacf1ee8eaabe04a6d5fc20d21c3e9e8d664c9c1..99735a70a8dd93957741b3c1c3a66e1cd3779171 100755 (executable)
--- a/rss2maildir.py
+++ b/rss2maildir.py
@@ -20,6 +20,7 @@
 import sys
 import os
 import stat
 import sys
 import os
 import stat

+import httplib

 import urllib
 
 import feedparser
 import urllib
 
 import feedparser


@@ -45,19 +46,18 @@ import dbm
 
 from HTMLParser import HTMLParser
 
 
 from HTMLParser import HTMLParser
 

-entities = {
-    "amp": "&",
-    "lt": "<",
-    "gt": ">",
-    "pound": "£",
-    "copy": "©",
-    "apos": "'",
-    "quote": "\"",
-    "nbsp": " ",
-    }
-

 class HTML2Text(HTMLParser):
 class HTML2Text(HTMLParser):

-    
+    entities = {
+        "amp": "&",
+        "lt": "<",
+        "gt": ">",
+        "pound": "£",
+        "copy": "©",
+        "apos": "'",
+        "quot": "\"",
+        "nbsp": " ",
+        }
+

     def __init__(self):
         self.inheadingone = False
         self.inheadingtwo = False
     def __init__(self):
         self.inheadingone = False
         self.inheadingtwo = False


@@ -88,23 +88,7 @@ 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":

-            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'
+            self.handle_br()

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


@@ -132,12 +116,16 @@ class HTML2Text(HTMLParser):
             else:
                 self.text = self.text \
                     + u' * ' \
             else:
                 self.text = self.text \
                     + u' * ' \

-                    + u'\n   '.join([a.strip() for a in textwrap.wrap(self.item, 67)]) \
+                    + 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":
                     + 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 \
                 + u'\n'.join( \
             if self.inparagraph:
                 self.text = self.text \
                 + u'\n'.join( \


@@ -242,15 +230,16 @@ class HTML2Text(HTMLParser):
 
     def handle_entityref(self, name):
         entity = name
 
     def handle_entityref(self, name):
         entity = name

-        if entities.has_key(name.lower()):
-            entity = entities[name.lower()]
+        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:
         elif name[0] == "#":
             entity = unichr(int(name[1:]))
         else:
             entity = "&" + name + ";"
 
         if self.inparagraph:

-            self.currentparagraph = self.currentparagraph + unicode(entity, "utf-8")
+            self.currentparagraph = self.currentparagraph \
+                + unicode(entity, "utf-8")

         elif self.inblockquote:
             self.blockquote = self.blockquote + unicode(entity, "utf-8")
         else:
         elif self.inblockquote:
             self.blockquote = self.blockquote + unicode(entity, "utf-8")
         else:


@@ -262,8 +251,79 @@ 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):

-    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?


@@ -276,9 +336,22 @@ def parse_and_deliver(maildir, url, statedir):
 
         md5sum = md5.md5(content.encode("utf-8")).hexdigest()
 
 
         md5sum = md5.md5(content.encode("utf-8")).hexdigest()
 

+        prevmessageid = None
+
+        # check if there's a guid too - if that exists and we match the md5,
+        # return
+        if item.has_key("guid"):
+            if db.has_key(url + "|" + item["guid"]):
+                data = db[url + "|" + item["guid"]]
+                data = cgi.parse_qs(data)
+                if data["contentmd5"][0] == md5sum:
+                    continue
+

         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
 


@@ -301,16 +374,31 @@ def parse_and_deliver(maildir, url, statedir):
         msg.set_unixfrom("\"%s\" <rss2maildir@localhost>" %(url))
         msg.add_header("From", "\"%s\" <rss2maildir@localhost>" %(author))
         msg.add_header("To", "\"%s\" <rss2maildir@localhost>" %(url))
         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]) \
+        if prevmessageid:
+            msg.add_header("References", prevmessageid)
+        createddate = datetime.datetime.now() \

             .strftime("%a, %e %b %Y %T -0000")
             .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("utf-8"), "html", "utf-8")
+        htmlcontent = content.encode("utf-8")
+        htmlcontent = "%s\n\n<p>Item URL: <a href='%s'>%s</a></p>" %( \
+            content, \
+            item["link"], \
+            item["link"] )
+        htmlpart = MIMEText(htmlcontent.encode("utf-8"), "html", "utf-8")

         textparser = HTML2Text()
         textparser.feed(content.encode("utf-8"))
         textcontent = textparser.gettext()
         textparser = HTML2Text()
         textparser.feed(content.encode("utf-8"))
         textcontent = textparser.gettext()

+        textcontent = "%s\n\nItem URL: %s" %( \
+            textcontent, \
+            item["link"] )

         textpart = MIMEText(textcontent.encode("utf-8"), "plain", "utf-8")
         msg.attach(textpart)
         msg.attach(htmlpart)
         textpart = MIMEText(textcontent.encode("utf-8"), "plain", "utf-8")
         msg.attach(textpart)
         msg.attach(htmlpart)


@@ -335,177 +423,222 @@ 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) \
-            ))
-        db[url + "|" + item["link"]] = data
+        if prevmessageid:
+            messageid = prevmessageid + " " + messageid
+        if item.has_key("guid") and item["guid"] != item["link"]:
+            data = urllib.urlencode(( \
+                ("message-id", messageid), \
+                ("created", createddate), \
+                ("contentmd5", md5sum) \
+                ))
+            db[url + "|" + item["guid"]] = data
+            try:
+                data = db[url + "|" + item["link"]]
+                data = cgi.parse_qs(data)
+                newdata = urllib.urlencode(( \
+                    ("message-id", messageid), \
+                    ("created", data["created"][0]), \
+                    ("contentmd5", data["contentmd5"][0]) \
+                    ))
+                db[url + "|" + item["link"]] = newdata
+            except:
+                db[url + "|" + item["link"]] = data
+        else:
+            data = urllib.urlencode(( \
+                ("message-id", messageid), \
+                ("created", createddate), \
+                ("contentmd5", md5sum) \
+                ))
+            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
+if __name__ == "__main__":
+    # This only gets executed if we really called the program
+    # first off, parse the command line arguments

 
 

-oparser = OptionParser()
-oparser.add_option(
-    "-c", "--conf", dest="conf",
-    help="location of config file"
-    )
-oparser.add_option(
-    "-s", "--statedir", dest="statedir",
-    help="location of directory to store state in"
-    )
+    oparser = OptionParser()
+    oparser.add_option(
+        "-c", "--conf", dest="conf",
+        help="location of config file"
+        )
+    oparser.add_option(
+        "-s", "--statedir", dest="statedir",
+        help="location of directory to store state in"
+        )

 
 

-(options, args) = oparser.parse_args()
+    (options, args) = oparser.parse_args()

 
 

-# check for the configfile
+    # check for the configfile

 
 

-configfile = None
+    configfile = None

 
 

-if options.conf != None:
-    # does the file exist?
-    try:
-        os.stat(options.conf)
-        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.exit(2)
-else:
-    # check through the default locations
-    try:
-        os.stat("%s/.rss2maildir.conf" %(os.environ["HOME"],))
-        configfile = "%s/.rss2maildir.conf" %(os.environ["HOME"],)
-    except:
+    if options.conf != None:
+        # does the file exist?

         try:
         try:

-            os.stat("/etc/rss2maildir.conf")
-            configfile = "/etc/rss2maildir.conf"
+            os.stat(options.conf)
+            configfile = options.conf

         except:
         except:

-            sys.stderr.write("No config file found. Exiting.\n")
+            # should exit here as the specified file doesn't exist
+            sys.stderr.write( \
+                "Config file %s does not exist. Exiting.\n" %(options.conf,))

             sys.exit(2)
             sys.exit(2)

+    else:
+        # check through the default locations
+        try:
+            os.stat("%s/.rss2maildir.conf" %(os.environ["HOME"],))
+            configfile = "%s/.rss2maildir.conf" %(os.environ["HOME"],)
+        except:
+            try:
+                os.stat("/etc/rss2maildir.conf")
+                configfile = "/etc/rss2maildir.conf"
+            except:
+                sys.stderr.write("No config file found. Exiting.\n")
+                sys.exit(2)

 
 

-# Right - if we've got this far, we've got a config file, now for the hard
-# bits...
+    # Right - if we've got this far, we've got a config file, now for the hard
+    # bits...

 
 

-scp = SafeConfigParser()
-scp.read(configfile)
+    scp = SafeConfigParser()
+    scp.read(configfile)

 
 

-maildir_root = "RSSMaildir"
-state_dir = "state"
+    maildir_root = "RSSMaildir"
+    state_dir = "state"

 
 

-if options.statedir != None:
-    state_dir = options.statedir
-    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 to make the directory
+    if options.statedir != None:
+        state_dir = options.statedir

         try:
         try:

-            os.mkdir(state_dir)
+            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:
         except:

-            sys.stderr.write("Couldn't create statedir %s" %(state_dir))
-            sys.exit(1)
-elif scp.has_option("general", "state_dir"):
-    new_state_dir = scp.get("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.exit(1)
-    except:
-        # try to create it
+            # try to make the directory
+            try:
+                os.mkdir(state_dir)
+            except:
+                sys.stderr.write("Couldn't create statedir %s" %(state_dir))
+                sys.exit(1)
+    elif scp.has_option("general", "state_dir"):
+        new_state_dir = scp.get("general", "state_dir")

         try:
         try:

-            os.mkdir(new_state_dir)
-            state_dir = new_state_dir
+            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:
         except:

-            sys.stderr.write("Couldn't create state directory %s\n" %(new_state_dir))
-            sys.exit(1)
-else:
+            # try to create it
+            try:
+                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.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")
+

     try:
     try:

-        mode = os.stat(state_dir)[stat.ST_MODE]
+        mode = os.stat(maildir_root)[stat.ST_MODE]

         if not stat.S_ISDIR(mode):
         if not stat.S_ISDIR(mode):

-            sys.stderr.write("State directory %s is not a directory\n" %(state_dir))
+            sys.stderr.write( \
+                "Maildir Root %s is not a directory\n" \
+                %(maildir_root))

             sys.exit(1)
     except:
         try:
             sys.exit(1)
     except:
         try:

-            os.mkdir(state_dir)
+            os.mkdir(maildir_root)

         except:
         except:

-            sys.stderr.write("State directory %s could not be created\n" %(state_dir))
+            sys.stderr.write("Couldn't create Maildir Root %s\n" \
+                %(maildir_root))

             sys.exit(1)
 
             sys.exit(1)
 

-if scp.has_option("general", "maildir_root"):
-    maildir_root = scp.get("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.exit(1)
-except:
+    feeds = scp.sections()

     try:
     try:

-        os.mkdir(maildir_root)
+        feeds.remove("general")

     except:
     except:

-        sys.stderr.write("Couldn't create Maildir Root %s\n" %(maildir_root))
-        sys.exit(1)
-
-feeds = scp.sections()
-try:
-    feeds.remove("general")
-except:
-    pass
-
-for section in feeds:
-    # check if the directory exists
-    maildir = None
-    try:
-        maildir = scp.get(section, "maildir")
-    except:
-        maildir = section
+        pass

 
 

-    maildir = urllib.urlencode(((section, maildir),)).split("=")[1]
-    maildir = os.path.join(maildir_root, maildir)
+    for section in feeds:
+        # check if the directory exists
+        maildir = None
+        try:
+            maildir = scp.get(section, "maildir")
+        except:
+            maildir = section

 
 

-    try:
-        exists = os.stat(maildir)
-        if stat.S_ISDIR(exists[stat.ST_MODE]):
-            # check if there's a new, cur and tmp directory
+        maildir = urllib.urlencode(((section, maildir),)).split("=")[1]
+        maildir = os.path.join(maildir_root, maildir)
+
+        try:
+            exists = os.stat(maildir)
+            if stat.S_ISDIR(exists[stat.ST_MODE]):
+                # check if there's a new, cur and tmp directory
+                try:
+                    mode = os.stat(os.path.join(maildir, "cur"))[stat.ST_MODE]
+                except:
+                    os.mkdir(os.path.join(maildir, "cur"))
+                    if not stat.S_ISDIR(mode):
+                        sys.stderr.write("Broken maildir: %s\n" %(maildir))
+                try:
+                    mode = os.stat(os.path.join(maildir, "tmp"))[stat.ST_MODE]
+                except:
+                    os.mkdir(os.path.join(maildir, "tmp"))
+                    if not stat.S_ISDIR(mode):
+                        sys.stderr.write("Broken maildir: %s\n" %(maildir))
+                try:
+                    mode = os.stat(os.path.join(maildir, "new"))[stat.ST_MODE]
+                    if not stat.S_ISDIR(mode):
+                        sys.stderr.write("Broken maildir: %s\n" %(maildir))
+                except:
+                    os.mkdir(os.path.join(maildir, "new"))
+            else:
+                sys.stderr.write("Broken maildir: %s\n" %(maildir))
+        except:

             try:
             try:

-                mode = os.stat(os.path.join(maildir, "cur"))[stat.ST_MODE]
+                os.mkdir(maildir)

             except:
             except:

-                os.mkdir(os.path.join(maildir, "cur"))
-                if not stat.S_ISDIR(mode):
-                    sys.stderr.write("Broken maildir: %s\n" %(maildir))
+                sys.stderr.write("Couldn't create root maildir %s\n" \
+                    %(maildir))
+                sys.exit(1)

             try:
             try:

-                mode = os.stat(os.path.join(maildir, "tmp"))[stat.ST_MODE]
-            except:
+                os.mkdir(os.path.join(maildir, "new"))
+                os.mkdir(os.path.join(maildir, "cur"))

                 os.mkdir(os.path.join(maildir, "tmp"))
                 os.mkdir(os.path.join(maildir, "tmp"))

-                if not stat.S_ISDIR(mode):
-                    sys.stderr.write("Broken maildir: %s\n" %(maildir))
-            try:
-                mode = os.stat(os.path.join(maildir, "new"))[stat.ST_MODE]
-                if not stat.S_ISDIR(mode):
-                    sys.stderr.write("Broken maildir: %s\n" %(maildir))

             except:
             except:

-                os.mkdir(os.path.join(maildir, "new"))
-        else:
-            sys.stderr.write("Broken maildir: %s\n" %(maildir))
-    except:
-        try:
-            os.mkdir(maildir)
-        except:
-            sys.stderr.write("Couldn't create root maildir %s\n" %(maildir))
-            sys.exit(1)
-        try:
-            os.mkdir(os.path.join(maildir, "new"))
-            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.exit(1)
+                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
-    # url... lets play!
+        # right - we've got the directories, we've got the section, we know the
+        # url... lets play!

 
 

-    parse_and_deliver(maildir, section, state_dir)
+        parse_and_deliver(maildir, section, state_dir)