X-Git-Url: https://git.sommitrealweird.co.uk/rss2maildir.git/blobdiff_plain/9a3c701ba08530e1a884ad3fea33f790feb0aa0a..HEAD:/rss2maildir.py

diff --git a/rss2maildir.py b/rss2maildir.py
index 6a24b94..314a9c1 100755
--- a/rss2maildir.py
+++ b/rss2maildir.py
@@ -39,11 +39,19 @@ from optparse import OptionParser
 from ConfigParser import SafeConfigParser
 
 from base64 import b64encode
-import md5
+
+import chardet
+
+if sys.version_info[0] == 2 and sys.version_info[1] >= 6:
+    import hashlib as md5
+else:
+    import md5
 
 import cgi
 import dbm
 
+import re
+
 from HTMLParser import HTMLParser
 
 class HTML2Text(HTMLParser):
@@ -137,6 +145,39 @@ class HTML2Text(HTMLParser):
         u'THORN': u'Þ',
         u'eth': u'ð',
         u'ETH': u'Ð',
+        u'mdash': u'—',
+        u'ndash': u'–',
+        u'sect': u'§',
+        u'para': u'¶',
+        u'uarr': u'↑',
+        u'darr': u'↓',
+        u'larr': u'←',
+        u'rarr': u'→',
+        u'dagger': u'†',
+        u'Dagger': u'‡',
+        u'permil': u'‰',
+        u'prod': u'∏',
+        u'infin': u'∞',
+        u'radic': u'√',
+        u'there4': u'∴',
+        u'int': u'∫',
+        u'asymp': u'≈',
+        u'ne': u'≠',
+        u'equiv': '≡',
+        u'le': u'≤',
+        u'ge': u'≥',
+        u'loz': u'⋄',
+        u'sum': u'∑',
+        u'part': u'∂',
+        u'prime': u'′',
+        u'Prime': u'″',
+        u'harr': u'↔',
+        u'micro': u'µ',
+        u'not': u'¬',
+        u'plusmn': u'±',
+        u'divide': u'÷',
+        u'cent': u'¢',
+        u'euro': u'€',
         }
 
     blockleveltags = [
@@ -155,7 +196,7 @@ class HTML2Text(HTMLParser):
         u'dt',
         u'dd',
         u'div',
-        #u'blockquote',
+        u'blockquote',
         ]
 
     liststarttags = [
@@ -236,7 +277,7 @@ class HTML2Text(HTMLParser):
             elif tag_name == u'a':
                 for attr in attrs:
                     if attr[0].lower() == u'href':
-                        self.urls.append(attr[1].decode('utf-8'))
+                        self.urls.append(attr[1])
                 self.curdata = self.curdata + u'`'
                 self.opentags.append(tag_name)
                 return
@@ -268,9 +309,15 @@ class HTML2Text(HTMLParser):
         url = u''
         for attr in attrs:
             if attr[0] == 'alt':
-                alt = attr[1].decode('utf-8')
+                if isinstance(attr[1], str):
+                    alt = u'%s' %(attr[1])
+                else:
+                    alt = attr[1]
             elif attr[0] == 'src':
-                url = attr[1].decode('utf-8')
+                if isinstance(attr[1], str):
+                    url = u'%s' %(attr[1])
+                else:
+                    url = attr[1]
         if url:
             if alt:
                 if self.images.has_key(alt):
@@ -280,12 +327,11 @@ class HTML2Text(HTMLParser):
                     else:
                         while self.images.has_key(alt):
                             alt = alt + "_"
-                        self.images[alt]["url"] = url
+                        self.images[alt] = {"url": url}
                         self.curdata = self.curdata \
                             + u'|%s|' %(alt,)
                 else:
-                    self.images[alt] = {}
-                    self.images[alt]["url"] = url
+                    self.images[alt] = {"url": url}
                     self.curdata = self.curdata \
                         + u'|%s|' %(alt,)
             else:
@@ -375,11 +421,11 @@ class HTML2Text(HTMLParser):
             quote = unicode( \
                 " ".join(self.curdata.encode("utf-8").strip().split()), \
                 "utf-8")
-            seperator = u'\n' + u' ' * self.indentlevel + u'> '
+            seperator = u'\n' + u' ' * self.indentlevel + u'    '
             if len(self.text) > 0 and self.text[-1] != u'\n':
                 self.text = self.text + u'\n'
             self.text = self.text \
-                + u'> ' \
+                + u'    ' \
                 + seperator.join( \
                     textwrap.wrap( \
                         quote, \
@@ -522,14 +568,26 @@ class HTML2Text(HTMLParser):
     def handle_data(self, data):
         if len(self.opentags) == 0:
             self.opentags.append(u'p')
-        self.curdata = self.curdata + data.decode("utf-8")
+        self.curdata = "%s%s" %(self.curdata, data)
+
+    def handle_charref(self, name):
+        try:
+            entity = unichr(int(name))
+        except:
+            if name[0] == 'x':
+                try:
+                    entity = unichr(int('0%s' %(name,), 16))
+                except:
+                    entity = u'#%s' %(name,)
+            else:
+                entity = u'#%s' %(name,)
+        self.curdata = self.curdata + unicode(entity.encode('utf-8'), \
+            "utf-8")
 
     def handle_entityref(self, name):
         entity = name
         if HTML2Text.entities.has_key(name):
             entity = HTML2Text.entities[name]
-        elif name[0] == "#":
-            entity = unichr(int(name[1:]))
         else:
             entity = "&" + name + ";"
 
@@ -550,7 +608,7 @@ class HTML2Text(HTMLParser):
             self.urls = []
         if len(self.images.keys()) > 0:
             self.text = self.text + u'\n.. ' \
-                + u'.. '.join( \
+                + u'\n.. '.join( \
                     ["|%s| image:: %s" %(a, self.images[a]["url"]) \
                 for a in self.images.keys()]) + u'\n'
             self.images = {}
@@ -562,10 +620,17 @@ def open_url(method, url):
         (type, rest) = urllib.splittype(url)
         (host, path) = urllib.splithost(rest)
         (host, port) = urllib.splitport(host)
-        if port == None:
+        if type == "https":
+            if port == None:
+                port = 443
+        elif port == None:
             port = 80
         try:
-            conn = httplib.HTTPConnection("%s:%s" %(host, port))
+            conn = None
+            if type == "http":
+                conn = httplib.HTTPConnection("%s:%s" %(host, port))
+            else:
+                conn = httplib.HTTPSConnection("%s:%s" %(host, port))
             conn.request(method, path)
             response = conn.getresponse()
             if response.status in [301, 302, 303, 307]:
@@ -637,23 +702,39 @@ def parse_and_deliver(maildir, url, statedir):
         if item.has_key("content"):
             content = item["content"][0]["value"]
         else:
-            content = item["summary"]
+            if item.has_key("description"):
+                content = item["description"]
+            else:
+                content = u''
 
         md5sum = md5.md5(content.encode("utf-8")).hexdigest()
 
+        # make sure content is unicode encoded
+        if not isinstance(content, unicode):
+            cd_res = chardet.detect(content)
+            chrset = cd_res['encoding']
+            print "detected charset %s for item %s" %(chrset, item["link"])
+            content = content.decode(chrset)
+
         prevmessageid = None
 
+        db_guid_key = None
+        if not item.has_key("link"):
+            item["link"] = u'#' + md5sum
+        db_link_key = (url + u'|' + item["link"]).encode("utf-8")
+
         # 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"]]
+            db_guid_key = (url + u'|' + item["guid"]).encode("utf-8")
+            if db.has_key(db_guid_key):
+                data = db[db_guid_key]
                 data = cgi.parse_qs(data)
                 if data["contentmd5"][0] == md5sum:
                     continue
 
-        if db.has_key(url + "|" + item["link"]):
-            data = db[url + "|" + item["link"]]
+        if db.has_key(db_link_key):
+            data = db[db_link_key]
             data = cgi.parse_qs(data)
             if data.has_key("message-id"):
                 prevmessageid = data["message-id"][0]
@@ -677,8 +758,8 @@ def parse_and_deliver(maildir, url, statedir):
                 ]) + "@" + 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("From", "\"%s\" <rss2maildir@localhost>" %(author.encode("utf-8")))
+        msg.add_header("To", "\"%s\" <rss2maildir@localhost>" %(url.encode("utf-8")))
         if prevmessageid:
             msg.add_header("References", prevmessageid)
         createddate = datetime.datetime.now() \
@@ -689,8 +770,13 @@ def parse_and_deliver(maildir, url, statedir):
         except:
             pass
         msg.add_header("Date", createddate)
+        msg.add_header("X-rss2maildir-rundate", datetime.datetime.now() \
+            .strftime("%a, %e %b %Y %T -0000"))
         subj_gen = HTML2Text()
-        subj_gen.feed(item["title"].encode("utf-8"))
+        title = item["title"]
+        title = re.sub(u'<', u'&lt;', title)
+        title = re.sub(u'>', u'&gt;', title)
+        subj_gen.feed(title)
         msg.add_header("Subject", subj_gen.gettext())
         msg.set_default_type("text/plain")
 
@@ -701,7 +787,7 @@ def parse_and_deliver(maildir, url, statedir):
             item["link"] )
         htmlpart = MIMEText(htmlcontent.encode("utf-8"), "html", "utf-8")
         textparser = HTML2Text()
-        textparser.feed(content.encode("utf-8"))
+        textparser.feed(content)
         textcontent = textparser.gettext()
         textcontent = "%s\n\nItem URL: %s" %( \
             textcontent, \
@@ -738,25 +824,25 @@ def parse_and_deliver(maildir, url, statedir):
                 ("created", createddate), \
                 ("contentmd5", md5sum) \
                 ))
-            db[url + "|" + item["guid"]] = data
+            db[db_guid_key] = data
             try:
-                data = db[url + "|" + item["link"]]
+                data = db[db_link_key]
                 data = cgi.parse_qs(data)
                 newdata = urllib.urlencode(( \
                     ("message-id", messageid), \
                     ("created", data["created"][0]), \
                     ("contentmd5", data["contentmd5"][0]) \
                     ))
-                db[url + "|" + item["link"]] = newdata
+                db[db_link_key] = newdata
             except:
-                db[url + "|" + item["link"]] = data
+                db[db_link_key] = data
         else:
             data = urllib.urlencode(( \
                 ("message-id", messageid), \
                 ("created", createddate), \
                 ("contentmd5", md5sum) \
                 ))
-            db[url + "|" + item["link"]] = data
+            db[db_link_key] = data
 
     if headers:
         data = []