Make sure that we feed the parser unicode data

[rss2maildir.git] / rss2maildir.py

diff --git a/rss2maildir.py b/rss2maildir.py
index b77c6d7d11ef14c2878fcdcb379829f4f9f82f1e..e327bc3f046e6b345cef7f7a2bdfffd87ace643f 100755 (executable)
--- a/rss2maildir.py
+++ b/rss2maildir.py
@@ -39,7 +39,13 @@ from optparse import OptionParser
 from ConfigParser import SafeConfigParser
 
 from base64 import b64encode
 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 cgi
 import dbm


@@ -271,7 +277,7 @@ class HTML2Text(HTMLParser):
             elif tag_name == u'a':
                 for attr in attrs:
                     if attr[0].lower() == u'href':
             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
                 self.curdata = self.curdata + u'`'
                 self.opentags.append(tag_name)
                 return


@@ -303,9 +309,15 @@ class HTML2Text(HTMLParser):
         url = u''
         for attr in attrs:
             if attr[0] == 'alt':
         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':
             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):
         if url:
             if alt:
                 if self.images.has_key(alt):


@@ -556,7 +568,7 @@ class HTML2Text(HTMLParser):
     def handle_data(self, data):
         if len(self.opentags) == 0:
             self.opentags.append(u'p')
     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:
 
     def handle_charref(self, name):
         try:


@@ -608,10 +620,17 @@ def open_url(method, url):
         (type, rest) = urllib.splittype(url)
         (host, path) = urllib.splithost(rest)
         (host, port) = urllib.splitport(host)
         (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:
             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]:
             conn.request(method, path)
             response = conn.getresponse()
             if response.status in [301, 302, 303, 307]:


@@ -690,19 +709,30 @@ def parse_and_deliver(maildir, url, statedir):
 
         md5sum = md5.md5(content.encode("utf-8")).hexdigest()
 
 
         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
 
         prevmessageid = None
 

+        db_guid_key = None
+        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"):
         # 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
 
                 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]
             data = cgi.parse_qs(data)
             if data.has_key("message-id"):
                 prevmessageid = data["message-id"][0]


@@ -726,8 +756,8 @@ def parse_and_deliver(maildir, url, statedir):
                 ]) + "@" + socket.gethostname() + ">"
         msg.add_header("Message-ID", messageid)
         msg.set_unixfrom("\"%s\" <rss2maildir@localhost>" %(url))
                 ]) + "@" + 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() \
         if prevmessageid:
             msg.add_header("References", prevmessageid)
         createddate = datetime.datetime.now() \


@@ -738,8 +768,10 @@ def parse_and_deliver(maildir, url, statedir):
         except:
             pass
         msg.add_header("Date", createddate)
         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 = HTML2Text()

-        title = 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)
         title = re.sub(u'<', u'&lt;', title)
         title = re.sub(u'>', u'&gt;', title)
         subj_gen.feed(title)


@@ -753,7 +785,7 @@ def parse_and_deliver(maildir, url, statedir):
             item["link"] )
         htmlpart = MIMEText(htmlcontent.encode("utf-8"), "html", "utf-8")
         textparser = HTML2Text()
             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, \
         textcontent = textparser.gettext()
         textcontent = "%s\n\nItem URL: %s" %( \
             textcontent, \


@@ -790,25 +822,25 @@ def parse_and_deliver(maildir, url, statedir):
                 ("created", createddate), \
                 ("contentmd5", md5sum) \
                 ))
                 ("created", createddate), \
                 ("contentmd5", md5sum) \
                 ))

-            db[url + "|" + item["guid"]] = data
+            db[db_guid_key] = data

             try:
             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]) \
                     ))
                 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:
             except:

-                db[url + "|" + item["link"]] = data
+                db[db_link_key] = data

         else:
             data = urllib.urlencode(( \
                 ("message-id", messageid), \
                 ("created", createddate), \
                 ("contentmd5", md5sum) \
                 ))
         else:
             data = urllib.urlencode(( \
                 ("message-id", messageid), \
                 ("created", createddate), \
                 ("contentmd5", md5sum) \
                 ))

-            db[url + "|" + item["link"]] = data
+            db[db_link_key] = data

 
     if headers:
         data = []
 
     if headers:
         data = []