Stop trying to decode image tags to utf-8

[rss2maildir.git] / rss2maildir.py

diff --git a/rss2maildir.py b/rss2maildir.py
index 59e936117223d5dcac522abd14eadbf974f8ca8e..dc0427a6d2e8ac9ee5a206a89ce5f0634364a3f8 100755 (executable)
--- a/rss2maildir.py
+++ b/rss2maildir.py
@@ -39,11 +39,17 @@ from optparse import OptionParser
 from ConfigParser import SafeConfigParser
 
 from base64 import b64encode
 from ConfigParser import SafeConfigParser
 
 from base64 import b64encode

-import md5
+
+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
 

+import re
+

 from HTMLParser import HTMLParser
 
 class HTML2Text(HTMLParser):
 from HTMLParser import HTMLParser
 
 class HTML2Text(HTMLParser):


@@ -137,6 +143,39 @@ class HTML2Text(HTMLParser):
         u'THORN': u'Þ',
         u'eth': u'ð',
         u'ETH': u'Ð',
         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 = [
         }
 
     blockleveltags = [


@@ -155,7 +194,7 @@ class HTML2Text(HTMLParser):
         u'dt',
         u'dd',
         u'div',
         u'dt',
         u'dd',
         u'div',

-        #u'blockquote',
+        u'blockquote',

         ]
 
     liststarttags = [
         ]
 
     liststarttags = [


@@ -197,6 +236,9 @@ class HTML2Text(HTMLParser):
                 self.listcount.append(1)
                 self.listlevel = len(self.listcount) - 1
 
                 self.listcount.append(1)
                 self.listlevel = len(self.listcount) - 1
 

+            if tag_name == u'dl':
+                self.indentlevel = self.indentlevel + 4
+

             if tag_name in self.liststarttags:
                 smallist = self.opentags[-3:-1]
                 smallist.reverse()
             if tag_name in self.liststarttags:
                 smallist = self.opentags[-3:-1]
                 smallist.reverse()


@@ -265,9 +307,9 @@ 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')
+                alt = attr[1]

             elif attr[0] == 'src':
             elif attr[0] == 'src':

-                url = attr[1].decode('utf-8')
+                url = attr[1]

         if url:
             if alt:
                 if self.images.has_key(alt):
         if url:
             if alt:
                 if self.images.has_key(alt):


@@ -277,12 +319,11 @@ class HTML2Text(HTMLParser):
                     else:
                         while self.images.has_key(alt):
                             alt = alt + "_"
                     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.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:
                     self.curdata = self.curdata \
                         + u'|%s|' %(alt,)
             else:


@@ -372,11 +413,11 @@ class HTML2Text(HTMLParser):
             quote = unicode( \
                 " ".join(self.curdata.encode("utf-8").strip().split()), \
                 "utf-8")
             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 \
             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, \
                 + seperator.join( \
                     textwrap.wrap( \
                         quote, \


@@ -430,12 +471,12 @@ class HTML2Text(HTMLParser):
                 self.text = self.text + u'\n\n'
             elif len(self.text) > 1 and self.text[-2] != u'\n':
                 self.text = self.text + u'\n'
                 self.text = self.text + u'\n\n'
             elif len(self.text) > 1 and self.text[-2] != u'\n':
                 self.text = self.text + u'\n'

-            definition = u' ' * self.indentlevel + definition + "::"
-            indentstring = u'\n' + u' ' * (self.indentlevel + 1)
+            definition = u' ' * (self.indentlevel - 4) + definition + "::"
+            indentstring = u'\n' + u' ' * (self.indentlevel - 3)

             self.text = self.text \
                 + indentstring.join(
                     textwrap.wrap(definition, \
             self.text = self.text \
                 + indentstring.join(
                     textwrap.wrap(definition, \

-                        self.textwidth - self.indentlevel - 1))
+                        self.textwidth - self.indentlevel - 4))

             self.curdata = u''
         elif tag_thats_done == u'dd':
             definition = unicode(" ".join( \
             self.curdata = u''
         elif tag_thats_done == u'dd':
             definition = unicode(" ".join( \


@@ -444,13 +485,13 @@ class HTML2Text(HTMLParser):
             if len(definition) > 0:
                 if len(self.text) > 0 and self.text[-1] != u'\n':
                     self.text = self.text + u'\n'
             if len(definition) > 0:
                 if len(self.text) > 0 and self.text[-1] != u'\n':
                     self.text = self.text + u'\n'

-                indentstring = u'\n' + u' ' * (self.indentlevel + 4)
+                indentstring = u'\n' + u' ' * self.indentlevel

                 self.text = self.text \
                 self.text = self.text \

-                    + u' ' * (self.indentlevel + 4) \
+                    + indentstring \

                     + indentstring.join( \
                         textwrap.wrap( \
                             definition, \
                     + indentstring.join( \
                         textwrap.wrap( \
                             definition, \

-                            self.textwidth - self.indentlevel - 4 \
+                            self.textwidth - self.indentlevel \

                             ) \
                         )
                 self.curdata = u''
                             ) \
                         )
                 self.curdata = u''


@@ -479,8 +520,11 @@ class HTML2Text(HTMLParser):
         if tag in [u'br', u'img']:
             return
 
         if tag in [u'br', u'img']:
             return
 

+        if tag == u'dl':
+            self.indentlevel = self.indentlevel - 4
+

         if tag in self.liststarttags:
         if tag in self.liststarttags:

-            if tag in [u'ol', u'dl', u'ul']:
+            if tag in [u'ol', u'dl', u'ul', u'dd']:

                 self.handle_curdata()
                 # find if there was a previous list level
                 smalllist = self.opentags[:-1]
                 self.handle_curdata()
                 # find if there was a previous list level
                 smalllist = self.opentags[:-1]


@@ -518,12 +562,24 @@ class HTML2Text(HTMLParser):
             self.opentags.append(u'p')
         self.curdata = self.curdata + data.decode("utf-8")
 
             self.opentags.append(u'p')
         self.curdata = self.curdata + data.decode("utf-8")
 

+    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]
     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 + ";"
 
         else:
             entity = "&" + name + ";"
 


@@ -544,7 +600,7 @@ class HTML2Text(HTMLParser):
             self.urls = []
         if len(self.images.keys()) > 0:
             self.text = self.text + u'\n.. ' \
             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 = {}
                     ["|%s| image:: %s" %(a, self.images[a]["url"]) \
                 for a in self.images.keys()]) + u'\n'
             self.images = {}


@@ -556,10 +612,17 @@ 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]:


@@ -631,23 +694,30 @@ def parse_and_deliver(maildir, url, statedir):
         if item.has_key("content"):
             content = item["content"][0]["value"]
         else:
         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()
 
         prevmessageid = None
 
 
         md5sum = md5.md5(content.encode("utf-8")).hexdigest()
 
         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]


@@ -671,8 +741,8 @@ 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() \


@@ -683,8 +753,13 @@ 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()

-        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.encode("utf-8"))

         msg.add_header("Subject", subj_gen.gettext())
         msg.set_default_type("text/plain")
 
         msg.add_header("Subject", subj_gen.gettext())
         msg.set_default_type("text/plain")
 


@@ -732,25 +807,25 @@ 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 = []