Fix for title parsing

[rss2maildir.git] / rss2maildir.py

diff --git a/rss2maildir.py b/rss2maildir.py
index 0dfd0f04b0f035224a900ab94d73e400aa2d82e9..ff2901c5ffab0414fa563975801c5e6fd76d315f 100755 (executable)
--- a/rss2maildir.py
+++ b/rss2maildir.py
@@ -44,6 +44,8 @@ import md5
 import cgi
 import dbm
 
+import re
+
 from HTMLParser import HTMLParser
 
 class HTML2Text(HTMLParser):
@@ -137,6 +139,39 @@ 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 +190,7 @@ class HTML2Text(HTMLParser):
         u'dt',
         u'dd',
         u'div',
-        #u'blockquote',
+        u'blockquote',
         ]
 
     liststarttags = [
@@ -197,6 +232,9 @@ class HTML2Text(HTMLParser):
                 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()
@@ -277,12 +315,11 @@ 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:
@@ -372,11 +409,11 @@ 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, \
@@ -430,12 +467,12 @@ 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'
-            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.textwidth - self.indentlevel - 1))
+                        self.textwidth - self.indentlevel - 4))
             self.curdata = u''
         elif tag_thats_done == u'dd':
             definition = unicode(" ".join( \
@@ -444,13 +481,13 @@ class HTML2Text(HTMLParser):
             if len(definition) > 0:
                 if len(self.text) > 0 and self.text[-1] != u'\n':
                     self.text = self.text + u'\n'
-                indentstring = u'\n' + u' ' * (self.indentlevel + 4)
+                indentstring = u'\n' + u' ' * self.indentlevel
                 self.text = self.text \
-                    + u' ' * (self.indentlevel + 4) \
+                    + indentstring \
                     + indentstring.join( \
                         textwrap.wrap( \
                             definition, \
-                            self.textwidth - self.indentlevel - 4 \
+                            self.textwidth - self.indentlevel \
                             ) \
                         )
                 self.curdata = u''
@@ -479,8 +516,11 @@ class HTML2Text(HTMLParser):
         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 [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]
@@ -518,12 +558,15 @@ class HTML2Text(HTMLParser):
             self.opentags.append(u'p')
         self.curdata = self.curdata + data.decode("utf-8")
 
+    def handle_charref(self, name):
+        entity = unichr(int(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 + ";"
 
@@ -544,7 +587,7 @@ 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 = {}
@@ -684,7 +727,10 @@ def parse_and_deliver(maildir, url, statedir):
             pass
         msg.add_header("Date", createddate)
         subj_gen = HTML2Text()
-        subj_gen.feed(item["title"].encod("utf-8"))
+        title = item["title"].encode("utf-8")
+        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")