* Begin fixes to list handling code - there's 2 unittests that are failing due

[rss2maildir.git] / rss2maildir.py

diff --git a/rss2maildir.py b/rss2maildir.py
index 910b8a913cf64ef9bc714cb2eefd72c28179a1ec..353787c178dc4610e71495edea31281ece9f0425 100755 (executable)
--- a/rss2maildir.py
+++ b/rss2maildir.py
@@ -70,6 +70,9 @@ class HTML2Text(HTMLParser):
         u'ul',
         u'ol',
         u'dl',
+        u'li',
+        u'dt',
+        u'dd',
         u'div',
         #u'blockquote',
         ]
@@ -180,21 +183,17 @@ class HTML2Text(HTMLParser):
         url = u''
         for attr in attrs:
             if attr[0] == 'alt':
-                alt = attr[1]
+                alt = attr[1].decode('utf-8')
             elif attr[0] == 'src':
-                url = attr[1]
+                url = attr[1].decode('utf-8')
         if url:
             self.curdata = self.curdata \
                 + u' [img:' \
-                + unicode( \
-                    url.encode('utf-8'), \
-                    'utf-8')
+                + url
             if alt:
                 self.curdata = self.curdata \
                     + u'(' \
-                    + unicode( \
-                        alt.encode('utf-8'), \
-                        'utf-8') \
+                    + alt \
                     + u')'
             self.curdata = self.curdata \
                 + u']'
@@ -223,10 +222,14 @@ class HTML2Text(HTMLParser):
             if self.ignorenodata:
                 newlinerequired = False
             self.ignorenodata = False
-            if newlinerequired \
-                and len(self.text) > 2 \
-                and self.text[-1] != u'\n' \
-                and self.text[-2] != u'\n':
+            if newlinerequired:
+                if tag_thats_done in [u'dt', u'dd', u'li'] \
+                    and len(self.text) > 1 \
+                    and self.text[-1] != u'\n':
+                        self.text = self.text + u'\n'
+                elif len(self.text) > 2 \
+                    and self.text[-1] != u'\n' \
+                    and self.text[-2] != u'\n':
                     self.text = self.text + u'\n\n'
 
         if tag_thats_done in ["h1", "h2", "h3", "h4", "h5", "h6"]: