Normalise spaces where they should be.

[rss2maildir.git] / rss2maildir.py

diff --git a/rss2maildir.py b/rss2maildir.py
index 113d931a1db7bf162f603ce1cc646137a1f97b9c..ce6c3426b54fff080b6639a09e85269375fb170a 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',
         ]
@@ -148,7 +151,7 @@ class HTML2Text(HTMLParser):
             elif tag_name == u'a':
                 for attr in attrs:
                     if attr[0].lower() == u'href':
-                        self.urls.append(attr[1])
+                        self.urls.append(attr[1].decode('utf-8'))
                 self.curdata = self.curdata + u'`'
                 self.opentags.append(tag_name)
                 return
@@ -219,17 +222,20 @@ 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"]:
             underline = u''
             underlinechar = u'='
-            headingtext = unicode( \
-                self.curdata.encode("utf-8").strip(), "utf-8")
+            headingtext = " ".join(self.curdata.split())
             seperator = u'\n' + u' '*self.indentlevel
             headingtext = seperator.join( \
                 textwrap.wrap( \
@@ -250,11 +256,12 @@ class HTML2Text(HTMLParser):
                 underline = u' ' * self.indentlevel \
                     + underlinechar * len(headingtext)
             self.text = self.text \
-                + headingtext.encode("utf-8") + u'\n' \
+                + headingtext + u'\n' \
                 + underline
         elif tag_thats_done in [u'p', u'div']:
             paragraph = unicode( \
-                self.curdata.strip().encode("utf-8"), "utf-8")
+                " ".join(self.curdata.strip().encode("utf-8").split()), \
+                "utf-8")
             seperator = u'\n' + u' ' * self.indentlevel
             self.text = self.text \
                 + u' ' * self.indentlevel \
@@ -263,10 +270,11 @@ class HTML2Text(HTMLParser):
                         paragraph, self.textwidth - self.indentlevel))
         elif tag_thats_done == "pre":
             self.text = self.text + unicode( \
-                self.curdata.encode("utf-8"), "utf-8")
+                " ".join(self.curdata.encode("utf-8").split()), "utf-8")
         elif tag_thats_done == u'blockquote':
             quote = unicode( \
-                self.curdata.encode("utf-8").strip(), "utf-8")
+                " ".join(self.curdata.encode("utf-8").strip().split()), \
+                "utf-8")
             seperator = u'\n' + u' ' * self.indentlevel + u'> '
             if len(self.text) > 0 and self.text[-1] != u'\n':
                 self.text = self.text + u'\n'
@@ -318,7 +326,9 @@ class HTML2Text(HTMLParser):
                 )
             self.curdata = u''
         elif tag_thats_done == u'dt':
-            definition = unicode(self.curdata.encode("utf-8").strip(), "utf-8")
+            definition = unicode(" ".join( \
+                    self.curdata.encode("utf-8").strip().split()), \
+                "utf-8")
             if len(self.text) > 0 and self.text[-1] != u'\n':
                 self.text = self.text + u'\n\n'
             elif len(self.text) > 1 and self.text[-2] != u'\n':
@@ -331,7 +341,9 @@ class HTML2Text(HTMLParser):
                         self.textwidth - self.indentlevel - 1))
             self.curdata = u''
         elif tag_thats_done == u'dd':
-            definition = unicode(self.curdata.encode("utf-8").strip(), "utf-8")
+            definition = unicode(" ".join( \
+                    self.curdata.encode("utf-8").strip().split()),
+                "utf-8")
             if len(definition) > 0:
                 if len(self.text) > 0 and self.text[-1] != u'\n':
                     self.text = self.text + u'\n'
@@ -407,7 +419,7 @@ class HTML2Text(HTMLParser):
     def handle_data(self, data):
         if len(self.opentags) == 0:
             self.opentags.append(u'p')
-        self.curdata = self.curdata + unicode(data, "utf-8")
+        self.curdata = self.curdata + data.decode("utf-8")
 
     def handle_entityref(self, name):
         entity = name
@@ -717,11 +729,13 @@ if __name__ == "__main__":
     elif scp.has_option("general", "state_dir"):
         new_state_dir = scp.get("general", "state_dir")
         try:
-            mode = os.stat(state_dir)[stat.ST_MODE]
+            mode = os.stat(new_state_dir)[stat.ST_MODE]
             if not stat.S_ISDIR(mode):
                 sys.stderr.write( \
                     "State directory (%s) is not a directory\n" %(state_dir))
                 sys.exit(1)
+            else:
+                state_dir = new_state_dir
         except:
             # try to create it
             try: