better utf-8 handling (though, we currently don't take in to account what

[rss2maildir.git] / rss2maildir.py

diff --git a/rss2maildir.py b/rss2maildir.py
index aaf883f81af394f74f64bb2a2271a4a20fcd3c3a..2d26217f9fb9d0df73f760926aea8946c10d421a 100755 (executable)
--- a/rss2maildir.py
+++ b/rss2maildir.py
@@ -1,5 +1,5 @@
 #!/usr/bin/python
-# coding=utf8
+# coding=utf-8
 
 # rss2maildir.py - RSS feeds to Maildir 1 email per item
 # Copyright (C) 2007  Brett Parker <iDunno@sommitrealweird.co.uk>
@@ -64,10 +64,13 @@ class HTML2Text(HTMLParser):
         self.inheadingtwo = False
         self.inotherheading = False
         self.inparagraph = True
+        self.inblockquote = False
         self.inlink = False
-        self.text = ""
-        self.currentparagraph = ""
-        self.headingtext = ""
+        self.text = u''
+        self.currentparagraph = u''
+        self.headingtext = u''
+        self.blockquote = u''
+        self.inpre = False
         HTMLParser.__init__(self)
 
     def handle_starttag(self, tag, attrs):
@@ -83,47 +86,91 @@ class HTML2Text(HTMLParser):
         elif tag.lower() == "a":
             self.inlink = True
         elif tag.lower() == "br":
+            if self.inparagraph:
+                self.text = self.text + "\n".join(textwrap.wrap(self.currentparagraph, 70)).encode('utf-8') + "\n"
+                self.currentparagraph = ""
+            elif self.inblockquote:
+                self.text = self.text + "\n> " + "\n> ".join([a.strip() for a in textwrap.wrap(self.blockquote, 68)]).encode("utf-8") + "\n"
+                self.blockquote = u''
+            else:
+                self.text = self.text + "\n"
+        elif tag.lower() == "blockquote":
+            self.inblockquote = True
             self.text = self.text + "\n"
         elif tag.lower() == "p":
             if self.text != "":
                 self.text = self.text + "\n\n"
-                self.currentparagraph = ""
-                self.inparagraph = True
+            if self.inparagraph:
+                self.text = self.text + "\n".join(textwrap.wrap(self.currentparagraph, 70)).encode("utf-8")
+            self.currentparagraph = u''
+            self.inparagraph = True
+        elif tag.lower() == "pre":
+            self.text = self.text + "\n"
+            self.inpre = True
+            self.inparagraph = False
+            self.inblockquote = False
 
     def handle_startendtag(self, tag, attrs):
         if tag.lower() == "br":
-            self.text = self.text + "\n"
+            if self.inparagraph:
+                self.text = self.text + "\n".join(textwrap.wrap(self.currentparagraph, 70)).encode("utf-8") + "\n"
+                self.currentparagraph = u''
+            elif self.inblockquote:
+                self.text = self.text + "\n> " + "\n> ".join([a.strip() for a in textwrap.wrap(self.blockquote, 68)]).encode("utf-8") + "\n"
+                self.blockquote = ""
+            else:
+                self.text = self.text + "\n"
 
     def handle_endtag(self, tag):
         if tag.lower() == "h1":
             self.inheadingone = False
-            self.text = self.text + self.headingtext + "\n" + "=" * len(self.headingtext)
-            self.headingtext = ""
+            self.text = self.text + "\n\n" + self.headingtext + "\n" + "=" * len(self.headingtext.strip())
+            self.headingtext = u''
         elif tag.lower() == "h2":
             self.inheadingtwo = False
-            self.text = self.text + self.headingtext + "\n" + "-" * len(self.headingtext)
-            self.headingtext = ""
+            self.text = self.text + "\n\n" + self.headingtext + "\n" + "-" * len(self.headingtext.strip())
+            self.headingtext = u''
         elif tag.lower() in ["h3", "h4", "h5", "h6"]:
             self.inotherheading = False
-            self.text = self.text + self.headingtext + "\n" + "~" * len(self.headingtext)
-            self.headingtext = ""
+            self.text = self.text + "\n\n" + self.headingtext + "\n" + "~" * len(self.headingtext.strip())
+            self.headingtext = u''
         elif tag.lower() == "p":
             self.text = self.text + "\n".join(textwrap.wrap(self.currentparagraph, 70))
             self.inparagraph = False
+        elif tag.lower() == "blockquote":
+            self.text = self.text + "\n> " + "\n> ".join([a.strip() for a in textwrap.wrap(self.blockquote, 68)]).encode("utf-8") + "\n"
+            self.inblockquote = False
+            self.blockquote = u''
+        elif tag.lower() == "pre":
+            self.inpre = False
 
     def handle_data(self, data):
-        if not self.inheadingone and not self.inheadingtwo and not self.inotherheading and not self.inparagraph:
-            self.text = self.text + data.strip() + " "
+        if self.inheadingone or self.inheadingtwo or self.inotherheading:
+            self.headingtext = self.headingtext + unicode(data, "utf-8").strip() + u' '
+        elif self.inblockquote:
+            self.blockquote = self.blockquote + unicode(data, "utf-8").strip() + u' '
         elif self.inparagraph:
-            self.currentparagraph = self.currentparagraph + data.strip() + " "
+            self.currentparagraph = self.currentparagraph + unicode(data, "utf-8").strip() + u' '
+        elif self.inpre:
+            self.text = self.text + data.encode("utf-8")
         else:
-            self.headingtext = self.headingtext + data.strip() + " "
+            self.text = self.text + unicode(data, "utf-8").strip() + u' '
 
     def handle_entityref(self, name):
+        entity = name
         if entities.has_key(name.lower()):
-            self.text = self.text + entities[name.lower()]
+            entity = entities[name.lower()]
+        elif name[0] == "#":
+            entity = unichr(int(name[1:]))
+        else:
+            entity = "&" + name + ";"
+
+        if self.inparagraph:
+            self.currentparagraph = self.currentparagraph + entity
+        elif self.inblockquote:
+            self.blockquote = self.blockquote + entity
         else:
-            self.text = self.text + "&" + name + ";"
+            self.text = self.text + entity
 
     def gettext(self):
         data = self.text
@@ -144,10 +191,10 @@ def parse_and_deliver(maildir, url, statedir):
         else:
             content = item["summary"]
 
-        md5sum = md5.md5(content.encode("utf8")).hexdigest()
+        md5sum = md5.md5(content.encode("utf-8")).hexdigest()
 
-        if db.has_key(item["link"]):
-            data = db[item["link"]]
+        if db.has_key(url + "|" + item["link"]):
+            data = db[url + "|" + item["link"]]
             data = cgi.parse_qs(data)
             if data["contentmd5"][0] == md5sum:
                 continue
@@ -169,11 +216,11 @@ def parse_and_deliver(maildir, url, statedir):
         msg.add_header("Subject", item["title"])
         msg.set_default_type("text/plain")
 
-        htmlpart = MIMEText(content.encode("utf8"), "html", "utf8")
+        htmlpart = MIMEText(content.encode("utf-8"), "html", "utf-8")
         textparser = HTML2Text()
-        textparser.feed(content.encode("utf8"))
+        textparser.feed(content.encode("utf-8"))
         textcontent = textparser.gettext()
-        textpart = MIMEText(textcontent, "plain", "utf8")
+        textpart = MIMEText(textcontent.encode("utf-8"), "plain", "utf-8")
         msg.attach(textpart)
         msg.attach(htmlpart)
 
@@ -191,7 +238,7 @@ def parse_and_deliver(maildir, url, statedir):
 
         # now add to the database about the item
         data = urllib.urlencode((("message-id", messageid), ("created", createddate), ("contentmd5", md5sum)))
-        db[item["link"]] = data
+        db[url + "|" + item["link"]] = data
 
     db.close()