* improved entity handling

[rss2maildir.git] / rss2maildir.py

diff --git a/rss2maildir.py b/rss2maildir.py
index 5d89340ba3da8f89a60d31fcd224764362f92620..641ffb196d488fbce95a6354242b55d80640b216 100755 (executable)
--- a/rss2maildir.py
+++ b/rss2maildir.py
@@ -31,6 +31,7 @@ from email.MIMEText import MIMEText
 import datetime
 import random
 import string
+import textwrap
 
 import socket
 
@@ -53,6 +54,7 @@ entities = {
     "copy": "©",
     "apos": "'",
     "quote": "\"",
+    "nbsp": " ",
     }
 
 class HTML2Text(HTMLParser):
@@ -61,58 +63,120 @@ class HTML2Text(HTMLParser):
         self.inheadingone = False
         self.inheadingtwo = False
         self.inotherheading = False
+        self.inparagraph = True
+        self.inblockquote = False
         self.inlink = False
         self.text = ""
+        self.currentparagraph = ""
         self.headingtext = ""
+        self.blockquote = ""
+        self.inpre = False
         HTMLParser.__init__(self)
 
     def handle_starttag(self, tag, attrs):
         if tag.lower() == "h1":
             self.inheadingone = True
+            self.inparagraph = False
         elif tag.lower() == "h2":
             self.inheadingtwo = True
+            self.inparagraph = False
         elif tag.lower() in ["h3", "h4", "h5", "h6"]:
             self.inotherheading = True
+            self.inparagraph = False
         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)) + "\n"
+                self.currentparagraph = ""
+            elif self.inblockquote:
+                self.text = self.text + "\n> " + "\n> ".join([a.strip() for a in textwrap.wrap(self.blockquote, 68)]) + "\n"
+                self.blockquote = ""
+            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"
+            if self.inparagraph:
+                self.text = self.text + "\n".join(textwrap.wrap(self.currentparagraph, 70))
+            self.currentparagraph = ""
+            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)) + "\n"
+                self.currentparagraph = ""
+            elif self.inblockquote:
+                self.text = self.text + "\n> " + "\n> ".join([a.strip() for a in textwrap.wrap(self.blockquote, 68)]) + "\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.text = self.text + "\n\n" + self.headingtext + "\n" + "=" * len(self.headingtext.strip())
             self.headingtext = ""
         elif tag.lower() == "h2":
             self.inheadingtwo = False
-            self.text = self.text + self.headingtext + "\n" + "-" * len(self.headingtext)
+            self.text = self.text + "\n\n" + self.headingtext + "\n" + "-" * len(self.headingtext.strip())
             self.headingtext = ""
         elif tag.lower() in ["h3", "h4", "h5", "h6"]:
             self.inotherheading = False
-            self.text = self.text + self.headingtext + "\n" + "~" * len(self.headingtext)
+            self.text = self.text + "\n\n" + self.headingtext + "\n" + "~" * len(self.headingtext.strip())
             self.headingtext = ""
+        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)]) + "\n"
+            self.inblockquote = False
+            self.blockquote = ""
+        elif tag.lower() == "pre":
+            self.inpre = False
 
     def handle_data(self, data):
-        if not self.inheadingone and not self.inheadingtwo and not self.inotherheading:
-            self.text = self.text + data.strip() + " "
-        else:
+        if self.inheadingone or self.inheadingtwo or self.inotherheading:
             self.headingtext = self.headingtext + data.strip() + " "
+        elif self.inblockquote:
+            self.blockquote = self.blockquote + data.strip() + " "
+        elif self.inparagraph:
+            self.currentparagraph = self.currentparagraph + data.strip() + " "
+        elif self.inpre:
+            self.text = self.text + data
+        else:
+            self.text = self.text + data.strip() + " "
 
     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):
-        return self.text
+        data = self.text
+        if self.inparagraph:
+            data = data + "\n".join(textwrap.wrap(self.currentparagraph, 70))
+        return data
 
 def parse_and_deliver(maildir, url, statedir):
     md = mailbox.Maildir(maildir)
@@ -129,8 +193,8 @@ def parse_and_deliver(maildir, url, statedir):
 
         md5sum = md5.md5(content.encode("utf8")).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
@@ -174,7 +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()