* make db key actually unique for feed url + link url

[rss2maildir.git] / rss2maildir.py

diff --git a/rss2maildir.py b/rss2maildir.py
index 5d89340ba3da8f89a60d31fcd224764362f92620..d721d0e49c19994c84b6e6407766e02c836ce5d4 100755 (executable)
--- a/rss2maildir.py
+++ b/rss2maildir.py
@@ -31,6 +31,7 @@ from email.MIMEText import MIMEText
 import datetime
 import random
 import string
 import datetime
 import random
 import string

+import textwrap

 
 import socket
 
 
 import socket
 


@@ -53,6 +54,7 @@ entities = {
     "copy": "©",
     "apos": "'",
     "quote": "\"",
     "copy": "©",
     "apos": "'",
     "quote": "\"",

+    "nbsp": " ",

     }
 
 class HTML2Text(HTMLParser):
     }
 
 class HTML2Text(HTMLParser):


@@ -61,18 +63,23 @@ class HTML2Text(HTMLParser):
         self.inheadingone = False
         self.inheadingtwo = False
         self.inotherheading = False
         self.inheadingone = False
         self.inheadingtwo = False
         self.inotherheading = False

+        self.inparagraph = True

         self.inlink = False
         self.text = ""
         self.inlink = False
         self.text = ""

+        self.currentparagraph = ""

         self.headingtext = ""
         HTMLParser.__init__(self)
 
     def handle_starttag(self, tag, attrs):
         if tag.lower() == "h1":
             self.inheadingone = True
         self.headingtext = ""
         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
         elif tag.lower() == "h2":
             self.inheadingtwo = True

+            self.inparagraph = False

         elif tag.lower() in ["h3", "h4", "h5", "h6"]:
             self.inotherheading = True
         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":
         elif tag.lower() == "a":
             self.inlink = True
         elif tag.lower() == "br":


@@ -80,6 +87,8 @@ class HTML2Text(HTMLParser):
         elif tag.lower() == "p":
             if self.text != "":
                 self.text = self.text + "\n\n"
         elif tag.lower() == "p":
             if self.text != "":
                 self.text = self.text + "\n\n"

+                self.currentparagraph = ""
+                self.inparagraph = True

 
     def handle_startendtag(self, tag, attrs):
         if tag.lower() == "br":
 
     def handle_startendtag(self, tag, attrs):
         if tag.lower() == "br":


@@ -98,10 +107,15 @@ class HTML2Text(HTMLParser):
             self.inotherheading = False
             self.text = self.text + self.headingtext + "\n" + "~" * len(self.headingtext)
             self.headingtext = ""
             self.inotherheading = False
             self.text = self.text + self.headingtext + "\n" + "~" * len(self.headingtext)
             self.headingtext = ""

+        elif tag.lower() == "p":
+            self.text = self.text + "\n".join(textwrap.wrap(self.currentparagraph, 70))
+            self.inparagraph = False

 
     def handle_data(self, data):
 
     def handle_data(self, data):

-        if not self.inheadingone and not self.inheadingtwo and not self.inotherheading:
+        if not self.inheadingone and not self.inheadingtwo and not self.inotherheading and not self.inparagraph:

             self.text = self.text + data.strip() + " "
             self.text = self.text + data.strip() + " "

+        elif self.inparagraph:
+            self.currentparagraph = self.currentparagraph + data.strip() + " "

         else:
             self.headingtext = self.headingtext + data.strip() + " "
 
         else:
             self.headingtext = self.headingtext + data.strip() + " "
 


@@ -112,7 +126,10 @@ class HTML2Text(HTMLParser):
             self.text = self.text + "&" + name + ";"
 
     def gettext(self):
             self.text = self.text + "&" + name + ";"
 
     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)
 
 def parse_and_deliver(maildir, url, statedir):
     md = mailbox.Maildir(maildir)


@@ -129,8 +146,8 @@ def parse_and_deliver(maildir, url, statedir):
 
         md5sum = md5.md5(content.encode("utf8")).hexdigest()
 
 
         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
             data = cgi.parse_qs(data)
             if data["contentmd5"][0] == md5sum:
                 continue


@@ -174,7 +191,7 @@ 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)))
 
         # 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()
 
 
     db.close()