Get comments only in case of rss-comments changes

modules/dlfp/backend.py

@@ -86,6 +86,7 @@ class DLFPBackend(BaseBackend, ICapMessages, ICapMessagesPost, ICapContent):
                     continue
                 thread = Thread(article.id)
                 thread.title = article.title
+                thread._rsscomment = article.rsscomment
                 if article.datetime:
                     thread.date = article.datetime
                 yield thread
@@ -97,6 +98,12 @@ class DLFPBackend(BaseBackend, ICapMessages, ICapMessagesPost, ICapContent):
         else:
             thread = None
 
+        oldhash = self.storage.get('hash', id, default="")
+        newhash = self.browser.get_hash(thread._rsscomment)
+        if not getseen and oldhash == newhash:
+            return None
+        self.storage.set('hash', id, newhash)
+        self.storage.save()
         with self.browser:
             content = self.browser.get_content(id)
 

modules/dlfp/browser.py

@@ -20,12 +20,14 @@
 
 import urllib
 import re
+import hashlib
+import lxml
 
 from weboob.tools.browser import BaseBrowser, BrowserHTTPNotFound, BrowserHTTPError, BrowserIncorrectPassword, BrokenPageError
 from weboob.capabilities.messages import CantSendMessage
 
 from .pages.index import IndexPage, LoginPage
-from .pages.news import ContentPage, NewCommentPage, NodePage, CommentPage, NewTagPage
+from .pages.news import ContentPage, NewCommentPage, NodePage, CommentPage, NewTagPage, RSSComment
 from .pages.board import BoardIndexPage
 from .pages.wiki import WikiEditPage
 from .tools import id2url, url2id
@@ -50,6 +52,7 @@ class DLFP(BaseBrowser):
              'https?://[^/]*linuxfr\.org/nodes/(\d+)/comments': NodePage,
              'https?://[^/]*linuxfr\.org/nodes/(\d+)/tags/nouveau': NewTagPage,
              'https?://[^/]*linuxfr\.org/board/index.xml': BoardIndexPage,
+             'https?://[^/]*linuxfr\.org/nodes/(\d+)/comments.atom': RSSComment,
             }
 
     last_board_msg_id = None
@@ -128,6 +131,11 @@ class DLFP(BaseBrowser):
         elif self.is_on_page(ContentPage):
             return self.page.get_article().body
 
+    def get_hash(self, url):
+        self.location(url)
+        myhash = hashlib.md5(lxml.etree.tostring(self.page.document)).hexdigest()
+        return myhash
+
     def get_content(self, _id):
         url, _id = self.parse_id(_id)
 

modules/dlfp/pages/news.py

@@ -26,6 +26,10 @@ from ..tools import url2id
 
 from .index import DLFPPage
 
+class RSSComment(DLFPPage):
+    def on_loaded(self): 
+        pass
+
 class Content(object):
     TAGGABLE = False