Do not disqualify subscriptions due to being not-well-formed

slixfeed/action.py

@@ -700,7 +700,7 @@ def is_feed_json(document):
     return value
 
 
-def is_feed(feed):
+def is_feed(url, feed):
     """
     Determine whether document is feed or not.
 
@@ -737,10 +737,9 @@ def is_feed(feed):
             #     "No entries nor title for {}"
             #     ).format(url)
     elif feed.bozo:
-        value = False
+        # NOTE Consider valid even when is not-well-formed
-        # message = (
+        value = True
-        #     "Bozo detected for {}"
+        logger.warning('Bozo detected for {}'.format(url))
-        #     ).format(url)
     else:
         value = True
         # message = (
@@ -984,8 +983,7 @@ async def add_feed(self, jid_bare, db_file, url, identifier):
                     document = result['content']
                     feed = parse(document)
                     # if document and status_code == 200:
-                    # if is_feed(url, feed):
+                    if is_feed(url, feed):
-                    if is_feed(feed):
                         if "title" in feed["feed"].keys():
                             title = feed["feed"]["title"]
                         else:

slixfeed/crawl.py

@@ -62,9 +62,12 @@ Link  : https://www.jsonfeed.org/feed.json/videos.xml
 from aiohttp import ClientError, ClientSession, ClientTimeout
 from feedparser import parse
 import logging
+from lxml import etree
 from lxml import html
+from lxml.etree import fromstring
 import slixfeed.config as config
 import slixfeed.fetch as fetch
+from slixfeed.log import Logger
 from slixfeed.url import complete_url, join_url, trim_url
 from urllib.parse import urlsplit, urlunsplit
 
@@ -107,6 +110,7 @@ from urllib.parse import urlsplit, urlunsplit
 #             else:
 #                 return await callback(url)
 
+logger = Logger(__name__)
 
 async def probe_page(url, document=None):
     """
@@ -130,8 +134,26 @@ async def probe_page(url, document=None):
         # tree = etree.fromstring(res[0]) # etree is for xml
         tree = html.fromstring(document)
         result = None
-    except:
+    except Exception as e:
-        logging.warning("Failed to parse URL as feed for {}.".format(url))
+        logger.error(str(e))
+        try:
+            # /questions/15830421/xml-unicode-strings-with-encoding-declaration-are-not-supported
+            # xml = html.fromstring(document.encode('utf-8'))
+            # parser = etree.XMLParser(ns_clean=True, recover=True, encoding='utf-8')
+            # tree = fromstring(xml, parser=parser)
+
+            # /questions/57833080/how-to-fix-unicode-strings-with-encoding-declaration-are-not-supported
+            #tree = html.fromstring(bytes(document, encoding='utf8'))
+
+            # https://twigstechtips.blogspot.com/2013/06/python-lxml-strings-with-encoding.html
+            #parser = etree.XMLParser(recover=True)
+            #tree = etree.fromstring(document, parser)
+
+            tree = html.fromstring(document.encode('utf-8'))
+            result = None
+        except Exception as e:
+            logger.error(str(e))
+            logger.warning("Failed to parse URL as feed for {}.".format(url))
             result = {'link' : None,
                       'index' : None,
                       'name' : None,
@@ -139,16 +161,16 @@ async def probe_page(url, document=None):
                       'error' : True,
                       'exist' : None}
     if not result:
-        logging.debug("Feed auto-discovery engaged for {}".format(url))
+        logger.debug("Feed auto-discovery engaged for {}".format(url))
         result = await feed_mode_auto_discovery(url, tree)
     if not result:
-        logging.debug("Feed link scan mode engaged for {}".format(url))
+        logger.debug("Feed link scan mode engaged for {}".format(url))
         result = await feed_mode_scan(url, tree)
     if not result:
-        logging.debug("Feed arbitrary mode engaged for {}".format(url))
+        logger.debug("Feed arbitrary mode engaged for {}".format(url))
         result = await feed_mode_guess(url, tree)
     if not result:
-        logging.debug("No feeds were found for {}".format(url))
+        logger.debug("No feeds were found for {}".format(url))
         result = None
     return result
 

slixfeed/version.py

@@ -1,2 +1,2 @@
-__version__ = '0.1.73'
+__version__ = '0.1.74'
-__version_info__ = (0, 1, 73)
+__version_info__ = (0, 1, 74)

slixfeed/xmpp/client.py

@@ -52,9 +52,9 @@ from slixfeed.config import Config
 import slixfeed.crawl as crawl
 import slixfeed.dt as dt
 import slixfeed.fetch as fetch
+from slixfeed.log import Logger
 import slixfeed.sqlite as sqlite
 import slixfeed.url as uri
-from slixfeed.log import Logger
 from slixfeed.version import __version__
 from slixfeed.xmpp.bookmark import XmppBookmark
 from slixfeed.xmpp.chat import Chat
@@ -1140,8 +1140,7 @@ class XmppClient(slixmpp.ClientXMPP):
             if not result['error']:
                 document = result['content']
                 feed = parse(document)
-                # if is_feed(url, feed):
+                if action.is_feed(url, feed):
-                if action.is_feed(feed):
                     form['instructions'] = 'Select entries to publish.'
                     options = form.add_field(desc='Select entries to post.',
                                              ftype='list-multi',

slixfeed/xmpp/commands.py

@@ -754,8 +754,7 @@ class XmppCommands:
                         if not result['error']:
                             document = result['content']
                             feed = parse(document)
-                            # if is_feed(url, feed):
+                            if action.is_feed(url, feed):
-                            if action.is_feed(feed):
                                 message = action.view_feed(url, feed)
                                 break
                             else:
@@ -791,8 +790,7 @@ class XmppCommands:
                             document = result['content']
                             status = result['status_code']
                             feed = parse(document)
-                            # if is_feed(url, feed):
+                            if action.is_feed(url, feed):
-                            if action.is_feed(feed):
                                 message = action.view_entry(url, feed, num)
                                 break
                             else: