Merge branch 'master' into social

* master: Adding build/ back to .gitignore. Updating .gitignore. Using dsa pem for retrieving rabbit update counts. Preserving feed fetch error histories for 2 weeks. Adding munin graph for known good feeds. Turning down verbosity on feed tasking. Adding known good feeds to munin to chart their progress. Fixing headers overflow in Chrome. Adding new feeds db column 'known_good', for forcing updates of feeds that were once good but are now throwing 500s. Adding blogging OT press story.
2025-09-18 21:43:31 +00:00 · 2012-02-02 17:22:32 -08:00 · 2012-02-02 17:22:32 -08:00 · b3fffab05a
commit b3fffab05a
parent 5b4c554445 06652b3110
10 changed files with 125 additions and 32 deletions
--- a/.gitignore
+++ b/.gitignore
@ -1,9 +1,5 @@
 logs/*.log
-*.pycmedia/release/
 *.pyc
-media/*/*-compressed-*.*
-media/css/*/*-compressed-*.*
-media/release
 static/*
 local_settings.py
 media/iphone/NewsBlur/build
@ -12,10 +8,7 @@ build/
 .DS_Store
 **/*.perspectivev*
 data/
-logs
-mongo/
 **/*.xcuserstate
-media/iphone/NewsBlur.xcodeproj/project.xcworkspace/xcuserdata/conesus.xcuserdatad/UserInterfaceState.xcuserstate
 UserInterfaceState.xcuserstate
 UserInterfaceState\.xcuserstate
 *.xcuserstate
@ -23,3 +16,4 @@ xcuserdata
 .xcodeproj/ push.xcodeproj/project.pbxproj
 *.mode1v3
 *.pbxuser
+media/maintenance.html
--- a/apps/rss_feeds/management/commands/task_feeds.py
+++ b/apps/rss_feeds/management/commands/task_feeds.py
@ -46,4 +46,12 @@ class Command(BaseCommand):
            active_subscribers__gte=1,
            active=True
        ).order_by('?')
+        if feeds: Feed.task_feeds(feeds)
+        
+        feeds = Feed.objects.filter(
+            last_update__lte=day,
+            active_subscribers__gte=1,
+            active=False,
+            known_good=True
+        ).order_by('?')
        if feeds: Feed.task_feeds(feeds)
--- a/apps/rss_feeds/migrations/0053_known_good.py
+++ b/apps/rss_feeds/migrations/0053_known_good.py
@ -0,0 +1,83 @@
+# encoding: utf-8
+import datetime
+from south.db import db
+from south.v2 import SchemaMigration
+from django.db import models
+
+class Migration(SchemaMigration):
+
+    def forwards(self, orm):
+        
+        # Adding field 'Feed.known_good'
+        db.add_column('feeds', 'known_good', self.gf('django.db.models.fields.BooleanField')(default=False, db_index=True), keep_default=False)
+
+
+    def backwards(self, orm):
+        
+        # Deleting field 'Feed.known_good'
+        db.delete_column('feeds', 'known_good')
+
+
+    models = {
+        'rss_feeds.duplicatefeed': {
+            'Meta': {'object_name': 'DuplicateFeed'},
+            'duplicate_address': ('django.db.models.fields.CharField', [], {'max_length': '255'}),
+            'duplicate_feed_id': ('django.db.models.fields.CharField', [], {'max_length': '255', 'null': 'True'}),
+            'feed': ('django.db.models.fields.related.ForeignKey', [], {'related_name': "'duplicate_addresses'", 'to': "orm['rss_feeds.Feed']"}),
+            'id': ('django.db.models.fields.AutoField', [], {'primary_key': 'True'})
+        },
+        'rss_feeds.feed': {
+            'Meta': {'ordering': "['feed_title']", 'object_name': 'Feed', 'db_table': "'feeds'"},
+            'active': ('django.db.models.fields.BooleanField', [], {'default': 'True', 'db_index': 'True'}),
+            'active_premium_subscribers': ('django.db.models.fields.IntegerField', [], {'default': '-1', 'db_index': 'True'}),
+            'active_subscribers': ('django.db.models.fields.IntegerField', [], {'default': '-1', 'db_index': 'True'}),
+            'average_stories_per_month': ('django.db.models.fields.IntegerField', [], {'default': '0'}),
+            'branch_from_feed': ('django.db.models.fields.related.ForeignKey', [], {'to': "orm['rss_feeds.Feed']", 'null': 'True', 'blank': 'True'}),
+            'creation': ('django.db.models.fields.DateField', [], {'auto_now_add': 'True', 'blank': 'True'}),
+            'days_to_trim': ('django.db.models.fields.IntegerField', [], {'default': '90'}),
+            'etag': ('django.db.models.fields.CharField', [], {'max_length': '255', 'null': 'True', 'blank': 'True'}),
+            'exception_code': ('django.db.models.fields.IntegerField', [], {'default': '0'}),
+            'favicon_color': ('django.db.models.fields.CharField', [], {'max_length': '6', 'null': 'True', 'blank': 'True'}),
+            'favicon_not_found': ('django.db.models.fields.BooleanField', [], {'default': 'False'}),
+            'feed_address': ('django.db.models.fields.URLField', [], {'max_length': '255'}),
+            'feed_address_locked': ('django.db.models.fields.NullBooleanField', [], {'default': 'False', 'null': 'True', 'blank': 'True'}),
+            'feed_link': ('django.db.models.fields.URLField', [], {'default': "''", 'max_length': '1000', 'null': 'True', 'blank': 'True'}),
+            'feed_link_locked': ('django.db.models.fields.BooleanField', [], {'default': 'False'}),
+            'feed_title': ('django.db.models.fields.CharField', [], {'default': "'[Untitled]'", 'max_length': '255', 'null': 'True', 'blank': 'True'}),
+            'fetched_once': ('django.db.models.fields.BooleanField', [], {'default': 'False'}),
+            'has_feed_exception': ('django.db.models.fields.BooleanField', [], {'default': 'False', 'db_index': 'True'}),
+            'has_page': ('django.db.models.fields.BooleanField', [], {'default': 'True'}),
+            'has_page_exception': ('django.db.models.fields.BooleanField', [], {'default': 'False', 'db_index': 'True'}),
+            'hash_address_and_link': ('django.db.models.fields.CharField', [], {'unique': 'True', 'max_length': '64', 'db_index': 'True'}),
+            'id': ('django.db.models.fields.AutoField', [], {'primary_key': 'True'}),
+            'known_good': ('django.db.models.fields.BooleanField', [], {'default': 'False', 'db_index': 'True'}),
+            'last_load_time': ('django.db.models.fields.IntegerField', [], {'default': '0'}),
+            'last_modified': ('django.db.models.fields.DateTimeField', [], {'null': 'True', 'blank': 'True'}),
+            'last_update': ('django.db.models.fields.DateTimeField', [], {'db_index': 'True'}),
+            'min_to_decay': ('django.db.models.fields.IntegerField', [], {'default': '0'}),
+            'next_scheduled_update': ('django.db.models.fields.DateTimeField', [], {'db_index': 'True'}),
+            'num_subscribers': ('django.db.models.fields.IntegerField', [], {'default': '-1'}),
+            'premium_subscribers': ('django.db.models.fields.IntegerField', [], {'default': '-1'}),
+            'queued_date': ('django.db.models.fields.DateTimeField', [], {'db_index': 'True'}),
+            'stories_last_month': ('django.db.models.fields.IntegerField', [], {'default': '0'})
+        },
+        'rss_feeds.feeddata': {
+            'Meta': {'object_name': 'FeedData'},
+            'feed': ('utils.fields.AutoOneToOneField', [], {'related_name': "'data'", 'unique': 'True', 'to': "orm['rss_feeds.Feed']"}),
+            'feed_classifier_counts': ('django.db.models.fields.TextField', [], {'null': 'True', 'blank': 'True'}),
+            'feed_tagline': ('django.db.models.fields.CharField', [], {'max_length': '1024', 'null': 'True', 'blank': 'True'}),
+            'id': ('django.db.models.fields.AutoField', [], {'primary_key': 'True'}),
+            'popular_authors': ('django.db.models.fields.CharField', [], {'max_length': '2048', 'null': 'True', 'blank': 'True'}),
+            'popular_tags': ('django.db.models.fields.CharField', [], {'max_length': '1024', 'null': 'True', 'blank': 'True'}),
+            'story_count_history': ('django.db.models.fields.TextField', [], {'null': 'True', 'blank': 'True'})
+        },
+        'rss_feeds.feedloadtime': {
+            'Meta': {'object_name': 'FeedLoadtime'},
+            'date_accessed': ('django.db.models.fields.DateTimeField', [], {'auto_now': 'True', 'blank': 'True'}),
+            'feed': ('django.db.models.fields.related.ForeignKey', [], {'to': "orm['rss_feeds.Feed']"}),
+            'id': ('django.db.models.fields.AutoField', [], {'primary_key': 'True'}),
+            'loadtime': ('django.db.models.fields.FloatField', [], {})
+        }
+    }
+
+    complete_apps = ['rss_feeds']
--- a/apps/rss_feeds/models.py
+++ b/apps/rss_feeds/models.py
@ -52,6 +52,7 @@ class Feed(models.Model):
    branch_from_feed = models.ForeignKey('Feed', blank=True, null=True, db_index=True)
    last_update = models.DateTimeField(db_index=True)
    fetched_once = models.BooleanField(default=False)
+    known_good = models.BooleanField(default=False, db_index=True)
    has_feed_exception = models.BooleanField(default=False, db_index=True)
    has_page_exception = models.BooleanField(default=False, db_index=True)
    has_page = models.BooleanField(default=True)
@ -1013,7 +1014,7 @@ class Feed(models.Model):
        return total, random_factor*2
        
    def set_next_scheduled_update(self):
-        total, random_factor = self.get_next_scheduled_update(force=True)
+        total, random_factor = self.get_next_scheduled_update(force=True, verbose=False)
        
        next_scheduled_update = datetime.datetime.utcnow() + datetime.timedelta(
                                minutes = total + random_factor)
--- a/apps/statistics/models.py
+++ b/apps/statistics/models.py
@ -52,17 +52,20 @@ class MStatistics(mongo.Document):
    def collect_statistics_feeds_fetched(cls, last_day=None):
        if not last_day:
            last_day = datetime.datetime.now() - datetime.timedelta(hours=24)
+        last_biweek = datetime.datetime.now() - datetime.timedelta(days=14)
        
-        feeds_fetched = MFeedFetchHistory.objects.count()
+        feeds_fetched = MFeedFetchHistory.objects.filter(fetch_date__lt=last_day).count()
        cls.objects(key='feeds_fetched').update_one(upsert=True, key='feeds_fetched', value=feeds_fetched)
-        pages_fetched = MPageFetchHistory.objects.count()
+        pages_fetched = MPageFetchHistory.objects.filter(fetch_date__lt=last_day).count()
        cls.objects(key='pages_fetched').update_one(upsert=True, key='pages_fetched', value=pages_fetched)
        
        from utils.feed_functions import timelimit, TimeoutError
        @timelimit(60)
        def delete_old_history():
-            MFeedFetchHistory.objects(fetch_date__lt=last_day).delete()
-            MPageFetchHistory.objects(fetch_date__lt=last_day).delete()
+            MFeedFetchHistory.objects(fetch_date__lt=last_day, status_code__in=[200, 304]).delete()
+            MPageFetchHistory.objects(fetch_date__lt=last_day, status_code__in=[200, 304]).delete()
+            MFeedFetchHistory.objects(fetch_date__lt=last_biweek).delete()
+            MPageFetchHistory.objects(fetch_date__lt=last_biweek).delete()
        try:
            delete_old_history()
        except TimeoutError:
--- a/media/css/reader.css
+++ b/media/css/reader.css
@ -3785,6 +3785,7 @@ background: transparent;
 .NB-module h5 {
    margin: 0 0 12px;
    padding: 8px 12px 6px;
+    overflow: hidden;
 }

 .NB-module .NB-module-header-left {
--- a/templates/static/press.xhtml
+++ b/templates/static/press.xhtml
@ -132,6 +132,10 @@
            Mar 8, 2011
          </span>
        </li>
+        <li>
+          <a href="http://www.bloggingot.com/blogging-tools/newsblur-rss-feed-client-for-feed-junkies/">NewsBlur: RSS Feed Client for Feed Junkies</a>
+          <span class="NB-press-publisher"><img src="http://www.bloggingot.com/favicon.ico"> Blogging OT</span>, <span class="NB-press-author">Panah</span>, <span class="NB-press-date">Jan 21, 2012</span>
+        </li>
        <li>
          <a href="http://www.genbeta.com/web/newsblur-una-excelente-alternativa-a-google-reader-que-filtra-los-posts-mas-relevantes">
            NewsBlur, una excelente alternativa a Google Reader que filtra los posts más relevantes
--- a/utils/feed_fetcher.py
+++ b/utils/feed_fetcher.py
@ -1,8 +1,13 @@
-# from apps.rss_feeds.models import FeedXML
+import time
+import datetime
+import traceback
+import multiprocessing
+import urllib2
+import xml.sax
+import redis
 from django.core.cache import cache
 from django.conf import settings
 from django.db import IntegrityError
-# from mongoengine.queryset import Q
 from apps.reader.models import UserSubscription, MUserStory
 from apps.rss_feeds.models import Feed, MStory
 from apps.rss_feeds.page_importer import PageImporter
@ -11,18 +16,10 @@ from utils import feedparser
 from utils.story_functions import pre_process_story
 from utils import log as logging
 from utils.feed_functions import timelimit, TimeoutError, mail_feed_error_to_admin, utf8encode
-import time
-import datetime
-import traceback
-import multiprocessing
-import urllib2
-import xml.sax
-import redis

 # Refresh feed code adapted from Feedjack.
 # http://feedjack.googlecode.com

-SLOWFEED_WARNING = 10
 ENTRY_NEW, ENTRY_UPDATED, ENTRY_SAME, ENTRY_ERR = range(4)
 FEED_OK, FEED_SAME, FEED_ERRPARSE, FEED_ERRHTTP, FEED_ERREXC = range(5)

@ -132,16 +129,16 @@ class ProcessFeed:
                if not self.feed.fetched_once:
                    self.feed.has_feed_exception = True
                    self.feed.fetched_once = True
+                    logging.debug("   ---> [%-30s] Feed is 302'ing, but it's not new. Refetching..." % (unicode(self.feed)[:30]))
                    self.feed.schedule_feed_fetch_immediately()
                if not self.fpf.entries:
                    self.feed.save()
                    self.feed.save_feed_history(self.fpf.status, "HTTP Redirect")
                    return FEED_ERRHTTP, ret_values
-                
            if self.fpf.status >= 400:
-                logging.debug("   ---> [%-30s] HTTP Status code: %s.%s Checking address..." % (unicode(self.feed)[:30], self.fpf.status, ' Not' if self.feed.fetched_once else ''))
+                logging.debug("   ---> [%-30s] HTTP Status code: %s.%s Checking address..." % (unicode(self.feed)[:30], self.fpf.status, ' Not' if self.feed.known_good else ''))
                fixed_feed = None
-                if not self.feed.fetched_once:
+                if not self.feed.known_good:
                    fixed_feed = self.feed.check_feed_link_for_feed_address()
                if not fixed_feed:
                    self.feed.save_feed_history(self.fpf.status, "HTTP Error")
@ -152,10 +149,10 @@ class ProcessFeed:
                return FEED_ERRHTTP, ret_values
                                    
        if self.fpf.bozo and isinstance(self.fpf.bozo_exception, feedparser.NonXMLContentType):
-            logging.debug("   ---> [%-30s] Feed is Non-XML. %s entries.%s Checking address..." % (unicode(self.feed)[:30], len(self.fpf.entries), ' Not' if self.fpf.entries else ''))
+            logging.debug("   ---> [%-30s] Feed is Non-XML. %s entries.%s Checking address..." % (unicode(self.feed)[:30], len(self.fpf.entries), ' Not' if self.feed.known_good and self.fpf.entries else ''))
            if not self.fpf.entries:
                fixed_feed = None
-                if not self.feed.fetched_once:
+                if not self.feed.known_good:
                    fixed_feed = self.feed.check_feed_link_for_feed_address()
                if not fixed_feed:
                    self.feed.save_feed_history(502, 'Non-xml feed', self.fpf.bozo_exception)
@ -168,7 +165,7 @@ class ProcessFeed:
            logging.debug("   ---> [%-30s] Feed has SAX/XML parsing issues. %s entries.%s Checking address..." % (unicode(self.feed)[:30], len(self.fpf.entries), ' Not' if self.fpf.entries else ''))
            if not self.fpf.entries:
                fixed_feed = None
-                if not self.feed.fetched_once:
+                if not self.feed.known_good:
                    fixed_feed = self.feed.check_feed_link_for_feed_address()
                if not fixed_feed:
                    self.feed.save_feed_history(503, 'SAX Exception', self.fpf.bozo_exception)
@ -306,9 +303,9 @@ class Dispatcher:
                    
                    feed = self.refresh_feed(feed_id)
                    
-                    if ret_entries.get(ENTRY_NEW) or self.options['force'] or not feed.fetched_once:
-                        if not feed.fetched_once:
-                            feed.fetched_once = True
+                    if ret_entries.get(ENTRY_NEW) or self.options['force']:
+                        if not feed.known_good:
+                            feed.known_good = True
                            feed.save()
                        MUserStory.delete_old_stories(feed_id=feed.pk)
                        try:
--- a/utils/munin/newsblur_feed_counts.py
+++ b/utils/munin/newsblur_feed_counts.py
@ -10,6 +10,7 @@ graph_config = {
    'inactive_feeds.label': 'inactive_feeds',
    'duplicate_feeds.label': 'duplicate_feeds',
    'active_feeds.label': 'active_feeds',
+    'known_good_feeds.label': 'known_good',
 }
 def calculate_metrics():
    from apps.rss_feeds.models import Feed, DuplicateFeed
@ -20,6 +21,7 @@ def calculate_metrics():
        'inactive_feeds': Feed.objects.filter(active=False).count(),
        'duplicate_feeds': DuplicateFeed.objects.count(),
        'active_feeds': Feed.objects.filter(active_subscribers__gt=0).count(),
+        'known_good_feeds': Feed.objects.filter(known_good=True).count(),
    }

 if __name__ == '__main__':
--- a/utils/munin/newsblur_updates.py
+++ b/utils/munin/newsblur_updates.py
@ -18,7 +18,7 @@ def calculate_metrics():
    from apps.rss_feeds.models import Feed
    
    hour_ago = datetime.datetime.utcnow() - datetime.timedelta(hours=1)
-    update_feeds_query = "ssh sclay@db01 \"sudo rabbitmqctl list_queues -p newsblurvhost | grep %s\" | awk '{print $2}'"
+    update_feeds_query = "ssh -i ~sclay/.ssh/id_dsa sclay@db01 \"sudo rabbitmqctl list_queues -p newsblurvhost | grep %s\" | awk '{print $2}'"
    
    return {
        'update_queue': Feed.objects.filter(queued_date__gte=hour_ago).count(),