Bug 133237

SVN commit 580572 by osterfeld:

make article selectable when the number of items in the list is reduced to one
BUG: 133237


 M  +11 -0     articlelistview.cpp  


--- trunk/KDE/kdepim/akregator/src/articlelistview.cpp #580571:580572
@@ -420,6 +420,10 @@
         setSelected(next, true);
         setCurrentItem(next);
     }
+    else
+    {
+        d->noneSelected = true;
+    }
 
     setUpdatesEnabled(true);
     triggerUpdate();
@@ -461,6 +465,10 @@
         setSelected(next, true);
         setCurrentItem(next);
     }
+    else
+    {
+        d->noneSelected = true;
+    }
 
     setUpdatesEnabled(true);
     triggerUpdate();
@@ -700,7 +708,10 @@
     if (ai)
         emit signalArticleChosen( ai->article() );
     else
+    {
+        d->noneSelected = true;
         emit signalArticleChosen( Article() );
+    }
 }
 
 

SVN commit 580574 by osterfeld:

backport: make single items in the list always selectable (i.e. emit signal)
CCBUG: 133237


 M  +12 -1     articlelistview.cpp  


--- branches/KDE/3.5/kdepim/akregator/src/articlelistview.cpp #580573:580574
@@ -424,6 +424,10 @@
         setSelected(next, true);
         setCurrentItem(next);
     }
+    else
+    {
+        d->noneSelected = true;
+    }
 
     setUpdatesEnabled(true);
     triggerUpdate();
@@ -465,6 +469,10 @@
         setSelected(next, true);
         setCurrentItem(next);
     }
+    else
+    {
+        d->noneSelected = true;
+    }
 
     setUpdatesEnabled(true);
     triggerUpdate();
@@ -745,8 +753,11 @@
     ArticleItem* ai = dynamic_cast<ArticleItem*> (item);
     if (ai)
         emit signalArticleChosen( ai->article() );
-    else 
+    else
+    {
+        d->noneSelected = true;
         emit signalArticleChosen( Article() );
+    }
 } 
 
 

Hrm, wrong bug number closed. Reopening.

Hmm, I can't see how this can happen - both switching nodes and updating the displayed node uses the same code.
Is there any pattern which article is left out? Like the very newest article, or the oldest of the new articles?

Ok, I can reproduce it now. It seems always the last article (the oldest is omitted)

I'm not sure if this is related at all but the total unread count also is sometimes 1 less than the actual count. it seems to get more accurate or something with the more unread you have.
such as yesterday I noticed All Feeds said I had 8 unread
my KDE section said I had 6 and another had 1 unread.
6+1 != 8
right now tho I have 121 unread and that is actually correct.

3.5.5 follow-up
still one article missing but there is another problem that might be related.

Under All Feeds the count there is almost always 1 off now too.
for example right now I see:
All Feeds (1)
But in a feed it has 2 unread.

Created attachment 21177 [details]
screenshot showing akregator and its systray icon with a bad feeds count.

I have a similar bug here.

The total count is 1 less than the sum of all sub-counts. It happens when I
let my computer on for some time (more than 1 week). If I stop and start
akgregator (well, more precisely kontact), the problem disappear.
The screenshot shows it clearly with only 1 sub-count as all other are already
read. It also happens when there are several other feeds not read.

The systray icon shows also the bad count.

I also experience wrong counts in the treeview (even negative counters!), but the original post was about items not visible. Is that still happening for you with a newer version of Akregator?

Looks to be working fine now so I'm going to close my own bug thanks!