David WE Roberts posted on Mon, 28 Jan 2013 16:02:06 +0000 as excerpted: > Yep - just tested a bit further. > For each numeric score, when sorted on score, the threads are sorted in > date order oldest first. > > So you get score 9999 - oldest to youngest Followed by score 100 - > oldest to youngest Followed by score 0 - oldest to youngest. > etc. > > So not an unstable sort but a sort policy which I don't fully > understand.
What happens if you click the score column again, to reverse the order? I said it before. It's using primary and secondary sort keys, the secondary one being date as that was the last sort before score. But there's just one order, so if you have scores sorted one way, the secondary key (date in this case) will be sorted the same way. You have score high at the top, which matches oldest date at the top as the secondary key. Reverse the sort and you should have score low at the top, with youngest date within that score first. What is there left to not fully understand? All that said, one /could/ argue that the earliest date should be the smallest, and corresponding to the smallest score. However, IMO that's simply arbitrary. It's ALSO possible that it's determined by the widget library's list sort function, *NOT* pan, which would make changing it in pan difficult as it's an upstream thing, but I don't know that. -- Duncan - List replies preferred. No HTML msgs. "Every nonfree program has a lord, a master -- and if you use the program, he is your master." Richard Stallman _______________________________________________ Pan-users mailing list Pan-users@nongnu.org https://lists.nongnu.org/mailman/listinfo/pan-users
