Recordings View is blank

[wnjj]

Ok. I had some time.

I updated the file in the same location as before: https://www.dropbox.com/s/h761q3pbbz...ring_sort?dl=0

Here's the new version:

          public int compare(Object o1, Object o2)
          {
            Show s1 = getShowObj(o1);
            Show s2 = getShowObj(o2);
            long l1 = (s1 == null) ? 0 : s1.getOriginalAirDate();
            long l2 = (s2 == null) ? 0 : s2.getOriginalAirDate();
            if (l1 < l2)
              return -1;
            else if (l1 > l2)
              return 1;
            if (s1 != null && s2 != null)
            {
              int n1 = s1.getSeasonNumber();
              int n2 = s2.getSeasonNumber();
              if (n1 > n2)  // larger season is older/earlier show
                return -1;
              else if (n1 < n2)
                return 1;
              n1 = s1.getEpisodeNumber();
              n2 = s2.getEpisodeNumber();
              if (n1 < n2)
                return -1;
              else if (n1 > n2)
                return 1;
            }
            Airing a1 = getAirObj(o1);
            Airing a2 = getAirObj(o2);
            l1 = (a1 == null) ? 0 : a1.getStartTime();
            l2 = (a2 == null) ? 0 : a2.getStartTime();
            return (l1 < l2) ? -1 : ((l1 > l2) ? 1 : 0);
          }

[wnjj]

I don't think the negative or large numbers were the issue. I now suspect it's shows that share an original airing date but have mixed availability of season and episode numbers.

I wonder if imported shows either don't have original airing dates or they get populated with the same date when imported. That would make my failing scenario occur far more often.

[jusjoken]

I dont follow your season logic... a larger season number is a Newer show. An older show could be the same season or a smaller season number.

?

[gary201]

Originally Posted by wnjj

Ok. I had some time.

I updated the file in the same location as before: https://www.dropbox.com/s/h761q3pbbz...ring_sort?dl=0

I downloaded the update you posted above and swapped it in a few minutes ago. It seems to be working. I'll report back if it fails again.

[wnjj]

Originally Posted by jusjoken

I dont follow your season logic... a larger season number is a Newer show. An older show could be the same season or a smaller season number.

?

Remember that season numbers are only compared if the original airing dates are the same. So my reasoning is that if 2 shows originslly aired on the same date, the older one is the one in its second season compared to the one in its first season (i.e. the series itself has been around longer). This logic only works if the shows are not in the same series.

If the shows are in the same series, then what you’re saying is true but then they should have really different original airing dates, shouldn’t they? I suppose the one place that’s tough to consider is when the original airing dates are both 0, they may or may not be from the same series.

In the end, it’s probably not a big enough deal to worry about and easiest to leave it as it was.

[jusjoken]

Originally Posted by wnjj

Remember that season numbers are only compared if the original airing dates are the same. So my reasoning is that if 2 shows originslly aired on the same date, the older one is the one in its second season compared to the one in its first season (i.e. the series itself has been around longer). This logic only works if the shows are not in the same series.

If the shows are in the same series, then what you’re saying is true but then they should have really different original airing dates, shouldn’t they? I suppose the one place that’s tough to consider is when the original airing dates are both 0, they may or may not be from the same series.

In the end, it’s probably not a big enough deal to worry about and easiest to leave it as it was.

Ok. I get your logic now. I often see bad or missing OAD so just need to plan for that as you say.

Great sleuthing BTW

k

[wnjj]

Originally Posted by jusjoken

Ok. I get your logic now. I often see bad or missing OAD so just need to plan for that as you say.

Great sleuthing BTW

k

Yeah, missing data makes sorting any of this tough but by definition the method doesn't matter if enough data is missing.

Thanks. I'm hoping this time it works. The last one was only a guess since I couldn't find anything else to blame. The multiple conditional compare logic makes it tricky.

[gary201]

I assume there isn't anything "confidential" in the wiz.bin (other than just all the stuff that was recorded/watched)? If not, you're welcome to it if you think that will help. Otherwise, we just wait and see since the problem seemed to appear/disappear periodically.

[wnjj]

Originally Posted by gary201

I assume there isn't anything "confidential" in the wiz.bin (other than just all the stuff that was recorded/watched)? If not, you're welcome to it if you think that will help. Otherwise, we just wait and see since the problem seemed to appear/disappear periodically.

It's your watched history and all of the photo metadata. I'm not sure what else to look for in there, other than building some kind of script to search the data for the situation I outlined. That could be a large task though. I think waiting for the next round of results is easiest.

[gary201]

Originally Posted by wnjj

I think waiting for the next round of results is easiest.

Okay. So far, it's still working.

[wnjj]

Originally Posted by gary201

Okay. So far, it's still working.

Still good on your end? Has it been long enough that you think it would have surfaced otherwise? If it looks fixed I'll try to get the changes submitted along with a different one I'm already doing.

[gary201]

Originally Posted by wnjj

Still good on your end? Has it been long enough that you think it would have surfaced otherwise? If it looks fixed I'll try to get the changes submitted along with a different one I'm already doing.

You bet! The problem hasn't returned at all since the last change. I can't thank you enough.

[wnjj]

Originally Posted by gary201

You bet! The problem hasn't returned at all since the last change. I can't thank you enough.

Good deal. I’ll sibmit the changes when I get the 64-bit Windows stuff submitted so you won’t have to stay on a temporary jar file.

[JustFred]

Originally Posted by wnjj

Good deal. I’ll sibmit the changes when I get the 64-bit Windows stuff submitted so you won’t have to stay on a temporary jar file.

Please submit the changes for this fix independent from the 64-bit stuff. Makes it much easier to track changes and regression issues. "One submit for one bug" is the mantra.

[wnjj]

Originally Posted by JustFred

Please submit the changes for this fix independent from the 64-bit stuff. Makes it much easier to track changes and regression issues. "One submit for one bug" is the mantra.

Will do. I guess what I meant was I'll do it when I get setup to do those. I haven't signed the open source agreement or ever submitted a pull request yet. I'll just submit them sequentially but separately. There's also another null pointer fix I have.

[KryptoNyte]

Originally Posted by wnjj

... when I get the 64-bit Windows stuff submitted ...

I can't tell you how happy this makes me to see a comment like that. Thought it might never come.

[wnjj]

This fix has been pulled into the main code so any future builds will include it.

[gary201]

Last night after watching/deleting a recording, the problem returned. I'm running v9.1.10.479. Is it possible that the fix was overwritten (and backed out) by someone else?

[Edit] Noticed the problem was intermittent as it did appear this morning, but after another show/two was watched/deleted the problem cleared up. As I recall, the issue was intermittent previously as well. I then upgraded to v9.2.1.485, so we'll see if the problem reappears again.

[wnjj]

Grab the sagetv_0.txt log file if it happens again. The code should still be there but there's always the possibility of another issue