I was looking back at my log from my activation of GM/ES-006 Ben Avon on 26th August 2022.
The database and sotl.as show 10 QSO’s on 2 m FM. sotamaps/activations also tells me I made 10 QSO’s but only 8 are shown in the table.
Is there a reason for this discrepancy, or is it user error?
Hi Fraser - when I look, I can see 9 entries in the table
What I find strange is that (without me doing any sorting of the columns) that last entry is out of sequence regarding time AND the missing entry for 2M0WTN took place at the same time as the “shifted to the bottom of the list” entry:
Was this log loaded in two parts?
If no one comes back with a better explanation, I’d be tempted to delete the log and re-upload or re-enter it.
73 Ed.
The order of data in the database is the order it was entered. If you enter manually you can enter activation QSOs in any order you like. CSV etc. requires ascending time order. When the data is fetched, the data normally comes back in the order it was entered. However, it is not guaranteed to come back in any order. What happens next is the DB web app displays data with specific ordering, i.e. sorted on ascending time so what you see displayed is not the same as the order it came from the DB. It would seem that Sotamaps is not applying any specific ordering other than how it gets it from the DB.
I was using vk port-a-log and it would have time stamped the QSO’s for me. I’ve since deleted the file from my phone.
Strange how there are 10, Ed can see 9 and I can only see 8!
I see 9
Hi Fraser,
Before you delete your log take a look on Sotl.as but I think you deleted this log !
Always I check both and always Sotl.as win with exact match 
73, Éric
F5JKK
Hi Eric, I can still see my log in the database:
https://www.sotadata.org.uk/en/log/activator/show/682342
and on sotl.as
Yes SMP for me is not sure
Hi Fraser,
As noted by @DD5LP and @M0VAZ, I can see in SMP 9 QSOs as well:
Seems like for some reason the ones at 15:52 are either both missing (for you) or only one is returned (SQL DISTINCT?).
73 Stephan
Just seen this thread. OK, looking at the code, I see that, because the SQL code statements on the back-end (i.e. on the SOTA server) for non-S2S and for S2S are SO different (because they access different database tables), I split the query into two, and get each part separately, each properly sorted on datetime, and add them both into an array. At the end of that, I have NOT sorted the final array, leaving each half sorted independently from each other … which is NOT good (and should be fixed, see below) but not necessarily a deal-breaker because there is sorting implemented in the browser-based Javascript code to handle sorting.
However, and unfortunately, the Javascript code will only sort the data returned from the server if there are 10 or more records in the QSOs list data. And in this case, there are, as we can see, only 9 records returned. Why on earth a limit of 10 ?? !!! … well, it’s tied to the table-sorting algorithms and to a (mostly) fixed height of the table which kind of guarantees (but not fully, since we’re dealing with HTML here) that no blank rows would also end up being included in the table sorting. It’s weak, I know, but when one is dealing with tens of thousands of lines of code (at the moment, the SMP has around 90,000 lines of hand-written code, excluding ~20,000 lines of essential internal documentation) it’s sometimes easy to overlook something like this kind of special case.
This is of course, a bug, and not at all a satisfactory situation, but it can be resolved if Andrew @VK3ARR can see his way to adding a few lines to the back-end code to sort the final array of data just before it’s returned to the Javascript calling code. I’ll PM Andrew and, if he so wishes, help him to execute any required changes.
Sorry for this error, but there is a fix,
Rob
One week later, and I have to report that, to the best of my knowledge, this problem is not going to be fixed. I myself have no access to the code on the server, and we must rely on the goodwill of those responsible for maintaining the SMP code to attend to such issues as they arise.
I’ve had good relations with Andrew VK3ARR in the past, with the odd hiccup to enliven our discussions. On this occasion, however, I have to report that I’ve received no reply at all from Andrew, despite my having linked to him directly in my above post, and I’ve also not received a reply to a Reflector PM I wrote him a week ago - not even a “I’m busy, I’ll get back to you” .
I know, of course, that the SMP is soon to be replaced, so Andrew’s not replying can probably be read as “don’t waste my time with asking for a sinking ship to be patched.” Well, I did try…
Well this is a disappointing message to come back to.
For the record, my last three weeks have been either:
In short, I’ve spent about 15 minutes thinking about SOTA in the past three or four weeks, most of which was “I must get back to that email/fix/update/whatever”. I acknowledge your PM. I will get to it, but it won’t be this week. My work calendar is saying I have 42.8 hours of meetings accepted this week already, but that excludes Friday which I’m off for a friend’s wedding, so you can do the maths - it’s not going to pretty this week either.
I will be discussing the need for additional management coverage underneath me with my boss this Wednesday, but in the meantime, if anyone can tell me what timezone my brain is currently in, it’ll be much appreciated.
Hi Andrew,
I guessed as much - since your having been promoted some while ago at work to [VP/Senior Manager/whatever] I have understood your time for SOTA to be extremely limited. No problems, take your time, and it’s understood that this particular glitch comes very far down the list of things to be attended to.
For my part, I’m relieved that nothing untoward has happened to you, and I’m sure I speak for everybody in this regard.
Thanks for getting back to this thread, it’s much appreciated.
Cheers, Rob
I pushed a fix just now based on code provided by @DM1CM . Testing welcome, but looks like it solves the problem to me.
