Error 6 Overflow

[jupe]

Not sure if this info is of any use to you, but I have been getting this error quite regularly since .0515, it is quite persistent, to get it to go away, you have to close the dialog at least once per file visible in the list, sometimes many, many, times per file to be able to recover back to a useable UI. Between dismissing the dialog and the next dialog reappearing, the icon and column text appears and disappears (Details view), unsure if that hints at anything. Among other things, SC refreshlist seems to trigger it.

Code: Select all

---------------------------
XYplorer 26.70.0515
---------------------------
Error:  	6 (0x00000006)
Desc:   	Overflow
Dll:    	0
Proc:   	pPtTx@250

idxCol=2, idxPos=0, idxNode=28, fMultiLine=False

Source: 	XYplorer
XY ver: 	XYplorer 26.70.0515 - Lifetime License Enterprise
OS:     	Windows 10 Pro, 64-bit, Version 22H2, Build 19045.5247 (10.0)
Locale: 	1033 (en-US)
ANSI:   	1252, ACP: 1252  (ANSI - Latin I)
Font:   	Segoe UI 11, Segoe UI 9, DBCS: No
DPI:    	120 (125%), 1920x1080

Date:   	2025-01-11 10:14:29

Copy to clipboard?
---------------------------
Yes   No   
---------------------------

2025-01-14_101713.png

[admin]

Can't make it happen here, but I can say it has to do with highlights in the list (live filter, spot...).

PS: There will be a better error message in the next beta.

[jupe]

Now that you have mentioned that it is related to Spot, I have investigated, and it seems it could be because the return from <get spotpatterns> has per pattern trailing extraneous info in it now, eg;

goto ">>X|Y|Z"; e <get spotpatterns>;

I was using the return as is with complex patterns, so I think that extra info could possibly cause the error. In any case the change that caused the error must have happened about .0515 or a few betas previous.

[admin]

I can't reproduce the error, and e <get spotpatterns>; does not return anything different in 0500. But it does return something different in 24.40.0100. Yes, I know what I'll do tomorrow.

[jupe]

Nothing different?

goto ">>X|Y|Z"; msg <get spotpatterns>;

2025-01-15_130752.png

[admin]

Yeah, that's what I saw too, but before 0500. You said "0515 or a few betas previous".

[jupe]

Oh, I didn't notice you'd edited your last post to say you saw it. I usually don't update immediately, so I guess the cause could be a few weeks old, but the error I have only seen within the last week, possibly it's related to the filenames I am using currently,

[admin]

Oh, I edited it very quickly after I had the idea to test earlier versions. I didn't think you'd seen the first version of the post.

[jupe]

It was sporadic, but I haven't seen the error for 2 days now, so seems resolved.

Not that it matters, but the date you mentioned the issue being since in the changelog has to be wrong, my screenshot above shows that in 26.70.0301 it was ok, and that is only 4 weeks old, so the <get> return changed within that timeframe max.

[admin]

Yes, you're right. The correct date is 20241213. I'll patch the change log.

[jupe]

Previous spot pattern report definitely still seems resolved.

But I have a new one, any idea what this is in relation to?

2025-01-29_141824.png

[admin]

No. You have any context?

[jupe]

Not really, it's only happened once (in my main conf), I think it was on a location change, since the last dialog suggest script, it made me think it could have been triggered by a cea browse, I haven't made any changes to those though, not even minor since at least Nov, going by file timestamp.
What happened is I got the first error like 30 times, then the 2nd like 5 times then the last one once. If pPaiCap@10 doesn't link to any feature I can investigate, we can just assume it was probably just a freak occurrence.

[admin]

pPaiCap@10 is when the rectangle (focus, selection) of an item in the list is calculated. Not something where bad things are likely to happen.
Since Count and CountAll are different, it's probably related to live filters.

Generally, Error 6 Overflow is one of those "meta" errors, where something broke so hard, that even the errors themselves broke and false errors are reported.