LiveCode Forums.

I am getting an eternal beach ball in a stack, but with no obvious culprit in the scripts. I have tried deleting the prefs (livecode7.rev), and a new pref is generated, but it has not helped.

I have opened the message watcher to see if I can track down the last handler before the screen seizure, but there is no consistent "last message".

Any techniques I could use to pin this down?

LC version? OS version?

So sorry:

Mojave 10.14.2
LC: latest stable 9.0.2

I sometimes get that when things get stuck in a LC IDE script. In that case there's not much you can do. Does it happen in LC 9.0.1?

Let me try 9.0.1 and see what happens. Thanks Jacque.

Hi,

this answer is for an old message but lot of people will read it for the first time. I do have that kind of problem often.

It happened to me last time when I wanted to set a stack as a substack of a stack that already had a substack. This is not permitted.

I would try opening the stack with the lockmessages to true and edit script of it to run it line by line. Maybe from a button from another stack. Maybe it could work from message box with multiline but such a button is often useful. Put it in a stack "MyPersonalTools". Then I would trace with step into. Take a screenshot after each step. Keep the variables visible.


Script of btn 1 of cd 1 of stack "MyPersonalTools"
on mouseup
set lockmessages to true --or simply --lock messages
open stack "offendingStackorwatherveritsname" --or get it's name with ask file and open it after
edit script of stack "offendingStackorwatherveritsname"
set lockmessages to false -- or -- unlock messages
end mouseup

on openstack -- if this is too late to do the bug, try on preopenstack after starting over. Also, check if there are frontscripts.
-- put red dot here (alternately a breakpoint instruction), clic Apply, click play button above, choose the handler if offered
-- step into or step over (but if it bugs in a sub, you will have to start again and step into)
-- if it passes the test, they try all the other handlers
end openstack


If you find my answer useful, tell me. Or if you need more support. I for sure will need some also.

I followed the History of Revolution, LiveCode, community edition and participated in the funding of that edition and the recent help call.


It seem that this phrase was not permitted: " I am a veteran of HC (which had very few bugs!)." when the full name of HC was written.
I try again "Hypercard" It was the 2 dot 0 the problem! It does not pass the url test.

ChristianBC wrote: ↑

Thu Jul 30, 2020 6:19 am

It happened to me last time when I wanted to set a stack as a substack of a stack that already had a substack. This is not permitted.

I think you must have phrased that incorrectly, any stack can have many substacks, and any stack can be set as the substack of a stack that already has a substack either through code or the editor.

I should add from the dictionary -

Caution! You *can* move a substack from one main stack to another by setting the substacks property of the destination stack. However, since this may cause a conflict if more than one substack with the same name is open or in memory, it is safer to move a substack by setting the substack's mainStack property.

Bogs.

I think the OP meant that he tried to set a substack to an existing substack.

Anyway, i get this now and then. The SE is always open, though I cannot remember if I am working in it or the stack itself. It just happens. I have learned, sometimes over and over, to save often.

We have spoken of this for years.

HC NEVER crashed. Of course, it was a rather smaller entity..

Craig

Let me try again: If a stack has a substack (child), that "parent" stack cannot be itself set as a substack of another stack ("grandparent" stack?).

I tried to change the mainstack property of a stack which had itself a substack (in a script) and an error message said It could not do that. (Hint:has a substack?)

Maybe this should be elsewhere... this is more specific than the subject.

ChristianBC wrote: ↑

Thu Jul 30, 2020 7:47 pm

Let me try again: If a stack has a substack (child), that "parent" stack cannot be itself set as a substack of another stack ("grandparent" stack?).

Exactly!

The second way you phrased it certainly makes more sense. Just out of curiosity, are you saying that back in the day, you *could* make a stack that had a sub-stack a sub-stack to a completely different main-stack, or are you simply saying this kind of thing didn't crash Hc ?

BTW, congrats on reaching the 'magic number' of posts, you can now go wild posting links and like that

ChristianBC wrote: ↑

Thu Jul 30, 2020 7:47 pm

Let me try again: If a stack has a substack (child), that "parent" stack cannot be itself set as a substack of another stack ("grandparent" stack?).

I tried to change the mainstack property of a stack which had itself a substack (in a script) and an error message said It could not do that. (Hint:has a substack?)

Yes, when attempting to reproduce the hang here I get the same error message. But I do not get a hang. What steps would I need to perform to see the hang?

Maybe this should be elsewhere... this is more specific than the subject.

TBH I'm tired of moving posts where they belong right now, so I'm going to be lazy and leave these messages where they are.

Going forward please make any reasonable attempt to look at the list of categories when starting a new thread, and please do indeed start a new thread for a new topic. Future generations of readers will thank you. And this tired moderator will thank you too.

Yeah, when I will come to a subject about stacks, substacks and files handling, I will surely bring this back.

So don't bother moving it!

Just out of curiosity, are you saying that back in the day, you *could* make a stack that had a sub-stack a sub-stack to a completely different main-stack, or are you simply saying this kind of thing didn't crash Hc ?

HC didn't have substacks at all.

So I guess the latter, however, if Hc didn't have substacks, I don't think the comparison floats well.

In comparing older vers. of Lc to newer ones, though, I did find one interesting difference (which might be considered a bug).

6.x plainly and immediately shows this as an error, with no ill effect I could tell (in other words, it just gracefully doesn't do it and moves on).
9.6, on the other hand doesn't give you any information about what happened, and here ('buntu 18.06 / XFCE), the message box froze up for some time. I don't know if it would have crashed or not (I suspect not), but it was still a bit disconcerting.

The more I think about it, though, the lack of any error message seems a bug in and of itself, anyone else think so?