Weird GNUPLOT problem (but *not* just Gnuplot)

[Vincent Adolph jr]

> On Sep 13, 2020, at 1:07 AM, Dave Horsfall <dave at horsfall.org> wrote:
> 
> No replies?  Am I the only person on the planet having this problem?
> 
> It's not confined to Gnuplot, but pretty much anything that wants to save a modified file, such as TextEdit (such as my prescriptions, which for me is pretty much a matter of life and death).
> 
> I note that this crap only started happening after last Monday's weekly update; up until then, my MacBook had performed flawlessly...  Could the perp responsible please raise their hand?
> 
> Should I back out that flawed update (if I can remember how) and stop doing any further updates until I am assured that they will actually work i.e. they were actually tested?  For some reason the Therac-25 (look it up; it's the story of an untested software update that went horribly wrong) springs to mind...
> 
> What can I do to provide more information?  Whoever in Cupertino is receiving the crash reports when I abort the hung processes must be getting pretty sick of them by now, and that's not looking too good for MacPorts...
> 
> Excuse my attitude, but after nearly 50 years in the IT game (probably longer than most people on this list have been alive) I've never seen anything like this.
> 
> On Thu, 10 Sep 2020, Dave Horsfall wrote:
> 
>> MacBook Pro (early) with Sierra (all updates); can't be more precise than that because "About This Mac" is no longer displaying anything (SMC reset required?) since a reboot, but that's not my immediate problem.
>> 
>> Gnuplot Version 5.2 patchlevel 8    last modified 2019-12-01
>> 
>> When trying to save a graph (any graph) I get as far as "Save As" and I get the dreaded spinning beachball instead of the file selection; I've waited up to an hour before killing it by doing in AquaTerm.  This started happening after my Monday "port upgrade outdated", so I'm guessing that I now have a bad library somewhere as a result of the upgrade?  There were a few library updates, as I recall...
>> 
>> I'm trying to figure out whether this is a Mac/Gnuplot problem or a user error, so I'm wondering whether others have seen this before I report it as a possible Gnuplot bug (but I can't see how it could be, because it wasn't touched as such).
>> 
>> Thanks (and yeah, I have a few outstanding replies to follow up)
>> 
>> -- Dave
>> 





> 
> -- Dave