Odd program interaction issue - Avalanche & Zap
Andrew McCarthy (3688) 605 posts |
Zap was crashing when opening a collection of files at a directory depth of 5, and I found out it happened when Avalanche was run. I also observed that the order in which I opened those directories was important. I thought it would be helpful to highlight that running Zap and Avalanche causes Zap to crash- I’m happy to say that Zap gracefully recovers itself. I hope knowing this is helpful to anyone who experiences a similar problem with Zap. A previous issue I’d experienced with Zap was partially resolved by updating LineEditor. I’m happy that I raised the problem at the last meetup because without those folks coming up with ideas around the issue, I don’t believe I would have tracked it down alone- Thank you. :)
Avalanche 0.28, Zap 1.49 Rick 07, RPi4, RISC OS 5.28-9 |
David J. Ruck (33) 1635 posts |
What was the Zap error? “Not a heap block”? |
Andrew McCarthy (3688) 605 posts |
Zap may have gone wrong – &FC014A20 is at offset &00014A28 in the kernel. The other message has been “not a heap block”.
|
Andrew McCarthy (3688) 605 posts |
On the back of David’s question, something started to bother me. I’ve now discovered what that was. I apologise to James P. for suggesting an odd interaction between Avalanche and Zap and anyone else I may have misled. I appear to have been focused on the new lines of inquiry to the detriment of the order of my actions. The error can be reproduced or avoided by choosing which directory of files you open first. In my example, I opened two directories of files and then opened both sets. Since the original post, I have discovered that if I open the directory with the higher alphabetical name (O) before the lower one (F), Zap crashes with the above error details. Surprisingly, when opening the directories (F→O) and then the files (^A), there’s no error. Both Edit and StrongED are unaffected. |