Search
Thursday, 2nd of April 2020, 7:50:01 UTC
13:11:43
kpoeck
Hello, having a question on how to handle ~/quicklisp/local-projects
13:12:10
kpoeck
I have there like 25 projects that I forked and made modifications too
13:12:34
kpoeck
Most changes are accepted, like you just merged my float-features pr
13:13:04
kpoeck
Once done, should I delete my fork in ~/quicklisp/local-projects?
13:13:29
Shinmera
Once Quicklisp's dist updates and the changes are public, yes.
13:13:37
Shinmera
Otherwise you risk running out of date with your local fork again.
13:14:06
Shinmera
There's no easy solution to it, I'm afraid.
13:14:20
kpoeck
Ok, so best practice to delete once quicklisp has the update
13:14:30
kpoeck
sounds better than what I do now
13:19:44
Shinmera
I use a script to automatically pull all of my repositories to sync across machines. I suppose that could also be extended to sync with upstream if available.
13:19:49
Shinmera
https://plaster.tymoon.eu/view/650
16:00:49
Bike
trying to figure out this can't recover from FPE thing. really don't get it. the stack looks like it should have all the frames, but it doesn't i guess?
16:01:37
Bike
maybe it's handled in the wrong thread or something? that doesn't seem right
16:01:49
Bike
since it happens in single threaded processes
18:13:49
Bike
i hacked something to print all the rbp frames and the return frames, and it looks like it's there, but it's not
18:13:59
Bike
frame check doesn't find it and neither does the actual exception handler. how
19:12:58
drmeister
Something that appears to have stopped working is backtraces when we suppress the repl.
19:13:22
drmeister
./iclasp-boehm -N -e '(error "an error")' --> no backtrace.
19:13:28
drmeister
it should though - it invokes (core:btcl)
19:13:40
drmeister
I'm debugging it now.
19:19:47
Bike
okay now i am seeing a truncated stack. what gives here.
19:44:45
drmeister
I'm moving core__btcl into C++ - I'm trying to eliminate depending on CL for backtraces.
Thursday, 2nd of April 2020, 19:50:01 UTC