Closed pb-cdunn closed 5 years ago
Whoa! I just told him to clear the nim-cache, and now it works!
Yeah. I clear my nim cache pretty obsessively. Not as robust as it could be, but easy to work around.
For anyone coming from google here is where the cache lies:
https://nim-lang.org/docs/nimc.html#generated-c-code-directory
Ran into this error as well after I removed nim 1.x
installed via brew and installed 1.4.0
via choosenim
. Tried to remove ~/.cache
but found no help. Switched back to 1.2.8
and working for now.
Do you have a stack trace with a line number like the original post here from 1.5 years ago? (I.e., the cligen.nim(730)
bit?). []
gets used a lot. If I cannot reproduce this, then more detailed bug reports are needed for me to be able to do much. :-/
Not much information:
=>> choosenim 1.4.0
Info: Version 1.4.0 already selected
=>> nim -v
Nim Compiler Version 1.4.0 [MacOSX: amd64]
Compiled at 2020-10-29
Copyright (c) 2006-2020 by Andreas Rumpf
active boot switches: -d:release
=>> nim c -r --hints:off tests/test_expr.nim
fatal.nim(49) sysFatal
Error: unhandled exception: index 1 not in 0 .. 0 [IndexDefect]
is there any other command to try for informations?
Yeah. That also isn't much to go on. Hmm. Is tests/test_expr.nim
small?
Yep, quite small https://github.com/Cirru/calcit-runner.nim/blob/master/tests/test_expr.nim .
Um...That test_expr.nim
does not use cligen
...Indeed, if I git clone
the whole repo and git log -p | grep -i cligen
I get empty results suggesting cligen
has never been used. So, I'm unsure why you think/thought this to be a cligen
issue...
Oh, my mistake, I attached the comment in the wrong repo...
A co-worker keeps seeing this. I cannot reproduce it. We managed to get rid of it by upgrading both Nim and cligen, but it keeps coming back.
I think it might be caused by Nimble not being very clear about what SHA1 is uses for various dependencies. But I was wondering if you've seen this, and if you know for sure how to avoid it.