I have determined I only see this problem when running 'make test'
inside a shell of a graphical emacs session. Running 'make test' in an
xterm or in a emacs session started with 'emacs -nw' complete successfully.
I have seen this failure on two separate machines.
Thanks,
Todd
On 02/05/2014 11:50 AM, [email protected] wrote:
Hi Todd,
If I change process.scm around line 124 like this:
(when #f
(cond-expand
;; can't set timeout on regular Java ports
(bigloo-jvm #f)
(else (test "&io-timeout-error"
(let* ((proc (run-process *bigloo-path* "-eval" "(sleep 4)"
output: pipe:))
(output (process-output-port proc)))
(input-port-timeout-set! output 1000)
(let ((result (with-handler
(lambda (e)
(isa? e&io-timeout-error))
(begin
(read-lines output)
#f))))
(close-input-port output)
(process-wait proc)
result))
#t))))
to keep this specific test from running, then 'make test' completes
(although I still get the three bignum fails). I don't understand enough
about this test to comment or suggest why.
last part of output:
All tests executed...
all succeeded
make[1]: Leaving directory
`/proj/samson/user/tdukes/Downloads/tmp/bigloo4.1a-beta04Feb14'
bash-4.2$
I will test with my application code tomorrow.
I will try to understand what's going on. What is your platform again
(os and arch)? I will try to test a configuration similar to yours.
I must admit that I'm annoyed but this problem.
I'm not able to reproduce it (I have tried three different machines without
success). It could be that this is something due to your virtual environment,
I could not tell. I have the vague feeling that this is something not that
important but still it would be much nicer to fix it.
Is there someone experiencing the same problem? Thanks in advance.
