On 08/16/2009 07:16 PM, Eduardo Cavazos wrote:
Below is a note I sent to the Ikarus Scheme list today regarding usage
of SRFI-64. If you have any input, let me know!
The reference implementation of srfi-64 contains some extra
Kawa-specific magic which makes things quite a bit nicer.
These are controlled by
(cond-expand (kawa ...) (else ...))
(There are also some conditionals that depend on other srfis.)
You might consider implementing similar hooks for Ikarus. If you
have any questions about what is needed, feel free to ask.
I'll be happy to merge them into the reference implementation.
For example, under Kawa failing tests report the line number:
%%%% Starting test numbers (Writing full log to "numbers.log")
./num-test.scm:104: FAIL
# of expected passes 1697
# of unexpected failures 1
Also, note that the output includes the
(Writing full log to "numbers.log")
This is controlled by the test-log-to-file parameter.
I'm surprised you're not seeing that, as it should be the default
for all implementations.
Under Kawa, the log file is pretty verbose:
%%%% Starting test numbers
Group begin: numbers
Test begin:
source-file: "./num-test.scm"
source-line: 3
source-form: (test-approximate 1.4 (sqrt 2) 0.02)
Test end:
result-kind: pass
actual-value: 1.4142135623730951
expected-value: 1.4
... etc ... etc ...
Test begin:
source-file: "./num-test.scm"
source-line: 104
source-form: (test-eqv 2 (logcount -2))
Test end:
result-kind: fail
actual-value: 1
expected-value: 2
... etc ... etc ...
Group end: numbers
# of expected passes 1697
# of unexpected failures 1
However, this depends on some Kawa-specific code; the generic
log file output doesn't have as much information.
--
--Per Bothner
[email protected] http://per.bothner.com/
