On 11/17/16 Nov 17 -1:21 PM, Elias Pipping wrote:
On 17 Nov 2016, at 19:55, Robert Goldman rpgoldman@sift.net wrote:
I show 3.1.7.35 passing all tests on Mac and Linux, failing test-require and test-run-program on ECL bytecodes on Windows.
test-require is a long-standing ECL bug on Windows that Jason Miller has confirmed. There's a bug report in to ECL, so we don't have to worry about this.
test-run-program shows a failure on run-program with a string argument versus with a list argument.
Attaching a snippet of the transcript. I'd like to get this either fixed or dismissed as an ECL problem before we release.
Cheers, R
<ecl-bytecodes-run-program.txt>
Thanks, Robert.
The test-run-program issue is a regression. And I must’ve introduced it with
https://gitlab.common-lisp.net/asdf/asdf/commit/b263ded6f57264dd2b36e97790e8...
which relied on
https://gitlab.common-lisp.net/asdf/asdf/commit/75c5bd04ee6b236c0c0b19807c1f...
I would’ve expected that to work after reading
https://gitlab.com/embeddable-common-lisp/ecl/blob/8f07cd58d87ea9abd37787b55...
but apparently I’m missing something. The simplest solution that will make things work again is to restrict the use of launch-program in %system further: Not just to recent versions but also to unix. That’ll give us a regression-free release. I’d much rather understand this quoting problem you mentioned in your other mail though and why this test would trigger it. The switch to launch-program in %system is what gives us reliable interactive output after all, and I’d prefer to have that on windows, too.
Elias
I'm busy today, but I'll try to squeeze in a little time to revert the use of launch-program and test.
I'll report back as soon as possible.
best, r