Re: SRFI-64 implementation for Guile 2.0

[Sunjoong Lee <sunjoong@gmail.com>]

[-- Attachment #1: Type: text/plain, Size: 1437 bytes --]

Hi, Ludo’;

Your implementation fails srfi-64-test.scm, a test suite for the SRFI 64.

Per:
I realize the test-error of the reference implementation has a bug;
It calls %test-error like this - (test-assert (%test-error etype expr))
but %test-error needs three arguments - (%test-error r etype expr) !!
I think you may fix it.
Humm... in %test-error (and srfi-34 srfi-35) case, (equal? type #t) is a
typo of (equal? etype #t), I think.

Ludo’:
"Store the test log in UTF-8" is a good idea.
I don't know "Display a backtrace upon error."
In my implementation on Guile and Per's reference implementation on
Kawa, actual-error is loged if exception occurs like this;
  Test begin:
    test-name: "3.3. test-begin with mismatched test-end"
    source-file: "srfi-64-test.scm"
    source-line: 236
    source-form: (test-error "3.3. test-begin with mismatched test-end" #t
(triv-runner (lambda () (test-begin "a") (test-assert "b" #t) (test-end
"x"))))
  Test end:
    result-kind: pass
    actual-error: (misc-error #f "~A" ("bad end grojup name a but expected
x") #f)
    expected-error: #t

2012/4/21 Ludovic Courtès <ludo@gnu.org>
>
> Per Bothner <per@bothner.com> skribis:
>
> > I think it would be great to get SRFI-64 into Guile.
>
> Me too.
>
> FWIW, I’ve been using an almost-unmodified version in several projects:
>
>  http://git.savannah.gnu.org/cgit/libchop.git/tree/guile2/srfi
>

[-- Attachment #2: Type: text/html, Size: 2233 bytes --]