[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: patch for io.tst
From: |
Rik |
Subject: |
Re: patch for io.tst |
Date: |
Sun, 22 Dec 2013 10:25:46 -0800 |
On 12/22/2013 07:15 AM, address@hidden wrote:
> Message: 1
> Date: Sun, 22 Dec 2013 15:46:05 +0100
> From: "c." <address@hidden>
> To: Daniel J Sebald <address@hidden>
> Cc: "John W. Eaton" <address@hidden>, octave-maintainers
> <address@hidden>
> Subject: Re: Octave 3.8.0-rc2 release candidate available for ftp
> Message-ID: <address@hidden>
> Content-Type: text/plain; charset="iso-8859-1"
>
>
> On 22 Dec 2013, at 10:00, c. <address@hidden> wrote:
>
>> > As I pointed out before, the test failures in "io.tst" are bogus,
>> > they only appear when running "__run_test_suite__" or "make check",
>> > otherwise (if running the code directly or through "test io.tst")
>> > there is no failure:
>> >
>>>> >>> test /opt/local/share/octave/3.8.0-rc2/etc/tests/fixed/io.tst
>> > PASSES 85 out of 85 tests
> The attached changeset solves the problem for me.
> Is it OK to push? Which branch should it go on?
Carlo,
Is this line of the patch necessary?
-%! eval (sprintf ("save %s %s %s", opts{i}, files{i}, vars));
+%! eval (sprintf ("save %s %s %s", files{i}, opts{i}, vars));
According to the documentation for save, options should come first.
-- Command: save file
-- Command: save options file
-- Command: save options file V1 V2 ...
-- Command: save options file -struct STRUCT F1 F2 ...
Second thought is that it might make the code look cleaner if you defined a
filename variable to hold the temporary filename for each test rather than
using the fullfile syntax every time.
Example:
fname = fullfile (P_tmpdir, "matrix.ascii");
save ("-ascii", fname, "matrix1")
matrix2 = load (fname);
assert (matrix1, matrix2)
delete (fname);
Cheers,
Rik
- Re: patch for io.tst,
Rik <=