Pádraig Brady <[email protected]> writes:
> On 27/10/2025 22:57, Bruno Haible via GNU coreutils General Discussion wrote:
>> On Solaris 11.4:
>> FAIL: tests/numfmt/numfmt
>> Was nonexistent on Oct 13, FAIL on Oct 20, 27.
>> Find attached the relevant part of test-suite.log.
>
> Interesting, it doesn't like the NBSP characters:
>
> # NBSP characters: U+00A0, U+2007, U+202F, U+2060
> ['lcl-suf-3', "--from=auto '2\xc2\xa0K'", {OUT => "2000"},
> {ENV=>"LC_ALL=$locale"}],
> ['lcl-suf-4', "--from=auto '2\xe2\x80\x87Ki'", {OUT => "2048"},
> {ENV=>"LC_ALL=$locale"}],
> ['lcl-suf-5', "--from=auto '2\xe2\x80\xafK'", {OUT => "2000"},
> {ENV=>"LC_ALL=$locale"}],
>
> Though this is OK:
>
> ['lcl-suf-6', "--from=auto '2\xe2\x81\xa0Ki'", {OUT => "2048"},
> {ENV=>"LC_ALL=$locale"}],
>
> I have access to the cfarm solari11 machines
> so will fix this tomorrow.
I looked into it a bit, but don't have a patch ready. The issue looks to
be the use of newline_or_blank in next_field(). As we ran into with the
'fold' changes, whether or not NBSP is considered a blank or space
varies across different systems.
I tried to use function like:
static bool
newline_or_blank (mcel_t g)
{
return g.ch == '\n' || c32isblank (g.ch);
}
But that didn't seem to work.
Collin
