Pierre Habouzit <madcoder@xxxxxxxxxx> writes:
> On Mon, Sep 24, 2007 at 07:57:31AM +0000, David Kastrup wrote:
>> "David Symonds" <dsymonds@xxxxxxxxx> writes:
>>
>> > On 24/09/2007, David Kastrup <dak@xxxxxxx> wrote:
>> >> Mike Hommey <mh@xxxxxxxxxxxx> writes:
>> >>
>> >> > On Sun, Sep 23, 2007 at 10:42:08PM +0200, David Kastrup wrote:
>> >> >> -while case $# in 0) break ;; esac
>> >> >> +while test $# != 0
>> >> >
>> >> > Wouldn't -ne be better ?
>> >>
>> >> Why?
>> >
>> > Because -ne does a numeric comparison, != does a string comparison,
>> > and it's a numeric comparison happening, semantically speaking.
>>
>> I don't see the point in converting $# and 0 into numbers before
>> comparing them. "!=" is quite more readable, and the old code also
>> compared the strings.
>
> Fwiw $# already is a number.
It isn't.
> Hence test $# -ne 0 is definitely a better test.
/* TEST/[ builtin. */
int
test_builtin (list)
WORD_LIST *list;
{
char **argv;
int argc, result;
/* We let Matthew Bradburn and Kevin Braunsdorf's code do the
actual test command. So turn the list of args into an array
of strings, since that is what their code wants. */
if (list == 0)
{
if (this_command_name[0] == '[' && !this_command_name[1])
{
builtin_error ("missing `]'");
return (EX_BADUSAGE);
}
return (EXECUTION_FAILURE);
}
argv = make_builtin_argv (list, &argc);
result = test_command (argc, argv);
free ((char *)argv);
return (result);
}
> $# != 0 would yield sth like (strcmp(sprintf("%d", argc), "0"))
> $# -ne 0 would yield sth like (argc != atoi("0")).
>
> Not that it matters much, but the latter looks better to me.
Not to me. The code does not support your argument, and all $x
expansions certainly are strings, according to manual and usage. I
will rework the patch this evening in order to get a commit message
more placable to Junio, and I will at his request remove all of the
(redundant) quoting. But removing quoting from $# does not turn it
into a number: it remains the same string '0'. If someone else feels
he should replace all "=" and "!=" tests for "numeric" comparisons
with the unreadable numeric tests, he can go ahead proposing a
separate patch that should not just cover $#.
You can have bash declare numeric variables, but even they are strings
(they just auto-evaluate on assignment):
declare -i nonsense
nonsense="3+$(echo 4)"
echo "$nonsense"
gives 7, even though everything has been "strings" here.
--
David Kastrup
-
To unsubscribe from this list: send the line "unsubscribe git" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
