[9fans] test(1) -older bug?

[9fans] test(1) -older bug?

[arisawa]

Hello,

How do you think?
I think test is buggy in "older" operation.
the test is from 9front. I have not tried bell-labs test.

term% ls -l
--rw-rw-r-- M 149 arisawa arisawa      3277 Apr  9 23:11 x
--rw-rw-r-- M 149 arisawa arisawa      4555 Apr  9 23:12 y
term% mtime x
 1365516710 x
term% mtime y
 1365516741 y
term% if(test x -ot y) echo OK
OK
term% if(test x -older 1365516741) echo OK
term% if(test x -older 1365516700) echo OK
term%
term% date -n
1370119926
term% dc
1370119926 1365516710 - p
4603216
term% if(test x -older 4603216) echo OK
OK
term% if(test x -older 4703216) echo OK
term%

term% man test
          f -older t True if file f is older than (modified before)
                     time t. If t is a integer followed by the letters
                     y(years), M(months), d(days), h(hours),
                     m(minutes), or s(seconds), it represents current
                     time minus the specified time.  If there is no
                     letter, it represents seconds since epoch.  You
                     can also concatenate mixed units.  For example,
                     3d12h means three days and twelve hours ago.

Kenji Arisawa

Re: [9fans] test(1) -older bug?

[erik quanstrom]

yup.  i think it's a bug:

/n/sources/patch/older
/n/atom/patch/older

- erik

Re: [9fans] test(1) -older bug?

[cinap_lenrek]

its /n/sources/patch/testolder, also leaks dir in the case:

	if(rel)
		n = time(0) - n;
	if(n < 0)
		return 0; <----- HERE
	r = dir->mtime < n;

	free(dir);
	return r;

--
cinap

Re: [9fans] test(1) -older bug?

[erik quanstrom]

On Sun Jun  2 17:31:05 EDT 2013, cinap_lenrek@gmx.de wrote:
> its /n/sources/patch/testolder, also leaks dir in the case:
>
> 	if(rel)
> 		n = time(0) - n;
> 	if(n < 0)
> 		return 0; <----- HERE
> 	r = dir->mtime < n;
>
> 	free(dir);
> 	return r;

thanks.  i appreciate the catch.

[sources] applied patch: /n/atom/patch/applied/testolder1

email
	quanstro@quanstro.net
readme
	fix memory leak pointed out by cinap
removed

files
	/sys/src/cmd/test.c	test.c

/sys/src/cmd/test.c	test.c
test.c.orig:373,380 - test.c:373,381
  	if(rel)
  		n = time(0) - n;
  	if(n < 0)
- 		return 0;
- 	r = dir->mtime < n;
+ 		r =  0;
+ 	else
+ 		r = dir->mtime < n;

  	free(dir);
  	return r;
------
merge...backup...copy...
cpfile test.c /n/dist/sys/src/cmd/test.c
done

Re: [9fans] test(1) -older bug?

[Richard Miller]

>> its /n/sources/patch/testolder, also leaks dir in the case:
>>
>> 	if(rel)
>> 		n = time(0) - n;
>> 	if(n < 0)
>> 		return 0; <----- HERE
>> 	r = dir->mtime < n;
>>
>> 	free(dir);
>> 	return r;

And the consequences of not freeing a few bytes of memory, in a command
which will exit a few microseconds later, would be ... ?

Re: [9fans] test(1) -older bug?

[arnold]

Richard Miller <9fans@hamnavoe.com> wrote:

> And the consequences of not freeing a few bytes of memory, in a command
> which will exit a few microseconds later, would be ... ?

The Code Correctness Police come and collect you and force you to
program on Windows...  <ducks>

:-)

Arnold

Re: [9fans] test(1) -older bug?

[erik quanstrom]

On Mon Jun  3 08:00:20 EDT 2013, arnold@skeeve.com wrote:
> Richard Miller <9fans@hamnavoe.com> wrote:
>
> > And the consequences of not freeing a few bytes of memory, in a command
> > which will exit a few microseconds later, would be ... ?
>
> The Code Correctness Police come and collect you and force you to
> program on Windows...  <ducks>

excellent points.  there is no consequence.  but conversely, there is
no penalty for free'ing the memory either.  given that, i prefer free'ing
the memory to be consistent with the rest of the source, and accept
cinap's bug as a bug.

of course you can say needless consistency is the hobgoblin of little
minds...

:-)

- erik

Re: [9fans] test(1) -older bug?

[Jacob Todd]

[-- Attachment #1: Type: text/plain, Size: 223 bytes --]

On Jun 3, 2013 6:49 AM, "Richard Miller" <9fans@hamnavoe.com> wrote:
> And the consequences of not freeing a few bytes of memory, in a command
> which will exit a few microseconds later, would be ... ?
>
>
bad taste.

[-- Attachment #2: Type: text/html, Size: 318 bytes --]

Re: [9fans] test(1) -older bug?

[Richard Miller]

> bad taste.

I agree - better style is to free memory always, or never.  But the
choice might not be as obvious as one first thinks.