Re: Should -et be available?

[Eli Schwartz]

On 5/25/20 9:55 AM, Dennis Williamson wrote:
> With an OS/filesystem which supports subsecond file modification times
> neither -ot or -nt may be true if the seconds are the same (so far as
> expected) but the fractional seconds are different (so the file times are
> not actually equal). This may not be what you expect.
> 
>  $ cat /etc/lsb-release
>  DISTRIB_ID=Ubuntu
>  DISTRIB_RELEASE=18.04
>  DISTRIB_CODENAME=bionic
>  DISTRIB_DESCRIPTION="Ubuntu 18.04.3 LTS"
>  $ <CTRL-X><CTRL-V>
>  GNU bash, version 4.4.20(1)-release (x86_64-pc-linux-gnu)
>  $ # touch with delay - modification times not equal
>  $ touch test1; sleep .001; touch test2; stat --format='%y %n' test1 test2
>  2020-05-25 08:38:26.209032136 -0500 test1
>  2020-05-25 08:38:26.221031929 -0500 test2
>  $ [[ test1 -ot test2 ]] && echo older
>  $ [[ test1 -nt test2 ]] && echo newer
>  $ # touch with no delay - modification times equal
>  $ touch test1 test2; stat --format='%y %n' test1 test2
>  2020-05-25 08:40:52.680722254 -0500 test1
>  2020-05-25 08:40:52.680722254 -0500 test2
>  $ [[ test1 -ot test2 ]] && echo older
>  $ [[ test1 -nt test2 ]] && echo newer

$ cat /etc/os-release
NAME="Arch Linux"
PRETTY_NAME="Arch Linux"
ID=arch
BUILD_ID=rolling
[...]
$ <CTRL-X><CTRL-V>
GNU bash, version 5.0.17(1)-release (x86_64-pc-linux-gnu)


$ touch test1; sleep .001; touch test2; stat --format='%y %n' test1 test2
2020-05-25 13:00:23.707169347 -0400 test1
2020-05-25 13:00:23.713836098 -0400 test2
$ [[ test1 -ot test2 ]] && echo older
older
$

$ touch test1 test2; stat --format='%y %n' test1 test2
2020-05-25 13:01:26.047939739 -0400 test1
2020-05-25 13:01:26.047939739 -0400 test2
$ [[ test1 -ot test2 ]] && echo older
$ [[ test1 -nt test2 ]] && echo newer
$



-- 
Eli Schwartz
Arch Linux Bug Wrangler and Trusted User

signature.asc
Description: OpenPGP digital signature