On Tue, Feb 07, 2017 at 10:24:58PM +1300, Chris Lamb wrote:
> Hi Leo,
>
> > Since I am comparing the entire output of Python 2.7's build process,
> > there are a lot of files, and they *all* have different ctimes (we set
> > mtimes to epoch).
>
> So, we already ignore mtimes on specified command-line arguments:
>
> $ touch --date="@0" epoch
>
> $ touch now
>
> $ diffoscope epoch now --no-progress
>
> $ echo $?
> 0
>
> Are you referring to {m,c}times, etc. within archives?
I was referring to {m,c}times of the files being compared themselves.
If by "within archives" you mean "of files within, for example, two zip
archives being compared", that's not what I meant.
I just tested diffoscope 77 and it ignores the different {m,c}times of
two otherwise identical test files. Great!
Now, it would be nice if diffoscope had an option to report these
differences :) I wondered if --fuzzy-threshold could be tweaked to make
diffoscope report them, but it did not for the values of 0, 1, and 400.
If I understood your question correctly, please feel free to close this
bug, or ask me to close it.
I'm sorry for the sloooooow response to your question. Thanks for the
ping!
signature.asc
Description: PGP signature
_______________________________________________ Reproducible-builds mailing list Reproducible-builds@lists.alioth.debian.org http://lists.alioth.debian.org/cgi-bin/mailman/listinfo/reproducible-builds
