Paul Eggert <[EMAIL PROTECTED]> wrote:
> I installed this minor code cleanup; it doesn't fix any bugs.
Thanks for the cleanup.
> I notice that md5sum's --string option isn't documented; is this
> intentional?
Yes.
It may be safely removed, now :-)
I actually removed it back in 1996, but restored
it right away, probably because of a test failure.
Instead, I undocumented it, but apparently didn't
mark it as to-be-removed in the source.
Thu Sep 26 21:02:54 1996 Jim Meyering <[EMAIL PROTECTED]>
* src/md5sum.c (usage): Remove references in --help output to
the now-deprecated --string option. Support for it will be
removed soon. Using this option can provoke a memory access
violation on some systems because of (unavoidable in an efficient
implementation) alignment assumptions made by functions in md5.c.
md5sum.c could leave the option in and accomodate the alignment
restriction, but this option was intended solely for testing, and
the tests are now all file-oriented, so it's not necessary.
> 2004-09-18 Paul Eggert <[EMAIL PROTECTED]>
>
> * src/md5sum.c (STATUS_OPTION, STRING_OPTION): New enums.
> (long_options, main): Use them instead of magic numbers 2 and 1.
> For --string, optarg can't possibly be NULL.
_______________________________________________
Bug-coreutils mailing list
[EMAIL PROTECTED]
http://lists.gnu.org/mailman/listinfo/bug-coreutils
