Description of problem: "mv -i --reply=no" is supposed to be equivalent to answering "no" to the prompt given by "mv -i". ie, it is supposed to NOT over-write existing files. but instead, it DOES over-write existing files. ie it behaves exactly like "mv -i --reply=yes" (or just "mv"). Version-Release number of selected component (if applicable): How reproducible: always. Steps to Reproduce: 1. touch a b 2. mv -i --reply=no a b 3. ls Actual results: only "b" exists. Expected results: both "a" and "b" should exist. Additional info: redhat-7.2: --reply is not understood by mv. redhat-8.0: has the bug. rehdat-9.0: has the bug. [edp@wpax edp]$ mkdir test [edp@wpax edp]$ cd test [edp@wpax test]$ touch a b [edp@wpax test]$ ls -l total 0 -rw-rw-r-- 1 edp edp 0 May 31 12:15 a -rw-rw-r-- 1 edp edp 0 May 31 12:15 b [edp@wpax test]$ mv -i a b mv: overwrite `b'? no [edp@wpax test]$ ls -l total 0 -rw-rw-r-- 1 edp edp 0 May 31 12:15 a -rw-rw-r-- 1 edp edp 0 May 31 12:15 b [edp@wpax test]$ mv -i --reply=no a b [edp@wpax test]$ ls -l total 0 -rw-rw-r-- 1 edp edp 0 May 31 12:15 b [edp@wpax test]$ rpm -q --whatprovides /bin/mv coreutils-5.0-34.1 [edp@wpax test]$
oops, it's a dupe. (i did search first, dont know why i didnt find the other one...) now i don't understand what "--reply" is supposed to do. the documentation should be improved, certainly. and possibly "--reply" should still be changed to work as users would reasonably expect... sorry for the dupe, good luck with upstream discussion :) *** This bug has been marked as a duplicate of 120742 ***
Changed to 'CLOSED' state since 'RESOLVED' has been deprecated.