Look at the source for mv and see what it's test is.... <br><br>I submitted a patch to cvs once that fixed an error that involved these semantics - oh because cvs was doing a rename when it should have been doing a move. Rename fails across filesystems, move works. It might be something as simple as 'if the rename system call fails - do a copy/delete". In point of fact, I think there is a rename command but I'm not sure if it is a literal wrapper for the rename syscall and has the same symantics or not... <br>
<br>Just giving ideas/spitballing. <br><br>-Rob<br><br> <br><div class="gmail_quote">On Tue, Nov 16, 2010 at 10:22 AM, gregwm <span dir="ltr"><<a href="mailto:tclugl@whitleymott.net">tclugl@whitleymott.net</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="margin: 0pt 0pt 0pt 0.8ex; border-left: 1px solid rgb(204, 204, 204); padding-left: 1ex;">can anyone show me a bash test to put before mv that will tell whether mv will do a simple rename or a copy and delete? for example for mv into a --bind mount, df reports both locations as within the same
fs, yet mv will copy and delete. the best test I can think of would create a file, mv and see what happens. can anyone concoct an accurate test that doesn't need to create a test file? (or perhaps reveal an mv alternative..)<br>
<br>_______________________________________________<br>
TCLUG Mailing List - Minneapolis/St. Paul, Minnesota<br>
<a href="mailto:tclug-list@mn-linux.org">tclug-list@mn-linux.org</a><br>
<a href="http://mailman.mn-linux.org/mailman/listinfo/tclug-list" target="_blank">http://mailman.mn-linux.org/mailman/listinfo/tclug-list</a><br>
<br></blockquote></div><br>