Re: "cp -p" does not preserve the mtime of symlinks

2015-12-17 Thread Philip Guenther
On Thu, 17 Dec 2015, Ted Unangst wrote: > Carsten Kunze wrote: > > Ted Unangst wrote: > > > > > Probably not intended. Just an artifact left over from the era before you > > > could set times on symlinks. (not likely that many people care, either.) > > > > I do care--I have a sync software (simi

Re: "cp -p" does not preserve the mtime of symlinks

2015-12-17 Thread Ted Unangst
Carsten Kunze wrote: > Ted Unangst wrote: > > > Probably not intended. Just an artifact left over from the era before you > > could set times on symlinks. (not likely that many people care, either.) > > I do care--I have a sync software (similar to rsync) which compares mtime and > length. lst

Re: "cp -p" does not preserve the mtime of symlinks

2015-12-17 Thread Carsten Kunze
Ted Unangst wrote: > Probably not intended. Just an artifact left over from the era before you > could set times on symlinks. (not likely that many people care, either.) I do care--I have a sync software (similar to rsync) which compares mtime and length. lstat needs do be done anyway so there

Re: "cp -p" does not preserve the mtime of symlinks

2015-12-17 Thread Ted Unangst
Carsten Kunze wrote: > Hello, > > "cp -p" seems to not preserve the modification time of symlinks. This is not > mentioned in the man page. Other systems I tested (others BSDs and Linux) do > preserve the mtime of symlinks with e.g. "cp -a". The OpenBSD kernel also > does support to set it w

"cp -p" does not preserve the mtime of symlinks

2015-12-17 Thread Carsten Kunze
Hello, "cp -p" seems to not preserve the modification time of symlinks. This is not mentioned in the man page. Other systems I tested (others BSDs and Linux) do preserve the mtime of symlinks with e.g. "cp -a". The OpenBSD kernel also does support to set it with e.g. utimensat(2). Is this i