On Thu, Mar 30, 2017 at 11:30 AM, Nick Kew <n...@apache.org> wrote: > On Thu, 2017-03-30 at 09:59 -0500, William A Rowe Jr wrote: > >> > Is this the expected behaviour? should documentation warn that if the >> > destination already exists, then permissions will not be copied even with >> > that flag? >> >> Since we make no statement either way I wouldn't call this 'expected'. >> >> The linux behavior of cp -p differs from the APR 1.5.x and earlier behavior. > > Is Linux in any sense definitive? Is there any standard > with unix roots (or even from the Windows world)? > Does the flag mirror anything from stdio & friends ?
There is no alternate implementation, the same file/unix/copy.c source is used on all architectures. Looking at a Win32 API for accomplishing the same, there are no flags to surpress copying the permissions/security attributes, these are always duplicated https://msdn.microsoft.com/en-us/library/windows/desktop/aa363852%28v=vs.85%29.aspx The CopyFile2 API is slightly richer (prepare for headspin); COPY_FILE_REQUEST_SECURITY_PRIVILEGES The copy is attempted, specifying ACCESS_SYSTEM_SECURITY for the source file and ACCESS_SYSTEM_SECURITY | WRITE_DAC | WRITE_OWNER for the destination file. If these requests are denied the access request will be reduced to the highest privilege level for which access is granted. For more information see SACL Access Right. This can be used to allow theCopyFile2ProgressRoutine callback to perform operations requiring higher privileges, such as copying the security attributes for the file. So in short, I don't think there is a good analog in the Win32 API, and I'm not aware a posix file copy facility. There are convoluted ioctl approaches, but nothing for general consumption, e.g. http://lists.gnu.org/archive/html/coreutils/2011-08/msg00006.html We should at least note that right now, permissions are only copied or assigned if the file is absent when apr_file_copy() is invoked.
