Re^4: Optimizing files' moving with File::Find & File::Copy modules

If he doesn't have that implementation of find, and there are spaces in filenames, the array form of system will handle them:

my @files = <$dir1/*>;
system("mv", @files, $dir2);
[download]

Comment on Re^4: Optimizing files' moving with File::Find & File::Copy modules Download Code

Replies are listed 'Best First'.
Re^5: Optimizing files' moving with File::Find & File::Copy modules by JavaFan (Canon) on Nov 17, 2009 at 17:27 UTC
Eh, no. `<$dir1/*>` doesn't distinguish between files and directories, it doesn't find files whose name start with a dot, and doesn't recurse into directories below $dir1. And if there are many files, you may run into system limits - most OSses have a limit on the number of arguments (and total size of the arguments) `exec` can take. (`E2BIG`).	[reply] [d/l] [select]
Re^6: Optimizing files' moving with File::Find & File::Copy modules by runrig (Abbot) on Nov 17, 2009 at 17:42 UTC
So I simplified "finding" the files :-) (as you did in your first answer...and before I read the followup about needing a recursive find...and if a single level was adequate but you needed to filter directories, then a simple grep could take care of that...and if you really wanted files with leading dots, then add a ".*" pattern to the glob) You could use File::Find to push the files onto an array, then splice off N files at a time and use system to move the files.	[reply]