treescan - Man Page

By default, treescan prints the full paths of all directories or files it finds. This option disables printing of filenames completely. This is useful if you want to run treescan solely for its side effects, such as pulling stat data into memory.

Instead of using newlines, use null characters after each filename. This is useful to avoid quoting problems when piping the result into other programs (for example, GNU grep, xargs and so on all have options to deal with this).

Normally, treescan will use heuristics to avoid most stat calls, which is what makes it so fast. This option forces it to stat every file.

This is only useful for the side effect of pulling the stat data into the cache. If your disk cache is big enough, it will be filled with file meta data after treescan is done, which can speed up subsequent commands considerably. Often, you can run treescan in parallel with other directory-scanning programs to speed them up.

Only lists directories, not file paths. This is useful if you quickly want a list of directories and their subdirectories.

Only list files, not directories. This is useful if you want to operate on all files in a hierarchy, and the directories would ony get in the way.

Regularly print some progress information to standard error. This is useful to get some progress information on long running tasks. Since the progress is printed to standard error, you can pipe the output of treescan into other programs as usual.

The --sync option can be used to make sure all the files/dirs in a tree are sync'ed to disk. For example this could be useful after unpacking an archive, to make sure the files hit the disk before deleting the archive file itself.

This applies a perl regular expression (see the perlre manpage) to all paths that would normally be printed and will only print matching paths.

The regular expression uses an /s (single line) modifier by default, so newlines are matched by ..