pyarrow.HadoopFileSystem.disk_usage

HadoopFileSystem.disk_usage(path)

Compute bytes used by all contents under indicated path in file tree.

Parameters:
pathstr

Can be a file path or directory.

Returns:
usageint