little-things/warc-size

#!/bin/bash
# Total size of all WARCs in the current directory (or subdirectories)
find -name '*.warc.gz' -printf '%s\n' | awk 'BEGIN { units[0] = "B"; units[1] = "KiB"; units[2] = "MiB"; units[3] = "GiB"; units[4] = "TiB"; units[5] = "PiB"; }  { size += $1 }  END { if (size > 0) { magnitude = int(log(size) / log(1024)); if (magnitude > 5) { magnitude = 5; } } else { magnitude = 0; } if (magnitude > 0) { sizeformat = "%.2f"; } else { sizeformat = "%d"; }  printf sizeformat " %s\n", size / (1024 ^ magnitude), units[magnitude]; }'