File size distribution: Difference between revisions

Content added Content deleted

Inline

@@ Line 1,605: / Line 1,605: @@
 {{works with|Bourne Shell}}
 Use POSIX conformant code unless the environment variable GNU is set to anything not empty.
-<lang sh>#!/bin/sh
+<lang sh>
+#!/bin/sh
+set -eu
+#!/bin/sh
 set -eu
 if [ ${GNU:-} ]
 then
-    find -- "${1:-.}" -type f -exec du -b -- {} +
+	find -- "${1:-.}" -type f -exec du -b -- {} +
 else
-    # Use a subshell to remove the last "total" line per each ARG_MAX
+	# Use a subshell to remove the last "total" line per each ARG_MAX
-    find -- "${1:-.}" -type f -exec sh -c 'wc -c -- "$@" | sed \$d' argv0 {} +
+	find -- "${1:-.}" -type f -exec sh -c 'wc -c -- "$@" | sed \$d' argv0 {} +
 fi | awk '
+	{
-    {
-        ++hist[$1 ? length($1) - 1 : -1]
+		++hist[$1 ? length($1) - 1 : -1]
-        total += $1
+		total += $1
+	}
-    }
-    END {
+	END {
+		print total, NR
-        for (i in hist)
+		for (i in hist)
-            print (i == -1 ? 0 : "1e" i) "\t" hist[i]
+			print i, hist[i]
-        print "Total: " total " bytes in " NR " files"
-    }' | sort</lang>
+	}' | \
+		{
+			read total
+			tabs -8
+			sort -n | awk -vtotal="$total" -vOFS='\t' '
+				BEGIN {
+					split("KB MB GB TB PB", u); u[0] = "B"
+					print "From", "To", "Count\n"
+				}
+				$1 == -1 {print "0B", "0B", $2; next}
+				{
+					print 10 ** ($1		  % 3) u[int($1		  / 3)],
+** (($1 + 1) % 3) u[int(($1 + 1) / 3)],
+						  $2
+				}
+				END {
+					$0 = total
+					l = length($1) - 1
+					printf "\nTotal: %.1f %s in %d files\n",
+						$1 / (10 ** l), u[int(l / 3)], $2}'
+		}</lang>
 {{out}}
 <pre>$ time ~/fsd.sh
-4
+From    To      Count
-e0     66
-e1     66
+B      0B      13
-e2     1418
+B      10B     74
-e3     1026
+B     100B    269
-e4     1564
+B    1KB     5894
-e5     60083
+KB     10KB    12727
-e6     16282
+KB    100KB   12755
-e7     3881
+KB   1MB     110922
-e8     1444
+MB     10MB    50019
-e9     16
+MB    100MB   17706
+MB   1GB     5056
-Total: 612404756079 bytes in 85850 files
+GB     10GB    1139
-~/fsd.sh  0.60s user 0.98s system 134% cpu 1.182 total
+GB    100GB   141
+GB   1TB     1
+Total: 8.9 TB in 216716 files
+~/fsd.sh  1.28s user 2.55s system 134% cpu 2.842 total
 $ time GNU=1 ~/fsd.sh
-4
+From    To      Count
-e0     66
-e1     66
+B      0B      13
-e2     1418
+B      10B     74
-e3     1026
+B     100B    269
-e4     1564
+B    1KB     5894
-e5     60083
+KB     10KB    12727
-e6     16282
+KB    100KB   12755
-e7     3881
+KB   1MB     110922
-e8     1444
+MB     10MB    50019
-e9     16
+MB    100MB   17706
+MB   1GB     5056
-Total: 612404756079 bytes in 85850 files
+GB     10GB    1139
-GNU=1 ~/fsd.sh  0.35s user 0.48s system 135% cpu 0.613 total</pre>
+GB    100GB   141
+GB   1TB     1
+Total: 8.9 TB in 216716 files
+GNU=1 ~/fsd.sh  0.81s user 1.33s system 135% cpu 1.586 total</pre>
 =={{header|zkl}}==