Unix command history distribution

histdist

#!/usr/bin/env python
from os.path import expanduser
from collections import Counter

distribution = Counter()

f = open(expanduser("~/.bash_history"))
commands = list()
for line in f.readlines():
    line = line.replace("\n","").split(" ")
    for i in range(1, len(line) + 1):
        commands.append(" ".join(line[:i]))
f.close()

for command in commands:
    distribution[command] += 1

print "Cnt:\tCmd:"
for cmd,cnt in sorted(distribution.items(), key=lambda x: x[1] * len(x[0])):
    print "%s\t'%s'" % (cnt, cmd)

Cool idea. If you can use Python2.7, collections.Counter can cut that code down a little.

from collections import Counter

distribution = Counter()

for command in commands:
    distribution[command] + 1

for cmd, cnt in distribution.most_common():
    print cnt, cmd

http://docs.python.org/2/library/collections.html#collections.Counter

Of course you need to use += instead of +, my mistake.

Thanks, I made an update!

In bash:

# weighted sort by command length and count
history | awk '{ print $2 }' | sort | uniq -c | awk '{ print $1+$1+length($2), $1, $2 }' | sort -n | awk '{ print $2, $3 }'

# sort by command count
history | awk '{ print $2 }' | sort | uniq -c | sort -n