anytop 0.3.0

Overview

anytop is a tool for viewing frequency distributions over streaming input. It reads input line by line, and shows the top elements in the distribution as they stream in. It is inspired by the excellent command-line tools available for the Varnish web accelerator.

anytop is designed to work within a shell flow, allowing you to easily modify the data streaming in with tools such as cut, sed and tr. Get usage help by typing anytop --help. To exit anytop, type CTRL-C.

Example: dictionary words

Get the distribution of word starting with each different letter from the dictionary:

cut -c 1-1 /usr/share/dict/words | tr [:upper:] [:lower:] | anytop

Looks like s, p and c have the most words.

Example: file types

Work out the relative distribution of file extensions in a source tree:

ack -f | fgrep . | awk -F . '{print $NF}' | anytop

Looks like a PHP codebase for a web app.

Example: shell commands

See what commands you use most often in bash:

cut -d ' ' -f 1 .bash_history | xargs -n 1 basename | anytop

Developer tools vim, git and j (autojump) feature pretty prominently. Maybe I type ls a little too much.

As you can see, anytop lends itself handily to shell pipelines, allowing it to be useful in a wide variety of situations.

Memory usage

Anytop uses memory proportional to the number of distinct lines in the input. If the input keyspace is bounded, then anytop will use limited memory, no matter how many lines or how long it runs.

When the input keyspace is not bounded, memory use can still be bounded by only displaying statistics on a fixed-size window of lines with the -l option.