histogram_timestamps

README

Histogram Timestamps

histogram_timestamps takes a series of timestamps (no need to be sorted) in any format and shows you an interactive histogram of those timestamps.

Use Case

You ever find yourself working with a dataset and you want to graph a particular aspect of that data over time, but due to lots of data and/or weird formats, there's no good tool to graph and visualize that data? This is the problem histogram_timestamps is meant to solve! If you can get the timestamps of anything, you can pipe them into histogram_timestamps to view trends in that data over time. Here are example questions answerable with histogram_timestamps:

• "Hmmmm, we had a lot of broken data created last night; how can I see when it started vs when it grew out of control? There's no metric for this broken data because it's caused by a bug, but I can find all the broken rows in the DB with a query. If only I could graph that somehow..."
  • Well with histogram_timestamps, you can! Run a select created_time from table ... command and pipe those timestamps straight into histogram_timestamps in order to instantly generate an interactive graph of all that data through time.

Installation

On any OS, you can install histogram_timestamps by downloading the latest binary for your platform from the latest release page, extracting that binary, then placing that binary into your $PATH or equivalent.

Linux (Bash)

# Installs in /usr/local/bin
curl --proto '=https' --tlsv1.2 -sSf https://raw.githubusercontent.com/lelandbatey/histogram_timestamps/master/install.sh | sudo bash

From source

You'll need Go 1.18+ and NPM installed to build from source, though the generated binary is totally stand-alone and can be transfered to other systems which do not have Go or NPM installed.

git clone https://github.com/lelandbatey/histogram_timestamps.git
cd histogram_timestamps
make install # Does require you have NPM installed to build the JS portion, but once pre-built the binary is totaly self-contained

Usage

As an example of usage, you can have histogram_timestamps generate some fake data which you feed back into histogram_timestamps. Examples:

./histogram_timestamps --generate-fake-data | ./histogram_timestamps
./histogram_timestamps --generate-fake-data | ./histogram_timestamps --unit minute
./histogram_timestamps --generate-fake-data | ./histogram_timestamps --unit hour