Skip to content

snejus/rich-tables

Repository files navigation

Coverage Status

Rich tables

JSON human-prettifier based on the brilliant rich python library.

Since my usual day involves building and interacting with various APIs, JSON data is constantly flying around the terminal. Reading JSON data (even when it's prettified by, say, jq) requires a fair bit of mental effort; and it's basically impossible to analyse / make sense of it just by looking at it.

This project initially started as a way to solve this issue – it takes JSON data as an input and prints it in (rich-)tables, making it somewhat more readable for humans. With time, it's become the main handler for most of structured data that gets displayed in my terminal and is now one of my core every day tools.

Releases

It's WIP but installable through pip: pip install rich-tables – feel free test it but do not expect it to be stable yet.

If we have enough interest, I am more than happy to prepare a release with a somewhat stable API.

Some bits to be aware of

Most commits are made by a daily cron job

You will find commits named

Automatic backup <date>

There is a daily cronjob which checks my local copy of the repository for changes, commits, and pushes them upstream.

It got setup because this entire thing got built without intention – whenever I came across data that was not handled by rich-tables, I opened the code, quickly added the logic and continued on with the original work without committing. Eventually, I'd have to deal with a huge diff with a lot of unrelated changes.

Tests simply attempt to render data in each test case without making any assertions
  • At least at this point I only want to know whether every piece of data is rendered fine.
For each tests/json/<name>.json test case, the rendered output is saved to svgs/<name>.svg file
  • Allows to see the visual difference a certain change makes
  • Helps to detect unintended side effects: if we're expecting an update of album.svg only but see that pr.svg is also updated, we know something's not right
  • In commit details, GitHub shows visual difference for svg file changes
    • This is very helpful when one is trying to track down the culprit behind some missing border
README is populated with pictures dynamically when tests pass
  • Once tests finish successfully, a subsection is added for every picture in the svgs/ folder (in the alphabetical order). case.svg would be found under the subsection Case.
  • This logic lives in a session-scoped pytest fixture, after the yield statement.

Examples

Album

image

Calendar

image

Diff

image

Emails

image

Hue

image

Jira diff

image

Nested JSON

image

Pr

image

Simple JSON

image

Sql

image

Tasks

image

Timed

image