README.md

Example gallery

examples.json is the whole Examples gallery. Edit it to add, remove or reorder examples; no code change is needed. Each entry is a small notebook:

{
  "id": "filter-rows",                  // unique, letters/digits/dashes; used in ?example= links
  "title": "Filtering Rows",
  "description": "Keep only rows that match a condition",
  "category": "filtering",              // must be one of the ids in "categories"
  "operations": ["where"],              // Table methods shown as badges on the card
  "markdown": "## Filtering rows ...",  // the note above the code (Markdown)
  "cells": [                            // one string per code cell, run top to bottom
    "from datascience import *",
    "students = Table().with_columns(...)\nstudents",
    "cs_students = students.where('Major', 'CS')\ncs_students"
  ]
}

Conventions that make examples read well in the tool:

• Start with the import cell (from datascience import *, plus import numpy as np and the matplotlib setup when the example plots), exactly as lecture notebooks do.
• End a cell with the bare table name (students) rather than print or .show(); the value of a trailing expression is shown under the cell, as in Jupyter.
• Keep tables to about 4 to 8 rows so every row fits in the visualization.
• For anything random (sample, shuffle, split) seed numpy in the import cell so the walkthrough matches what students see.
• Do not use em dashes in notes.

npm test validates the file (unique ids, known categories, non-empty cells).