janapada

Track how administrative boundaries split, merge, and rename over time. Each unit's lineage is color-coded forward from its earliest known boundary.

How it works

1. Provide a transitions CSV: a wide table where each row is one unit's journey through time, columns are census years.
2. Provide GeoJSON boundary files: one per year, or shared across years.
3. Run janapada process with a janapada.yml config. It matches the CSV to the GeoJSON features and writes evolution.json.
4. Drop EvolutionMap.tsx into a React app and pass it evolution.json and the GeoJSON URLs.

Requirements

Node.js 20+. Install via nvm:

nvm install --lts
nvm use --lts

Or download directly from nodejs.org.

Quickstart

npm install
node cli/process.cjs examples/india-districts/janapada.yml
npm run dev

npm run dev copies the processed data into viewer/public/ and starts the Vite dev server.

Using your own dataset

1. Copy the example config:

   cp janapada.yml.example mydata/janapada.yml

2. Edit mydata/janapada.yml to set levels, years, transitions, and sources.
3. Process:

   node cli/process.cjs mydata/janapada.yml
   npm run dev -- mydata/janapada.yml

Transition CSV format

One row per lineage chain, columns follow {year}-{level1} / {year}-{level2}:

1951-State,1951-District,1961-State,1961-District,2001-State,2001-District,2011-State,2011-District,2024-State,2024-District
Madras,Srikakulam,Andhra Pradesh,Srikakulam,Andhra Pradesh,Srikakulam,Andhra Pradesh,Srikakulam,Andhra Pradesh,Srikakulam
Madras,Srikakulam,Andhra Pradesh,Srikakulam,Andhra Pradesh,Srikakulam,Andhra Pradesh,Srikakulam,Andhra Pradesh,Parvathipuram Manyam

The two rows above both originate from Srikakulam in 1951. By 2024 it split: one row stays Srikakulam, the other becomes Parvathipuram Manyam.

• Empty cells mean the unit didn't exist that year.
• Splits: one source row, multiple destination rows sharing that source.
• Merges: multiple source rows converging to the same destination.
• Level names (state, district) are set in the config and can be anything.

Config reference (janapada.yml)

name: "India Districts 1951–2024"

levels:
  level1: state      # parent unit (province, region, …)
  level2: district   # child unit (municipality, county, …)

years: [1951, 1961, 1971, 1981, 1991, 2001, 2011, 2024]

transitions: ./data/transitions.csv

sources:
  - year: 1951
    geojson: ./data/1951.geojson
    key: district_name      # GeoJSON property matching level-2 values
    parent_key: state_name  # GeoJSON property matching level-1 values (optional)
    parent_filter: "Bengal" # restrict to features where parent_key contains this (optional)
  - year: 2024
    geojson: ./data/2011.geojson  # reuse a file for multiple years

fuzzy_match: true  # tolerate spelling variations

level1_aliases:
  "nct of delhi": "delhi"

level2_corrections:
  2011:
    "north twenty four parganas": "North 24 Parganas"

output: ./evolution.json

palette:
  - "#e63946"
  - "#f4a261"

Viewer

import { EvolutionMap } from './EvolutionMap';

<EvolutionMap
  evolutionFile="/evolution.json"
  geojsonSources={[
    { year: 1951, url: '/India-1951-districts.geojson', key: 'district_name', parentKey: 'state_name' },
    { year: 1961, url: '/India-1961-districts.geojson', key: 'district_name', parentKey: 'state_name' },
    { year: 2024, url: '/India-2011-districts.geojson', key: 'district_name', parentKey: 'state_name' },
  ]}
  darkMode={false}
/>

Fetch evolution.json and GeoJSON at runtime. Serve them from public/ or any static host.

Props

Prop	Type	Default	Description
evolutionFile	string		URL to evolution.json
geojsonSources	GeoJSONSource[]		{ year, url, key, parentKey?, parentFilter? } per year
darkMode	boolean	false
initialMode	string	'grid'	Opening tab: 'grid', 'single', or 'district'
defaultDistrict	string		Pre-loaded district in the "By district" tab
noOriginLabel	string		Legend label for units with no origin-year ancestor

License

MIT

जनपद (Sanskrit janpada), "foothold of a clan"