Raja Ongkir Destination

Crawl the Raja Ongkir Komerce destination API and emit PostgreSQL schema.sql + seed.sql for Indonesian provinces, cities/regencies, districts, and sub-districts.

Requirements

• Bun
• Raja Ongkir API key (Key header)
• PostgreSQL (to load the generated SQL)

Setup

bun install
cp .env.example .env
# set RAJAONGKIR_API_KEY in .env

Extract

bun run extract

Outputs (default ./sql-out/):

File	Contents
schema.sql	DDL for your configured table names
seed.sql	TRUNCATE those tables, then batched INSERTs

Load into Postgres:

psql "$DATABASE_URL" -f sql-out/schema.sql
psql "$DATABASE_URL" -f sql-out/seed.sql

seed.sql only truncates the four destination tables (names from env) before inserting.

Options

bun run extract -- --dry-run              # smoke test (shallow crawl)
bun run extract -- --max-provinces 2
bun run extract -- --province-start 11 --province-end 11
bun run scripts/extract-raja-ongkir.ts --help

Variable	Default	Purpose
RAJAONGKIR_API_KEY	—	API key (required)
OUT_DIR	./sql-out	Output directory
REQUEST_DELAY_MS	100	Delay between API requests
BATCH_SIZE	750	Rows per INSERT in seed.sql
MAX_RETRIES	3	Retries per failed request
TABLE_PREFIX	raja_	Prefix for default table names
TABLE_PROVINCES	{prefix}provinces	Override province table name
TABLE_CITIES	{prefix}cities	Override city table name
TABLE_DISTRICTS	{prefix}districts	Override district table name
TABLE_SUB_DISTRICTS	{prefix}sub_districts	Override sub-district table name

Notes

• Full crawl issues thousands of sequential requests; expect a long run. If you hit rate limits, raise REQUEST_DELAY_MS.
• API flow: province → city/:provinceId → district/:cityId → sub-district/:districtId.

bun run typecheck