Export a D1 database to a .sql file and re-import it (backup, restore, or migrate data between databases) via wrangler and/or the REST API.

Documented steps

1. Full export (schema+data) of a remote database: npx wrangler d1 export <database_name> --remote --output=./database.sql. Doc: https://developers.cloudflare.com/d1/best-practices/import-export-data/
2. Narrow the export with flags: --table=<table_name> to scope to one table, --no-data for schema only, --no-schema for data only (combine as needed, e.g. schema-only single table).
3. Re-import the exported (or any) .sql file into a database: npx wrangler d1 execute <database_name> --remote --file=./database.sql (drop --remote, i.e. use --local, to load into your local dev DB instead).
4. If starting from a raw SQLite3 dump, convert it first with `sqlite3 db_dump.sqlite3 .dump > db.sql`, then remove the BEGIN TRANSACTION/COMMIT lines and any `CREATE TABLE _cf_KV (...)` statement before running d1 execute --file.
5. To export programmatically without wrangler: POST /accounts/{account_id}/d1/database/{database_id}/export with {"output_format":"polling"}; the response includes at_bookmark and status. Doc: https://developers.cloudflare.com/api/resources/d1/subresources/database/methods/export/
6. Poll the same export endpoint, passing back {"output_format":"polling","current_bookmark": at_bookmark}, until result.status is "complete"; then download the file from result.result.signed_url (valid for one hour).
7. To import programmatically via REST: POST .../import with {"action":"init","etag":"<md5-of-file>"} to receive an upload_url, PUT your file there, POST {"action":"ingest","etag":...,"filename":...}, then poll with {"action":"poll","current_bookmark":...} until status is "complete". Doc: https://developers.cloudflare.com/api/resources/d1/subresources/database/methods/import/
8. Verify the result: npx wrangler d1 execute <database_name> --remote --command="SELECT name FROM sqlite_schema WHERE type='table' ORDER BY name;"