IntoSQL for Python

Convert spreadsheets (XLSX, XLS, ODS), CSV and JSON into SQL for PostgreSQL, MySQL, MariaDB, SQLite, SQL Server or Oracle with the IntoSQL API. Column types are inferred and dirty values fixed, so the script loads on the first try.

pip install intosql

Python 3.9 or newer. The API is part of IntoSQL Pro; make a key in Settings, API keys.

Convert a file

import os
from intosql import IntoSQL, spec

intosql = IntoSQL(os.environ["INTOSQL_API_KEY"])

conversion = intosql.convert(
    "ventas.xlsx",
    name="ventas.xlsx",
    spec=spec("postgresql", table="ventas"),
)

print(conversion["report"]["rows"]["converted"], conversion["output"]["url"])

The file streams to storage as it is; the output URL downloads the script for ten minutes. Read the conversion again (intosql.conversions.retrieve(id)) for a fresh one.

Convert a file by URL

conversion = intosql.convert_url(
    "https://example.com/exports/ventas.csv",
    spec=spec("mysql"),
)

Check a webhook

Save a webhook address in Settings, API keys, and IntoSQL calls it when a conversion is done, failed or canceled.

from intosql import parse_webhook

# Raises WebhookSignatureError when the signature doesn't match.
event = parse_webhook(
    request.get_data(),
    request.headers.get("IntoSQL-Signature", ""),
    os.environ["INTOSQL_WEBHOOK_SECRET"],
)

Errors and retries

Failed requests raise IntoSQLError with the HTTP status, a code that doesn't change (needs_pro, over_quota, too_big, invalid_spec, invalid_url and more) and the request_id. A conversion that ends as failed raises ConversionFailedError with its code and the engine's problems. Rate limits, a full account, server errors and lost connections are retried twice with the same idempotency key (max_retries changes it); already_started never is.

Every endpoint: intosql.com/developers. This SDK: intosql.com/docs/sdks/python.