Skip to content
DataSquaresDocs

Connection flows explained

Every connector in the wizard uses one of a small set of connection flows — the form you fill in. Once you’ve connected one host/port database, you’ve effectively learned them all. This page explains each flow once, so the per-connector pages can stay short.

Host / port databases

Section titled “Host / port databases”

The classic form: host, port, database, username, password, plus a TLS toggle. It covers nearly every traditional database and warehouse — PostgreSQL, MySQL, MariaDB, SQL Server, Azure SQL, Oracle, Redshift, ClickHouse, Trino, IBM Db2, and many more. The port is prefilled with the engine’s default when you pick the connector.

Many engines are wire-compatible with a major protocol, so they connect exactly like their parent: TimescaleDB, CockroachDB, Greenplum, and YugabyteDB speak PostgreSQL; SingleStore and TiDB speak MySQL; Azure Synapse and Fabric Warehouse speak TDS like SQL Server. The connector page tells you when this applies.

Cloud warehouses with their own forms

Section titled “Cloud warehouses with their own forms”

A few warehouses authenticate their own way and get a dedicated short form:

  • Snowflake — account identifier + virtual warehouse (+ optional role).
  • Databricks — workspace host + SQL Warehouse HTTP path + access token.
  • BigQuery — GCP project + service-account JSON key.
  • Amazon Athena — AWS region + IAM keys + S3 result staging location.
  • DuckDB / MotherDuck — a DSN (:memory: or md:<database>) + token.

File uploads

Section titled “File uploads”

CSV, Excel (.xlsx), SQLite, Parquet, and JSON/NDJSON files upload directly — no credentials. The file is imported into a staging area in your workspace and becomes queryable immediately: one table for CSV/Parquet/JSON, one table per sheet for Excel, and every table for SQLite.

API pulls

Section titled “API pulls”
  • Google Sheets — share the sheet with a service account, and every tab imports as a table.
  • REST API — point at a JSON endpoint (with optional auth headers and a records path) and the response is staged as a table. Re-sync to refresh.

Credentials & security

Section titled “Credentials & security”
  • Connection credentials are encrypted at rest and never leave the server.
  • DataSquares only reads from your sources — grant the connection user read-only access as a matter of hygiene.
  • Prefer TLS whenever your server supports it, and always for connections over the public internet.
Section titled “Related”
© 2026 DataSquares