DataCat

Getting Started

This guide takes you from a new DataCat installation to a first local SQLite dataset.

Before you begin

You need:

  1. A Mac running a supported macOS version.
  2. A writable folder for DataCat datasets.
  3. One supported Provider configuration.
  4. Provider credentials only when the chosen Provider requires them.

DataCat supports two Providers:

  1. CoinAPI — enter your CoinAPI credentials.
  2. Binance Global Spot — uses public Spot market data and does not require a Binance account or credentials.

Choose a dataset folder

Open Settings and choose the default dataset folder. New Jobs inherit this location unless you choose a per-Job override.

Keep the generated dataset folder intact. A working SQLite dataset can include the main .db file and SQLite -wal and -shm companion files. Moving or deleting only one file while a Job is active can make the dataset unavailable or inconsistent.

Add a Provider

Open Providers and add the source you want to use.

CoinAPI

Enter your own CoinAPI credentials, select the CoinAPI tier that matches your provider account, and review the Finality delay. DataCat uses the tier for request orchestration. The Finality delay controls when recent CoinAPI REST OHLCV is eligible for confirmation; it does not delay WebSocket Live delivery.

Binance Global Spot

Add Binance Global Spot without credentials. This Provider is limited to public Spot Market Data. DataCat does not use Binance trading, account, Margin, Futures, Options, or regional-platform endpoints in the current integration.

Wait until the Provider has loaded the metadata needed by the Job editor. If Venue, Symbol, or Timeframe controls are unavailable, finish or retry Provider setup before creating the Job.

Optionally create an Indicator Set

Open Indicator Sets if you want calculated technical indicators in the dataset.

  1. Create a named set.
  2. Add one or more indicator types.
  3. Review each Indicator Instance and its parameters.
  4. Save the set.

An empty indicator selection is valid. Choose No Indicator Set when you want an OHLCV-only dataset.

Create a Job

Open Jobs and create a new Draft. Configure the available fields:

  1. Provider.
  2. Venue when the Provider exposes a meaningful choice.
  3. Symbol.
  4. Timeframe.
  5. Mode.
  6. History dates when the selected mode includes History.
  7. Indicator Set or No Indicator Set.
  8. Dataset folder or size-limit overrides when needed.

All Job date and time values are UTC. History ranges use closed, timeframe-aligned bars. From identifies the first requested bar open and To identifies the last requested bar close.

Choose a mode

History

Use History for a bounded historical range. DataCat fetches the selected range, writes it to SQLite, performs the applicable provider confirmation and recovery work, and stops.

Live

Use Live to receive each newly completed bar from the Provider and write it to the dataset. Open or in-progress WebSocket updates are not written as completed dataset bars. Recently completed bars may remain pending until provider-specific REST confirmation.

History + Live

Use History + Live when you need historical context and ongoing updates in one dataset. While DataCat loads the selected historical range, it also keeps receiving newly completed bars so a long History request does not intentionally leave a gap.

Start the Job

Resolve any validation messages and select Start Job. At start, DataCat freezes the Job configuration, including the selected Indicator Set instances and provider-specific settings captured by the Job.

Later edits to the reusable Indicator Set or Provider configuration do not mutate a Job that has already started.

Monitor progress and data quality

Job Cards and the Inspector show operational progress and issue states.

  1. History progress describes fetch and write progress, not only provider request count.
  2. Awaiting final bars means recent provider data still has confirmation work.
  3. Gaps identifies actionable completeness work under the active provider policy.
  4. Last Bar identifies the latest persisted bar available to the Job presentation.

A History Job can reach 100.0% historical loading while still showing Awaiting final bars. Loading the requested range and confirming recent provider data are separate stages.

Open the dataset

After the Job starts and DataCat receives its first data, the dataset appears in the selected folder. Use the dataset-folder action to reveal it. Start with these tables:

  1. meta for provenance and requested coverage.
  2. bars for canonical OHLCV periods and provider-data state.
  3. gaps for missing-period and recovery history.
  4. indicator_definitions and indicators tables when the Job uses indicators.

Continue with Datasets for artifact handling and Dataset Reference for SQL examples.

For a useful first check, choose a liquid symbol, a moderate timeframe, a short History range, and an Indicator Set with several familiar indicators, such as SMA and RSI. Confirm that rows appear in bars, use indicator_definitions to identify the generated output columns, and inspect the indicators table after the initial warmup rows. Then expand the range or Indicator Set in a new Job.

Search across 46 documentation pages.