Jobs

A Job is DataCat’s unit of dataset work. It combines one Provider context, symbol, timeframe, mode, output location, and optional frozen Indicator Set configuration.

Draft configuration

A new Job starts as an editable Draft with its own canonical Job ID. Its display name and Draft sequence are for orientation only; they do not define persistent identity.

The available controls depend on Provider capabilities. A fixed single-venue Provider can omit Venue without removing venue provenance from the eventual dataset.

Job modes

History

History requests a bounded UTC range and stops after the requested loading, finality, and applicable recovery work reaches a terminal result. Use it for backtests, research snapshots, and reproducible bounded imports.

Live

Live begins from the current provider context and continues until stopped or failed. It persists completed bars, not every open-period update. Recently completed bars can remain pending until REST confirmation.

History + Live

History + Live combines ongoing collection with a historical backfill. DataCat starts the Live side first, records the anchor, and then requests History. After backfill, the Card switches to Live presentation even if provider finality work is still draining.

History ranges

DataCat displays and stores Job dates in UTC. History-capable modes require timeframe-aligned closed periods.

1. From is the open time of the first requested bar.
2. To is the close time of the last requested bar.
3. A History range must include at least one complete timeframe.
4. The UI excludes a still-forming current bar.
5. Provider availability metadata may be informative but is not always a precise OHLCV boundary.

Large ranges can require multiple provider requests and substantial local writes. Receiving the first data quickly does not imply that an arbitrary historical range will finish in the same amount of time.

Indicator snapshot

When a Job starts, DataCat copies the selected Indicator Set configuration into the Job. That copy stays unchanged for the lifetime of the Job.

Later edits to the reusable Indicator Set affect new Jobs only. Jobs that have already started keep the indicator configuration they started with, so each dataset continues to describe the calculations used for its values.

Lifecycle

Draft

The Job can be edited and validated but has not started ingestion.

In Progress

The Job is loading History, collecting Live data, performing History + Live work, or completing allowed stopping/finality operations.

Stopping

After you select Stop, DataCat stops the normal History or Live data flow. If a dataset write is already in progress, DataCat finishes it safely. It then completes the provider-finality checks required at the stopping boundary before changing the Job to Stopped.

Stopped

The Job no longer ingests. Its dataset remains available for inspection. Stop is not Pause: DataCat does not resume the same running process in place.

Failed

A provider, dataset, configuration, or runtime error prevented the Job from continuing. Open Details to read the error. If the cause is unclear or the problem happens again, email [email protected] or open Settings → Feedback & Diagnostics → Send Feedback in DataCat.

Reading progress and issues

History progress covers fetch and write work. It can show 100.0% while the Job still has Awaiting final bars, because historical loading and provider-data finality are separate dimensions.

Common issue states include:

1. Awaiting final bars: recent periods still require provider confirmation.
2. Gaps: actionable completeness work remains.
3. Rate limit: provider pacing or cooldown affects progress.
4. Dataset missing: the dataset artifact cannot be found.
5. Dataset access: DataCat cannot currently access the artifact.
6. Size limit or Size limit reached: the dataset is near or at its configured limit.

Continue in New Job

Continue in New Job appears at the top of the selected Job’s Inspector when the source Job is Stopping, Stopped, or Failed. Select it to create a new editable Draft based on the source Job. The source Job and its dataset stay unchanged.

The new Draft receives a new Job ID. When you start it, DataCat creates a new dataset with its own Dataset ID. The new dataset records the Job and Dataset it was continued from. If you continue several times, each new dataset points to its immediate source.

For a historical range, the new Draft begins at the first timeframe-aligned period after the source Job’s requested coverage. DataCat uses the source Job’s saved request range for this boundary rather than guessing from the last row written to SQLite.

Dataset availability

An active Job requires its dataset to remain available and writable. Avoid moving, renaming, deleting, or denying access to the dataset folder while the Job is active. Older stopped Jobs can still be inspected through the filesystem even when DataCat can no longer resolve the original location.

Removing a Job

Delete Job removes the Job from the Jobs list. It does not delete its dataset folder or SQLite file. Before deleting the Job, reveal the dataset in Finder or note its location if you plan to keep using it. Delete the dataset files separately in Finder only when you no longer need them.