Import & Export

Import turns rows from a spreadsheet or PDF into artefacts inside a source. Export takes those artefacts back out as a downloadable file. Both run as background jobs, so large files don't freeze your browser.

Import runs in the import workbench - a full page inside Modules, not a pop-up dialog. Because it's a page, you can refresh it, link to it, and leave it to do something else without losing your place: the workbench autosaves your choices and keeps the import waiting for you under Unfinished imports in the Modules drawer.

The import workbench

When you start an import, Integrity opens the workbench at its own address (for example …/sources/imports/{job}). While you work:

• The Modules tree stays available from the left drawer, so you can see your sources and switch away at any time.
• A pinned header shows the import title; a pinned footer shows the autosave status.
• The page is refreshable and shareable - reloading resumes the same step, and you land back where you left off.

Your decisions are saved as you go, so there's no "Save draft" button to remember. An import in progress is just an import you haven't finished yet.

Start an import AdminA EditorE

There are two ways in, depending on whether the data belongs in a brand-new source or an existing one.

Import as a new source

1. In Modules, use the root or folder table action row to add to the Source Library.
2. Choose the Import option and pick your file.
3. The workbench opens and walks you through naming the source and shaping its fields.

Import into an existing source

1. Open the source you want to add rows to.
2. In the source action controls, choose Import.
3. The workbench opens already pointed at that source, so you only map your columns onto its existing fields.

Both paths accept Excel (.xlsx), CSV (.csv), and PDF (.pdf) files.

Work through the import steps

The workbench keeps the same ordered steps for every import:

1. Upload — choose the file. Once a file is attached, it's bound to this import (see the FAQ on changing files).
2. Sheet select — for multi-sheet Excel files, pick the sheet to import.
3. Configure — tell Integrity which row holds the column headers and choose how artefacts are identified (auto-generated from a prefix, or taken from one of the columns).
4. Field types (new source) — choose which columns to import, rename them, and review the type Integrity inferred for each (text, number, date, select, and so on).
5. Map columns (existing source) — match each incoming column to a field that already exists in the source, create a new field for it, or mark it "Don't import".
6. Import preview (existing source) — see how many rows will be created, updated, or skipped before anything changes.
7. Readiness — review any warnings (for example, unsupported values) and acknowledge them.
8. Confirm — start the import. Progress runs in the background, and you can watch it on the source.

PDF imports replace the spreadsheet-specific steps with a PDF progress step while the parser extracts a table from the document; the rest of the flow is the same.

Readiness reference states

When readiness, preview, or completed import reports can resolve an existing Integrity source or artefact, they use the same status vocabulary as the rest of the app: Active, In Trash, Deleted, Restricted, or Unknown. Restricted references do not reveal hidden names, identifiers, colors, or source context.

External package IDs, old-system URIs, and lineage references are treated as external evidence. Integrity does not guess that an external reference is in Trash or Deleted unless it maps to a known Integrity entity with explicit status data.

Identifier matching and blocking

Artefact identifiers are unique across the whole workspace. Import preview and readiness use that rule before an import runs, so unsafe rows are blocked while the data is still easy to fix.

When importing into an existing source:

• A row whose identifier already exists in the target source can be treated as an update candidate.
• A row whose identifier exists in another source is a workspace collision and cannot be created as a new artefact.
• The Create all option is unavailable when it would duplicate target-source or other-source identifiers.
• Readiness findings show the file row and imported identifier that need attention for the active importer.

When importing as a new source, any imported identifier that already exists in the workspace is a collision, because there is no target-source artefact to update.

Leave and resume unfinished imports AdminA EditorE

You can stop in the middle of any import and come back later.

• Leaving is safe. Once the footer shows Saved, you can navigate away or close the tab without losing your choices. Integrity only warns you about leaving while a save is still Saving or has failed - the two cases where unsaved changes could be lost.
• Find it again under the Unfinished imports card in the Modules drawer. Each row shows the import's name (the target source for an existing-source import, the new source's name once you've set it, or the file name otherwise) and the step you're on.
• Resume by clicking the row. The workbench reopens at the saved step with all your choices restored.

The footer reports the autosave status as you work - Not saved yet, Saving, Saved, Save failed, Retrying save, or Recovered - and offers a Retry action if a save fails.

Discard an import AdminA EditorE

If you don't want to finish an import, discard it:

• From the workbench, click Discard in the header.
• From the Modules drawer, open an Unfinished imports row's actions menu and choose Discard.

Either way, Integrity asks you to confirm, then removes the unfinished import and cleans up everything behind it - the uploaded file, cached previews, and any parser by-products. Discarded imports don't come back after a refresh.

When an import completes

Finishing an import is a clean handoff from "in progress" to "real data":

• The import disappears from Unfinished imports automatically - there's nothing to discard afterward.
• Integrity selects the source you imported into. For a new-source import, it selects the newly created source once it's ready.
• The completed import stays on the record as durable evidence through the import job/report trail and linked artefact versions, so completion never erases the audit trail - it only clears the working draft.

If you're removed from a workspace, your unfinished imports for that workspace are discarded and cleaned up automatically.

Export your data AdminA EditorE ViewerV

Export takes artefacts out of a source as a file:

1. Open the source and choose Export.
2. Pick a format — Excel (.xlsx), CSV (.csv), JSON (.json), or Markdown (.md).
3. Choose which columns to include and, optionally, narrow the rows with the source's filters.
4. Start the export. Integrity generates the file in the background and shows Export ready with the file size when it's done.
5. Click to download.

Export files are available for 24 hours after they're generated, and each download link is signed for a few minutes for security. Run a fresh export if an old link has expired.

FAQ

Will I lose my progress if I close the tab?

No, as long as the footer shows Saved. The workbench autosaves your choices, so you can leave and resume from Unfinished imports. Integrity only warns you about leaving while a save is in progress or has failed.

Can my teammates or admins see my unfinished imports?

No. Unfinished imports are private to the person who started them. There's no shared or admin-wide queue of other people's in-progress imports.

Can I change the file after I've started an import?

No - once a file is attached to an import, it's bound to that import, so the Back button on the first step is disabled. To use a different file, start a new import; you can discard the old one if you no longer need it.

What file types can I import?

Excel (.xlsx), CSV (.csv), and PDF (.pdf). PDF import extracts a table from the document and is best-effort — check the preview before confirming.

What happens to my unfinished imports if I lose access?

If you lose edit access to a restricted target source, the unfinished import stops being resumable and won't reveal that source's details. If you're removed from the workspace, your unfinished imports there are discarded and cleaned up.

Why is Create all disabled?

Create all is disabled when preview or readiness found identifiers that would duplicate existing workspace artefacts. Use update/skip for target-source matches, or change the imported identifier before creating a new artefact.

Why can't a viewer import?

Importing adds or changes rows, which is an editor action. Viewers can read and export a source, but they can't import into it.

Related