When you already have test data in a spreadsheet, importing loads it into a pool in one pass, and the same flow updates records that already exist. It is also the quickest way to give a new pool its columns. This article covers preparing the file, running the import, reading the preview, fixing problems, and exporting.
Importing records into an existing pool is open to everyone in the project; creating a new pool from a spreadsheet requires the Manage Data Pools permission.
Import data
An import is a short sequence: prepare the file, preview the plan, fix anything that was skipped, and apply.
Prepare the file
Use an .xlsx file. The first row holds the column names and each later row is one record. A few rules keep the import clean:
- Include a column named Data Class. It is required, with one value per record, because every record must have a class. It does not need to be unique. Records with same Data Class are "equivalent" from test consumption point of view.
- If the pool uses Record Scope & Variability, include an Environment column, a Variant column, or both, to match the dimensions that are on. In each cell, give one or more values separated by commas, or Any for a record that is the same everywhere.
- Match the remaining column names to the pool’s columns.
- Note: For encrypted columns, put the plain value in the file; ACCELQ encrypts it on import (see Encrypted columns).
Run the import
Open the pool and choose Import, then select your file. ACCELQ reads and validates it, then shows a plan: how many records will be added, how many updated, and how many skipped. Nothing has changed yet; the plan is a preview. When you are satisfied, choose Apply import to commit.
Nothing changes until you applyBecause the preview is a dry run, you can review it, download the problem rows, fix the file, and import again without having changed anything in the pool.
Read the preview
The preview leads with a one-line summary of new, updated and skipped counts, then shows the problems first and a sample of the accepted rows, with encrypted columns masked.
Fix skipped rows
Skipped rows are grouped by reason, with a Download rejected rows option so you can correct them in the file and import again. There are four reasons a row is skipped, which also serve as your troubleshooting reference:
| Reason | What it means | How to fix |
|---|---|---|
| Blank Data Class | A new record has no Data Class, which is required. | Add a Data Class value to the row. |
| ID not found | A row meant to update an existing record points to an ID that is not in the pool. | Correct the ID, or remove it to add the row as new. |
| Duplicate ID in file | Two rows in the same file share an ID. | Keep one row per ID. |
| Value over 1,000 characters | A single cell value exceeds the maximum length. | Shorten the value. |
Rows that are completely empty are ignored quietly and are not counted as skipped.
Update existing records or add new ones
The ID column decides what each row does. A row with an ID updates the record that has that ID; a row with a blank ID is added as a new record. One file can do both. Records that are currently in flight are never changed.
Updating keeps the record itself: its ID and its status stay as they are, so correcting a value never puts consumed data back into use by accident. When an import updates records in a single-use pool, applying it asks how to handle their status: keep their current status, or return them to Ready, which suits reissued records.
| Situation | What to do |
|---|---|
| Correcting or changing values on records already in the pool, such as an expired card date, a changed plan, or one column across many records | Export the pool, edit the file, and import it. The export includes the ID column, so each row updates its own record. |
| Adding new items, such as a fresh batch of SIMs | Import rows without an ID. Each one becomes a new record. |
| Replacing the whole data set, for example after a lower environment has been refreshed | Delete the pool’s existing records in the grid, then import the new data. Matching IDs does not help when the old records no longer apply. |
Keep the ID column when you mean to update. If the ID column is removed, or the file is a copy that never had it, every row is added as new. The pool then holds two records for the same item, and in a single-use pool two tests can be given the same one.
Encrypted columns on import
Supply encrypted columns as plain values in the file; ACCELQ encrypts them as they are stored. Exports never contain usable encrypted values; an encrypted column exports as a header only, with no values.
Handle the source file carefullyBecause you supply encrypted values as plain text, the import file, and any rejected-rows file you download from the preview, contains sensitive data. Store and dispose of it according to your organization’s policy.
Export data
Export a pool to .xlsx to review or edit its data offline, and reimport if needed with corrections. The export includes the ID column, so it can be imported back to update the same records. Encrypted columns export as headers only, so an export is safe to share but cannot be used to copy secrets.
Comments
0 comments
Please sign in to leave a comment.