Operations and insights

Importing data

Load JSON, CSV or Excel files into any table through a three-step wizard with preview, per-column type mapping, renaming, batch sizing and live progress.

Supported files

Format	Rules
JSON	An array of objects (a single object is treated as one row). Columns are the union of keys across the first 200 rows. Nested objects and arrays are kept as maps and lists.
CSV	Comma-separated, UTF-8, first line as headers, double-quote quoting with doubled quotes for escaping. Blank lines are skipped.
Excel (.xlsx, .xls)	Row 1 as headers. Numbers and booleans are preserved and formula cells use their computed result. Multi-sheet workbooks show a sheet picker. Parsed by the embedded backend; files up to 100 MB.

Plain JSON, not DynamoDB JSON

Typed attribute maps such as {"S": "value"} are not interpreted; they would be imported as nested maps. Export from DynamoOps produces plain JSON that imports back cleanly.

Click ⬆ Import. Drop a file onto the zone or browse for it. A chip confirms the file name with the detected row and column counts. Choose the Target table, which defaults to the open table, pick a Sheet if the workbook has several, then click Next →.

Import Data step 1 with a CSV file selected — Step 1: the file chip shows the detected rows and columns.

Step 2 — Map columns

A preview of the first five rows sits above the mapping grid, which has one line per column.

Import Data step 2 with preview and column mapping — Step 2: preview, per-column mapping and write options.

Includecheckbox: Untick to leave the column out of the import.
Type in DynamoDBselect: Auto, String, Number, Boolean, JSON (parse) or Skip. DynamoOps pre-selects Number or Boolean when more than 90% of the sampled values look like one.
Rename totext: Write the column under a different attribute name. Use this to map file headers onto the table’s key attribute names.

Below the grid, three options control the write:

Skip empty / null values (default on) drops empty, null and undefined attributes from each item.
Batch size per request (default 500, range 25 to 5000) is how many rows are sent to the backend per request; the backend splits them into DynamoDB batches of 25.
Overwrite existing items reflects the write semantics: every row is written with a Put, so an item with the same key is replaced.

Click Import. The footer confirms “N rows → table”.

How values are cast

Type	Cast rule
Auto	true/false become booleans, null becomes null, numeric text becomes a number, JSON objects and arrays are parsed, everything else stays a string
String	Kept as text
Number	Converted with Number(); non-numeric text is kept as a string
Boolean	true, 1 and yes become true; anything else false
JSON (parse)	Parsed as JSON; invalid JSON is kept as the raw string

Step 3 — Progress and errors

Import Data step 3 showing a completed import — Step 3: a finished import with no errors.

Watch the progress bar
The label counts “Importing… x / N” and the stats show Imported, Total and Errors.
Review errors
Up to ten batch errors are listed, for example a validation error when a row lacks the partition key.
Finish
Abort stops before the next chunk; Done closes the wizard and refreshes the grid.

Keys are required in the data

Every row must contain the table’s partition key (and sort key) under the exact attribute names and types. Use Rename to for the names and the type selector for the types. A row that violates this fails its DynamoDB batch of 25 and the whole chunk is reported as errors. Unprocessed items from throttling are counted but not retried; re-import the failed rows after the burst subsides.

Consumed write capacity is reported to Cost Analytics.

Example files

[
  { "userId": "usr_001", "email": "alice@example.com", "active": true, "plan": { "name": "pro", "seats": 3 } },
  { "userId": "usr_002", "email": "bob@example.com",   "active": false }
]

Exporting data

Export the loaded items or everything behind your current query as JSON or CSV, choosing exactly which fields to include.

Creating items and templates

Insert items as free-form JSON or through a structured form driven by a per-table template. Clone existing items to create variations in seconds.

DynamoDB Local and LocalStack

Develop fully offline. Point DynamoOps at DynamoDB Local, LocalStack or any DynamoDB-compatible endpoint, with dummy credentials injected automatically.

Live Streams viewer