On April 8th, we released the Azure DocumentDB Data Migration Tool, an open source solution that imports data to DocumentDB from a variety of sources, including:
- JSON files
- MongoDB
- SQL Server
- CSV files
- DocumentDB collections
Today, we’re pleased to announce two new import options: Azure Table storage and URL support for JSON, CSV and MongoDB export files.
Azure Table Storage
You can now import data from Azure Table storage to DocumentDB. Simply provide a connection string and table name to import all entities from an Azure Table. For finer control over the data you wish to import, you can optionally specify the following:
- Filter
- Specify a filter to limit the entities you wish to import (e.g. PartitionKey eq ‘Partition1’ and RowKey gt ‘00001’)
- Include Internal Fields
- All – Include all internal fields (PartitionKey, RowKey, and Timestamp)
- None – Exclude all internal fields
- RowKey – Only include the RowKey field
- Select Columns
- If you want to only import specific Azure Table entity properties, add them to the Select Columns list. All other entity properties will be ignored.
URL support for JSON, CSV and MongoDB export files
When importing JSON, CSV, and MongoDB export files, you’ll notice a new option in the Data Migration Tool: namely, the ability to add one or more HTTP(s) URL(s) to source files. Whether you have files in Azure Blob storage, or you want to directly import CSV data from data.gov, importing data from HTTP accessible files couldn’t be easier.
The updated migration tool source code is available on GitHub and an updated compiled version is available from Microsoft Download Center. You may either compile the solution or simply download and extract the compiled version to a directory of your choice. For full details on how to use the tool, including command line samples for each data source option, head over to my article here. We’d also love to hear about additional sources you would like to see in the Data Migration Tool, so please submit your feedback. To learn more about DocumentDB, please visit our service page.