What the Development Team worked on in August
During August, the team have been really busy working on testing the API to ensure that all the ‘test’ and ‘development’ environments work in exactly the same way through full and automated regression testing. There has also been a lot of work to change over environments following an update to tenancies. A number of development changes have still been made alongside the provision of new support materials.
Changes have been made to:
- Allow users to detect specific errors and fix those issues.
- Implementing a limit of 10,000 records per submission.
- Enabling the rejection of submissions that have not been changed or committed in 72 hours.
Supporting documentation and resources
A lot of work has also been done in setting up the right support documentation and processes for our Vendors and Sample Resources, such as:
- The Data Model
- Vendor API User Guides
- Vendor testing user guides
- Trello tool to allow for testing feedback
- Glossary of Terms to ensure consistency in our terminology
- Swagger technical documentation
- Postman example scripts
We have currently been working on volumes of up to 10,000 records at a time but are busy working on the next steps to be able to handle more, and quicker which is currently being planned in for September’s workloads.
We have addressed the following bugs from our prototype:
- Errors occurring when posting to the gateway
- Submitted data not being rejected after 72 hours when not updated/committed.
- Errors occurring when rejecting certain data.
- Issues when processing multiple invalid samples in a batch
- Issues with environments due to tenancy move.
Overview of What is Available Now on the API
We are working in agile sprints so that we can learn, adapt and develop the product throughout the project and this has worked well. We have learnt and improved a lot during the journey so far.
The final sprint for this release enables vendors using our API to;
- Upload data through API (currently max 10,000 at a time until processed and committed).
- Monitor that data is validated and processed.
- Access clear error messaging with the ability for users to fix issues and resubmit.
- Access a clear status summary of submission results of what has been successful and committed.
- Access clear user guidance on how to submit data, check for errors and commit has been completed.
The team are already busy working on the upcoming changes to the API to improve performance and its usability of the API.
- Allow for passed records to be committed so that error records can be fixed and resubmitted/committed.
- The next release will see us working on putting in place a ‘queue function’ so that you can submit multiples of 10,000 records at a time and then go off to leave it updating while you do something else, then be presented with a final summary of the results once they have all been processed.
- It will also see us make a few changes to the data model to allow us to automatically aggregate and translate sample data into the collections ready to be ‘searchable’ in the Directory’