ADR : Dataset Service (File Collection generic) - retrievalInstructions to show count of files and details of individual files in the collection

Status

• Proposed
• Trialing
• Under review
• Approved
• Retired

Context

Take the case of FileCollection.Generic for handling 3 files. If the same user does end-to-end operation (ex: upload files, create dataset record, later retrieve files) then the steps are fine. However, if some user gets a dataset record ID and is expected to retrieve the file(s) then there are two gaps -

1. User does not know how many files are in the collection
2. User does not know what names were used at the time of uploading files. Without this information, it is not possible to retrieve any file.

See this documented test case for current approach. For this, I used AWS provided Postman Collection for Preship testers. (cc @ydzeng ) https://community.opengroup.org/osdu/platform/pre-shipping/-/blob/main/R3-M22/Test_plan_Results_M22/Core%20Services/M22-AWS-Dataset-FileCollection-steps-and-data-Debasis.zip

The issue also impacts EDS development team and is blocking them from properly implementing "naturalization" feature for File collection. (cc @AshishSaxenaAccenture and @nisha.thakran )

Scope

Make suitable change in code :

• Dataset service (retrievalInstruction). Get and Post.
• Alternatively, add new service if required.

Decision

Analyze impact (if adverse) caused by this suggested change :

1. Approve change to Dataset service retrievalInstruction
2. Alternatively propose new service

Rationale

As discussed in special meeting on 28-March-2024 when @thulasi_dass and @deepapathak presented the current approach from Azure/M22 and Postman Collection, there is gap in the current approach.

You can find recording here. https://opengroup.webex.com/opengroup/ldr.php?RCID=9ffc2cbe5c81d9f4c3624c04eea33e85

cc @Yauhen_Shaliou and @vikasrana for information

Consequences

• Revise API Swagger documentation