Spike: how will we store and present issue messages?

[GeorgeGoodall]

Problem Statement

the dataset task list needs to present a list of tasks, this list of tasks are dependent on what issues are present in the dataset for that lpa. for example

2 missing values -> 2 fields have missing values 1 missing value -> 1 field has a missing value 4 invalid geometries -> 4 rows had errors processing the geometry etc

See bellow for an image of what this would look like on the dataset task list:

all the error types can be found here: https://datasette.planning.data.gov.uk/digital-land/issue_type

Acceptance Criteria

We need a solution that

• allows our code to be simple (no long switch statements)
• facilitates easy addition of new error types and their accompanying messages in the future without needing an update in code
• offers both plural and singular versions of error messages

Initial proposed solution

store singular and plural messages in our database's. either in an additional table (keyed by error type), or in additional columns in the error type table. these messages would have tokens that could be replaced programmatically based on the issue count and could be read by either the frontend. or backend performance api

[GeorgeGoodall]

Following a meeting with Infa team

• We aren't going to aim to have this done for the 14th
• Infa want a better understanding of what we want and how we will use it
• infa want a complete list of messages
• infa want data management to review these message

cc: @Darren-CheckPlanningData

[CharliePatterson]

@GeorgeGoodall-GovUk is this still needed?