Skip to content

Datastream observations row format takes too long #393

@Maurier

Description

@Maurier

Steps to Reproduce

GET /api/data/datastreams/:id/observations (row format, default) takes > 90 s for 35k rows and times out. Columnar format (?format=column) returns the same dataset in a fraction of a second, but only includes phenomenonTime + result, none of the other ObservationSummaryResponse fields. For the QC app we need resultQualifierCodes and in the future per-point ids (for value tracking), both of which are only available in row responses today.

I think for now we could add a dedicated endpoint to fetch resultQualifierCodes applied to a datastream in dictionary form of index and codes applied.

Expected Behavior

Faster response.

Actual Behavior

Response is too slow and timeouts.

Any additional comments?

No response

Metadata

Metadata

Assignees

No one assigned

    Labels

    QC appAssociated with the quality control appbackendAssociated with the backend repository

    Type

    No type

    Projects

    No projects

    Milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions