Bug: Timeout Error when Processing Many Explores

[kashira202111]

How to reproduce the problem

When there are too many Explores stored in the BigQuery table ${datasetName}.explore_assistant_examples —as in our environment, where there are 663 Explores and 51 models — the fetchSemanticModel function in useLookerFields.ts times out.

By default, error details do not appear in the debug log, so I added a log for better visibility.

    const fetchSemanticModel = async (
     ...
    ): Promise<SemanticModel | undefined> => {
      try {
        ....
      } catch (error) {
        showBoundary({
          // message: `Failed to fetch semantic model for ${modelName}::${exploreId}`,
          message: `Failed to fetch semantic model for ${modelName}::${exploreId}, ${error}`, // add a detail error
        })
        return undefined
      }

My Environments

• Looker SDK versions: ^24.2.0
• Looker versions: 25.0.23
• Mac OS: 15.3.1 (Sequoia)
• Memory: 16GB
• CPU: Apple M1
• The num of my Explores: 663

Expecting Behavior

It should handle a large number of Explores without causing errors.

For example, instead of loading all dimensions and measures during initialization, load them when switching Explores. This design also helps reduce the Assistant's startup time.