Test Scenarios for Single config changes

[bhanvimenghani]

Describe the feature

Test Scenario for Single model Single term config changes

A. /createExperiment api call

Model Test Case:

1. Valid Input:

• If input is “cost” → trigger cost model
• If input is “performance” → trigger performance model
• Input is case insensitive

2. Invalid input:

• If input is anything other than these eg. “abc” results in an error saying only cost and performance models support.

3. Missing input/Empty input/Whitespaces only:

• throw error msg

4. Multiple models:

• If input has multiple models, then it should say we support only single model

Term Test case:

1. Valid inputs:

• If input is “short” → trigger short term
• If input is “medium” → trigger medium term
• If input is “long” → trigger long term

2. Invalid input:

• If input is anything other than these eg. “abc” results in an error saying unsupported.

3. Missing input/Empty/Whitespace:

• throw error

4. Multiple terms:

• If multiple terms are mentioned then it should say we support only a single term

B./generateRecommendations api call

• monitoring flow should work as it is.
• If user chooses to specify the settings then we should provide everything as per the specification (even if this multi model / multi term requirement)
• If mode is auto or recreate then check if settings are present or not, if not use defaults. If settings are present, check if they are single or multiple, show error in case of multiple settings. Single settings do as directed.
• edge case test: If data is available till medium term, and user wants medium term details then it should not show short term recommendations even if they are available.