public async void Handle_WhenLeftPayLoadIsNotFound_ShouldReturnFalse(AnalyzeDiffCommand command, byte[] content)
{
var payLoads = new List <PayLoad>
{
new PayLoad(command.CorrelationId, content, SideEnum.Right)
};
_payLoadRepository.GetByCorrelationId(command.CorrelationId).Returns(payLoads);
var result = await _sut.Handle(command, new CancellationToken());
result.Should().BeFalse();
}
/// <summary>
/// It analyzes the diff and raises a <see cref="DiffAnalyzedEvent"/> if it goes right.
/// The diff will be analyze only if two payloads with same correlation id exists.
/// </summary>
/// <param name="request"></param>
/// <param name="cancellationToken"></param>
/// <returns></returns>
public async Task <bool> Handle(AnalyzeDiffCommand request, CancellationToken cancellationToken)
{
var payLoads = await _payLoadRepository.GetByCorrelationId(request.CorrelationId);
if (payLoads.Count() != 2)
{
return(false);
}
var diff = _diffEngine.ProcessDiff(request.CorrelationId,
payLoads.FirstOrDefault(x => x.Side == SideEnum.Left).Content,
payLoads.FirstOrDefault(x => x.Side == SideEnum.Right).Content);
/* Reading the assignment, looks like to me the diff result is a read-only data and I believe is
* something that doesn't need to be stored forever. With that in mind I assumed a cache solution would
* be a good approach to maintain the data available.
* I am using a memory cache with 1 day expiration just for demonstrations purposes, but I created an interface
* that could fit for distributed cache solutions, like redis. */
await _cache.SetAsync($"diff_{request.CorrelationId}", diff, 86400);
await _bus.RaiseEvent(new DiffAnalyzedEvent(diff.Id, diff.CorrelationId));
return(true);
}