I was playing around converting this to slices.ContainsFunc, but then I realized maybe it's not optimal to break here? If we run into something that isn't yet cached, everything after it won't be updated from cache. Maybe that's fine though?

(also dunno what I think about the Func functions in slices; I guess the main use case is to not dupe logic in loops, but that doesn't apply here)

Yeah, this is just kinda unfortunate. If we find something that's not cached, we have to rerun FromData, which will return all the results for this chunk - whether they're cached or not. So there's no reason to read from the cache. (Did I successfully explain that?)

So there's no reason to read from the cache. (Did I successfully explain that?)

Essentially, this is a stopgap solution to handle duplicate chunks (e.g., scanning orgs + members that have multiple forks of kubernetes/kubernetes.)

Proper caching on a granular level will require splitting FromData and Verify into discrete functions.

// pseudo-code
cache := cache.New()

for _, chunk := range chunks {
    for _, d := range detectors {
        results := d.FromData(ctx, chunk)
        
        for result := range results {
            if val, ok := cache.Get(detector.Type, result.Raw); ok {
                result.Update(val)
            } else {
                detector.Verify(result)
                cache.Set(detector.Type, result.Raw)
            }
        }
    }
}