refactor AWS URL mapping with a default URL and don't depend on region list from CIDR data

cmd/archeio/app/buckets.go

@@ -23,10 +23,10 @@ import (
 	"k8s.io/klog/v2"
 )
 
-// awsRegionToS3URL returns the base S3 bucket URL for an OCI layer blob given the AWS region
+// awsRegionToHostURL returns the base S3 bucket URL for an OCI layer blob given the AWS region
 //
 // blobs in the buckets should be stored at /containers/images/sha256:$hash
-func awsRegionToS3URL(region string) string {
+func awsRegionToHostURL(region, defaultURL string) string {
 	switch region {
 	// each of these has the region in which we have a bucket listed first
 	// and then additional regions we're mapping to that bucket
@@ -66,15 +66,7 @@ func awsRegionToS3URL(region string) string {
 	case "eu-west-2", "eu-west-3", "eu-north-1":
 		return "https://prod-registry-k8s-io-eu-west-2.s3.dualstack.eu-west-2.amazonaws.com"
 	default:
-		// TestRegionToAWSRegionToS3URL checks we return a non-empty result for all regions
-		// that this app knows about
-		//
-		// we will not attempt to route to a region we do now know about
-		//
-		// if we see empty string returned, then we've failed to account for all regions
-		//
-		// we want to precompute the mapping for all regions
-		return ""
+		return defaultURL
 	}
 }
 

cmd/archeio/app/buckets_integration_test.go

@@ -24,7 +24,7 @@ import (
 )
 
 func TestCachedBlobChecker(t *testing.T) {
-	bucket := awsRegionToS3URL("us-east-1")
+	bucket := awsRegionToHostURL("us-east-1", "")
 	blobs := newCachedBlobChecker()
 	testCases := []struct {
 		Name         string

cmd/archeio/app/buckets_test.go

@@ -18,22 +18,29 @@ package app
 
 import (
 	"testing"
-
-	"k8s.io/registry.k8s.io/pkg/net/cloudcidrs"
 )
 
-func TestRegionToAWSRegionToS3URL(t *testing.T) {
-	// TODO: replace / fix this
-	// ensure all known regions return a configured bucket
-	regions := cloudcidrs.Regions()
-	for region := range regions {
-		url := awsRegionToS3URL(region)
+func TestRegionToAWSRegionToHostURL(t *testing.T) {
+	// ensure known regions return a configured bucket
+	regions := []string{
+		"GLOBAL", "af-south-1", "ap-east-1",
+		"ap-northeast-1", "ap-northeast-2", "ap-northeast-3",
+		"ap-south-1", "ap-south-2", "ap-southeast-1",
+		"ap-southeast-2", "ap-southeast-3", "ap-southeast-4",
+		"ap-southeast-6", "ca-central-1", "ca-west-1", "cn-north-1",
+		"cn-northwest-1", "eu-central-1", "eu-central-2", "eu-north-1",
+		"eu-south-1", "eu-south-2", "eu-west-1", "eu-west-2", "eu-west-3",
+		"il-central-1", "me-central-1", "me-south-1", "sa-east-1", "us-east-1",
+		"us-east-2", "us-gov-east-1", "us-gov-west-1", "us-west-1", "us-west-2",
+	}
+	for _, region := range regions {
+		url := awsRegionToHostURL(region, "")
 		if url == "" {
-			t.Fatalf("received empty string for known region %q url", region)
+			t.Fatalf("received empty string for known region %q", region)
 		}
 	}
-	// ensure bogus region would return "" so we know above test is valid
-	if url := awsRegionToS3URL("nonsensical-region"); url != "" {
+	// test default region
+	if url := awsRegionToHostURL("nonsensical-region", "____default____"); url != "____default____" {
 		t.Fatalf("received non-empty URL string for made up region \"nonsensical-region\": %q", url)
 	}
 }

cmd/archeio/app/handlers.go

@@ -33,6 +33,7 @@ type RegistryConfig struct {
 	UpstreamRegistryPath     string
 	InfoURL                  string
 	PrivacyURL               string
+	DefaultAWSBaseURL        string
 }
 
 // MakeHandler returns the root archeio HTTP handler
@@ -131,7 +132,7 @@ func makeV2Handler(rc RegistryConfig, blobs blobChecker) func(w http.ResponseWri
 		}
 
 		// check if blob is available in our S3 bucket for the region
-		bucketURL := awsRegionToS3URL(ipInfo.Region)
+		bucketURL := awsRegionToHostURL(ipInfo.Region, rc.DefaultAWSBaseURL)
 		// this matches GCR's GCS layout, which we will use for other buckets
 		blobURL := bucketURL + "/containers/images/sha256%3A" + hash
 		if blobs.BlobExists(blobURL, bucketURL, hash) {

cmd/archeio/main.go

@@ -45,6 +45,7 @@ func main() {
 		UpstreamRegistryPath:     getEnv("UPSTREAM_REGISTRY_PATH", "k8s-artifacts-prod/images"),
 		InfoURL:                  "https://github.com/kubernetes/registry.k8s.io",
 		PrivacyURL:               "https://www.linuxfoundation.org/privacy-policy/",
+		DefaultAWSBaseURL:        getEnv("DEFAULT_AWS_BASE_URL", "https://prod-registry-k8s-io-us-east-1.s3.dualstack.us-east-1.amazonaws.com"),
 	}
 
 	// configure server with reasonable timeout