chain: add implementation of ChainID

The specification defines an algorithm to calculate a `ChainID`, which
can be used to identify the result of subsequent applications of layers.
Because this algorithm is subtle and only needs to implemented in a
single place, we provide a reference implementation.

For convenience, we provide functions that calculate all the chain ids
and just the top-level one. It is is integrated with the
distribution/digest type for safety and convenience.

Tests are formulated based on pre-calculation of chain identifiers to
ensure correctness.

Signed-off-by: Stephen J Day <[email protected]>

Author: stevvooe

specs-go/chain/chainid.go

@@ -0,0 +1,46 @@
+package chain
+
+import "github.com/docker/distribution/digest"
+
+// ChainID takes a slice of digests and returns the ChainID corresponding to
+// the last entry. Typically, these are a list of layer DiffIDs, with the
+// result providing the ChainID identifying the result of sequential
+// application of the preceeding layers.
+func ChainID(dgsts []digest.Digest) digest.Digest {
+	chainIDs := make([]digest.Digest, len(dgsts))
+	copy(chainIDs, dgsts)
+	ChainIDs(chainIDs)
+
+	if len(chainIDs) == 0 {
+		return ""
+	}
+	return chainIDs[len(chainIDs)-1]
+}
+
+// ChainIDs calculates the recursively applied chain id for each identifier in
+// the slice. The result is written direcly back into the slice such that the
+// ChainID for each item will be in the respective position.
+//
+// By definition of ChainID, the zeroth element will always be the same before
+// and after the call.
+//
+// As an exmaple, given the chain of ids `[A, B, C]`, the result `[A,
+// ChainID(A|B), ChainID(A|B|C)]` will be written back to the slice.
+//
+// The input is provided as a return value for convenience.
+//
+// Typically, these are a list of layer DiffIDs, with the
+// result providing the ChainID for each the result of each layer application
+// sequentially.
+func ChainIDs(dgsts []digest.Digest) []digest.Digest {
+	if len(dgsts) < 2 {
+		return dgsts
+	}
+
+	parent := digest.FromBytes([]byte(dgsts[0] + " " + dgsts[1]))
+	next := dgsts[1:]
+	next[0] = parent
+	ChainIDs(next)
+
+	return dgsts
+}

specs-go/chain/chainid_test.go

@@ -0,0 +1,81 @@
+package chain
+
+import (
+	_ "crypto/sha256" // required to install sha256 digest support
+	"reflect"
+	"testing"
+
+	"github.com/docker/distribution/digest"
+)
+
+func TestChainID(t *testing.T) {
+	// To provide a good testing base, we define the individual links in a
+	// chain recursively, illustrating the calculations for each chain.
+	//
+	// Note that we use invalid digests for the unmodified identifiers here to
+	// make the computation more readable.
+	chainDigestAB := digest.FromString("sha256:a" + " " + "sha256:b")              // chain for A|B
+	chainDigestABC := digest.FromString(chainDigestAB.String() + " " + "sha256:c") // chain for A|B|C
+
+	for _, testcase := range []struct {
+		Name     string
+		Digests  []digest.Digest
+		Expected []digest.Digest
+	}{
+		{
+			Name: "nil",
+		},
+		{
+			Name:     "empty",
+			Digests:  []digest.Digest{},
+			Expected: []digest.Digest{},
+		},
+		{
+			Name:     "identity",
+			Digests:  []digest.Digest{"sha256:a"},
+			Expected: []digest.Digest{"sha256:a"},
+		},
+		{
+			Name:     "two",
+			Digests:  []digest.Digest{"sha256:a", "sha256:b"},
+			Expected: []digest.Digest{"sha256:a", chainDigestAB},
+		},
+		{
+			Name:     "three",
+			Digests:  []digest.Digest{"sha256:a", "sha256:b", "sha256:c"},
+			Expected: []digest.Digest{"sha256:a", chainDigestAB, chainDigestABC},
+		},
+	} {
+		t.Run(testcase.Name, func(t *testing.T) {
+			t.Log("before", testcase.Digests)
+
+			var ids []digest.Digest
+
+			if testcase.Digests != nil {
+				ids = make([]digest.Digest, len(testcase.Digests))
+				copy(ids, testcase.Digests)
+			}
+
+			ids = ChainIDs(ids)
+			t.Log("after", ids)
+			if !reflect.DeepEqual(ids, testcase.Expected) {
+				t.Errorf("unexpected chain: %v != %v", ids, testcase.Expected)
+			}
+
+			if len(testcase.Digests) == 0 {
+				return
+			}
+
+			// Make sure parent stays stable
+			if ids[0] != testcase.Digests[0] {
+				t.Errorf("parent changed: %v != %v", ids[0], testcase.Digests[0])
+			}
+
+			// make sure that the ChainID function takes the last element
+			id := ChainID(testcase.Digests)
+			if id != ids[len(ids)-1] {
+				t.Errorf("incorrect chain id returned from ChainID: %v != %v", id, ids[len(ids)-1])
+			}
+		})
+	}
+}