diff --git a/reference/docker/helpers.go b/reference/docker/helpers.go
new file mode 100644
index 000000000..386025104
--- /dev/null
+++ b/reference/docker/helpers.go
@@ -0,0 +1,58 @@
+/*
+   Copyright The containerd Authors.
+
+   Licensed under the Apache License, Version 2.0 (the "License");
+   you may not use this file except in compliance with the License.
+   You may obtain a copy of the License at
+
+       http://www.apache.org/licenses/LICENSE-2.0
+
+   Unless required by applicable law or agreed to in writing, software
+   distributed under the License is distributed on an "AS IS" BASIS,
+   WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+   See the License for the specific language governing permissions and
+   limitations under the License.
+*/
+
+package docker
+
+import "path"
+
+// IsNameOnly returns true if reference only contains a repo name.
+func IsNameOnly(ref Named) bool {
+	if _, ok := ref.(NamedTagged); ok {
+		return false
+	}
+	if _, ok := ref.(Canonical); ok {
+		return false
+	}
+	return true
+}
+
+// FamiliarName returns the familiar name string
+// for the given named, familiarizing if needed.
+func FamiliarName(ref Named) string {
+	if nn, ok := ref.(normalizedNamed); ok {
+		return nn.Familiar().Name()
+	}
+	return ref.Name()
+}
+
+// FamiliarString returns the familiar string representation
+// for the given reference, familiarizing if needed.
+func FamiliarString(ref Reference) string {
+	if nn, ok := ref.(normalizedNamed); ok {
+		return nn.Familiar().String()
+	}
+	return ref.String()
+}
+
+// FamiliarMatch reports whether ref matches the specified pattern.
+// See https://godoc.org/path#Match for supported patterns.
+func FamiliarMatch(pattern string, ref Reference) (bool, error) {
+	matched, err := path.Match(pattern, FamiliarString(ref))
+	if namedRef, isNamed := ref.(Named); isNamed && !matched {
+		matched, _ = path.Match(pattern, FamiliarName(namedRef))
+	}
+	return matched, err
+}
diff --git a/reference/docker/normalize.go b/reference/docker/normalize.go
new file mode 100644
index 000000000..fbb6e1cf4
--- /dev/null
+++ b/reference/docker/normalize.go
@@ -0,0 +1,197 @@
+/*
+   Copyright The containerd Authors.
+
+   Licensed under the Apache License, Version 2.0 (the "License");
+   you may not use this file except in compliance with the License.
+   You may obtain a copy of the License at
+
+       http://www.apache.org/licenses/LICENSE-2.0
+
+   Unless required by applicable law or agreed to in writing, software
+   distributed under the License is distributed on an "AS IS" BASIS,
+   WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+   See the License for the specific language governing permissions and
+   limitations under the License.
+*/
+
+package docker
+
+import (
+	"errors"
+	"fmt"
+	"strings"
+
+	"github.com/opencontainers/go-digest"
+)
+
+var (
+	legacyDefaultDomain = "index.docker.io"
+	defaultDomain       = "docker.io"
+	officialRepoName    = "library"
+	defaultTag          = "latest"
+)
+
+// normalizedNamed represents a name which has been
+// normalized and has a familiar form. A familiar name
+// is what is used in Docker UI. An example normalized
+// name is "docker.io/library/ubuntu" and corresponding
+// familiar name of "ubuntu".
+type normalizedNamed interface {
+	Named
+	Familiar() Named
+}
+
+// ParseNormalizedNamed parses a string into a named reference
+// transforming a familiar name from Docker UI to a fully
+// qualified reference. If the value may be an identifier
+// use ParseAnyReference.
+func ParseNormalizedNamed(s string) (Named, error) {
+	if ok := anchoredIdentifierRegexp.MatchString(s); ok {
+		return nil, fmt.Errorf("invalid repository name (%s), cannot specify 64-byte hexadecimal strings", s)
+	}
+	domain, remainder := splitDockerDomain(s)
+	var remoteName string
+	if tagSep := strings.IndexRune(remainder, ':'); tagSep > -1 {
+		remoteName = remainder[:tagSep]
+	} else {
+		remoteName = remainder
+	}
+	if strings.ToLower(remoteName) != remoteName {
+		return nil, errors.New("invalid reference format: repository name must be lowercase")
+	}
+
+	ref, err := Parse(domain + "/" + remainder)
+	if err != nil {
+		return nil, err
+	}
+	named, isNamed := ref.(Named)
+	if !isNamed {
+		return nil, fmt.Errorf("reference %s has no name", ref.String())
+	}
+	return named, nil
+}
+
+// ParseDockerRef normalizes the image reference following the docker convention. This is added
+// mainly for backward compatibility.
+// The reference returned can only be either tagged or digested. For reference contains both tag
+// and digest, the function returns digested reference, e.g. docker.io/library/busybox:latest@
+// sha256:7cc4b5aefd1d0cadf8d97d4350462ba51c694ebca145b08d7d41b41acc8db5aa will be returned as
+// docker.io/library/busybox@sha256:7cc4b5aefd1d0cadf8d97d4350462ba51c694ebca145b08d7d41b41acc8db5aa.
+func ParseDockerRef(ref string) (Named, error) {
+	named, err := ParseNormalizedNamed(ref)
+	if err != nil {
+		return nil, err
+	}
+	if _, ok := named.(NamedTagged); ok {
+		if canonical, ok := named.(Canonical); ok {
+			// The reference is both tagged and digested, only
+			// return digested.
+			newNamed, err := WithName(canonical.Name())
+			if err != nil {
+				return nil, err
+			}
+			newCanonical, err := WithDigest(newNamed, canonical.Digest())
+			if err != nil {
+				return nil, err
+			}
+			return newCanonical, nil
+		}
+	}
+	return TagNameOnly(named), nil
+}
+
+// splitDockerDomain splits a repository name to domain and remotename string.
+// If no valid domain is found, the default domain is used. Repository name
+// needs to be already validated before.
+func splitDockerDomain(name string) (domain, remainder string) {
+	i := strings.IndexRune(name, '/')
+	if i == -1 || (!strings.ContainsAny(name[:i], ".:") && name[:i] != "localhost") {
+		domain, remainder = defaultDomain, name
+	} else {
+		domain, remainder = name[:i], name[i+1:]
+	}
+	if domain == legacyDefaultDomain {
+		domain = defaultDomain
+	}
+	if domain == defaultDomain && !strings.ContainsRune(remainder, '/') {
+		remainder = officialRepoName + "/" + remainder
+	}
+	return
+}
+
+// familiarizeName returns a shortened version of the name familiar
+// to to the Docker UI. Familiar names have the default domain
+// "docker.io" and "library/" repository prefix removed.
+// For example, "docker.io/library/redis" will have the familiar
+// name "redis" and "docker.io/dmcgowan/myapp" will be "dmcgowan/myapp".
+// Returns a familiarized named only reference.
+func familiarizeName(named namedRepository) repository {
+	repo := repository{
+		domain: named.Domain(),
+		path:   named.Path(),
+	}
+
+	if repo.domain == defaultDomain {
+		repo.domain = ""
+		// Handle official repositories which have the pattern "library/<official repo name>"
+		if split := strings.Split(repo.path, "/"); len(split) == 2 && split[0] == officialRepoName {
+			repo.path = split[1]
+		}
+	}
+	return repo
+}
+
+func (r reference) Familiar() Named {
+	return reference{
+		namedRepository: familiarizeName(r.namedRepository),
+		tag:             r.tag,
+		digest:          r.digest,
+	}
+}
+
+func (r repository) Familiar() Named {
+	return familiarizeName(r)
+}
+
+func (t taggedReference) Familiar() Named {
+	return taggedReference{
+		namedRepository: familiarizeName(t.namedRepository),
+		tag:             t.tag,
+	}
+}
+
+func (c canonicalReference) Familiar() Named {
+	return canonicalReference{
+		namedRepository: familiarizeName(c.namedRepository),
+		digest:          c.digest,
+	}
+}
+
+// TagNameOnly adds the default tag "latest" to a reference if it only has
+// a repo name.
+func TagNameOnly(ref Named) Named {
+	if IsNameOnly(ref) {
+		namedTagged, err := WithTag(ref, defaultTag)
+		if err != nil {
+			// Default tag must be valid, to create a NamedTagged
+			// type with non-validated input the WithTag function
+			// should be used instead
+			panic(err)
+		}
+		return namedTagged
+	}
+	return ref
+}
+
+// ParseAnyReference parses a reference string as a possible identifier,
+// full digest, or familiar name.
+func ParseAnyReference(ref string) (Reference, error) {
+	if ok := anchoredIdentifierRegexp.MatchString(ref); ok {
+		return digestReference("sha256:" + ref), nil
+	}
+	if dgst, err := digest.Parse(ref); err == nil {
+		return digestReference(dgst), nil
+	}
+
+	return ParseNormalizedNamed(ref)
+}
diff --git a/reference/docker/reference.go b/reference/docker/reference.go
index 25436b645..d8f75e158 100644
--- a/reference/docker/reference.go
+++ b/reference/docker/reference.go
@@ -43,8 +43,6 @@ package docker
 import (
 	"errors"
 	"fmt"
-	"path"
-	"regexp"
 	"strings"
 
 	"github.com/opencontainers/go-digest"
@@ -451,349 +449,3 @@ func (c canonicalReference) String() string {
 func (c canonicalReference) Digest() digest.Digest {
 	return c.digest
 }
-
-var (
-	// alphaNumericRegexp defines the alpha numeric atom, typically a
-	// component of names. This only allows lower case characters and digits.
-	alphaNumericRegexp = match(`[a-z0-9]+`)
-
-	// separatorRegexp defines the separators allowed to be embedded in name
-	// components. This allow one period, one or two underscore and multiple
-	// dashes.
-	separatorRegexp = match(`(?:[._]|__|[-]*)`)
-
-	// nameComponentRegexp restricts registry path component names to start
-	// with at least one letter or number, with following parts able to be
-	// separated by one period, one or two underscore and multiple dashes.
-	nameComponentRegexp = expression(
-		alphaNumericRegexp,
-		optional(repeated(separatorRegexp, alphaNumericRegexp)))
-
-	// domainComponentRegexp restricts the registry domain component of a
-	// repository name to start with a component as defined by DomainRegexp
-	// and followed by an optional port.
-	domainComponentRegexp = match(`(?:[a-zA-Z0-9]|[a-zA-Z0-9][a-zA-Z0-9-]*[a-zA-Z0-9])`)
-
-	// DomainRegexp defines the structure of potential domain components
-	// that may be part of image names. This is purposely a subset of what is
-	// allowed by DNS to ensure backwards compatibility with Docker image
-	// names.
-	DomainRegexp = expression(
-		domainComponentRegexp,
-		optional(repeated(literal(`.`), domainComponentRegexp)),
-		optional(literal(`:`), match(`[0-9]+`)))
-
-	// TagRegexp matches valid tag names. From docker/docker:graph/tags.go.
-	TagRegexp = match(`[\w][\w.-]{0,127}`)
-
-	// anchoredTagRegexp matches valid tag names, anchored at the start and
-	// end of the matched string.
-	anchoredTagRegexp = anchored(TagRegexp)
-
-	// DigestRegexp matches valid digests.
-	DigestRegexp = match(`[A-Za-z][A-Za-z0-9]*(?:[-_+.][A-Za-z][A-Za-z0-9]*)*[:][[:xdigit:]]{32,}`)
-
-	// anchoredDigestRegexp matches valid digests, anchored at the start and
-	// end of the matched string.
-	anchoredDigestRegexp = anchored(DigestRegexp)
-
-	// NameRegexp is the format for the name component of references. The
-	// regexp has capturing groups for the domain and name part omitting
-	// the separating forward slash from either.
-	NameRegexp = expression(
-		optional(DomainRegexp, literal(`/`)),
-		nameComponentRegexp,
-		optional(repeated(literal(`/`), nameComponentRegexp)))
-
-	// anchoredNameRegexp is used to parse a name value, capturing the
-	// domain and trailing components.
-	anchoredNameRegexp = anchored(
-		optional(capture(DomainRegexp), literal(`/`)),
-		capture(nameComponentRegexp,
-			optional(repeated(literal(`/`), nameComponentRegexp))))
-
-	// ReferenceRegexp is the full supported format of a reference. The regexp
-	// is anchored and has capturing groups for name, tag, and digest
-	// components.
-	ReferenceRegexp = anchored(capture(NameRegexp),
-		optional(literal(":"), capture(TagRegexp)),
-		optional(literal("@"), capture(DigestRegexp)))
-
-	// IdentifierRegexp is the format for string identifier used as a
-	// content addressable identifier using sha256. These identifiers
-	// are like digests without the algorithm, since sha256 is used.
-	IdentifierRegexp = match(`([a-f0-9]{64})`)
-
-	// ShortIdentifierRegexp is the format used to represent a prefix
-	// of an identifier. A prefix may be used to match a sha256 identifier
-	// within a list of trusted identifiers.
-	ShortIdentifierRegexp = match(`([a-f0-9]{6,64})`)
-
-	// anchoredIdentifierRegexp is used to check or match an
-	// identifier value, anchored at start and end of string.
-	anchoredIdentifierRegexp = anchored(IdentifierRegexp)
-)
-
-// match compiles the string to a regular expression.
-var match = regexp.MustCompile
-
-// literal compiles s into a literal regular expression, escaping any regexp
-// reserved characters.
-func literal(s string) *regexp.Regexp {
-	re := match(regexp.QuoteMeta(s))
-
-	if _, complete := re.LiteralPrefix(); !complete {
-		panic("must be a literal")
-	}
-
-	return re
-}
-
-// expression defines a full expression, where each regular expression must
-// follow the previous.
-func expression(res ...*regexp.Regexp) *regexp.Regexp {
-	var s string
-	for _, re := range res {
-		s += re.String()
-	}
-
-	return match(s)
-}
-
-// optional wraps the expression in a non-capturing group and makes the
-// production optional.
-func optional(res ...*regexp.Regexp) *regexp.Regexp {
-	return match(group(expression(res...)).String() + `?`)
-}
-
-// repeated wraps the regexp in a non-capturing group to get one or more
-// matches.
-func repeated(res ...*regexp.Regexp) *regexp.Regexp {
-	return match(group(expression(res...)).String() + `+`)
-}
-
-// group wraps the regexp in a non-capturing group.
-func group(res ...*regexp.Regexp) *regexp.Regexp {
-	return match(`(?:` + expression(res...).String() + `)`)
-}
-
-// capture wraps the expression in a capturing group.
-func capture(res ...*regexp.Regexp) *regexp.Regexp {
-	return match(`(` + expression(res...).String() + `)`)
-}
-
-// anchored anchors the regular expression by adding start and end delimiters.
-func anchored(res ...*regexp.Regexp) *regexp.Regexp {
-	return match(`^` + expression(res...).String() + `$`)
-}
-
-var (
-	legacyDefaultDomain = "index.docker.io"
-	defaultDomain       = "docker.io"
-	officialRepoName    = "library"
-	defaultTag          = "latest"
-)
-
-// normalizedNamed represents a name which has been
-// normalized and has a familiar form. A familiar name
-// is what is used in Docker UI. An example normalized
-// name is "docker.io/library/ubuntu" and corresponding
-// familiar name of "ubuntu".
-type normalizedNamed interface {
-	Named
-	Familiar() Named
-}
-
-// ParseNormalizedNamed parses a string into a named reference
-// transforming a familiar name from Docker UI to a fully
-// qualified reference. If the value may be an identifier
-// use ParseAnyReference.
-func ParseNormalizedNamed(s string) (Named, error) {
-	if ok := anchoredIdentifierRegexp.MatchString(s); ok {
-		return nil, fmt.Errorf("invalid repository name (%s), cannot specify 64-byte hexadecimal strings", s)
-	}
-	domain, remainder := splitDockerDomain(s)
-	var remoteName string
-	if tagSep := strings.IndexRune(remainder, ':'); tagSep > -1 {
-		remoteName = remainder[:tagSep]
-	} else {
-		remoteName = remainder
-	}
-	if strings.ToLower(remoteName) != remoteName {
-		return nil, errors.New("invalid reference format: repository name must be lowercase")
-	}
-
-	ref, err := Parse(domain + "/" + remainder)
-	if err != nil {
-		return nil, err
-	}
-	named, isNamed := ref.(Named)
-	if !isNamed {
-		return nil, fmt.Errorf("reference %s has no name", ref.String())
-	}
-	return named, nil
-}
-
-// ParseDockerRef normalizes the image reference following the docker convention. This is added
-// mainly for backward compatibility.
-// The reference returned can only be either tagged or digested. For reference contains both tag
-// and digest, the function returns digested reference, e.g. docker.io/library/busybox:latest@
-// sha256:7cc4b5aefd1d0cadf8d97d4350462ba51c694ebca145b08d7d41b41acc8db5aa will be returned as
-// docker.io/library/busybox@sha256:7cc4b5aefd1d0cadf8d97d4350462ba51c694ebca145b08d7d41b41acc8db5aa.
-func ParseDockerRef(ref string) (Named, error) {
-	named, err := ParseNormalizedNamed(ref)
-	if err != nil {
-		return nil, err
-	}
-	if _, ok := named.(NamedTagged); ok {
-		if canonical, ok := named.(Canonical); ok {
-			// The reference is both tagged and digested, only
-			// return digested.
-			newNamed, err := WithName(canonical.Name())
-			if err != nil {
-				return nil, err
-			}
-			newCanonical, err := WithDigest(newNamed, canonical.Digest())
-			if err != nil {
-				return nil, err
-			}
-			return newCanonical, nil
-		}
-	}
-	return TagNameOnly(named), nil
-}
-
-// splitDockerDomain splits a repository name to domain and remotename string.
-// If no valid domain is found, the default domain is used. Repository name
-// needs to be already validated before.
-func splitDockerDomain(name string) (domain, remainder string) {
-	i := strings.IndexRune(name, '/')
-	if i == -1 || (!strings.ContainsAny(name[:i], ".:") && name[:i] != "localhost") {
-		domain, remainder = defaultDomain, name
-	} else {
-		domain, remainder = name[:i], name[i+1:]
-	}
-	if domain == legacyDefaultDomain {
-		domain = defaultDomain
-	}
-	if domain == defaultDomain && !strings.ContainsRune(remainder, '/') {
-		remainder = officialRepoName + "/" + remainder
-	}
-	return
-}
-
-// familiarizeName returns a shortened version of the name familiar
-// to to the Docker UI. Familiar names have the default domain
-// "docker.io" and "library/" repository prefix removed.
-// For example, "docker.io/library/redis" will have the familiar
-// name "redis" and "docker.io/dmcgowan/myapp" will be "dmcgowan/myapp".
-// Returns a familiarized named only reference.
-func familiarizeName(named namedRepository) repository {
-	repo := repository{
-		domain: named.Domain(),
-		path:   named.Path(),
-	}
-
-	if repo.domain == defaultDomain {
-		repo.domain = ""
-		// Handle official repositories which have the pattern "library/<official repo name>"
-		if split := strings.Split(repo.path, "/"); len(split) == 2 && split[0] == officialRepoName {
-			repo.path = split[1]
-		}
-	}
-	return repo
-}
-
-func (r reference) Familiar() Named {
-	return reference{
-		namedRepository: familiarizeName(r.namedRepository),
-		tag:             r.tag,
-		digest:          r.digest,
-	}
-}
-
-func (r repository) Familiar() Named {
-	return familiarizeName(r)
-}
-
-func (t taggedReference) Familiar() Named {
-	return taggedReference{
-		namedRepository: familiarizeName(t.namedRepository),
-		tag:             t.tag,
-	}
-}
-
-func (c canonicalReference) Familiar() Named {
-	return canonicalReference{
-		namedRepository: familiarizeName(c.namedRepository),
-		digest:          c.digest,
-	}
-}
-
-// TagNameOnly adds the default tag "latest" to a reference if it only has
-// a repo name.
-func TagNameOnly(ref Named) Named {
-	if IsNameOnly(ref) {
-		namedTagged, err := WithTag(ref, defaultTag)
-		if err != nil {
-			// Default tag must be valid, to create a NamedTagged
-			// type with non-validated input the WithTag function
-			// should be used instead
-			panic(err)
-		}
-		return namedTagged
-	}
-	return ref
-}
-
-// ParseAnyReference parses a reference string as a possible identifier,
-// full digest, or familiar name.
-func ParseAnyReference(ref string) (Reference, error) {
-	if ok := anchoredIdentifierRegexp.MatchString(ref); ok {
-		return digestReference("sha256:" + ref), nil
-	}
-	if dgst, err := digest.Parse(ref); err == nil {
-		return digestReference(dgst), nil
-	}
-
-	return ParseNormalizedNamed(ref)
-}
-
-// IsNameOnly returns true if reference only contains a repo name.
-func IsNameOnly(ref Named) bool {
-	if _, ok := ref.(NamedTagged); ok {
-		return false
-	}
-	if _, ok := ref.(Canonical); ok {
-		return false
-	}
-	return true
-}
-
-// FamiliarName returns the familiar name string
-// for the given named, familiarizing if needed.
-func FamiliarName(ref Named) string {
-	if nn, ok := ref.(normalizedNamed); ok {
-		return nn.Familiar().Name()
-	}
-	return ref.Name()
-}
-
-// FamiliarString returns the familiar string representation
-// for the given reference, familiarizing if needed.
-func FamiliarString(ref Reference) string {
-	if nn, ok := ref.(normalizedNamed); ok {
-		return nn.Familiar().String()
-	}
-	return ref.String()
-}
-
-// FamiliarMatch reports whether ref matches the specified pattern.
-// See https://godoc.org/path#Match for supported patterns.
-func FamiliarMatch(pattern string, ref Reference) (bool, error) {
-	matched, err := path.Match(pattern, FamiliarString(ref))
-	if namedRef, isNamed := ref.(Named); isNamed && !matched {
-		matched, _ = path.Match(pattern, FamiliarName(namedRef))
-	}
-	return matched, err
-}
diff --git a/reference/docker/regexp.go b/reference/docker/regexp.go
new file mode 100644
index 000000000..50e6e7758
--- /dev/null
+++ b/reference/docker/regexp.go
@@ -0,0 +1,154 @@
+/*
+   Copyright The containerd Authors.
+
+   Licensed under the Apache License, Version 2.0 (the "License");
+   you may not use this file except in compliance with the License.
+   You may obtain a copy of the License at
+
+       http://www.apache.org/licenses/LICENSE-2.0
+
+   Unless required by applicable law or agreed to in writing, software
+   distributed under the License is distributed on an "AS IS" BASIS,
+   WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+   See the License for the specific language governing permissions and
+   limitations under the License.
+*/
+
+package docker
+
+import "regexp"
+
+var (
+	// alphaNumericRegexp defines the alpha numeric atom, typically a
+	// component of names. This only allows lower case characters and digits.
+	alphaNumericRegexp = match(`[a-z0-9]+`)
+
+	// separatorRegexp defines the separators allowed to be embedded in name
+	// components. This allow one period, one or two underscore and multiple
+	// dashes.
+	separatorRegexp = match(`(?:[._]|__|[-]*)`)
+
+	// nameComponentRegexp restricts registry path component names to start
+	// with at least one letter or number, with following parts able to be
+	// separated by one period, one or two underscore and multiple dashes.
+	nameComponentRegexp = expression(
+		alphaNumericRegexp,
+		optional(repeated(separatorRegexp, alphaNumericRegexp)))
+
+	// domainComponentRegexp restricts the registry domain component of a
+	// repository name to start with a component as defined by DomainRegexp
+	// and followed by an optional port.
+	domainComponentRegexp = match(`(?:[a-zA-Z0-9]|[a-zA-Z0-9][a-zA-Z0-9-]*[a-zA-Z0-9])`)
+
+	// DomainRegexp defines the structure of potential domain components
+	// that may be part of image names. This is purposely a subset of what is
+	// allowed by DNS to ensure backwards compatibility with Docker image
+	// names.
+	DomainRegexp = expression(
+		domainComponentRegexp,
+		optional(repeated(literal(`.`), domainComponentRegexp)),
+		optional(literal(`:`), match(`[0-9]+`)))
+
+	// TagRegexp matches valid tag names. From docker/docker:graph/tags.go.
+	TagRegexp = match(`[\w][\w.-]{0,127}`)
+
+	// anchoredTagRegexp matches valid tag names, anchored at the start and
+	// end of the matched string.
+	anchoredTagRegexp = anchored(TagRegexp)
+
+	// DigestRegexp matches valid digests.
+	DigestRegexp = match(`[A-Za-z][A-Za-z0-9]*(?:[-_+.][A-Za-z][A-Za-z0-9]*)*[:][[:xdigit:]]{32,}`)
+
+	// anchoredDigestRegexp matches valid digests, anchored at the start and
+	// end of the matched string.
+	anchoredDigestRegexp = anchored(DigestRegexp)
+
+	// NameRegexp is the format for the name component of references. The
+	// regexp has capturing groups for the domain and name part omitting
+	// the separating forward slash from either.
+	NameRegexp = expression(
+		optional(DomainRegexp, literal(`/`)),
+		nameComponentRegexp,
+		optional(repeated(literal(`/`), nameComponentRegexp)))
+
+	// anchoredNameRegexp is used to parse a name value, capturing the
+	// domain and trailing components.
+	anchoredNameRegexp = anchored(
+		optional(capture(DomainRegexp), literal(`/`)),
+		capture(nameComponentRegexp,
+			optional(repeated(literal(`/`), nameComponentRegexp))))
+
+	// ReferenceRegexp is the full supported format of a reference. The regexp
+	// is anchored and has capturing groups for name, tag, and digest
+	// components.
+	ReferenceRegexp = anchored(capture(NameRegexp),
+		optional(literal(":"), capture(TagRegexp)),
+		optional(literal("@"), capture(DigestRegexp)))
+
+	// IdentifierRegexp is the format for string identifier used as a
+	// content addressable identifier using sha256. These identifiers
+	// are like digests without the algorithm, since sha256 is used.
+	IdentifierRegexp = match(`([a-f0-9]{64})`)
+
+	// ShortIdentifierRegexp is the format used to represent a prefix
+	// of an identifier. A prefix may be used to match a sha256 identifier
+	// within a list of trusted identifiers.
+	ShortIdentifierRegexp = match(`([a-f0-9]{6,64})`)
+
+	// anchoredIdentifierRegexp is used to check or match an
+	// identifier value, anchored at start and end of string.
+	anchoredIdentifierRegexp = anchored(IdentifierRegexp)
+)
+
+// match compiles the string to a regular expression.
+var match = regexp.MustCompile
+
+// literal compiles s into a literal regular expression, escaping any regexp
+// reserved characters.
+func literal(s string) *regexp.Regexp {
+	re := match(regexp.QuoteMeta(s))
+
+	if _, complete := re.LiteralPrefix(); !complete {
+		panic("must be a literal")
+	}
+
+	return re
+}
+
+// expression defines a full expression, where each regular expression must
+// follow the previous.
+func expression(res ...*regexp.Regexp) *regexp.Regexp {
+	var s string
+	for _, re := range res {
+		s += re.String()
+	}
+
+	return match(s)
+}
+
+// optional wraps the expression in a non-capturing group and makes the
+// production optional.
+func optional(res ...*regexp.Regexp) *regexp.Regexp {
+	return match(group(expression(res...)).String() + `?`)
+}
+
+// repeated wraps the regexp in a non-capturing group to get one or more
+// matches.
+func repeated(res ...*regexp.Regexp) *regexp.Regexp {
+	return match(group(expression(res...)).String() + `+`)
+}
+
+// group wraps the regexp in a non-capturing group.
+func group(res ...*regexp.Regexp) *regexp.Regexp {
+	return match(`(?:` + expression(res...).String() + `)`)
+}
+
+// capture wraps the expression in a capturing group.
+func capture(res ...*regexp.Regexp) *regexp.Regexp {
+	return match(`(` + expression(res...).String() + `)`)
+}
+
+// anchored anchors the regular expression by adding start and end delimiters.
+func anchored(res ...*regexp.Regexp) *regexp.Regexp {
+	return match(`^` + expression(res...).String() + `$`)
+}