mirror of
https://github.com/ceph/ceph-csi.git
synced 2024-11-23 06:40:23 +00:00
1b23d78113
Updated kubernetes packages to latest release. resizefs package has been included into k8s.io/mount-utils package. updated code to use the same. Updates: #1968 Signed-off-by: Rakshith R <rar@redhat.com>
506 lines
14 KiB
Go
506 lines
14 KiB
Go
/*
|
|
Copyright 2018 The Kubernetes 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 fieldpath
|
|
|
|
import (
|
|
"sort"
|
|
"strings"
|
|
|
|
"sigs.k8s.io/structured-merge-diff/v4/schema"
|
|
)
|
|
|
|
// Set identifies a set of fields.
|
|
type Set struct {
|
|
// Members lists fields that are part of the set.
|
|
// TODO: will be serialized as a list of path elements.
|
|
Members PathElementSet
|
|
|
|
// Children lists child fields which themselves have children that are
|
|
// members of the set. Appearance in this list does not imply membership.
|
|
// Note: this is a tree, not an arbitrary graph.
|
|
Children SetNodeMap
|
|
}
|
|
|
|
// NewSet makes a set from a list of paths.
|
|
func NewSet(paths ...Path) *Set {
|
|
s := &Set{}
|
|
for _, p := range paths {
|
|
s.Insert(p)
|
|
}
|
|
return s
|
|
}
|
|
|
|
// Insert adds the field identified by `p` to the set. Important: parent fields
|
|
// are NOT added to the set; if that is desired, they must be added separately.
|
|
func (s *Set) Insert(p Path) {
|
|
if len(p) == 0 {
|
|
// Zero-length path identifies the entire object; we don't
|
|
// track top-level ownership.
|
|
return
|
|
}
|
|
for {
|
|
if len(p) == 1 {
|
|
s.Members.Insert(p[0])
|
|
return
|
|
}
|
|
s = s.Children.Descend(p[0])
|
|
p = p[1:]
|
|
}
|
|
}
|
|
|
|
// Union returns a Set containing elements which appear in either s or s2.
|
|
func (s *Set) Union(s2 *Set) *Set {
|
|
return &Set{
|
|
Members: *s.Members.Union(&s2.Members),
|
|
Children: *s.Children.Union(&s2.Children),
|
|
}
|
|
}
|
|
|
|
// Intersection returns a Set containing leaf elements which appear in both s
|
|
// and s2. Intersection can be constructed from Union and Difference operations
|
|
// (example in the tests) but it's much faster to do it in one pass.
|
|
func (s *Set) Intersection(s2 *Set) *Set {
|
|
return &Set{
|
|
Members: *s.Members.Intersection(&s2.Members),
|
|
Children: *s.Children.Intersection(&s2.Children),
|
|
}
|
|
}
|
|
|
|
// Difference returns a Set containing elements which:
|
|
// * appear in s
|
|
// * do not appear in s2
|
|
//
|
|
// In other words, for leaf fields, this acts like a regular set difference
|
|
// operation. When non leaf fields are compared with leaf fields ("parents"
|
|
// which contain "children"), the effect is:
|
|
// * parent - child = parent
|
|
// * child - parent = {empty set}
|
|
func (s *Set) Difference(s2 *Set) *Set {
|
|
return &Set{
|
|
Members: *s.Members.Difference(&s2.Members),
|
|
Children: *s.Children.Difference(s2),
|
|
}
|
|
}
|
|
|
|
// RecursiveDifference returns a Set containing elements which:
|
|
// * appear in s
|
|
// * do not appear in s2
|
|
//
|
|
// Compared to a regular difference,
|
|
// this removes every field **and its children** from s that is contained in s2.
|
|
//
|
|
// For example, with s containing `a.b.c` and s2 containing `a.b`,
|
|
// a RecursiveDifference will result in `a`, as the entire node `a.b` gets removed.
|
|
func (s *Set) RecursiveDifference(s2 *Set) *Set {
|
|
return &Set{
|
|
Members: *s.Members.Difference(&s2.Members),
|
|
Children: *s.Children.RecursiveDifference(s2),
|
|
}
|
|
}
|
|
|
|
// EnsureNamedFieldsAreMembers returns a Set that contains all the
|
|
// fields in s, as well as all the named fields that are typically not
|
|
// included. For example, a set made of "a.b.c" will end-up also owning
|
|
// "a" if it's a named fields but not "a.b" if it's a map.
|
|
func (s *Set) EnsureNamedFieldsAreMembers(sc *schema.Schema, tr schema.TypeRef) *Set {
|
|
members := PathElementSet{
|
|
members: make(sortedPathElements, 0, s.Members.Size()+len(s.Children.members)),
|
|
}
|
|
atom, _ := sc.Resolve(tr)
|
|
members.members = append(members.members, s.Members.members...)
|
|
for _, node := range s.Children.members {
|
|
// Only insert named fields.
|
|
if node.pathElement.FieldName != nil && atom.Map != nil {
|
|
if _, has := atom.Map.FindField(*node.pathElement.FieldName); has {
|
|
members.Insert(node.pathElement)
|
|
}
|
|
}
|
|
}
|
|
return &Set{
|
|
Members: members,
|
|
Children: *s.Children.EnsureNamedFieldsAreMembers(sc, tr),
|
|
}
|
|
}
|
|
|
|
// Size returns the number of members of the set.
|
|
func (s *Set) Size() int {
|
|
return s.Members.Size() + s.Children.Size()
|
|
}
|
|
|
|
// Empty returns true if there are no members of the set. It is a separate
|
|
// function from Size since it's common to check whether size > 0, and
|
|
// potentially much faster to return as soon as a single element is found.
|
|
func (s *Set) Empty() bool {
|
|
if s.Members.Size() > 0 {
|
|
return false
|
|
}
|
|
return s.Children.Empty()
|
|
}
|
|
|
|
// Has returns true if the field referenced by `p` is a member of the set.
|
|
func (s *Set) Has(p Path) bool {
|
|
if len(p) == 0 {
|
|
// No one owns "the entire object"
|
|
return false
|
|
}
|
|
for {
|
|
if len(p) == 1 {
|
|
return s.Members.Has(p[0])
|
|
}
|
|
var ok bool
|
|
s, ok = s.Children.Get(p[0])
|
|
if !ok {
|
|
return false
|
|
}
|
|
p = p[1:]
|
|
}
|
|
}
|
|
|
|
// Equals returns true if s and s2 have exactly the same members.
|
|
func (s *Set) Equals(s2 *Set) bool {
|
|
return s.Members.Equals(&s2.Members) && s.Children.Equals(&s2.Children)
|
|
}
|
|
|
|
// String returns the set one element per line.
|
|
func (s *Set) String() string {
|
|
elements := []string{}
|
|
s.Iterate(func(p Path) {
|
|
elements = append(elements, p.String())
|
|
})
|
|
return strings.Join(elements, "\n")
|
|
}
|
|
|
|
// Iterate calls f once for each field that is a member of the set (preorder
|
|
// DFS). The path passed to f will be reused so make a copy if you wish to keep
|
|
// it.
|
|
func (s *Set) Iterate(f func(Path)) {
|
|
s.iteratePrefix(Path{}, f)
|
|
}
|
|
|
|
func (s *Set) iteratePrefix(prefix Path, f func(Path)) {
|
|
s.Members.Iterate(func(pe PathElement) { f(append(prefix, pe)) })
|
|
s.Children.iteratePrefix(prefix, f)
|
|
}
|
|
|
|
// WithPrefix returns the subset of paths which begin with the given prefix,
|
|
// with the prefix not included.
|
|
func (s *Set) WithPrefix(pe PathElement) *Set {
|
|
subset, ok := s.Children.Get(pe)
|
|
if !ok {
|
|
return NewSet()
|
|
}
|
|
return subset
|
|
}
|
|
|
|
// Leaves returns a set containing only the leaf paths
|
|
// of a set.
|
|
func (s *Set) Leaves() *Set {
|
|
leaves := PathElementSet{}
|
|
im := 0
|
|
ic := 0
|
|
|
|
// any members that are not also children are leaves
|
|
outer:
|
|
for im < len(s.Members.members) {
|
|
member := s.Members.members[im]
|
|
|
|
for ic < len(s.Children.members) {
|
|
d := member.Compare(s.Children.members[ic].pathElement)
|
|
if d == 0 {
|
|
ic++
|
|
im++
|
|
continue outer
|
|
} else if d < 0 {
|
|
break
|
|
} else /* if d > 0 */ {
|
|
ic++
|
|
}
|
|
}
|
|
leaves.members = append(leaves.members, member)
|
|
im++
|
|
}
|
|
|
|
return &Set{
|
|
Members: leaves,
|
|
Children: *s.Children.Leaves(),
|
|
}
|
|
}
|
|
|
|
// setNode is a pair of PathElement / Set, for the purpose of expressing
|
|
// nested set membership.
|
|
type setNode struct {
|
|
pathElement PathElement
|
|
set *Set
|
|
}
|
|
|
|
// SetNodeMap is a map of PathElement to subset.
|
|
type SetNodeMap struct {
|
|
members sortedSetNode
|
|
}
|
|
|
|
type sortedSetNode []setNode
|
|
|
|
// Implement the sort interface; this would permit bulk creation, which would
|
|
// be faster than doing it one at a time via Insert.
|
|
func (s sortedSetNode) Len() int { return len(s) }
|
|
func (s sortedSetNode) Less(i, j int) bool { return s[i].pathElement.Less(s[j].pathElement) }
|
|
func (s sortedSetNode) Swap(i, j int) { s[i], s[j] = s[j], s[i] }
|
|
|
|
// Descend adds pe to the set if necessary, returning the associated subset.
|
|
func (s *SetNodeMap) Descend(pe PathElement) *Set {
|
|
loc := sort.Search(len(s.members), func(i int) bool {
|
|
return !s.members[i].pathElement.Less(pe)
|
|
})
|
|
if loc == len(s.members) {
|
|
s.members = append(s.members, setNode{pathElement: pe, set: &Set{}})
|
|
return s.members[loc].set
|
|
}
|
|
if s.members[loc].pathElement.Equals(pe) {
|
|
return s.members[loc].set
|
|
}
|
|
s.members = append(s.members, setNode{})
|
|
copy(s.members[loc+1:], s.members[loc:])
|
|
s.members[loc] = setNode{pathElement: pe, set: &Set{}}
|
|
return s.members[loc].set
|
|
}
|
|
|
|
// Size returns the sum of the number of members of all subsets.
|
|
func (s *SetNodeMap) Size() int {
|
|
count := 0
|
|
for _, v := range s.members {
|
|
count += v.set.Size()
|
|
}
|
|
return count
|
|
}
|
|
|
|
// Empty returns false if there's at least one member in some child set.
|
|
func (s *SetNodeMap) Empty() bool {
|
|
for _, n := range s.members {
|
|
if !n.set.Empty() {
|
|
return false
|
|
}
|
|
}
|
|
return true
|
|
}
|
|
|
|
// Get returns (the associated set, true) or (nil, false) if there is none.
|
|
func (s *SetNodeMap) Get(pe PathElement) (*Set, bool) {
|
|
loc := sort.Search(len(s.members), func(i int) bool {
|
|
return !s.members[i].pathElement.Less(pe)
|
|
})
|
|
if loc == len(s.members) {
|
|
return nil, false
|
|
}
|
|
if s.members[loc].pathElement.Equals(pe) {
|
|
return s.members[loc].set, true
|
|
}
|
|
return nil, false
|
|
}
|
|
|
|
// Equals returns true if s and s2 have the same structure (same nested
|
|
// child sets).
|
|
func (s *SetNodeMap) Equals(s2 *SetNodeMap) bool {
|
|
if len(s.members) != len(s2.members) {
|
|
return false
|
|
}
|
|
for i := range s.members {
|
|
if !s.members[i].pathElement.Equals(s2.members[i].pathElement) {
|
|
return false
|
|
}
|
|
if !s.members[i].set.Equals(s2.members[i].set) {
|
|
return false
|
|
}
|
|
}
|
|
return true
|
|
}
|
|
|
|
// Union returns a SetNodeMap with members that appear in either s or s2.
|
|
func (s *SetNodeMap) Union(s2 *SetNodeMap) *SetNodeMap {
|
|
out := &SetNodeMap{}
|
|
|
|
i, j := 0, 0
|
|
for i < len(s.members) && j < len(s2.members) {
|
|
if s.members[i].pathElement.Less(s2.members[j].pathElement) {
|
|
out.members = append(out.members, s.members[i])
|
|
i++
|
|
} else {
|
|
if !s2.members[j].pathElement.Less(s.members[i].pathElement) {
|
|
out.members = append(out.members, setNode{pathElement: s.members[i].pathElement, set: s.members[i].set.Union(s2.members[j].set)})
|
|
i++
|
|
} else {
|
|
out.members = append(out.members, s2.members[j])
|
|
}
|
|
j++
|
|
}
|
|
}
|
|
|
|
if i < len(s.members) {
|
|
out.members = append(out.members, s.members[i:]...)
|
|
}
|
|
if j < len(s2.members) {
|
|
out.members = append(out.members, s2.members[j:]...)
|
|
}
|
|
return out
|
|
}
|
|
|
|
// Intersection returns a SetNodeMap with members that appear in both s and s2.
|
|
func (s *SetNodeMap) Intersection(s2 *SetNodeMap) *SetNodeMap {
|
|
out := &SetNodeMap{}
|
|
|
|
i, j := 0, 0
|
|
for i < len(s.members) && j < len(s2.members) {
|
|
if s.members[i].pathElement.Less(s2.members[j].pathElement) {
|
|
i++
|
|
} else {
|
|
if !s2.members[j].pathElement.Less(s.members[i].pathElement) {
|
|
res := s.members[i].set.Intersection(s2.members[j].set)
|
|
if !res.Empty() {
|
|
out.members = append(out.members, setNode{pathElement: s.members[i].pathElement, set: res})
|
|
}
|
|
i++
|
|
}
|
|
j++
|
|
}
|
|
}
|
|
return out
|
|
}
|
|
|
|
// Difference returns a SetNodeMap with members that appear in s but not in s2.
|
|
func (s *SetNodeMap) Difference(s2 *Set) *SetNodeMap {
|
|
out := &SetNodeMap{}
|
|
|
|
i, j := 0, 0
|
|
for i < len(s.members) && j < len(s2.Children.members) {
|
|
if s.members[i].pathElement.Less(s2.Children.members[j].pathElement) {
|
|
out.members = append(out.members, setNode{pathElement: s.members[i].pathElement, set: s.members[i].set})
|
|
i++
|
|
} else {
|
|
if !s2.Children.members[j].pathElement.Less(s.members[i].pathElement) {
|
|
|
|
diff := s.members[i].set.Difference(s2.Children.members[j].set)
|
|
// We aren't permitted to add nodes with no elements.
|
|
if !diff.Empty() {
|
|
out.members = append(out.members, setNode{pathElement: s.members[i].pathElement, set: diff})
|
|
}
|
|
|
|
i++
|
|
}
|
|
j++
|
|
}
|
|
}
|
|
|
|
if i < len(s.members) {
|
|
out.members = append(out.members, s.members[i:]...)
|
|
}
|
|
return out
|
|
}
|
|
|
|
// RecursiveDifference returns a SetNodeMap with members that appear in s but not in s2.
|
|
//
|
|
// Compared to a regular difference,
|
|
// this removes every field **and its children** from s that is contained in s2.
|
|
//
|
|
// For example, with s containing `a.b.c` and s2 containing `a.b`,
|
|
// a RecursiveDifference will result in `a`, as the entire node `a.b` gets removed.
|
|
func (s *SetNodeMap) RecursiveDifference(s2 *Set) *SetNodeMap {
|
|
out := &SetNodeMap{}
|
|
|
|
i, j := 0, 0
|
|
for i < len(s.members) && j < len(s2.Children.members) {
|
|
if s.members[i].pathElement.Less(s2.Children.members[j].pathElement) {
|
|
if !s2.Members.Has(s.members[i].pathElement) {
|
|
out.members = append(out.members, setNode{pathElement: s.members[i].pathElement, set: s.members[i].set})
|
|
}
|
|
i++
|
|
} else {
|
|
if !s2.Children.members[j].pathElement.Less(s.members[i].pathElement) {
|
|
if !s2.Members.Has(s.members[i].pathElement) {
|
|
diff := s.members[i].set.RecursiveDifference(s2.Children.members[j].set)
|
|
if !diff.Empty() {
|
|
out.members = append(out.members, setNode{pathElement: s.members[i].pathElement, set: diff})
|
|
}
|
|
}
|
|
i++
|
|
}
|
|
j++
|
|
}
|
|
}
|
|
|
|
if i < len(s.members) {
|
|
for _, c := range s.members[i:] {
|
|
if !s2.Members.Has(c.pathElement) {
|
|
out.members = append(out.members, c)
|
|
}
|
|
}
|
|
}
|
|
|
|
return out
|
|
}
|
|
|
|
// EnsureNamedFieldsAreMembers returns a set that contains all the named fields along with the leaves.
|
|
func (s *SetNodeMap) EnsureNamedFieldsAreMembers(sc *schema.Schema, tr schema.TypeRef) *SetNodeMap {
|
|
out := make(sortedSetNode, 0, s.Size())
|
|
atom, _ := sc.Resolve(tr)
|
|
for _, member := range s.members {
|
|
tr := schema.TypeRef{}
|
|
if member.pathElement.FieldName != nil && atom.Map != nil {
|
|
tr = atom.Map.ElementType
|
|
if sf, ok := atom.Map.FindField(*member.pathElement.FieldName); ok {
|
|
tr = sf.Type
|
|
}
|
|
} else if member.pathElement.Key != nil && atom.List != nil {
|
|
tr = atom.List.ElementType
|
|
}
|
|
out = append(out, setNode{
|
|
pathElement: member.pathElement,
|
|
set: member.set.EnsureNamedFieldsAreMembers(sc, tr),
|
|
})
|
|
}
|
|
|
|
return &SetNodeMap{
|
|
members: out,
|
|
}
|
|
}
|
|
|
|
// Iterate calls f for each PathElement in the set.
|
|
func (s *SetNodeMap) Iterate(f func(PathElement)) {
|
|
for _, n := range s.members {
|
|
f(n.pathElement)
|
|
}
|
|
}
|
|
|
|
func (s *SetNodeMap) iteratePrefix(prefix Path, f func(Path)) {
|
|
for _, n := range s.members {
|
|
pe := n.pathElement
|
|
n.set.iteratePrefix(append(prefix, pe), f)
|
|
}
|
|
}
|
|
|
|
// Leaves returns a SetNodeMap containing
|
|
// only setNodes with leaf PathElements.
|
|
func (s *SetNodeMap) Leaves() *SetNodeMap {
|
|
out := &SetNodeMap{}
|
|
out.members = make(sortedSetNode, len(s.members))
|
|
for i, n := range s.members {
|
|
out.members[i] = setNode{
|
|
pathElement: n.pathElement,
|
|
set: n.set.Leaves(),
|
|
}
|
|
}
|
|
return out
|
|
}
|