You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
boundary/internal/perms/acl.go

141 lines
3.8 KiB

package perms
/*
A really useful page to be aware of when looking at ACLs is
https://hashicorp.atlassian.net/wiki/spaces/ICU/pages/866976600/API+Actions+and+Permissions
speaking of which: TODO: put that chart in public docs.
Anyways, from that page you can see that there are really only a few patterns of
ACLs that are ever allowed; see the Allowed function for a description along
with the code.
This makes it actually quite simple to perform the ACL checking. Much of ACL
construction is thus synthesizing something reasonable from a set of Grants.
*/
import (
"github.com/hashicorp/boundary/internal/types/action"
"github.com/hashicorp/boundary/internal/types/resource"
)
// ACL provides an entry point into the permissions engine for determining if an
// action is allowed on a resource based on a principal's (user or group) grants.
type ACL struct {
scopeMap map[string][]Grant
}
// ACLResults provides a type for the permission's engine results so that we can
// pass more detailed information along in the future if we want. It was useful
// in Vault, may be useful here.
type ACLResults struct {
Allowed bool
// This is included but unexported for testing/debugging
scopeMap map[string][]Grant
}
// Resource defines something within boundary that requires authorization
// capabilities. Resources must have a ScopeId.
type Resource struct {
// ScopeId is the scope that contains the Resource.
ScopeId string
// Id is the public id of the resource.
Id string
// Type of resource.
Type resource.Type
// Pin if defined would constrain the resource within the collection of the
// pin id.
Pin string
}
// NewACL creates an ACL from the grants provided.
func NewACL(grants ...Grant) ACL {
ret := ACL{
scopeMap: make(map[string][]Grant, len(grants)),
}
for _, grant := range grants {
ret.scopeMap[grant.scope.Id] = append(ret.scopeMap[grant.scope.Id], grant)
}
return ret
}
// Allowed determines if the grants for an ACL allow an action for a resource.
func (a ACL) Allowed(r Resource, aType action.Type) (results ACLResults) {
// First, get the grants within the specified scope
grants := a.scopeMap[r.ScopeId]
results.scopeMap = a.scopeMap
// Now, go through and check the cases indicated above
for _, grant := range grants {
if !(grant.actions[aType] || grant.actions[action.All]) {
continue
}
switch {
// id=<resource.id>;actions=<action> where ID cannot be a wildcard
case grant.id == r.Id &&
grant.id != "" &&
grant.id != "*" &&
grant.typ == resource.Unknown:
results.Allowed = true
return
// type=<resource.type>;actions=<action> when action is list or create.
// Must be a top level collection, otherwise must be one of the two
// formats specified below.
case grant.id == "" &&
r.Id == "" &&
grant.typ == r.Type &&
grant.typ != resource.Unknown &&
topLevelType(r.Type) &&
(aType == action.List || aType == action.Create):
results.Allowed = true
return
// id=*;type=<resource.type>;actions=<action> where type cannot be
// unknown but can be a wildcard to allow any resource at all
case grant.id == "*" &&
grant.typ != resource.Unknown &&
(grant.typ == r.Type ||
grant.typ == resource.All):
results.Allowed = true
return
// id=<pin>;type=<resource.type>;actions=<action> where type can be a
// wildcard and this this is operating on a non-top-level type
case grant.id != "" &&
grant.id == r.Pin &&
grant.typ != resource.Unknown &&
(grant.typ == r.Type || grant.typ == resource.All) &&
!topLevelType(r.Type):
results.Allowed = true
return
}
}
return
}
func topLevelType(typ resource.Type) bool {
switch typ {
case resource.AuthMethod,
resource.AuthToken,
resource.Group,
resource.HostCatalog,
resource.Role,
resource.Scope,
resource.Session,
resource.Target,
resource.User:
return true
}
return false
}