terraform/states/module.go

286 lines
9.9 KiB
Go
Raw Normal View History

states: New package with modern models for Terraform state Our previous state models in the "terraform" package had a few limitations that are addressed here: - Instance attributes were stored as map[string]string with dot-separated keys representing traversals through a data structure. Now that we have a full type system, it's preferable to store it as a real data structure. - The existing state structures skipped over the "resource" concept and went straight to resource instance, requiring heuristics to decide whether a particular resource should appear as a single object or as a list of objects when used in configuration expressions. - Related to the previous point, the state models also used incorrect terminology where "ResourceState" was really a resource instance state and "InstanceState" was really the state of a particular remote object associated with an instance. These new models use the correct names for each of these, introducing the idea of a "ResourceInstanceObject" as the local record of a remote object associated with an instance. This is a first pass at fleshing out a new model for state. Undoubtedly there will be further iterations of this as we work on integrating these new models into the "terraform" package. These new model types no longer serve double-duty as a description of the JSON state file format, since they are for in-memory use only. A subsequent commit will introduce a separate package that deals with persisting state to files and reloading those files later.
2018-06-08 02:27:57 +02:00
package states
import (
"github.com/zclconf/go-cty/cty"
"github.com/hashicorp/terraform/addrs"
)
// Module is a container for the states of objects within a particular module.
type Module struct {
Addr addrs.ModuleInstance
// Resources contains the state for each resource. The keys in this map are
// an implementation detail and must not be used by outside callers.
Resources map[string]*Resource
// OutputValues contains the state for each output value. The keys in this
// map are output value names.
OutputValues map[string]*OutputValue
// LocalValues contains the value for each named output value. The keys
// in this map are local value names.
LocalValues map[string]cty.Value
}
// NewModule constructs an empty module state for the given module address.
func NewModule(addr addrs.ModuleInstance) *Module {
return &Module{
Addr: addr,
Resources: map[string]*Resource{},
OutputValues: map[string]*OutputValue{},
LocalValues: map[string]cty.Value{},
}
}
// Resource returns the state for the resource with the given address within
// the receiving module state, or nil if the requested resource is not tracked
// in the state.
func (ms *Module) Resource(addr addrs.Resource) *Resource {
return ms.Resources[addr.String()]
}
// ResourceInstance returns the state for the resource instance with the given
// address within the receiving module state, or nil if the requested instance
// is not tracked in the state.
func (ms *Module) ResourceInstance(addr addrs.ResourceInstance) *ResourceInstance {
rs := ms.Resource(addr.Resource)
if rs == nil {
return nil
}
return rs.Instance(addr.Key)
}
// SetResourceMeta updates the resource-level metadata for the resource
// with the given address, creating the resource state for it if it doesn't
// already exist.
func (ms *Module) SetResourceMeta(addr addrs.Resource, eachMode EachMode, provider addrs.AbsProviderConfig) {
rs := ms.Resource(addr)
if rs == nil {
rs = &Resource{
Addr: addr,
Instances: map[addrs.InstanceKey]*ResourceInstance{},
}
ms.Resources[addr.String()] = rs
}
rs.EachMode = eachMode
rs.ProviderConfig = provider
}
// RemoveResource removes the entire state for the given resource, taking with
// it any instances associated with the resource. This should generally be
// called only for resource objects whose instances have all been destroyed.
func (ms *Module) RemoveResource(addr addrs.Resource) {
delete(ms.Resources, addr.String())
}
// SetResourceInstanceCurrent saves the given instance object as the current
// generation of the resource instance with the given address, simulataneously
// updating the recorded provider configuration address, dependencies, and
// resource EachMode.
//
// Any existing current instance object for the given resource is overwritten.
// Set obj to nil to remove the primary generation object altogether. If there
// are no deposed objects then the instance will be removed altogether.
//
// The provider address and "each mode" are resource-wide settings and so they
// are updated for all other instances of the same resource as a side-effect of
// this call.
func (ms *Module) SetResourceInstanceCurrent(addr addrs.ResourceInstance, obj *ResourceInstanceObjectSrc, provider addrs.AbsProviderConfig) {
states: New package with modern models for Terraform state Our previous state models in the "terraform" package had a few limitations that are addressed here: - Instance attributes were stored as map[string]string with dot-separated keys representing traversals through a data structure. Now that we have a full type system, it's preferable to store it as a real data structure. - The existing state structures skipped over the "resource" concept and went straight to resource instance, requiring heuristics to decide whether a particular resource should appear as a single object or as a list of objects when used in configuration expressions. - Related to the previous point, the state models also used incorrect terminology where "ResourceState" was really a resource instance state and "InstanceState" was really the state of a particular remote object associated with an instance. These new models use the correct names for each of these, introducing the idea of a "ResourceInstanceObject" as the local record of a remote object associated with an instance. This is a first pass at fleshing out a new model for state. Undoubtedly there will be further iterations of this as we work on integrating these new models into the "terraform" package. These new model types no longer serve double-duty as a description of the JSON state file format, since they are for in-memory use only. A subsequent commit will introduce a separate package that deals with persisting state to files and reloading those files later.
2018-06-08 02:27:57 +02:00
ms.SetResourceMeta(addr.Resource, eachModeForInstanceKey(addr.Key), provider)
rs := ms.Resource(addr.Resource)
is := rs.EnsureInstance(addr.Key)
is.Current = obj
if !is.HasObjects() {
// If we have no objects at all then we'll clean up.
delete(rs.Instances, addr.Key)
}
if rs.EachMode == NoEach && len(rs.Instances) == 0 {
// Also clean up if we only expect to have one instance anyway
// and there are none. We leave the resource behind if an each mode
// is active because an empty list or map of instances is a valid state.
delete(ms.Resources, addr.Resource.String())
}
}
// SetResourceInstanceDeposed saves the given instance object as a deposed
// generation of the resource instance with the given address and deposed key.
//
// Call this method only for pre-existing deposed objects that already have
// a known DeposedKey. For example, this method is useful if reloading objects
// that were persisted to a state file. To mark the current object as deposed,
// use DeposeResourceInstanceObject instead.
//
// The resource that contains the given instance must already exist in the
// state, or this method will panic. Use Resource to check first if its
// presence is not already guaranteed.
//
// Any existing current instance object for the given resource and deposed key
// is overwritten. Set obj to nil to remove the deposed object altogether. If
// the instance is left with no objects after this operation then it will
// be removed from its containing resource altogether.
func (ms *Module) SetResourceInstanceDeposed(addr addrs.ResourceInstance, key DeposedKey, obj *ResourceInstanceObjectSrc, provider addrs.AbsProviderConfig) {
ms.SetResourceMeta(addr.Resource, eachModeForInstanceKey(addr.Key), provider)
states: New package with modern models for Terraform state Our previous state models in the "terraform" package had a few limitations that are addressed here: - Instance attributes were stored as map[string]string with dot-separated keys representing traversals through a data structure. Now that we have a full type system, it's preferable to store it as a real data structure. - The existing state structures skipped over the "resource" concept and went straight to resource instance, requiring heuristics to decide whether a particular resource should appear as a single object or as a list of objects when used in configuration expressions. - Related to the previous point, the state models also used incorrect terminology where "ResourceState" was really a resource instance state and "InstanceState" was really the state of a particular remote object associated with an instance. These new models use the correct names for each of these, introducing the idea of a "ResourceInstanceObject" as the local record of a remote object associated with an instance. This is a first pass at fleshing out a new model for state. Undoubtedly there will be further iterations of this as we work on integrating these new models into the "terraform" package. These new model types no longer serve double-duty as a description of the JSON state file format, since they are for in-memory use only. A subsequent commit will introduce a separate package that deals with persisting state to files and reloading those files later.
2018-06-08 02:27:57 +02:00
rs := ms.Resource(addr.Resource)
is := rs.EnsureInstance(addr.Key)
if obj != nil {
is.Deposed[key] = obj
} else {
delete(is.Deposed, key)
}
states: New package with modern models for Terraform state Our previous state models in the "terraform" package had a few limitations that are addressed here: - Instance attributes were stored as map[string]string with dot-separated keys representing traversals through a data structure. Now that we have a full type system, it's preferable to store it as a real data structure. - The existing state structures skipped over the "resource" concept and went straight to resource instance, requiring heuristics to decide whether a particular resource should appear as a single object or as a list of objects when used in configuration expressions. - Related to the previous point, the state models also used incorrect terminology where "ResourceState" was really a resource instance state and "InstanceState" was really the state of a particular remote object associated with an instance. These new models use the correct names for each of these, introducing the idea of a "ResourceInstanceObject" as the local record of a remote object associated with an instance. This is a first pass at fleshing out a new model for state. Undoubtedly there will be further iterations of this as we work on integrating these new models into the "terraform" package. These new model types no longer serve double-duty as a description of the JSON state file format, since they are for in-memory use only. A subsequent commit will introduce a separate package that deals with persisting state to files and reloading those files later.
2018-06-08 02:27:57 +02:00
if !is.HasObjects() {
// If we have no objects at all then we'll clean up.
delete(rs.Instances, addr.Key)
}
if rs.EachMode == NoEach && len(rs.Instances) == 0 {
// Also clean up if we only expect to have one instance anyway
// and there are none. We leave the resource behind if an each mode
// is active because an empty list or map of instances is a valid state.
delete(ms.Resources, addr.Resource.String())
}
}
// ForgetResourceInstanceAll removes the record of all objects associated with
// the specified resource instance, if present. If not present, this is a no-op.
func (ms *Module) ForgetResourceInstanceAll(addr addrs.ResourceInstance) {
rs := ms.Resource(addr.Resource)
if rs == nil {
return
}
delete(rs.Instances, addr.Key)
if rs.EachMode == NoEach && len(rs.Instances) == 0 {
// Also clean up if we only expect to have one instance anyway
// and there are none. We leave the resource behind if an each mode
// is active because an empty list or map of instances is a valid state.
delete(ms.Resources, addr.Resource.String())
}
}
// ForgetResourceInstanceDeposed removes the record of the deposed object with
// the given address and key, if present. If not present, this is a no-op.
func (ms *Module) ForgetResourceInstanceDeposed(addr addrs.ResourceInstance, key DeposedKey) {
rs := ms.Resource(addr.Resource)
if rs == nil {
return
}
is := rs.Instance(addr.Key)
if is == nil {
return
}
delete(is.Deposed, key)
states: New package with modern models for Terraform state Our previous state models in the "terraform" package had a few limitations that are addressed here: - Instance attributes were stored as map[string]string with dot-separated keys representing traversals through a data structure. Now that we have a full type system, it's preferable to store it as a real data structure. - The existing state structures skipped over the "resource" concept and went straight to resource instance, requiring heuristics to decide whether a particular resource should appear as a single object or as a list of objects when used in configuration expressions. - Related to the previous point, the state models also used incorrect terminology where "ResourceState" was really a resource instance state and "InstanceState" was really the state of a particular remote object associated with an instance. These new models use the correct names for each of these, introducing the idea of a "ResourceInstanceObject" as the local record of a remote object associated with an instance. This is a first pass at fleshing out a new model for state. Undoubtedly there will be further iterations of this as we work on integrating these new models into the "terraform" package. These new model types no longer serve double-duty as a description of the JSON state file format, since they are for in-memory use only. A subsequent commit will introduce a separate package that deals with persisting state to files and reloading those files later.
2018-06-08 02:27:57 +02:00
if !is.HasObjects() {
// If we have no objects at all then we'll clean up.
delete(rs.Instances, addr.Key)
}
if rs.EachMode == NoEach && len(rs.Instances) == 0 {
// Also clean up if we only expect to have one instance anyway
// and there are none. We leave the resource behind if an each mode
// is active because an empty list or map of instances is a valid state.
delete(ms.Resources, addr.Resource.String())
}
}
// deposeResourceInstanceObject is the real implementation of
// SyncState.DeposeResourceInstanceObject.
core: Be more explicit in how we handle create_before_destroy Previously our handling of create_before_destroy -- and of deposed objects in particular -- was rather "implicit" and spread over various different subsystems. We'd quietly just destroy every deposed object during a destroy operation, without any user-visible plan to do so. Here we make things more explicit by tracking each deposed object individually by its pseudorandomly-allocated key. There are two different mechanisms at play here, building on the same concepts: - During a replace operation with create_before_destroy, we *pre-allocate* a DeposedKey to use for the prior object in the "apply" node and then pass that exact id to the destroy node, ensuring that we only destroy the single object we planned to destroy. In the happy path here the user never actually sees the allocated deposed key because we use it and then immediately destroy it within the same operation. However, that destroy may fail, which brings us to the second mechanism: - If any deposed objects are already present in state during _plan_, we insert a destroy change for them into the plan so that it's explicit to the user that we are going to destroy these additional objects, and then create an individual graph node for each one in DiffTransformer. The main motivation here is to be more careful in how we handle these destroys so that from a user's standpoint we never destroy something without the user knowing about it ahead of time. However, this new organization also hopefully makes the code itself a little easier to follow because the connection between the create and destroy steps of a Replace is reprseented in a single place (in DiffTransformer) and deposed instances each have their own explicit graph node rather than being secretly handled as part of the main instance-level graph node.
2018-09-20 21:30:52 +02:00
func (ms *Module) deposeResourceInstanceObject(addr addrs.ResourceInstance, forceKey DeposedKey) DeposedKey {
states: New package with modern models for Terraform state Our previous state models in the "terraform" package had a few limitations that are addressed here: - Instance attributes were stored as map[string]string with dot-separated keys representing traversals through a data structure. Now that we have a full type system, it's preferable to store it as a real data structure. - The existing state structures skipped over the "resource" concept and went straight to resource instance, requiring heuristics to decide whether a particular resource should appear as a single object or as a list of objects when used in configuration expressions. - Related to the previous point, the state models also used incorrect terminology where "ResourceState" was really a resource instance state and "InstanceState" was really the state of a particular remote object associated with an instance. These new models use the correct names for each of these, introducing the idea of a "ResourceInstanceObject" as the local record of a remote object associated with an instance. This is a first pass at fleshing out a new model for state. Undoubtedly there will be further iterations of this as we work on integrating these new models into the "terraform" package. These new model types no longer serve double-duty as a description of the JSON state file format, since they are for in-memory use only. A subsequent commit will introduce a separate package that deals with persisting state to files and reloading those files later.
2018-06-08 02:27:57 +02:00
is := ms.ResourceInstance(addr)
if is == nil {
return NotDeposed
}
core: Be more explicit in how we handle create_before_destroy Previously our handling of create_before_destroy -- and of deposed objects in particular -- was rather "implicit" and spread over various different subsystems. We'd quietly just destroy every deposed object during a destroy operation, without any user-visible plan to do so. Here we make things more explicit by tracking each deposed object individually by its pseudorandomly-allocated key. There are two different mechanisms at play here, building on the same concepts: - During a replace operation with create_before_destroy, we *pre-allocate* a DeposedKey to use for the prior object in the "apply" node and then pass that exact id to the destroy node, ensuring that we only destroy the single object we planned to destroy. In the happy path here the user never actually sees the allocated deposed key because we use it and then immediately destroy it within the same operation. However, that destroy may fail, which brings us to the second mechanism: - If any deposed objects are already present in state during _plan_, we insert a destroy change for them into the plan so that it's explicit to the user that we are going to destroy these additional objects, and then create an individual graph node for each one in DiffTransformer. The main motivation here is to be more careful in how we handle these destroys so that from a user's standpoint we never destroy something without the user knowing about it ahead of time. However, this new organization also hopefully makes the code itself a little easier to follow because the connection between the create and destroy steps of a Replace is reprseented in a single place (in DiffTransformer) and deposed instances each have their own explicit graph node rather than being secretly handled as part of the main instance-level graph node.
2018-09-20 21:30:52 +02:00
return is.deposeCurrentObject(forceKey)
states: New package with modern models for Terraform state Our previous state models in the "terraform" package had a few limitations that are addressed here: - Instance attributes were stored as map[string]string with dot-separated keys representing traversals through a data structure. Now that we have a full type system, it's preferable to store it as a real data structure. - The existing state structures skipped over the "resource" concept and went straight to resource instance, requiring heuristics to decide whether a particular resource should appear as a single object or as a list of objects when used in configuration expressions. - Related to the previous point, the state models also used incorrect terminology where "ResourceState" was really a resource instance state and "InstanceState" was really the state of a particular remote object associated with an instance. These new models use the correct names for each of these, introducing the idea of a "ResourceInstanceObject" as the local record of a remote object associated with an instance. This is a first pass at fleshing out a new model for state. Undoubtedly there will be further iterations of this as we work on integrating these new models into the "terraform" package. These new model types no longer serve double-duty as a description of the JSON state file format, since they are for in-memory use only. A subsequent commit will introduce a separate package that deals with persisting state to files and reloading those files later.
2018-06-08 02:27:57 +02:00
}
// maybeRestoreResourceInstanceDeposed is the real implementation of
// SyncState.MaybeRestoreResourceInstanceDeposed.
func (ms *Module) maybeRestoreResourceInstanceDeposed(addr addrs.ResourceInstance, key DeposedKey) bool {
rs := ms.Resource(addr.Resource)
if rs == nil {
return false
}
is := rs.Instance(addr.Key)
if is == nil {
return false
}
if is.Current != nil {
return false
}
if len(is.Deposed) == 0 {
return false
}
is.Current = is.Deposed[key]
delete(is.Deposed, key)
return true
}
states: New package with modern models for Terraform state Our previous state models in the "terraform" package had a few limitations that are addressed here: - Instance attributes were stored as map[string]string with dot-separated keys representing traversals through a data structure. Now that we have a full type system, it's preferable to store it as a real data structure. - The existing state structures skipped over the "resource" concept and went straight to resource instance, requiring heuristics to decide whether a particular resource should appear as a single object or as a list of objects when used in configuration expressions. - Related to the previous point, the state models also used incorrect terminology where "ResourceState" was really a resource instance state and "InstanceState" was really the state of a particular remote object associated with an instance. These new models use the correct names for each of these, introducing the idea of a "ResourceInstanceObject" as the local record of a remote object associated with an instance. This is a first pass at fleshing out a new model for state. Undoubtedly there will be further iterations of this as we work on integrating these new models into the "terraform" package. These new model types no longer serve double-duty as a description of the JSON state file format, since they are for in-memory use only. A subsequent commit will introduce a separate package that deals with persisting state to files and reloading those files later.
2018-06-08 02:27:57 +02:00
// SetOutputValue writes an output value into the state, overwriting any
// existing value of the same name.
func (ms *Module) SetOutputValue(name string, value cty.Value, sensitive bool) *OutputValue {
os := &OutputValue{
Value: value,
Sensitive: sensitive,
}
ms.OutputValues[name] = os
return os
}
// RemoveOutputValue removes the output value of the given name from the state,
// if it exists. This method is a no-op if there is no value of the given
// name.
func (ms *Module) RemoveOutputValue(name string) {
delete(ms.OutputValues, name)
}
// SetLocalValue writes a local value into the state, overwriting any
// existing value of the same name.
func (ms *Module) SetLocalValue(name string, value cty.Value) {
ms.LocalValues[name] = value
}
// RemoveLocalValue removes the local value of the given name from the state,
// if it exists. This method is a no-op if there is no value of the given
// name.
func (ms *Module) RemoveLocalValue(name string) {
delete(ms.LocalValues, name)
}
// PruneResourceHusks is a specialized method that will remove any Resource
// objects that do not contain any instances, even if they have an EachMode.
//
// You probably shouldn't call this! See the method of the same name on
// type State for more information on what this is for and the rare situations
// where it is safe to use.
func (ms *Module) PruneResourceHusks() {
for _, rs := range ms.Resources {
if len(rs.Instances) == 0 {
ms.RemoveResource(rs.Addr)
}
}
}
// empty returns true if the receving module state is contributing nothing
// to the state. In other words, it returns true if the module could be
// removed from the state altogether without changing the meaning of the state.
//
// In practice a module containing no objects is the same as a non-existent
// module, and so we can opportunistically clean up once a module becomes
// empty on the assumption that it will be re-added if needed later.
func (ms *Module) empty() bool {
if ms == nil {
return true
}
// This must be updated to cover any new collections added to Module
// in future.
return (len(ms.Resources) == 0 &&
len(ms.OutputValues) == 0 &&
len(ms.LocalValues) == 0)
}