2016-09-13 12:56:37 -05:00
|
|
|
package terraform
|
|
|
|
|
|
|
|
import (
|
2021-05-17 14:00:50 -05:00
|
|
|
"github.com/hashicorp/terraform/internal/addrs"
|
2021-05-17 14:17:09 -05:00
|
|
|
"github.com/hashicorp/terraform/internal/configs"
|
2021-05-17 11:30:37 -05:00
|
|
|
"github.com/hashicorp/terraform/internal/dag"
|
2021-05-17 14:33:17 -05:00
|
|
|
"github.com/hashicorp/terraform/internal/plans"
|
2021-05-17 14:43:35 -05:00
|
|
|
"github.com/hashicorp/terraform/internal/states"
|
2021-05-17 12:11:06 -05:00
|
|
|
"github.com/hashicorp/terraform/internal/tfdiags"
|
2016-09-13 12:56:37 -05:00
|
|
|
)
|
|
|
|
|
|
|
|
// ApplyGraphBuilder implements GraphBuilder and is responsible for building
|
|
|
|
// a graph for applying a Terraform diff.
|
|
|
|
//
|
|
|
|
// Because the graph is built from the diff (vs. the config or state),
|
|
|
|
// this helps ensure that the apply-time graph doesn't modify any resources
|
|
|
|
// that aren't explicitly in the diff. There are other scenarios where the
|
|
|
|
// diff can be deviated, so this is just one layer of protection.
|
|
|
|
type ApplyGraphBuilder struct {
|
terraform: ugly huge change to weave in new HCL2-oriented types
Due to how deeply the configuration types go into Terraform Core, there
isn't a great way to switch out to HCL2 gradually. As a consequence, this
huge commit gets us from the old state to a _compilable_ new state, but
does not yet attempt to fix any tests and has a number of known missing
parts and bugs. We will continue to iterate on this in forthcoming
commits, heading back towards passing tests and making Terraform
fully-functional again.
The three main goals here are:
- Use the configuration models from the "configs" package instead of the
older models in the "config" package, which is now deprecated and
preserved only to help us write our migration tool.
- Do expression inspection and evaluation using the functionality of the
new "lang" package, instead of the Interpolator type and related
functionality in the main "terraform" package.
- Represent addresses of various objects using types in the addrs package,
rather than hand-constructed strings. This is not critical to support
the above, but was a big help during the implementation of these other
points since it made it much more explicit what kind of address is
expected in each context.
Since our new packages are built to accommodate some future planned
features that are not yet implemented (e.g. the "for_each" argument on
resources, "count"/"for_each" on modules), and since there's still a fair
amount of functionality still using old-style APIs, there is a moderate
amount of shimming here to connect new assumptions with old, hopefully in
a way that makes it easier to find and eliminate these shims later.
I apologize in advance to the person who inevitably just found this huge
commit while spelunking through the commit history.
2018-04-30 12:33:53 -05:00
|
|
|
// Config is the configuration tree that the diff was built from.
|
|
|
|
Config *configs.Config
|
2016-09-13 12:56:37 -05:00
|
|
|
|
terraform: Ugly huge change to weave in new State and Plan types
Due to how often the state and plan types are referenced throughout
Terraform, there isn't a great way to switch them out gradually. As a
consequence, this huge commit gets us from the old world to a _compilable_
new world, but still has a large number of known test failures due to
key functionality being stubbed out.
The stubs here are for anything that interacts with providers, since we
now need to do the follow-up work to similarly replace the old
terraform.ResourceProvider interface with its replacement in the new
"providers" package. That work, along with work to fix the remaining
failing tests, will follow in subsequent commits.
The aim here was to replace all references to terraform.State and its
downstream types with states.State, terraform.Plan with plans.Plan,
state.State with statemgr.State, and switch to the new implementations of
the state and plan file formats. However, due to the number of times those
types are used, this also ended up affecting numerous other parts of core
such as terraform.Hook, the backend.Backend interface, and most of the CLI
commands.
Just as with 5861dbf3fc49b19587a31816eb06f511ab861bb4 before, I apologize
in advance to the person who inevitably just found this huge commit while
spelunking through the commit history.
2018-08-14 16:24:45 -05:00
|
|
|
// Changes describes the changes that we need apply.
|
|
|
|
Changes *plans.Changes
|
2016-09-13 12:56:37 -05:00
|
|
|
|
2016-09-13 19:52:09 -05:00
|
|
|
// State is the current state
|
terraform: Ugly huge change to weave in new State and Plan types
Due to how often the state and plan types are referenced throughout
Terraform, there isn't a great way to switch them out gradually. As a
consequence, this huge commit gets us from the old world to a _compilable_
new world, but still has a large number of known test failures due to
key functionality being stubbed out.
The stubs here are for anything that interacts with providers, since we
now need to do the follow-up work to similarly replace the old
terraform.ResourceProvider interface with its replacement in the new
"providers" package. That work, along with work to fix the remaining
failing tests, will follow in subsequent commits.
The aim here was to replace all references to terraform.State and its
downstream types with states.State, terraform.Plan with plans.Plan,
state.State with statemgr.State, and switch to the new implementations of
the state and plan file formats. However, due to the number of times those
types are used, this also ended up affecting numerous other parts of core
such as terraform.Hook, the backend.Backend interface, and most of the CLI
commands.
Just as with 5861dbf3fc49b19587a31816eb06f511ab861bb4 before, I apologize
in advance to the person who inevitably just found this huge commit while
spelunking through the commit history.
2018-08-14 16:24:45 -05:00
|
|
|
State *states.State
|
2016-09-13 19:52:09 -05:00
|
|
|
|
core: Handle root and child module input variables consistently
Previously we had a significant discrepancy between these two situations:
we wrote the raw root module variables directly into the EvalContext and
then applied type conversions only at expression evaluation time, while
for child modules we converted and validated the values while visiting
the variable graph node and wrote only the _final_ value into the
EvalContext.
This confusion seems to have been the root cause for #29899, where
validation rules for root module variables were being applied at the wrong
point in the process, prior to type conversion.
To fix that bug and also make similar mistakes less likely in the future,
I've made the root module variable handling more like the child module
variable handling in the following ways:
- The "raw value" (exactly as given by the user) lives only in the graph
node representing the variable, which mirrors how the _expression_
for a child module variable lives in its graph node. This means that
the flow for the two is the same except that there's no expression
evaluation step for root module variables, because they arrive as
constant values from the caller.
- The set of variable values in the EvalContext is always only "final"
values, after type conversion is complete. That in turn means we no
longer need to do "just in time" conversion in
evaluationStateData.GetInputVariable, and can just return the value
exactly as stored, which is consistent with how we handle all other
references between objects.
This diff is noisier than I'd like because of how much it takes to wire
a new argument (the raw variable values) through to the plan graph builder,
but those changes are pretty mechanical and the interesting logic lives
inside the plan graph builder itself, in NodeRootVariable, and
the shared helper functions in eval_variable.go.
While here I also took the opportunity to fix a historical API wart in
EvalContext, where SetModuleCallArguments was built to take a set of
variable values all at once but our current caller always calls with only
one at a time. That is now just SetModuleCallArgument singular, to match
with the new SetRootModuleArgument to deal with root module variables.
2021-11-10 19:29:45 -06:00
|
|
|
// RootVariableValues are the root module input variables captured as
|
|
|
|
// part of the plan object, which we must reproduce in the apply step
|
|
|
|
// to get a consistent result.
|
|
|
|
RootVariableValues InputValues
|
|
|
|
|
2021-08-31 12:58:05 -05:00
|
|
|
// Plugins is a library of the plug-in components (providers and
|
2018-05-02 22:16:22 -05:00
|
|
|
// provisioners) available for use.
|
2021-08-31 12:58:05 -05:00
|
|
|
Plugins *contextPlugins
|
2016-09-22 13:03:03 -05:00
|
|
|
|
2017-02-13 14:52:45 -06:00
|
|
|
// Targets are resources to target. This is only required to make sure
|
|
|
|
// unnecessary outputs aren't included in the apply graph. The plan
|
|
|
|
// builder successfully handles targeting resources. In the future,
|
|
|
|
// outputs should go into the diff so that this is unnecessary.
|
terraform: ugly huge change to weave in new HCL2-oriented types
Due to how deeply the configuration types go into Terraform Core, there
isn't a great way to switch out to HCL2 gradually. As a consequence, this
huge commit gets us from the old state to a _compilable_ new state, but
does not yet attempt to fix any tests and has a number of known missing
parts and bugs. We will continue to iterate on this in forthcoming
commits, heading back towards passing tests and making Terraform
fully-functional again.
The three main goals here are:
- Use the configuration models from the "configs" package instead of the
older models in the "config" package, which is now deprecated and
preserved only to help us write our migration tool.
- Do expression inspection and evaluation using the functionality of the
new "lang" package, instead of the Interpolator type and related
functionality in the main "terraform" package.
- Represent addresses of various objects using types in the addrs package,
rather than hand-constructed strings. This is not critical to support
the above, but was a big help during the implementation of these other
points since it made it much more explicit what kind of address is
expected in each context.
Since our new packages are built to accommodate some future planned
features that are not yet implemented (e.g. the "for_each" argument on
resources, "count"/"for_each" on modules), and since there's still a fair
amount of functionality still using old-style APIs, there is a moderate
amount of shimming here to connect new assumptions with old, hopefully in
a way that makes it easier to find and eliminate these shims later.
I apologize in advance to the person who inevitably just found this huge
commit while spelunking through the commit history.
2018-04-30 12:33:53 -05:00
|
|
|
Targets []addrs.Targetable
|
2017-02-13 14:52:45 -06:00
|
|
|
|
2021-04-06 19:37:38 -05:00
|
|
|
// ForceReplace are the resource instance addresses that the user
|
|
|
|
// requested to force replacement for when creating the plan, if any.
|
|
|
|
// The apply step refers to these as part of verifying that the planned
|
|
|
|
// actions remain consistent between plan and apply.
|
|
|
|
ForceReplace []addrs.AbsResourceInstance
|
2022-10-18 14:34:37 -05:00
|
|
|
|
|
|
|
// Plan Operation this graph will be used for.
|
|
|
|
Operation walkOperation
|
2016-09-13 12:56:37 -05:00
|
|
|
}
|
|
|
|
|
|
|
|
// See GraphBuilder
|
terraform: ugly huge change to weave in new HCL2-oriented types
Due to how deeply the configuration types go into Terraform Core, there
isn't a great way to switch out to HCL2 gradually. As a consequence, this
huge commit gets us from the old state to a _compilable_ new state, but
does not yet attempt to fix any tests and has a number of known missing
parts and bugs. We will continue to iterate on this in forthcoming
commits, heading back towards passing tests and making Terraform
fully-functional again.
The three main goals here are:
- Use the configuration models from the "configs" package instead of the
older models in the "config" package, which is now deprecated and
preserved only to help us write our migration tool.
- Do expression inspection and evaluation using the functionality of the
new "lang" package, instead of the Interpolator type and related
functionality in the main "terraform" package.
- Represent addresses of various objects using types in the addrs package,
rather than hand-constructed strings. This is not critical to support
the above, but was a big help during the implementation of these other
points since it made it much more explicit what kind of address is
expected in each context.
Since our new packages are built to accommodate some future planned
features that are not yet implemented (e.g. the "for_each" argument on
resources, "count"/"for_each" on modules), and since there's still a fair
amount of functionality still using old-style APIs, there is a moderate
amount of shimming here to connect new assumptions with old, hopefully in
a way that makes it easier to find and eliminate these shims later.
I apologize in advance to the person who inevitably just found this huge
commit while spelunking through the commit history.
2018-04-30 12:33:53 -05:00
|
|
|
func (b *ApplyGraphBuilder) Build(path addrs.ModuleInstance) (*Graph, tfdiags.Diagnostics) {
|
2016-09-13 12:56:37 -05:00
|
|
|
return (&BasicGraphBuilder{
|
2022-03-11 08:31:24 -06:00
|
|
|
Steps: b.Steps(),
|
|
|
|
Name: "ApplyGraphBuilder",
|
2016-09-13 12:56:37 -05:00
|
|
|
}).Build(path)
|
|
|
|
}
|
|
|
|
|
|
|
|
// See GraphBuilder
|
|
|
|
func (b *ApplyGraphBuilder) Steps() []GraphTransformer {
|
2016-09-15 01:58:16 -05:00
|
|
|
// Custom factory for creating providers.
|
2016-12-03 17:27:38 -06:00
|
|
|
concreteProvider := func(a *NodeAbstractProvider) dag.Vertex {
|
2016-09-15 01:58:16 -05:00
|
|
|
return &NodeApplyableProvider{
|
2016-12-03 17:27:38 -06:00
|
|
|
NodeAbstractProvider: a,
|
2016-09-15 01:58:16 -05:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2018-09-11 16:22:25 -05:00
|
|
|
concreteResource := func(a *NodeAbstractResource) dag.Vertex {
|
2020-03-20 14:45:39 -05:00
|
|
|
return &nodeExpandApplyableResource{
|
2018-09-11 16:22:25 -05:00
|
|
|
NodeAbstractResource: a,
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
concreteResourceInstance := func(a *NodeAbstractResourceInstance) dag.Vertex {
|
terraform: ugly huge change to weave in new HCL2-oriented types
Due to how deeply the configuration types go into Terraform Core, there
isn't a great way to switch out to HCL2 gradually. As a consequence, this
huge commit gets us from the old state to a _compilable_ new state, but
does not yet attempt to fix any tests and has a number of known missing
parts and bugs. We will continue to iterate on this in forthcoming
commits, heading back towards passing tests and making Terraform
fully-functional again.
The three main goals here are:
- Use the configuration models from the "configs" package instead of the
older models in the "config" package, which is now deprecated and
preserved only to help us write our migration tool.
- Do expression inspection and evaluation using the functionality of the
new "lang" package, instead of the Interpolator type and related
functionality in the main "terraform" package.
- Represent addresses of various objects using types in the addrs package,
rather than hand-constructed strings. This is not critical to support
the above, but was a big help during the implementation of these other
points since it made it much more explicit what kind of address is
expected in each context.
Since our new packages are built to accommodate some future planned
features that are not yet implemented (e.g. the "for_each" argument on
resources, "count"/"for_each" on modules), and since there's still a fair
amount of functionality still using old-style APIs, there is a moderate
amount of shimming here to connect new assumptions with old, hopefully in
a way that makes it easier to find and eliminate these shims later.
I apologize in advance to the person who inevitably just found this huge
commit while spelunking through the commit history.
2018-04-30 12:33:53 -05:00
|
|
|
return &NodeApplyableResourceInstance{
|
|
|
|
NodeAbstractResourceInstance: a,
|
2021-04-06 19:37:38 -05:00
|
|
|
forceReplace: b.ForceReplace,
|
2016-09-21 16:30:41 -05:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2016-09-13 12:56:37 -05:00
|
|
|
steps := []GraphTransformer{
|
2018-09-11 16:22:25 -05:00
|
|
|
// Creates all the resources represented in the config. During apply,
|
|
|
|
// we use this just to ensure that the whole-resource metadata is
|
|
|
|
// updated to reflect things such as whether the count argument is
|
|
|
|
// set in config, or which provider configuration manages each resource.
|
|
|
|
&ConfigTransformer{
|
2016-09-21 16:30:41 -05:00
|
|
|
Concrete: concreteResource,
|
2018-09-11 16:22:25 -05:00
|
|
|
Config: b.Config,
|
|
|
|
},
|
|
|
|
|
2020-10-06 16:39:53 -05:00
|
|
|
// Add dynamic values
|
core: Handle root and child module input variables consistently
Previously we had a significant discrepancy between these two situations:
we wrote the raw root module variables directly into the EvalContext and
then applied type conversions only at expression evaluation time, while
for child modules we converted and validated the values while visiting
the variable graph node and wrote only the _final_ value into the
EvalContext.
This confusion seems to have been the root cause for #29899, where
validation rules for root module variables were being applied at the wrong
point in the process, prior to type conversion.
To fix that bug and also make similar mistakes less likely in the future,
I've made the root module variable handling more like the child module
variable handling in the following ways:
- The "raw value" (exactly as given by the user) lives only in the graph
node representing the variable, which mirrors how the _expression_
for a child module variable lives in its graph node. This means that
the flow for the two is the same except that there's no expression
evaluation step for root module variables, because they arrive as
constant values from the caller.
- The set of variable values in the EvalContext is always only "final"
values, after type conversion is complete. That in turn means we no
longer need to do "just in time" conversion in
evaluationStateData.GetInputVariable, and can just return the value
exactly as stored, which is consistent with how we handle all other
references between objects.
This diff is noisier than I'd like because of how much it takes to wire
a new argument (the raw variable values) through to the plan graph builder,
but those changes are pretty mechanical and the interesting logic lives
inside the plan graph builder itself, in NodeRootVariable, and
the shared helper functions in eval_variable.go.
While here I also took the opportunity to fix a historical API wart in
EvalContext, where SetModuleCallArguments was built to take a set of
variable values all at once but our current caller always calls with only
one at a time. That is now just SetModuleCallArgument singular, to match
with the new SetRootModuleArgument to deal with root module variables.
2021-11-10 19:29:45 -06:00
|
|
|
&RootVariableTransformer{Config: b.Config, RawValues: b.RootVariableValues},
|
2020-10-06 16:39:53 -05:00
|
|
|
&ModuleVariableTransformer{Config: b.Config},
|
|
|
|
&LocalTransformer{Config: b.Config},
|
2022-10-18 14:34:37 -05:00
|
|
|
&OutputTransformer{
|
|
|
|
Config: b.Config,
|
2022-10-20 08:36:10 -05:00
|
|
|
ApplyDestroy: b.Operation == walkDestroy,
|
2022-10-18 14:34:37 -05:00
|
|
|
},
|
2020-10-06 16:39:53 -05:00
|
|
|
|
2018-09-11 16:22:25 -05:00
|
|
|
// Creates all the resource instances represented in the diff, along
|
|
|
|
// with dependency edges against the whole-resource nodes added by
|
|
|
|
// ConfigTransformer above.
|
|
|
|
&DiffTransformer{
|
|
|
|
Concrete: concreteResourceInstance,
|
2018-09-20 14:30:52 -05:00
|
|
|
State: b.State,
|
terraform: Ugly huge change to weave in new State and Plan types
Due to how often the state and plan types are referenced throughout
Terraform, there isn't a great way to switch them out gradually. As a
consequence, this huge commit gets us from the old world to a _compilable_
new world, but still has a large number of known test failures due to
key functionality being stubbed out.
The stubs here are for anything that interacts with providers, since we
now need to do the follow-up work to similarly replace the old
terraform.ResourceProvider interface with its replacement in the new
"providers" package. That work, along with work to fix the remaining
failing tests, will follow in subsequent commits.
The aim here was to replace all references to terraform.State and its
downstream types with states.State, terraform.Plan with plans.Plan,
state.State with statemgr.State, and switch to the new implementations of
the state and plan file formats. However, due to the number of times those
types are used, this also ended up affecting numerous other parts of core
such as terraform.Hook, the backend.Backend interface, and most of the CLI
commands.
Just as with 5861dbf3fc49b19587a31816eb06f511ab861bb4 before, I apologize
in advance to the person who inevitably just found this huge commit while
spelunking through the commit history.
2018-08-14 16:24:45 -05:00
|
|
|
Changes: b.Changes,
|
2022-10-26 11:04:08 -05:00
|
|
|
Config: b.Config,
|
2016-09-13 19:52:09 -05:00
|
|
|
},
|
2016-09-13 16:34:15 -05:00
|
|
|
|
2020-10-06 16:39:53 -05:00
|
|
|
// Attach the state
|
|
|
|
&AttachStateTransformer{State: b.State},
|
|
|
|
|
2016-09-18 23:45:19 -05:00
|
|
|
// Create orphan output nodes
|
terraform: ugly huge change to weave in new HCL2-oriented types
Due to how deeply the configuration types go into Terraform Core, there
isn't a great way to switch out to HCL2 gradually. As a consequence, this
huge commit gets us from the old state to a _compilable_ new state, but
does not yet attempt to fix any tests and has a number of known missing
parts and bugs. We will continue to iterate on this in forthcoming
commits, heading back towards passing tests and making Terraform
fully-functional again.
The three main goals here are:
- Use the configuration models from the "configs" package instead of the
older models in the "config" package, which is now deprecated and
preserved only to help us write our migration tool.
- Do expression inspection and evaluation using the functionality of the
new "lang" package, instead of the Interpolator type and related
functionality in the main "terraform" package.
- Represent addresses of various objects using types in the addrs package,
rather than hand-constructed strings. This is not critical to support
the above, but was a big help during the implementation of these other
points since it made it much more explicit what kind of address is
expected in each context.
Since our new packages are built to accommodate some future planned
features that are not yet implemented (e.g. the "for_each" argument on
resources, "count"/"for_each" on modules), and since there's still a fair
amount of functionality still using old-style APIs, there is a moderate
amount of shimming here to connect new assumptions with old, hopefully in
a way that makes it easier to find and eliminate these shims later.
I apologize in advance to the person who inevitably just found this huge
commit while spelunking through the commit history.
2018-04-30 12:33:53 -05:00
|
|
|
&OrphanOutputTransformer{Config: b.Config, State: b.State},
|
2016-09-18 23:45:19 -05:00
|
|
|
|
2016-09-20 12:23:48 -05:00
|
|
|
// Attach the configuration to any resources
|
terraform: ugly huge change to weave in new HCL2-oriented types
Due to how deeply the configuration types go into Terraform Core, there
isn't a great way to switch out to HCL2 gradually. As a consequence, this
huge commit gets us from the old state to a _compilable_ new state, but
does not yet attempt to fix any tests and has a number of known missing
parts and bugs. We will continue to iterate on this in forthcoming
commits, heading back towards passing tests and making Terraform
fully-functional again.
The three main goals here are:
- Use the configuration models from the "configs" package instead of the
older models in the "config" package, which is now deprecated and
preserved only to help us write our migration tool.
- Do expression inspection and evaluation using the functionality of the
new "lang" package, instead of the Interpolator type and related
functionality in the main "terraform" package.
- Represent addresses of various objects using types in the addrs package,
rather than hand-constructed strings. This is not critical to support
the above, but was a big help during the implementation of these other
points since it made it much more explicit what kind of address is
expected in each context.
Since our new packages are built to accommodate some future planned
features that are not yet implemented (e.g. the "for_each" argument on
resources, "count"/"for_each" on modules), and since there's still a fair
amount of functionality still using old-style APIs, there is a moderate
amount of shimming here to connect new assumptions with old, hopefully in
a way that makes it easier to find and eliminate these shims later.
I apologize in advance to the person who inevitably just found this huge
commit while spelunking through the commit history.
2018-04-30 12:33:53 -05:00
|
|
|
&AttachResourceConfigTransformer{Config: b.Config},
|
2016-09-20 12:23:48 -05:00
|
|
|
|
2018-05-25 18:27:11 -05:00
|
|
|
// add providers
|
2021-08-31 12:16:44 -05:00
|
|
|
transformProviders(concreteProvider, b.Config),
|
2018-05-25 18:27:11 -05:00
|
|
|
|
2017-11-09 09:34:56 -06:00
|
|
|
// Remove modules no longer present in the config
|
terraform: ugly huge change to weave in new HCL2-oriented types
Due to how deeply the configuration types go into Terraform Core, there
isn't a great way to switch out to HCL2 gradually. As a consequence, this
huge commit gets us from the old state to a _compilable_ new state, but
does not yet attempt to fix any tests and has a number of known missing
parts and bugs. We will continue to iterate on this in forthcoming
commits, heading back towards passing tests and making Terraform
fully-functional again.
The three main goals here are:
- Use the configuration models from the "configs" package instead of the
older models in the "config" package, which is now deprecated and
preserved only to help us write our migration tool.
- Do expression inspection and evaluation using the functionality of the
new "lang" package, instead of the Interpolator type and related
functionality in the main "terraform" package.
- Represent addresses of various objects using types in the addrs package,
rather than hand-constructed strings. This is not critical to support
the above, but was a big help during the implementation of these other
points since it made it much more explicit what kind of address is
expected in each context.
Since our new packages are built to accommodate some future planned
features that are not yet implemented (e.g. the "for_each" argument on
resources, "count"/"for_each" on modules), and since there's still a fair
amount of functionality still using old-style APIs, there is a moderate
amount of shimming here to connect new assumptions with old, hopefully in
a way that makes it easier to find and eliminate these shims later.
I apologize in advance to the person who inevitably just found this huge
commit while spelunking through the commit history.
2018-04-30 12:33:53 -05:00
|
|
|
&RemovedModuleTransformer{Config: b.Config, State: b.State},
|
2017-11-09 09:34:56 -06:00
|
|
|
|
2018-06-01 15:00:52 -05:00
|
|
|
// Must attach schemas before ReferenceTransformer so that we can
|
|
|
|
// analyze the configuration to find references.
|
2021-08-31 18:36:27 -05:00
|
|
|
&AttachSchemaTransformer{Plugins: b.Plugins, Config: b.Config},
|
2018-06-01 15:00:52 -05:00
|
|
|
|
2019-11-21 20:45:43 -06:00
|
|
|
// Create expansion nodes for all of the module calls. This must
|
|
|
|
// come after all other transformers that create nodes representing
|
|
|
|
// objects that can belong to modules.
|
|
|
|
&ModuleExpansionTransformer{Config: b.Config},
|
|
|
|
|
2016-09-15 20:30:11 -05:00
|
|
|
// Connect references so ordering is correct
|
|
|
|
&ReferenceTransformer{},
|
2019-10-17 15:05:27 -05:00
|
|
|
&AttachDependenciesTransformer{},
|
2016-09-15 20:30:11 -05:00
|
|
|
|
2020-05-14 13:08:33 -05:00
|
|
|
// Detect when create_before_destroy must be forced on for a particular
|
|
|
|
// node due to dependency edges, to avoid graph cycles during apply.
|
|
|
|
&ForcedCBDTransformer{},
|
|
|
|
|
2019-09-25 12:20:24 -05:00
|
|
|
// Destruction ordering
|
2022-10-20 09:59:08 -05:00
|
|
|
&DestroyEdgeTransformer{
|
|
|
|
Changes: b.Changes,
|
|
|
|
},
|
2019-09-26 11:36:44 -05:00
|
|
|
&CBDEdgeTransformer{
|
2021-08-31 18:36:27 -05:00
|
|
|
Config: b.Config,
|
|
|
|
State: b.State,
|
2019-09-26 11:36:44 -05:00
|
|
|
},
|
2019-09-25 12:20:24 -05:00
|
|
|
|
2020-05-20 12:45:31 -05:00
|
|
|
// We need to remove configuration nodes that are not used at all, as
|
|
|
|
// they may not be able to evaluate, especially during destroy.
|
|
|
|
// These include variables, locals, and instance expanders.
|
|
|
|
&pruneUnusedNodesTransformer{},
|
2019-12-18 13:37:28 -06:00
|
|
|
|
2020-06-23 16:22:44 -05:00
|
|
|
// Target
|
|
|
|
&TargetsTransformer{Targets: b.Targets},
|
|
|
|
|
2017-04-12 16:25:15 -05:00
|
|
|
// Close opened plugin connections
|
|
|
|
&CloseProviderTransformer{},
|
|
|
|
|
2020-04-02 14:49:32 -05:00
|
|
|
// close the root module
|
|
|
|
&CloseRootModuleTransformer{},
|
2016-09-19 11:28:24 -05:00
|
|
|
|
|
|
|
// Perform the transitive reduction to make our graph a bit
|
2020-10-18 11:56:51 -05:00
|
|
|
// more understandable if possible (it usually is possible).
|
2020-08-14 13:13:33 -05:00
|
|
|
&TransitiveReductionTransformer{},
|
2016-09-13 12:56:37 -05:00
|
|
|
}
|
|
|
|
|
|
|
|
return steps
|
|
|
|
}
|