2015-02-03 19:46:11 -06:00
|
|
|
package terraform
|
|
|
|
|
|
|
|
import (
|
2016-12-22 13:33:26 -06:00
|
|
|
"context"
|
2015-02-03 19:46:11 -06:00
|
|
|
"fmt"
|
2015-02-13 11:05:09 -06:00
|
|
|
"log"
|
2015-02-03 19:46:11 -06:00
|
|
|
"sync"
|
|
|
|
|
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
|
|
|
"github.com/hashicorp/terraform/plans"
|
2018-08-16 10:40:08 -05:00
|
|
|
"github.com/hashicorp/terraform/providers"
|
2018-08-17 11:42:07 -05:00
|
|
|
"github.com/hashicorp/terraform/provisioners"
|
2018-08-16 10:40:08 -05:00
|
|
|
"github.com/hashicorp/terraform/version"
|
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
|
|
|
|
|
|
|
"github.com/hashicorp/terraform/states"
|
|
|
|
|
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
|
|
|
"github.com/hashicorp/hcl2/hcl"
|
2018-07-05 12:33:29 -05:00
|
|
|
"github.com/hashicorp/terraform/configs/configschema"
|
2018-05-03 22:41:46 -05:00
|
|
|
"github.com/hashicorp/terraform/lang"
|
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
|
|
|
"github.com/hashicorp/terraform/tfdiags"
|
|
|
|
|
|
|
|
"github.com/hashicorp/terraform/addrs"
|
|
|
|
"github.com/zclconf/go-cty/cty"
|
2015-02-03 19:46:11 -06:00
|
|
|
)
|
|
|
|
|
|
|
|
// BuiltinEvalContext is an EvalContext implementation that is used by
|
|
|
|
// Terraform by default.
|
|
|
|
type BuiltinEvalContext struct {
|
2016-12-22 13:33:26 -06:00
|
|
|
// StopContext is the context used to track whether we're complete
|
|
|
|
StopContext context.Context
|
|
|
|
|
2015-05-04 12:51:34 -05:00
|
|
|
// PathValue is the Path that this context is operating within.
|
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
|
|
|
PathValue addrs.ModuleInstance
|
|
|
|
|
|
|
|
// Evaluator is used for evaluating expressions within the scope of this
|
|
|
|
// eval context.
|
|
|
|
Evaluator *Evaluator
|
|
|
|
|
2018-06-01 14:36:55 -05:00
|
|
|
// Schemas is a repository of all of the schemas we should need to
|
|
|
|
// decode configuration blocks and expressions. This must be constructed by
|
|
|
|
// the caller to include schemas for all of the providers, resource types,
|
|
|
|
// data sources and provisioners used by the given configuration and
|
|
|
|
// state.
|
|
|
|
//
|
|
|
|
// This must not be mutated during evaluation.
|
|
|
|
Schemas *Schemas
|
|
|
|
|
2018-05-03 20:25:16 -05:00
|
|
|
// VariableValues contains the variable values across all modules. This
|
|
|
|
// structure is shared across the entire containing context, and so it
|
|
|
|
// may be accessed only when holding VariableValuesLock.
|
|
|
|
// The keys of the first level of VariableValues are the string
|
|
|
|
// representations of addrs.ModuleInstance values. The second-level keys
|
|
|
|
// are variable names within each module instance.
|
|
|
|
VariableValues map[string]map[string]cty.Value
|
|
|
|
VariableValuesLock *sync.Mutex
|
2015-05-04 12:51:34 -05:00
|
|
|
|
2017-10-13 09:34:03 -05:00
|
|
|
Components contextComponentFactory
|
|
|
|
Hooks []Hook
|
|
|
|
InputValue UIInput
|
2018-08-16 10:40:08 -05:00
|
|
|
ProviderCache map[string]providers.Interface
|
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
|
|
|
ProviderInputConfig map[string]map[string]cty.Value
|
2015-02-10 01:32:28 -06:00
|
|
|
ProviderLock *sync.Mutex
|
2018-08-17 11:42:07 -05:00
|
|
|
ProvisionerCache map[string]provisioners.Interface
|
2015-02-10 01:32:28 -06:00
|
|
|
ProvisionerLock *sync.Mutex
|
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
|
|
|
ChangesValue *plans.ChangesSync
|
|
|
|
StateValue *states.SyncState
|
2015-02-03 19:46:11 -06:00
|
|
|
|
2015-05-01 18:29:19 -05:00
|
|
|
once sync.Once
|
2015-02-03 19:46:11 -06:00
|
|
|
}
|
|
|
|
|
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
|
|
|
// BuiltinEvalContext implements EvalContext
|
|
|
|
var _ EvalContext = (*BuiltinEvalContext)(nil)
|
|
|
|
|
2016-12-22 13:33:26 -06:00
|
|
|
func (ctx *BuiltinEvalContext) Stopped() <-chan struct{} {
|
|
|
|
// This can happen during tests. During tests, we just block forever.
|
|
|
|
if ctx.StopContext == nil {
|
|
|
|
return nil
|
|
|
|
}
|
|
|
|
|
|
|
|
return ctx.StopContext.Done()
|
|
|
|
}
|
|
|
|
|
2015-02-11 15:43:07 -06:00
|
|
|
func (ctx *BuiltinEvalContext) Hook(fn func(Hook) (HookAction, error)) error {
|
|
|
|
for _, h := range ctx.Hooks {
|
|
|
|
action, err := fn(h)
|
|
|
|
if err != nil {
|
|
|
|
return err
|
|
|
|
}
|
|
|
|
|
|
|
|
switch action {
|
|
|
|
case HookActionContinue:
|
|
|
|
continue
|
|
|
|
case HookActionHalt:
|
|
|
|
// Return an early exit error to trigger an early exit
|
2015-02-13 11:05:09 -06:00
|
|
|
log.Printf("[WARN] Early exit triggered by hook: %T", h)
|
2015-02-11 15:43:07 -06:00
|
|
|
return EvalEarlyExitError{}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
return nil
|
|
|
|
}
|
|
|
|
|
2015-02-13 19:59:54 -06:00
|
|
|
func (ctx *BuiltinEvalContext) Input() UIInput {
|
|
|
|
return ctx.InputValue
|
|
|
|
}
|
|
|
|
|
2018-08-16 10:40:08 -05:00
|
|
|
func (ctx *BuiltinEvalContext) InitProvider(typeName string, addr addrs.ProviderConfig) (providers.Interface, error) {
|
2015-02-03 19:46:11 -06:00
|
|
|
ctx.once.Do(ctx.init)
|
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
|
|
|
absAddr := addr.Absolute(ctx.Path())
|
2015-02-03 19:46:11 -06:00
|
|
|
|
2015-02-08 19:58:02 -06:00
|
|
|
// If we already initialized, it is an error
|
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
|
|
|
if p := ctx.Provider(absAddr); p != nil {
|
|
|
|
return nil, fmt.Errorf("%s is already initialized", addr)
|
2015-02-03 19:46:11 -06:00
|
|
|
}
|
|
|
|
|
2015-02-08 19:58:02 -06:00
|
|
|
// Warning: make sure to acquire these locks AFTER the call to Provider
|
|
|
|
// above, since it also acquires locks.
|
|
|
|
ctx.ProviderLock.Lock()
|
|
|
|
defer ctx.ProviderLock.Unlock()
|
|
|
|
|
2018-05-11 18:53:52 -05:00
|
|
|
key := absAddr.String()
|
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
|
|
|
|
|
|
|
p, err := ctx.Components.ResourceProvider(typeName, key)
|
2015-02-04 19:02:18 -06:00
|
|
|
if err != nil {
|
|
|
|
return nil, err
|
|
|
|
}
|
|
|
|
|
2018-05-11 18:53:52 -05:00
|
|
|
log.Printf("[TRACE] BuiltinEvalContext: Initialized %q provider for %s", typeName, absAddr)
|
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
|
|
|
ctx.ProviderCache[key] = p
|
2018-04-19 16:47:36 -05:00
|
|
|
|
2015-02-04 19:02:18 -06:00
|
|
|
return p, nil
|
2015-02-03 19:46:11 -06:00
|
|
|
}
|
|
|
|
|
2018-08-16 10:40:08 -05:00
|
|
|
func (ctx *BuiltinEvalContext) Provider(addr addrs.AbsProviderConfig) providers.Interface {
|
2015-02-03 19:46:11 -06:00
|
|
|
ctx.once.Do(ctx.init)
|
2015-02-08 19:58:02 -06:00
|
|
|
|
|
|
|
ctx.ProviderLock.Lock()
|
|
|
|
defer ctx.ProviderLock.Unlock()
|
|
|
|
|
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 ctx.ProviderCache[addr.String()]
|
2015-02-10 01:32:28 -06:00
|
|
|
}
|
|
|
|
|
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 (ctx *BuiltinEvalContext) ProviderSchema(addr addrs.AbsProviderConfig) *ProviderSchema {
|
2018-04-19 16:47:36 -05:00
|
|
|
ctx.once.Do(ctx.init)
|
|
|
|
|
2018-06-01 14:36:55 -05:00
|
|
|
return ctx.Schemas.ProviderSchema(addr.ProviderConfig.Type)
|
2018-04-19 16:47:36 -05:00
|
|
|
}
|
|
|
|
|
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 (ctx *BuiltinEvalContext) CloseProvider(addr addrs.ProviderConfig) error {
|
2015-06-19 14:52:50 -05:00
|
|
|
ctx.once.Do(ctx.init)
|
|
|
|
|
|
|
|
ctx.ProviderLock.Lock()
|
|
|
|
defer ctx.ProviderLock.Unlock()
|
|
|
|
|
2018-06-01 21:05:08 -05:00
|
|
|
key := addr.Absolute(ctx.Path()).String()
|
2018-08-16 10:40:08 -05:00
|
|
|
provider := ctx.ProviderCache[key]
|
2015-06-19 14:52:50 -05:00
|
|
|
if provider != nil {
|
2018-08-16 10:40:08 -05:00
|
|
|
delete(ctx.ProviderCache, key)
|
|
|
|
return provider.Close()
|
2015-06-19 14:52:50 -05:00
|
|
|
}
|
|
|
|
|
|
|
|
return nil
|
|
|
|
}
|
|
|
|
|
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 (ctx *BuiltinEvalContext) ConfigureProvider(addr addrs.ProviderConfig, cfg cty.Value) tfdiags.Diagnostics {
|
|
|
|
var diags tfdiags.Diagnostics
|
|
|
|
absAddr := addr.Absolute(ctx.Path())
|
|
|
|
p := ctx.Provider(absAddr)
|
2015-02-10 01:32:28 -06:00
|
|
|
if p == nil {
|
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
|
|
|
diags = diags.Append(fmt.Errorf("%s not initialized", addr))
|
|
|
|
return diags
|
2015-02-10 01:32:28 -06:00
|
|
|
}
|
2018-05-04 21:43:28 -05:00
|
|
|
|
|
|
|
providerSchema := ctx.ProviderSchema(absAddr)
|
|
|
|
if providerSchema == nil {
|
|
|
|
diags = diags.Append(fmt.Errorf("schema for %s is not available", absAddr))
|
|
|
|
return diags
|
|
|
|
}
|
|
|
|
|
2018-08-16 10:40:08 -05:00
|
|
|
req := providers.ConfigureRequest{
|
|
|
|
TerraformVersion: version.String(),
|
|
|
|
Config: cfg,
|
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
|
|
|
}
|
2018-08-16 10:40:08 -05:00
|
|
|
|
|
|
|
resp := p.Configure(req)
|
|
|
|
return resp.Diagnostics
|
2015-04-09 10:48:08 -05:00
|
|
|
}
|
|
|
|
|
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 (ctx *BuiltinEvalContext) ProviderInput(pc addrs.ProviderConfig) map[string]cty.Value {
|
2015-02-13 19:59:54 -06:00
|
|
|
ctx.ProviderLock.Lock()
|
|
|
|
defer ctx.ProviderLock.Unlock()
|
|
|
|
|
2018-05-31 22:03:03 -05:00
|
|
|
if !ctx.Path().IsRoot() {
|
|
|
|
// Only root module provider configurations can have input.
|
|
|
|
return nil
|
2015-06-24 11:34:21 -05:00
|
|
|
}
|
|
|
|
|
2018-05-31 22:03:03 -05:00
|
|
|
return ctx.ProviderInputConfig[pc.String()]
|
2015-02-13 19:59:54 -06:00
|
|
|
}
|
|
|
|
|
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 (ctx *BuiltinEvalContext) SetProviderInput(pc addrs.ProviderConfig, c map[string]cty.Value) {
|
|
|
|
absProvider := pc.Absolute(ctx.Path())
|
2015-02-13 19:59:54 -06:00
|
|
|
|
2018-06-01 17:08:56 -05:00
|
|
|
if !ctx.Path().IsRoot() {
|
|
|
|
// Only root module provider configurations can have input.
|
|
|
|
log.Printf("[WARN] BuiltinEvalContext: attempt to SetProviderInput for non-root module")
|
|
|
|
return
|
|
|
|
}
|
|
|
|
|
2015-06-24 11:34:21 -05:00
|
|
|
// Save the configuration
|
|
|
|
ctx.ProviderLock.Lock()
|
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
|
|
|
ctx.ProviderInputConfig[absProvider.String()] = c
|
2015-06-24 11:34:21 -05:00
|
|
|
ctx.ProviderLock.Unlock()
|
2015-02-13 19:59:54 -06:00
|
|
|
}
|
|
|
|
|
2018-08-17 11:42:07 -05:00
|
|
|
func (ctx *BuiltinEvalContext) InitProvisioner(n string) (provisioners.Interface, error) {
|
2015-02-09 13:15:54 -06:00
|
|
|
ctx.once.Do(ctx.init)
|
|
|
|
|
|
|
|
// If we already initialized, it is an error
|
|
|
|
if p := ctx.Provisioner(n); p != nil {
|
|
|
|
return nil, fmt.Errorf("Provisioner '%s' already initialized", n)
|
|
|
|
}
|
|
|
|
|
|
|
|
// Warning: make sure to acquire these locks AFTER the call to Provisioner
|
|
|
|
// above, since it also acquires locks.
|
|
|
|
ctx.ProvisionerLock.Lock()
|
|
|
|
defer ctx.ProvisionerLock.Unlock()
|
|
|
|
|
2019-08-21 18:06:25 -05:00
|
|
|
p, err := ctx.Components.ResourceProvisioner(n, "")
|
2016-10-03 21:34:07 -05:00
|
|
|
if err != nil {
|
|
|
|
return nil, err
|
|
|
|
}
|
2015-02-20 12:50:36 -06:00
|
|
|
|
2019-08-21 18:06:25 -05:00
|
|
|
ctx.ProvisionerCache[n] = p
|
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
|
|
|
|
2015-02-09 13:15:54 -06:00
|
|
|
return p, nil
|
|
|
|
}
|
|
|
|
|
2018-08-17 11:42:07 -05:00
|
|
|
func (ctx *BuiltinEvalContext) Provisioner(n string) provisioners.Interface {
|
2015-02-09 13:15:54 -06:00
|
|
|
ctx.once.Do(ctx.init)
|
|
|
|
|
|
|
|
ctx.ProvisionerLock.Lock()
|
|
|
|
defer ctx.ProvisionerLock.Unlock()
|
|
|
|
|
2019-08-21 18:06:25 -05:00
|
|
|
return ctx.ProvisionerCache[n]
|
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 (ctx *BuiltinEvalContext) ProvisionerSchema(n string) *configschema.Block {
|
|
|
|
ctx.once.Do(ctx.init)
|
2015-02-20 12:50:36 -06:00
|
|
|
|
2018-06-01 14:36:55 -05:00
|
|
|
return ctx.Schemas.ProvisionerConfig(n)
|
2015-02-09 13:15:54 -06:00
|
|
|
}
|
|
|
|
|
2015-06-19 14:52:50 -05:00
|
|
|
func (ctx *BuiltinEvalContext) CloseProvisioner(n string) error {
|
|
|
|
ctx.once.Do(ctx.init)
|
|
|
|
|
|
|
|
ctx.ProvisionerLock.Lock()
|
|
|
|
defer ctx.ProvisionerLock.Unlock()
|
|
|
|
|
2019-08-21 18:06:25 -05:00
|
|
|
prov := ctx.ProvisionerCache[n]
|
2015-06-19 14:52:50 -05:00
|
|
|
if prov != nil {
|
2018-08-17 11:42:07 -05:00
|
|
|
return prov.Close()
|
2015-06-19 14:52:50 -05:00
|
|
|
}
|
|
|
|
|
|
|
|
return nil
|
|
|
|
}
|
|
|
|
|
2018-06-22 15:04:19 -05:00
|
|
|
func (ctx *BuiltinEvalContext) EvaluateBlock(body hcl.Body, schema *configschema.Block, self addrs.Referenceable, keyData InstanceKeyEvalData) (cty.Value, hcl.Body, tfdiags.Diagnostics) {
|
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
|
|
|
var diags tfdiags.Diagnostics
|
2018-06-22 15:04:19 -05:00
|
|
|
scope := ctx.EvaluationScope(self, keyData)
|
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
|
|
|
body, evalDiags := scope.ExpandBlock(body, schema)
|
|
|
|
diags = diags.Append(evalDiags)
|
|
|
|
val, evalDiags := scope.EvalBlock(body, schema)
|
|
|
|
diags = diags.Append(evalDiags)
|
|
|
|
return val, body, diags
|
2017-10-13 09:11:10 -05:00
|
|
|
}
|
|
|
|
|
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 (ctx *BuiltinEvalContext) EvaluateExpr(expr hcl.Expression, wantType cty.Type, self addrs.Referenceable) (cty.Value, tfdiags.Diagnostics) {
|
2018-06-22 15:04:19 -05:00
|
|
|
scope := ctx.EvaluationScope(self, EvalDataForNoInstanceKey)
|
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 scope.EvalExpr(expr, wantType)
|
2015-02-03 19:46:11 -06:00
|
|
|
}
|
|
|
|
|
2018-06-22 15:04:19 -05:00
|
|
|
func (ctx *BuiltinEvalContext) EvaluationScope(self addrs.Referenceable, keyData InstanceKeyEvalData) *lang.Scope {
|
2018-05-03 22:41:46 -05:00
|
|
|
data := &evaluationStateData{
|
2018-06-22 15:04:19 -05:00
|
|
|
Evaluator: ctx.Evaluator,
|
|
|
|
ModulePath: ctx.PathValue,
|
|
|
|
InstanceKeyData: keyData,
|
2018-12-16 20:59:44 -06:00
|
|
|
Operation: ctx.Evaluator.Operation,
|
2018-05-03 19:24:49 -05:00
|
|
|
}
|
2018-05-03 22:41:46 -05:00
|
|
|
return ctx.Evaluator.Scope(data, self)
|
2018-05-03 19:24:49 -05:00
|
|
|
}
|
|
|
|
|
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 (ctx *BuiltinEvalContext) Path() addrs.ModuleInstance {
|
2015-02-08 16:00:13 -06:00
|
|
|
return ctx.PathValue
|
|
|
|
}
|
|
|
|
|
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 (ctx *BuiltinEvalContext) SetModuleCallArguments(n addrs.ModuleCallInstance, vals map[string]cty.Value) {
|
2018-05-03 20:25:16 -05:00
|
|
|
ctx.VariableValuesLock.Lock()
|
|
|
|
defer ctx.VariableValuesLock.Unlock()
|
2015-05-01 18:29:19 -05:00
|
|
|
|
2018-05-03 20:25:16 -05:00
|
|
|
childPath := n.ModuleInstance(ctx.PathValue)
|
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
|
|
|
key := childPath.String()
|
2015-05-01 18:29:19 -05:00
|
|
|
|
2018-05-03 20:25:16 -05:00
|
|
|
args := ctx.VariableValues[key]
|
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
|
|
|
if args == nil {
|
|
|
|
args = make(map[string]cty.Value)
|
2018-05-03 20:25:16 -05:00
|
|
|
ctx.VariableValues[key] = vals
|
|
|
|
return
|
2015-05-01 18:29:19 -05:00
|
|
|
}
|
2015-05-01 16:10:41 -05:00
|
|
|
|
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
|
|
|
for k, v := range vals {
|
|
|
|
args[k] = v
|
2015-02-11 19:01:08 -06:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
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
|
|
|
func (ctx *BuiltinEvalContext) Changes() *plans.ChangesSync {
|
|
|
|
return ctx.ChangesValue
|
2015-02-11 17:22:03 -06:00
|
|
|
}
|
|
|
|
|
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
|
|
|
func (ctx *BuiltinEvalContext) State() *states.SyncState {
|
|
|
|
return ctx.StateValue
|
2015-02-11 10:48:45 -06:00
|
|
|
}
|
|
|
|
|
2015-02-03 19:46:11 -06:00
|
|
|
func (ctx *BuiltinEvalContext) init() {
|
2015-02-08 19:58:02 -06:00
|
|
|
}
|