terraform/command/import.go

399 lines
13 KiB
Go
Raw Normal View History

2016-05-04 19:06:16 +02:00
package command
import (
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 19:33:53 +02:00
"errors"
2016-05-04 19:06:16 +02:00
"fmt"
"log"
"os"
2016-05-04 19:06:16 +02:00
"strings"
2016-05-04 19:32:08 +02:00
"github.com/hashicorp/hcl/v2"
"github.com/hashicorp/hcl/v2/hclsyntax"
"github.com/hashicorp/terraform/addrs"
2017-01-19 05:50:45 +01:00
"github.com/hashicorp/terraform/backend"
"github.com/hashicorp/terraform/configs"
2016-05-04 19:32:08 +02:00
"github.com/hashicorp/terraform/terraform"
"github.com/hashicorp/terraform/tfdiags"
2016-05-04 19:06:16 +02:00
)
// ImportCommand is a cli.Command implementation that imports resources
// into the Terraform state.
type ImportCommand struct {
Meta
}
func (c *ImportCommand) Run(args []string) int {
// Get the pwd since its our default -config flag value
pwd, err := os.Getwd()
if err != nil {
c.Ui.Error(fmt.Sprintf("Error getting pwd: %s", err))
return 1
}
var configPath string
args, err = c.Meta.process(args, true)
if err != nil {
return 1
}
2016-05-04 19:06:16 +02:00
cmdFlags := c.Meta.extendedFlagSet("import")
cmdFlags.IntVar(&c.Meta.parallelism, "parallelism", DefaultParallelism, "parallelism")
cmdFlags.StringVar(&c.Meta.statePath, "state", "", "path")
2016-05-04 19:06:16 +02:00
cmdFlags.StringVar(&c.Meta.stateOutPath, "state-out", "", "path")
cmdFlags.StringVar(&c.Meta.backupPath, "backup", "", "path")
cmdFlags.StringVar(&configPath, "config", pwd, "path")
cmdFlags.StringVar(&c.Meta.provider, "provider", "", "provider")
cmdFlags.BoolVar(&c.Meta.stateLock, "lock", true, "lock state")
cmdFlags.DurationVar(&c.Meta.stateLockTimeout, "lock-timeout", 0, "lock timeout")
cmdFlags.BoolVar(&c.Meta.allowMissingConfig, "allow-missing-config", false, "allow missing config")
2016-05-04 19:06:16 +02:00
cmdFlags.Usage = func() { c.Ui.Error(c.Help()) }
if err := cmdFlags.Parse(args); err != nil {
return 1
}
args = cmdFlags.Args()
if len(args) != 2 {
c.Ui.Error("The import command expects two arguments.")
cmdFlags.Usage()
return 1
}
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 19:33:53 +02:00
var diags tfdiags.Diagnostics
// Parse the provided resource address.
traversalSrc := []byte(args[0])
traversal, travDiags := hclsyntax.ParseTraversalAbs(traversalSrc, "<import-address>", hcl.Pos{Line: 1, Column: 1})
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 19:33:53 +02:00
diags = diags.Append(travDiags)
if travDiags.HasErrors() {
c.registerSynthConfigSource("<import-address>", traversalSrc) // so we can include a source snippet
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 19:33:53 +02:00
c.showDiagnostics(diags)
c.Ui.Info(importCommandInvalidAddressReference)
return 1
}
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 19:33:53 +02:00
addr, addrDiags := addrs.ParseAbsResourceInstance(traversal)
diags = diags.Append(addrDiags)
if addrDiags.HasErrors() {
c.registerSynthConfigSource("<import-address>", traversalSrc) // so we can include a source snippet
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 19:33:53 +02:00
c.showDiagnostics(diags)
c.Ui.Info(importCommandInvalidAddressReference)
return 1
}
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 19:33:53 +02:00
if addr.Resource.Resource.Mode != addrs.ManagedResourceMode {
diags = diags.Append(errors.New("A managed resource address is required. Importing into a data resource is not allowed."))
c.showDiagnostics(diags)
return 1
}
if !c.dirIsConfigPath(configPath) {
diags = diags.Append(&hcl.Diagnostic{
Severity: hcl.DiagError,
Summary: "No Terraform configuration files",
Detail: fmt.Sprintf(
"The directory %s does not contain any Terraform configuration files (.tf or .tf.json). To specify a different configuration directory, use the -config=\"...\" command line option.",
configPath,
),
})
c.showDiagnostics(diags)
return 1
}
// Load the full config, so we can verify that the target resource is
// already configured.
config, configDiags := c.loadConfig(configPath)
diags = diags.Append(configDiags)
if configDiags.HasErrors() {
c.showDiagnostics(diags)
return 1
2017-01-19 05:50:45 +01:00
}
// Verify that the given address points to something that exists in config.
// This is to reduce the risk that a typo in the resource address will
// import something that Terraform will want to immediately destroy on
// the next plan, and generally acts as a reassurance of user intent.
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 19:33:53 +02:00
targetConfig := config.DescendentForInstance(addr.Module)
if targetConfig == 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 19:33:53 +02:00
modulePath := addr.Module.String()
diags = diags.Append(&hcl.Diagnostic{
Severity: hcl.DiagError,
Summary: "Import to non-existent module",
Detail: fmt.Sprintf(
"%s is not defined in the configuration. Please add configuration for this module before importing into it.",
modulePath,
),
})
c.showDiagnostics(diags)
return 1
}
targetMod := targetConfig.Module
rcs := targetMod.ManagedResources
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 19:33:53 +02:00
var rc *configs.Resource
resourceRelAddr := addr.Resource.Resource
for _, thisRc := range rcs {
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 19:33:53 +02:00
if resourceRelAddr.Type == thisRc.Type && resourceRelAddr.Name == thisRc.Name {
rc = thisRc
break
}
}
if !c.Meta.allowMissingConfig && rc == 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 19:33:53 +02:00
modulePath := addr.Module.String()
if modulePath == "" {
modulePath = "the root module"
}
c.showDiagnostics(diags)
// This is not a diagnostic because currently our diagnostics printer
// doesn't support having a code example in the detail, and there's
// a code example in this message.
// TODO: Improve the diagnostics printer so we can use it for this
// message.
c.Ui.Error(fmt.Sprintf(
importCommandMissingResourceFmt,
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 19:33:53 +02:00
addr, modulePath, resourceRelAddr.Type, resourceRelAddr.Name,
))
return 1
}
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 19:33:53 +02:00
// Also parse the user-provided provider address, if any.
var providerAddr addrs.AbsProviderConfig
if c.Meta.provider != "" {
traversal, travDiags := hclsyntax.ParseTraversalAbs([]byte(c.Meta.provider), `-provider=...`, hcl.Pos{Line: 1, Column: 1})
diags = diags.Append(travDiags)
if travDiags.HasErrors() {
c.showDiagnostics(diags)
c.Ui.Info(importCommandInvalidAddressReference)
return 1
}
relAddr, addrDiags := configs.ParseProviderConfigCompact(traversal)
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 19:33:53 +02:00
diags = diags.Append(addrDiags)
if addrDiags.HasErrors() {
c.showDiagnostics(diags)
return 1
}
providerAddr = relAddr.Absolute(addrs.RootModuleInstance)
} else {
// Use a default address inferred from the resource type.
// We assume the same module as the resource address here, which
// may get resolved to an inherited provider when we construct the
// import graph inside ctx.Import, called below.
if rc != nil && rc.ProviderConfigRef != nil {
providerAddr = rc.ProviderConfigAddr().Absolute(addr.Module)
} else {
Initial steps towards AbsProviderConfig/LocalProviderConfig separation (#23978) * Introduce "Local" terminology for non-absolute provider config addresses In a future change AbsProviderConfig and LocalProviderConfig are going to become two entirely distinct types, rather than Abs embedding Local as written here. This naming change is in preparation for that subsequent work, which will also include introducing a new "ProviderConfig" type that is an interface that AbsProviderConfig and LocalProviderConfig both implement. This is intended to be largely just a naming change to get started, so we can deal with all of the messy renaming. However, this did also require a slight change in modeling where the Resource.DefaultProviderConfig method has become Resource.DefaultProvider returning a Provider address directly, because this method doesn't have enough information to construct a true and accurate LocalProviderConfig -- it would need to refer to the configuration to know what this module is calling the provider it has selected. In order to leave a trail to follow for subsequent work, all of the changes here are intended to ensure that remaining work will become obvious via compile-time errors when all of the following changes happen: - The concept of "legacy" provider addresses is removed from the addrs package, including removing addrs.NewLegacyProvider and addrs.Provider.LegacyString. - addrs.AbsProviderConfig stops having addrs.LocalProviderConfig embedded in it and has an addrs.Provider and a string alias directly instead. - The provider-schema-handling parts of Terraform core are updated to work with addrs.Provider to identify providers, rather than legacy strings. In particular, there are still several codepaths here making legacy provider address assumptions (in order to limit the scope of this change) but I've made sure each one is doing something that relies on at least one of the above changes not having been made yet. * addrs: ProviderConfig interface In a (very) few special situations in the main "terraform" package we need to make runtime decisions about whether a provider config is absolute or local. We currently do that by exploiting the fact that AbsProviderConfig has LocalProviderConfig nested inside of it and so in the local case we can just ignore the wrapping AbsProviderConfig and use the embedded value. In a future change we'll be moving away from that embedding and making these two types distinct in order to represent that mapping between them requires consulting a lookup table in the configuration, and so here we introduce a new interface type ProviderConfig that can represent either AbsProviderConfig or LocalProviderConfig decided dynamically at runtime. This also includes the Config.ResolveAbsProviderAddr method that will eventually be responsible for that local-to-absolute translation, so that callers with access to the configuration can normalize to an addrs.AbsProviderConfig given a non-nil addrs.ProviderConfig. That's currently unused because existing callers are still relying on the simplistic structural transform, but we'll switch them over in a later commit. * rename LocalType to LocalName Co-authored-by: Kristin Laemmert <mildwonkey@users.noreply.github.com>
2020-01-31 14:23:07 +01:00
providerType := resourceRelAddr.DefaultProvider()
providerAddr = addrs.NewDefaultLocalProviderConfig(providerType.LegacyString()).Absolute(addr.Module)
}
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 19:33:53 +02:00
}
// Check for user-supplied plugin path
if c.pluginPath, err = c.loadPluginPath(); err != nil {
c.Ui.Error(fmt.Sprintf("Error loading plugin path: %s", err))
return 1
}
2017-01-19 05:50:45 +01:00
// Load the backend
b, backendDiags := c.Backend(&BackendOpts{
Config: config.Module.Backend,
})
diags = diags.Append(backendDiags)
if backendDiags.HasErrors() {
c.showDiagnostics(diags)
2017-01-19 05:50:45 +01:00
return 1
}
// We require a backend.Local to build a context.
// This isn't necessarily a "local.Local" backend, which provides local
// operations, however that is the only current implementation. A
// "local.Local" backend also doesn't necessarily provide local state, as
// that may be delegated to a "remotestate.Backend".
2017-01-19 05:50:45 +01:00
local, ok := b.(backend.Local)
if !ok {
c.Ui.Error(ErrUnsupportedLocalOp)
return 1
}
// Build the operation
opReq := c.Operation(b)
opReq.ConfigDir = configPath
opReq.ConfigLoader, err = c.initConfigLoader()
if err != nil {
diags = diags.Append(err)
c.showDiagnostics(diags)
return 1
}
2018-10-15 01:54:23 +02:00
{
var moreDiags tfdiags.Diagnostics
opReq.Variables, moreDiags = c.collectVariableValues()
diags = diags.Append(moreDiags)
if moreDiags.HasErrors() {
c.showDiagnostics(diags)
return 1
}
}
2017-01-19 05:50:45 +01:00
// Get the context
ctx, state, ctxDiags := local.Context(opReq)
diags = diags.Append(ctxDiags)
if ctxDiags.HasErrors() {
c.showDiagnostics(diags)
2016-05-04 19:06:16 +02:00
return 1
}
// Make sure to unlock the state
defer func() {
err := opReq.StateLocker.Unlock(nil)
if err != nil {
c.Ui.Error(err.Error())
}
}()
2016-05-04 19:32:08 +02:00
// Perform the import. Note that as you can see it is possible for this
// API to import more than one resource at once. For now, we only allow
// one while we stabilize this feature.
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 19:33:53 +02:00
newState, importDiags := ctx.Import(&terraform.ImportOpts{
2016-05-04 19:32:08 +02:00
Targets: []*terraform.ImportTarget{
&terraform.ImportTarget{
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 19:33:53 +02:00
Addr: addr,
ID: args[1],
ProviderAddr: providerAddr,
2016-05-04 19:32:08 +02:00
},
},
})
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 19:33:53 +02:00
diags = diags.Append(importDiags)
if diags.HasErrors() {
c.showDiagnostics(diags)
2016-05-04 19:06:16 +02:00
return 1
}
2016-05-04 19:32:08 +02:00
// Persist the final state
2016-05-04 19:06:16 +02:00
log.Printf("[INFO] Writing state output to: %s", c.Meta.StateOutPath())
2017-01-19 05:50:45 +01:00
if err := state.WriteState(newState); err != nil {
c.Ui.Error(fmt.Sprintf("Error writing state file: %s", err))
return 1
}
if err := state.PersistState(); err != nil {
2016-05-04 19:06:16 +02:00
c.Ui.Error(fmt.Sprintf("Error writing state file: %s", err))
return 1
}
c.Ui.Output(c.Colorize().Color("[reset][green]\n" + importCommandSuccessMsg))
2016-05-04 21:05:42 +02:00
if c.Meta.allowMissingConfig && rc == nil {
c.Ui.Output(c.Colorize().Color("[reset][yellow]\n" + importCommandAllowMissingResourceMsg))
}
c.showDiagnostics(diags)
if diags.HasErrors() {
return 1
}
2016-05-04 19:06:16 +02:00
return 0
}
func (c *ImportCommand) Help() string {
helpText := `
Usage: terraform import [options] ADDR ID
Import existing infrastructure into your Terraform state.
This will find and import the specified resource into your Terraform
state, allowing existing infrastructure to come under Terraform
management without having to be initially created by Terraform.
The ADDR specified is the address to import the resource to. Please
see the documentation online for resource addresses. The ID is a
resource-specific ID to identify that resource being imported. Please
reference the documentation for the resource type you're importing to
determine the ID syntax to use. It typically matches directly to the ID
that the provider uses.
The current implementation of Terraform import can only import resources
into the state. It does not generate configuration. A future version of
Terraform will also generate configuration.
Because of this, prior to running terraform import it is necessary to write
a resource configuration block for the resource manually, to which the
imported object will be attached.
2016-05-04 19:06:16 +02:00
This command will not modify your infrastructure, but it will make
network requests to inspect parts of your infrastructure relevant to
the resource being imported.
Options:
-backup=path Path to backup the existing state file before
modifying. Defaults to the "-state-out" path with
".backup" extension. Set to "-" to disable backup.
2016-05-04 19:06:16 +02:00
-config=path Path to a directory of Terraform configuration files
to use to configure the provider. Defaults to pwd.
If no config files are present, they must be provided
via the input prompts or env vars.
-allow-missing-config Allow import when no resource configuration block exists.
2016-05-04 19:06:16 +02:00
-input=true Ask for input for variables if not directly set.
-lock=true Lock the state file when locking is supported.
-lock-timeout=0s Duration to retry a state lock.
2016-05-04 19:06:16 +02:00
-no-color If specified, output won't contain any color.
2016-11-23 18:40:11 +01:00
-provider=provider Deprecated: Override the provider configuration to use
when importing the object. By default, Terraform uses the
provider specified in the configuration for the target
resource, and that is the best behavior in most cases.
2016-05-04 19:06:16 +02:00
-state=PATH Path to the source state file. Defaults to the configured
backend, or "terraform.tfstate"
2016-05-04 19:06:16 +02:00
-state-out=PATH Path to the destination state file to write to. If this
isn't specified, the source state file will be used. This
can be a new or existing path.
-var 'foo=bar' Set a variable in the Terraform configuration. This
flag can be set multiple times. This is only useful
with the "-config" flag.
-var-file=foo Set variables in the Terraform configuration from
a file. If "terraform.tfvars" or any ".auto.tfvars"
files are present, they will be automatically loaded.
2016-05-04 19:06:16 +02:00
`
return strings.TrimSpace(helpText)
}
func (c *ImportCommand) Synopsis() string {
return "Import existing infrastructure into Terraform"
}
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 19:33:53 +02:00
const importCommandInvalidAddressReference = `For information on valid syntax, see:
https://www.terraform.io/docs/internals/resource-addressing.html`
const importCommandMissingResourceFmt = `[reset][bold][red]Error:[reset][bold] resource address %q does not exist in the configuration.[reset]
Before importing this resource, please create its configuration in %s. For example:
resource %q %q {
# (resource arguments)
}
`
const importCommandSuccessMsg = `Import successful!
The resources that were imported are shown above. These resources are now in
your Terraform state and will henceforth be managed by Terraform.
`
const importCommandAllowMissingResourceMsg = `Import does not generate resource configuration, you must create a resource
configuration block that matches the current or desired state manually.
If there is no matching resource configuration block for the imported
resource, Terraform will delete the resource on the next "terraform apply".
It is recommended that you run "terraform plan" to verify that the
configuration is correct and complete.
`