terraform/lang/eval_test.go

845 lines
20 KiB
Go
Raw Normal View History

package lang
import (
"bytes"
"encoding/json"
"testing"
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
"github.com/hashicorp/terraform/addrs"
"github.com/hashicorp/terraform/configs/configschema"
"github.com/hashicorp/terraform/internal/instances"
"github.com/hashicorp/hcl/v2"
"github.com/hashicorp/hcl/v2/hclsyntax"
"github.com/zclconf/go-cty/cty"
ctyjson "github.com/zclconf/go-cty/cty/json"
)
func TestScopeEvalContext(t *testing.T) {
data := &dataForTests{
CountAttrs: map[string]cty.Value{
"index": cty.NumberIntVal(0),
},
2019-06-12 17:07:32 +02:00
ForEachAttrs: map[string]cty.Value{
"key": cty.StringVal("a"),
"value": cty.NumberIntVal(1),
},
Resources: map[string]cty.Value{
"null_resource.foo": cty.ObjectVal(map[string]cty.Value{
"attr": cty.StringVal("bar"),
}),
"data.null_data_source.foo": cty.ObjectVal(map[string]cty.Value{
"attr": cty.StringVal("bar"),
}),
"null_resource.multi": cty.TupleVal([]cty.Value{
cty.ObjectVal(map[string]cty.Value{
"attr": cty.StringVal("multi0"),
}),
cty.ObjectVal(map[string]cty.Value{
"attr": cty.StringVal("multi1"),
}),
}),
"null_resource.each": cty.ObjectVal(map[string]cty.Value{
"each0": cty.ObjectVal(map[string]cty.Value{
"attr": cty.StringVal("each0"),
}),
"each1": cty.ObjectVal(map[string]cty.Value{
"attr": cty.StringVal("each1"),
}),
}),
"null_resource.multi[1]": cty.ObjectVal(map[string]cty.Value{
"attr": cty.StringVal("multi1"),
}),
},
LocalValues: map[string]cty.Value{
"foo": cty.StringVal("bar"),
},
Modules: map[string]cty.Value{
"module.foo": cty.ObjectVal(map[string]cty.Value{
"output0": cty.StringVal("bar0"),
"output1": cty.StringVal("bar1"),
}),
},
PathAttrs: map[string]cty.Value{
"module": cty.StringVal("foo/bar"),
},
TerraformAttrs: map[string]cty.Value{
"workspace": cty.StringVal("default"),
},
InputVariables: map[string]cty.Value{
"baz": cty.StringVal("boop"),
},
}
tests := []struct {
Expr string
Want map[string]cty.Value
}{
{
`12`,
map[string]cty.Value{},
},
{
`count.index`,
map[string]cty.Value{
"count": cty.ObjectVal(map[string]cty.Value{
"index": cty.NumberIntVal(0),
}),
},
},
2019-06-12 17:07:32 +02:00
{
`each.key`,
map[string]cty.Value{
"each": cty.ObjectVal(map[string]cty.Value{
"key": cty.StringVal("a"),
}),
},
},
{
`each.value`,
map[string]cty.Value{
"each": cty.ObjectVal(map[string]cty.Value{
"value": cty.NumberIntVal(1),
}),
},
},
{
`local.foo`,
map[string]cty.Value{
"local": cty.ObjectVal(map[string]cty.Value{
"foo": cty.StringVal("bar"),
}),
},
},
{
`null_resource.foo`,
map[string]cty.Value{
"null_resource": cty.ObjectVal(map[string]cty.Value{
"foo": cty.ObjectVal(map[string]cty.Value{
"attr": cty.StringVal("bar"),
}),
}),
lang: Allow "resource." prefix as another way to refer to resources The current way to refer to a managed resource is to use its resource type name as a top-level symbol in the reference. This is convenient and makes sense given that managed resources are the primary kind of object in Terraform. However, it does mean that an externally-extensible namespace (the set of all possible resource type names) overlaps with a reserved word namespace (the special prefixes like "path", "var", etc), and thus introducing any new reserved prefix in future risks masking an existing resource type so it can't be used anymore. We only intend to introduce new reserved symbols as part of future language editions that each module can opt into separately, and when doing so we will always research to try to choose a name that doesn't overlap with commonly-used providers, but not all providers are visible to us and so there is always a small chance that the name we choose will already be in use by a third-party provider. In preparation for that event, this introduces an alternative way to refer to managed resources that mimics the reference style used for data resources: resource.type.name . When using this form, the second portion is _always_ a resource type name and never a reserved word. There is currently no need to use this because all of the already-reserved symbol names are effectively blocked from use by existing Terraform versions that lack this escape hatch. Therefore there's no explicit documentation about it yet. The intended use for this is that a module upgrade tool for a future language edition would detect references to resource types that have now become reserved words and add the "resource." prefix to keep that functionality working. Existing modules that aren't opted in to the new language edition would keep working without that prefix, thus keeping to compatibility promises.
2021-05-15 01:42:30 +02:00
"resource": cty.ObjectVal(map[string]cty.Value{
"null_resource": cty.ObjectVal(map[string]cty.Value{
"foo": cty.ObjectVal(map[string]cty.Value{
"attr": cty.StringVal("bar"),
}),
}),
}),
},
},
{
`null_resource.foo.attr`,
map[string]cty.Value{
"null_resource": cty.ObjectVal(map[string]cty.Value{
"foo": cty.ObjectVal(map[string]cty.Value{
"attr": cty.StringVal("bar"),
}),
}),
lang: Allow "resource." prefix as another way to refer to resources The current way to refer to a managed resource is to use its resource type name as a top-level symbol in the reference. This is convenient and makes sense given that managed resources are the primary kind of object in Terraform. However, it does mean that an externally-extensible namespace (the set of all possible resource type names) overlaps with a reserved word namespace (the special prefixes like "path", "var", etc), and thus introducing any new reserved prefix in future risks masking an existing resource type so it can't be used anymore. We only intend to introduce new reserved symbols as part of future language editions that each module can opt into separately, and when doing so we will always research to try to choose a name that doesn't overlap with commonly-used providers, but not all providers are visible to us and so there is always a small chance that the name we choose will already be in use by a third-party provider. In preparation for that event, this introduces an alternative way to refer to managed resources that mimics the reference style used for data resources: resource.type.name . When using this form, the second portion is _always_ a resource type name and never a reserved word. There is currently no need to use this because all of the already-reserved symbol names are effectively blocked from use by existing Terraform versions that lack this escape hatch. Therefore there's no explicit documentation about it yet. The intended use for this is that a module upgrade tool for a future language edition would detect references to resource types that have now become reserved words and add the "resource." prefix to keep that functionality working. Existing modules that aren't opted in to the new language edition would keep working without that prefix, thus keeping to compatibility promises.
2021-05-15 01:42:30 +02:00
"resource": cty.ObjectVal(map[string]cty.Value{
"null_resource": cty.ObjectVal(map[string]cty.Value{
"foo": cty.ObjectVal(map[string]cty.Value{
"attr": cty.StringVal("bar"),
}),
}),
}),
},
},
{
`null_resource.multi`,
map[string]cty.Value{
"null_resource": cty.ObjectVal(map[string]cty.Value{
"multi": cty.TupleVal([]cty.Value{
cty.ObjectVal(map[string]cty.Value{
"attr": cty.StringVal("multi0"),
}),
cty.ObjectVal(map[string]cty.Value{
"attr": cty.StringVal("multi1"),
}),
}),
}),
lang: Allow "resource." prefix as another way to refer to resources The current way to refer to a managed resource is to use its resource type name as a top-level symbol in the reference. This is convenient and makes sense given that managed resources are the primary kind of object in Terraform. However, it does mean that an externally-extensible namespace (the set of all possible resource type names) overlaps with a reserved word namespace (the special prefixes like "path", "var", etc), and thus introducing any new reserved prefix in future risks masking an existing resource type so it can't be used anymore. We only intend to introduce new reserved symbols as part of future language editions that each module can opt into separately, and when doing so we will always research to try to choose a name that doesn't overlap with commonly-used providers, but not all providers are visible to us and so there is always a small chance that the name we choose will already be in use by a third-party provider. In preparation for that event, this introduces an alternative way to refer to managed resources that mimics the reference style used for data resources: resource.type.name . When using this form, the second portion is _always_ a resource type name and never a reserved word. There is currently no need to use this because all of the already-reserved symbol names are effectively blocked from use by existing Terraform versions that lack this escape hatch. Therefore there's no explicit documentation about it yet. The intended use for this is that a module upgrade tool for a future language edition would detect references to resource types that have now become reserved words and add the "resource." prefix to keep that functionality working. Existing modules that aren't opted in to the new language edition would keep working without that prefix, thus keeping to compatibility promises.
2021-05-15 01:42:30 +02:00
"resource": cty.ObjectVal(map[string]cty.Value{
"null_resource": cty.ObjectVal(map[string]cty.Value{
"multi": cty.TupleVal([]cty.Value{
cty.ObjectVal(map[string]cty.Value{
"attr": cty.StringVal("multi0"),
}),
cty.ObjectVal(map[string]cty.Value{
"attr": cty.StringVal("multi1"),
}),
}),
}),
}),
},
},
{
// at this level, all instance references return the entire resource
`null_resource.multi[1]`,
map[string]cty.Value{
"null_resource": cty.ObjectVal(map[string]cty.Value{
"multi": cty.TupleVal([]cty.Value{
cty.ObjectVal(map[string]cty.Value{
"attr": cty.StringVal("multi0"),
}),
cty.ObjectVal(map[string]cty.Value{
"attr": cty.StringVal("multi1"),
}),
}),
}),
lang: Allow "resource." prefix as another way to refer to resources The current way to refer to a managed resource is to use its resource type name as a top-level symbol in the reference. This is convenient and makes sense given that managed resources are the primary kind of object in Terraform. However, it does mean that an externally-extensible namespace (the set of all possible resource type names) overlaps with a reserved word namespace (the special prefixes like "path", "var", etc), and thus introducing any new reserved prefix in future risks masking an existing resource type so it can't be used anymore. We only intend to introduce new reserved symbols as part of future language editions that each module can opt into separately, and when doing so we will always research to try to choose a name that doesn't overlap with commonly-used providers, but not all providers are visible to us and so there is always a small chance that the name we choose will already be in use by a third-party provider. In preparation for that event, this introduces an alternative way to refer to managed resources that mimics the reference style used for data resources: resource.type.name . When using this form, the second portion is _always_ a resource type name and never a reserved word. There is currently no need to use this because all of the already-reserved symbol names are effectively blocked from use by existing Terraform versions that lack this escape hatch. Therefore there's no explicit documentation about it yet. The intended use for this is that a module upgrade tool for a future language edition would detect references to resource types that have now become reserved words and add the "resource." prefix to keep that functionality working. Existing modules that aren't opted in to the new language edition would keep working without that prefix, thus keeping to compatibility promises.
2021-05-15 01:42:30 +02:00
"resource": cty.ObjectVal(map[string]cty.Value{
"null_resource": cty.ObjectVal(map[string]cty.Value{
"multi": cty.TupleVal([]cty.Value{
cty.ObjectVal(map[string]cty.Value{
"attr": cty.StringVal("multi0"),
}),
cty.ObjectVal(map[string]cty.Value{
"attr": cty.StringVal("multi1"),
}),
}),
}),
}),
},
},
{
// at this level, all instance references return the entire resource
`null_resource.each["each1"]`,
map[string]cty.Value{
"null_resource": cty.ObjectVal(map[string]cty.Value{
"each": cty.ObjectVal(map[string]cty.Value{
"each0": cty.ObjectVal(map[string]cty.Value{
"attr": cty.StringVal("each0"),
}),
"each1": cty.ObjectVal(map[string]cty.Value{
"attr": cty.StringVal("each1"),
}),
}),
}),
lang: Allow "resource." prefix as another way to refer to resources The current way to refer to a managed resource is to use its resource type name as a top-level symbol in the reference. This is convenient and makes sense given that managed resources are the primary kind of object in Terraform. However, it does mean that an externally-extensible namespace (the set of all possible resource type names) overlaps with a reserved word namespace (the special prefixes like "path", "var", etc), and thus introducing any new reserved prefix in future risks masking an existing resource type so it can't be used anymore. We only intend to introduce new reserved symbols as part of future language editions that each module can opt into separately, and when doing so we will always research to try to choose a name that doesn't overlap with commonly-used providers, but not all providers are visible to us and so there is always a small chance that the name we choose will already be in use by a third-party provider. In preparation for that event, this introduces an alternative way to refer to managed resources that mimics the reference style used for data resources: resource.type.name . When using this form, the second portion is _always_ a resource type name and never a reserved word. There is currently no need to use this because all of the already-reserved symbol names are effectively blocked from use by existing Terraform versions that lack this escape hatch. Therefore there's no explicit documentation about it yet. The intended use for this is that a module upgrade tool for a future language edition would detect references to resource types that have now become reserved words and add the "resource." prefix to keep that functionality working. Existing modules that aren't opted in to the new language edition would keep working without that prefix, thus keeping to compatibility promises.
2021-05-15 01:42:30 +02:00
"resource": cty.ObjectVal(map[string]cty.Value{
"null_resource": cty.ObjectVal(map[string]cty.Value{
"each": cty.ObjectVal(map[string]cty.Value{
"each0": cty.ObjectVal(map[string]cty.Value{
"attr": cty.StringVal("each0"),
}),
"each1": cty.ObjectVal(map[string]cty.Value{
"attr": cty.StringVal("each1"),
}),
}),
}),
}),
},
},
{
// at this level, all instance references return the entire resource
`null_resource.each["each1"].attr`,
map[string]cty.Value{
"null_resource": cty.ObjectVal(map[string]cty.Value{
"each": cty.ObjectVal(map[string]cty.Value{
"each0": cty.ObjectVal(map[string]cty.Value{
"attr": cty.StringVal("each0"),
}),
"each1": cty.ObjectVal(map[string]cty.Value{
"attr": cty.StringVal("each1"),
}),
}),
}),
lang: Allow "resource." prefix as another way to refer to resources The current way to refer to a managed resource is to use its resource type name as a top-level symbol in the reference. This is convenient and makes sense given that managed resources are the primary kind of object in Terraform. However, it does mean that an externally-extensible namespace (the set of all possible resource type names) overlaps with a reserved word namespace (the special prefixes like "path", "var", etc), and thus introducing any new reserved prefix in future risks masking an existing resource type so it can't be used anymore. We only intend to introduce new reserved symbols as part of future language editions that each module can opt into separately, and when doing so we will always research to try to choose a name that doesn't overlap with commonly-used providers, but not all providers are visible to us and so there is always a small chance that the name we choose will already be in use by a third-party provider. In preparation for that event, this introduces an alternative way to refer to managed resources that mimics the reference style used for data resources: resource.type.name . When using this form, the second portion is _always_ a resource type name and never a reserved word. There is currently no need to use this because all of the already-reserved symbol names are effectively blocked from use by existing Terraform versions that lack this escape hatch. Therefore there's no explicit documentation about it yet. The intended use for this is that a module upgrade tool for a future language edition would detect references to resource types that have now become reserved words and add the "resource." prefix to keep that functionality working. Existing modules that aren't opted in to the new language edition would keep working without that prefix, thus keeping to compatibility promises.
2021-05-15 01:42:30 +02:00
"resource": cty.ObjectVal(map[string]cty.Value{
"null_resource": cty.ObjectVal(map[string]cty.Value{
"each": cty.ObjectVal(map[string]cty.Value{
"each0": cty.ObjectVal(map[string]cty.Value{
"attr": cty.StringVal("each0"),
}),
"each1": cty.ObjectVal(map[string]cty.Value{
"attr": cty.StringVal("each1"),
}),
}),
}),
}),
},
},
{
`foo(null_resource.multi, null_resource.multi[1])`,
map[string]cty.Value{
"null_resource": cty.ObjectVal(map[string]cty.Value{
"multi": cty.TupleVal([]cty.Value{
cty.ObjectVal(map[string]cty.Value{
"attr": cty.StringVal("multi0"),
}),
cty.ObjectVal(map[string]cty.Value{
"attr": cty.StringVal("multi1"),
}),
}),
}),
lang: Allow "resource." prefix as another way to refer to resources The current way to refer to a managed resource is to use its resource type name as a top-level symbol in the reference. This is convenient and makes sense given that managed resources are the primary kind of object in Terraform. However, it does mean that an externally-extensible namespace (the set of all possible resource type names) overlaps with a reserved word namespace (the special prefixes like "path", "var", etc), and thus introducing any new reserved prefix in future risks masking an existing resource type so it can't be used anymore. We only intend to introduce new reserved symbols as part of future language editions that each module can opt into separately, and when doing so we will always research to try to choose a name that doesn't overlap with commonly-used providers, but not all providers are visible to us and so there is always a small chance that the name we choose will already be in use by a third-party provider. In preparation for that event, this introduces an alternative way to refer to managed resources that mimics the reference style used for data resources: resource.type.name . When using this form, the second portion is _always_ a resource type name and never a reserved word. There is currently no need to use this because all of the already-reserved symbol names are effectively blocked from use by existing Terraform versions that lack this escape hatch. Therefore there's no explicit documentation about it yet. The intended use for this is that a module upgrade tool for a future language edition would detect references to resource types that have now become reserved words and add the "resource." prefix to keep that functionality working. Existing modules that aren't opted in to the new language edition would keep working without that prefix, thus keeping to compatibility promises.
2021-05-15 01:42:30 +02:00
"resource": cty.ObjectVal(map[string]cty.Value{
"null_resource": cty.ObjectVal(map[string]cty.Value{
"multi": cty.TupleVal([]cty.Value{
cty.ObjectVal(map[string]cty.Value{
"attr": cty.StringVal("multi0"),
}),
cty.ObjectVal(map[string]cty.Value{
"attr": cty.StringVal("multi1"),
}),
}),
}),
}),
},
},
{
`data.null_data_source.foo`,
map[string]cty.Value{
"data": cty.ObjectVal(map[string]cty.Value{
"null_data_source": cty.ObjectVal(map[string]cty.Value{
"foo": cty.ObjectVal(map[string]cty.Value{
"attr": cty.StringVal("bar"),
}),
}),
}),
},
},
{
`module.foo`,
map[string]cty.Value{
"module": cty.ObjectVal(map[string]cty.Value{
"foo": cty.ObjectVal(map[string]cty.Value{
"output0": cty.StringVal("bar0"),
"output1": cty.StringVal("bar1"),
}),
}),
},
},
// any module reference returns the entire module
{
`module.foo.output1`,
map[string]cty.Value{
"module": cty.ObjectVal(map[string]cty.Value{
"foo": cty.ObjectVal(map[string]cty.Value{
"output0": cty.StringVal("bar0"),
"output1": cty.StringVal("bar1"),
}),
}),
},
},
{
`path.module`,
map[string]cty.Value{
"path": cty.ObjectVal(map[string]cty.Value{
"module": cty.StringVal("foo/bar"),
}),
},
},
{
`self.baz`,
map[string]cty.Value{
"self": cty.ObjectVal(map[string]cty.Value{
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
"attr": cty.StringVal("multi1"),
}),
},
},
{
`terraform.workspace`,
map[string]cty.Value{
"terraform": cty.ObjectVal(map[string]cty.Value{
"workspace": cty.StringVal("default"),
}),
},
},
{
`var.baz`,
map[string]cty.Value{
"var": cty.ObjectVal(map[string]cty.Value{
"baz": cty.StringVal("boop"),
}),
},
},
}
for _, test := range tests {
t.Run(test.Expr, func(t *testing.T) {
expr, parseDiags := hclsyntax.ParseExpression([]byte(test.Expr), "", hcl.Pos{Line: 1, Column: 1})
if len(parseDiags) != 0 {
t.Errorf("unexpected diagnostics during parse")
for _, diag := range parseDiags {
t.Errorf("- %s", diag)
}
return
}
refs, refsDiags := ReferencesInExpr(expr)
if refsDiags.HasErrors() {
t.Fatal(refsDiags.Err())
}
scope := &Scope{
Data: data,
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
// "self" will just be an arbitrary one of the several resource
// instances we have in our test dataset.
SelfAddr: addrs.ResourceInstance{
Resource: addrs.Resource{
Mode: addrs.ManagedResourceMode,
Type: "null_resource",
Name: "multi",
},
Key: addrs.IntKey(1),
},
}
ctx, ctxDiags := scope.EvalContext(refs)
if ctxDiags.HasErrors() {
t.Fatal(ctxDiags.Err())
}
// For easier test assertions we'll just remove any top-level
// empty objects from our variables map.
for k, v := range ctx.Variables {
if v.RawEquals(cty.EmptyObjectVal) {
delete(ctx.Variables, k)
}
}
gotVal := cty.ObjectVal(ctx.Variables)
wantVal := cty.ObjectVal(test.Want)
if !gotVal.RawEquals(wantVal) {
// We'll JSON-ize our values here just so it's easier to
// read them in the assertion output.
gotJSON := formattedJSONValue(gotVal)
wantJSON := formattedJSONValue(wantVal)
t.Errorf(
"wrong result\nexpr: %s\ngot: %s\nwant: %s",
test.Expr, gotJSON, wantJSON,
)
}
})
}
}
func TestScopeExpandEvalBlock(t *testing.T) {
nestedObjTy := cty.Object(map[string]cty.Type{
"boop": cty.String,
})
schema := &configschema.Block{
Attributes: map[string]*configschema.Attribute{
"foo": {Type: cty.String, Optional: true},
"list_of_obj": {Type: cty.List(nestedObjTy), Optional: true},
},
BlockTypes: map[string]*configschema.NestedBlock{
"bar": {
Nesting: configschema.NestingMap,
Block: configschema.Block{
Attributes: map[string]*configschema.Attribute{
"baz": {Type: cty.String, Optional: true},
},
},
},
},
}
data := &dataForTests{
LocalValues: map[string]cty.Value{
"greeting": cty.StringVal("howdy"),
"list": cty.ListVal([]cty.Value{
cty.StringVal("elem0"),
cty.StringVal("elem1"),
}),
"map": cty.MapVal(map[string]cty.Value{
"key1": cty.StringVal("val1"),
"key2": cty.StringVal("val2"),
}),
},
}
tests := map[string]struct {
Config string
Want cty.Value
}{
"empty": {
`
`,
cty.ObjectVal(map[string]cty.Value{
"foo": cty.NullVal(cty.String),
"list_of_obj": cty.NullVal(cty.List(nestedObjTy)),
"bar": cty.MapValEmpty(cty.Object(map[string]cty.Type{
"baz": cty.String,
})),
}),
},
"literal attribute": {
`
foo = "hello"
`,
cty.ObjectVal(map[string]cty.Value{
"foo": cty.StringVal("hello"),
"list_of_obj": cty.NullVal(cty.List(nestedObjTy)),
"bar": cty.MapValEmpty(cty.Object(map[string]cty.Type{
"baz": cty.String,
})),
}),
},
"variable attribute": {
`
foo = local.greeting
`,
cty.ObjectVal(map[string]cty.Value{
"foo": cty.StringVal("howdy"),
"list_of_obj": cty.NullVal(cty.List(nestedObjTy)),
"bar": cty.MapValEmpty(cty.Object(map[string]cty.Type{
"baz": cty.String,
})),
}),
},
"one static block": {
`
bar "static" {}
`,
cty.ObjectVal(map[string]cty.Value{
"foo": cty.NullVal(cty.String),
"list_of_obj": cty.NullVal(cty.List(nestedObjTy)),
"bar": cty.MapVal(map[string]cty.Value{
"static": cty.ObjectVal(map[string]cty.Value{
"baz": cty.NullVal(cty.String),
}),
}),
}),
},
"two static blocks": {
`
bar "static0" {
baz = 0
}
bar "static1" {
baz = 1
}
`,
cty.ObjectVal(map[string]cty.Value{
"foo": cty.NullVal(cty.String),
"list_of_obj": cty.NullVal(cty.List(nestedObjTy)),
"bar": cty.MapVal(map[string]cty.Value{
"static0": cty.ObjectVal(map[string]cty.Value{
"baz": cty.StringVal("0"),
}),
"static1": cty.ObjectVal(map[string]cty.Value{
"baz": cty.StringVal("1"),
}),
}),
}),
},
"dynamic blocks from list": {
`
dynamic "bar" {
for_each = local.list
labels = [bar.value]
content {
baz = bar.key
}
}
`,
cty.ObjectVal(map[string]cty.Value{
"foo": cty.NullVal(cty.String),
"list_of_obj": cty.NullVal(cty.List(nestedObjTy)),
"bar": cty.MapVal(map[string]cty.Value{
"elem0": cty.ObjectVal(map[string]cty.Value{
"baz": cty.StringVal("0"),
}),
"elem1": cty.ObjectVal(map[string]cty.Value{
"baz": cty.StringVal("1"),
}),
}),
}),
},
"dynamic blocks from map": {
`
dynamic "bar" {
for_each = local.map
labels = [bar.key]
content {
baz = bar.value
}
}
`,
cty.ObjectVal(map[string]cty.Value{
"foo": cty.NullVal(cty.String),
"list_of_obj": cty.NullVal(cty.List(nestedObjTy)),
"bar": cty.MapVal(map[string]cty.Value{
"key1": cty.ObjectVal(map[string]cty.Value{
"baz": cty.StringVal("val1"),
}),
"key2": cty.ObjectVal(map[string]cty.Value{
"baz": cty.StringVal("val2"),
}),
}),
}),
},
"list-of-object attribute": {
`
list_of_obj = [
{
boop = local.greeting
},
]
`,
cty.ObjectVal(map[string]cty.Value{
"foo": cty.NullVal(cty.String),
"list_of_obj": cty.ListVal([]cty.Value{
cty.ObjectVal(map[string]cty.Value{
"boop": cty.StringVal("howdy"),
}),
}),
"bar": cty.MapValEmpty(cty.Object(map[string]cty.Type{
"baz": cty.String,
})),
}),
},
"list-of-object attribute as blocks": {
`
list_of_obj {
boop = local.greeting
}
`,
cty.ObjectVal(map[string]cty.Value{
"foo": cty.NullVal(cty.String),
"list_of_obj": cty.ListVal([]cty.Value{
cty.ObjectVal(map[string]cty.Value{
"boop": cty.StringVal("howdy"),
}),
}),
"bar": cty.MapValEmpty(cty.Object(map[string]cty.Type{
"baz": cty.String,
})),
}),
},
"lots of things at once": {
`
foo = "whoop"
bar "static0" {
baz = "s0"
}
dynamic "bar" {
for_each = local.list
labels = [bar.value]
content {
baz = bar.key
}
}
bar "static1" {
baz = "s1"
}
dynamic "bar" {
for_each = local.map
labels = [bar.key]
content {
baz = bar.value
}
}
bar "static2" {
baz = "s2"
}
`,
cty.ObjectVal(map[string]cty.Value{
"foo": cty.StringVal("whoop"),
"list_of_obj": cty.NullVal(cty.List(nestedObjTy)),
"bar": cty.MapVal(map[string]cty.Value{
"key1": cty.ObjectVal(map[string]cty.Value{
"baz": cty.StringVal("val1"),
}),
"key2": cty.ObjectVal(map[string]cty.Value{
"baz": cty.StringVal("val2"),
}),
"elem0": cty.ObjectVal(map[string]cty.Value{
"baz": cty.StringVal("0"),
}),
"elem1": cty.ObjectVal(map[string]cty.Value{
"baz": cty.StringVal("1"),
}),
"static0": cty.ObjectVal(map[string]cty.Value{
"baz": cty.StringVal("s0"),
}),
"static1": cty.ObjectVal(map[string]cty.Value{
"baz": cty.StringVal("s1"),
}),
"static2": cty.ObjectVal(map[string]cty.Value{
"baz": cty.StringVal("s2"),
}),
}),
}),
},
}
for name, test := range tests {
t.Run(name, func(t *testing.T) {
file, parseDiags := hclsyntax.ParseConfig([]byte(test.Config), "", hcl.Pos{Line: 1, Column: 1})
if len(parseDiags) != 0 {
t.Errorf("unexpected diagnostics during parse")
for _, diag := range parseDiags {
t.Errorf("- %s", diag)
}
return
}
body := file.Body
scope := &Scope{
Data: data,
}
body, expandDiags := scope.ExpandBlock(body, schema)
if expandDiags.HasErrors() {
t.Fatal(expandDiags.Err())
}
got, valDiags := scope.EvalBlock(body, schema)
if valDiags.HasErrors() {
t.Fatal(valDiags.Err())
}
if !got.RawEquals(test.Want) {
// We'll JSON-ize our values here just so it's easier to
// read them in the assertion output.
gotJSON := formattedJSONValue(got)
wantJSON := formattedJSONValue(test.Want)
t.Errorf(
"wrong result\nconfig: %s\ngot: %s\nwant: %s",
test.Config, gotJSON, wantJSON,
)
}
})
}
}
func formattedJSONValue(val cty.Value) string {
val = cty.UnknownAsNull(val) // since JSON can't represent unknowns
j, err := ctyjson.Marshal(val, val.Type())
if err != nil {
panic(err)
}
var buf bytes.Buffer
json.Indent(&buf, j, "", " ")
return buf.String()
}
func TestScopeEvalSelfBlock(t *testing.T) {
data := &dataForTests{
PathAttrs: map[string]cty.Value{
"module": cty.StringVal("foo/bar"),
"cwd": cty.StringVal("/home/foo/bar"),
"root": cty.StringVal("/home/foo"),
},
TerraformAttrs: map[string]cty.Value{
"workspace": cty.StringVal("default"),
},
}
schema := &configschema.Block{
Attributes: map[string]*configschema.Attribute{
"attr": {
Type: cty.String,
},
"num": {
Type: cty.Number,
},
},
}
tests := []struct {
Config string
Self cty.Value
KeyData instances.RepetitionData
Want map[string]cty.Value
}{
{
Config: `attr = self.foo`,
Self: cty.ObjectVal(map[string]cty.Value{
"foo": cty.StringVal("bar"),
}),
KeyData: instances.RepetitionData{
CountIndex: cty.NumberIntVal(0),
},
Want: map[string]cty.Value{
"attr": cty.StringVal("bar"),
"num": cty.NullVal(cty.Number),
},
},
{
Config: `num = count.index`,
KeyData: instances.RepetitionData{
CountIndex: cty.NumberIntVal(0),
},
Want: map[string]cty.Value{
"attr": cty.NullVal(cty.String),
"num": cty.NumberIntVal(0),
},
},
{
Config: `attr = each.key`,
KeyData: instances.RepetitionData{
EachKey: cty.StringVal("a"),
},
Want: map[string]cty.Value{
"attr": cty.StringVal("a"),
"num": cty.NullVal(cty.Number),
},
},
{
Config: `attr = path.cwd`,
Want: map[string]cty.Value{
"attr": cty.StringVal("/home/foo/bar"),
"num": cty.NullVal(cty.Number),
},
},
{
Config: `attr = path.module`,
Want: map[string]cty.Value{
"attr": cty.StringVal("foo/bar"),
"num": cty.NullVal(cty.Number),
},
},
{
Config: `attr = path.root`,
Want: map[string]cty.Value{
"attr": cty.StringVal("/home/foo"),
"num": cty.NullVal(cty.Number),
},
},
{
Config: `attr = terraform.workspace`,
Want: map[string]cty.Value{
"attr": cty.StringVal("default"),
"num": cty.NullVal(cty.Number),
},
},
}
for _, test := range tests {
t.Run(test.Config, func(t *testing.T) {
file, parseDiags := hclsyntax.ParseConfig([]byte(test.Config), "", hcl.Pos{Line: 1, Column: 1})
if len(parseDiags) != 0 {
t.Errorf("unexpected diagnostics during parse")
for _, diag := range parseDiags {
t.Errorf("- %s", diag)
}
return
}
body := file.Body
scope := &Scope{
Data: data,
}
gotVal, ctxDiags := scope.EvalSelfBlock(body, test.Self, schema, test.KeyData)
if ctxDiags.HasErrors() {
t.Fatal(ctxDiags.Err())
}
wantVal := cty.ObjectVal(test.Want)
if !gotVal.RawEquals(wantVal) {
t.Errorf(
"wrong result\nexpr: %s\ngot: %#v\nwant: %#v",
test.Config, gotVal, wantVal,
)
}
})
}
}