[mdatagen] Enable configurable attributes for metrics (#14281)

Reopens
[13900](#13900)

#### Description
Modified `cmd/mdatagen` to allow for two new fields in the configuration
yaml of metrics. These new configuration options will allow for a user
to enable or disable attributes (i.e. reduce dimensionality of metrics
being generated) from their collector configuration. The modified
`MetricsBuilder` generated by `mdatagen` does not record disabled
attributes and automatically re-aggregates metrics based on the
resulting enabled set of attributes. There are four different
aggregation strategies supported (`sum, average, min, and max`) which
can be specified as a setting in `config.yaml`.

The changes to the config.yaml are:
```yaml
receiver:
  someMetricReceiver:
    ...
    metrics:
    ...
      attributes: [<list of attributes to include for metric at runtime>] # new
      aggregation_strategy: <sum|avg|min|max> # new
```
Attribute `requirement_level` takes president over this re-aggregation
and any metric attribute with a `requirement_level` of `required` cannot
be disabled in this way. For more information on `requirement_level`
please see
[13913](#13913)

The set of attributes provided by the user _must_ be contained in the
set of attributes defined in the metadata.yaml, including an attribute
which is not defined in the metrics `metadata` file will fail. Also,
attempting to omit an attribute defined as `requirement_level: required`
will cause the configuration to fail and the collector not to start.
Omitting the field entirely will default to the defined set of metric
attributes in the `metadata` file with a requirement level of
`recommended` or above.

Due to its large scope and foundational nature, a featuregate has been
added for this change. This gate is represented by a new field in the
`metadata.yaml`, `reaggregation_enabled: <bool>`. Setting this to `true`
will enable the new codegen with spatial reaggregation features.

<!-- Issue number if applicable -->
#### Link to tracking issue
Fixes
[10726](#10726)

<!--Describe what testing was performed and which tests were added.-->
#### Testing
Generated tests have been updated to test new aggregation behavior.

<!--Describe the documentation added.-->
#### Documentation
ran `make gogenerate`
ran `make generate`

<!--Please delete paragraphs that you did not use before submitting.-->

Author: shalper2

.chloggen/10726-reag.yaml

@@ -0,0 +1,25 @@
+# Use this changelog template to create an entry for release notes.
+
+# One of 'breaking', 'deprecation', 'new_component', 'enhancement', 'bug_fix'
+change_type: 'enhancement'
+
+# The name of the component, or a single word describing the area of concern, (e.g. receiver/otlp)
+component: 'cmd/mdatagen'
+
+# A brief description of the change.  Surround your text with quotes ("") if it needs to start with a backtick (`).
+note: Add the ability to disable attributes at the metric level and re-aggregate data points based off of these new dimensions
+
+# One or more tracking issues or pull requests related to the change
+issues: [10726]
+
+# (Optional) One or more lines of additional information to render under the primary note.
+# These lines will be padded with 2 spaces and then inserted directly into the document.
+# Use pipe (|) for multiline entries.
+subtext:
+
+# Optional: The change log or logs in which this entry should be included.
+# e.g. '[user]' or '[user, api]'
+# Include 'user' if the change is relevant to end users.
+# Include 'api' if there is a change to a library API.
+# Default: '[user]'
+change_logs: [api, user]

.github/workflows/utils/cspell.json

@@ -402,6 +402,7 @@
       "protos",
       "ptraceotlp",
       "queuebatch",
+      "reaggregate",
       "receiverhelper",
       "receiverprofiles",
       "receivertest",

cmd/mdatagen/internal/command.go

@@ -231,6 +231,24 @@ func templatize(tmplFile string, md Metadata) *template.Template {
 				"attributeInfo": func(an AttributeName) Attribute {
 					return md.Attributes[an]
 				},
+				"defaultAttributes": func(ans []AttributeName) []string {
+					var atts []string
+					for _, an := range ans {
+						if md.Attributes[an].IsNotOptIn() {
+							atts = append(atts, string(md.Attributes[an].Name()))
+						}
+					}
+					return atts
+				},
+				"requiredAttributes": func(ans []AttributeName) []string {
+					var atts []string
+					for _, an := range ans {
+						if md.Attributes[an].IsRequired() {
+							atts = append(atts, string(md.Attributes[an].Name()))
+						}
+					}
+					return atts
+				},
 				"getEventConditionalAttributes": func(attrs map[AttributeName]Attribute) []AttributeName {
 					seen := make(map[AttributeName]bool)
 					used := make([]AttributeName, 0)

cmd/mdatagen/internal/loader_test.go

@@ -51,6 +51,7 @@ func TestLoadMetadata(t *testing.T) {
 				Description:          "This receiver is used for testing purposes to check the output of mdatagen.",
 				SemConvVersion:       "1.38.0",
 				PackageName:          "go.opentelemetry.io/collector/cmd/mdatagen/internal/samplereceiver",
+				ReaggregationEnabled: true,
 				Status: &Status{
 					DisableCodeCov: true,
 					Class:          "receiver",
@@ -263,6 +264,18 @@ func TestLoadMetadata(t *testing.T) {
 							Mono:                   Mono{Monotonic: true},
 						},
 					},
+					"reaggregate.metric": {
+						Signal: Signal{
+							Enabled:     true,
+							Description: "Metric for testing spacial reaggregation",
+							Stability:   Stability{Level: component.StabilityLevelBeta},
+							Attributes:  []AttributeName{"string_attr", "boolean_attr"},
+						},
+						Unit: strPtr("1"),
+						Gauge: &Gauge{
+							MetricValueType: MetricValueType{pmetric.NumberDataPointValueTypeDouble},
+						},
+					},
 					"system.cpu.time": {
 						Signal: Signal{
 							Enabled:               true,

cmd/mdatagen/internal/metadata.go

@@ -26,6 +26,8 @@ type Metadata struct {
 	Parent string `mapstructure:"parent"`
 	// Status information for the component.
 	Status *Status `mapstructure:"status"`
+	// Spatial Re-aggregation featuregate.
+	ReaggregationEnabled bool `mapstructure:"reaggregation_enabled"`
 	// The name of the package that will be generated.
 	GeneratedPackageName string `mapstructure:"generated_package_name"`
 	// Telemetry information for the component.
@@ -412,6 +414,17 @@ func (a Attribute) IsConditional() bool {
 	return a.RequirementLevel == AttributeRequirementLevelConditionallyRequired
 }
 
+// IsRequired returns true if the attribute is required.
+func (a Attribute) IsRequired() bool {
+	return a.RequirementLevel == AttributeRequirementLevelRequired
+}
+
+// IsNotOptIn returns true if the attribute is any requirement_level above
+// opt_in
+func (a Attribute) IsNotOptIn() bool {
+	return a.RequirementLevel != AttributeRequirementLevelOptIn
+}
+
 // UnmarshalText implements the encoding.TextUnmarshaler interface.
 func (rl *AttributeRequirementLevel) UnmarshalText(text []byte) error {
 	switch string(text) {
@@ -476,6 +489,31 @@ func (a Attribute) TestValue() string {
 	return ""
 }
 
+func (a Attribute) TestValueTwo() string {
+	if a.Enum != nil {
+		return fmt.Sprintf(`%q`, a.Enum[1])
+	}
+	switch a.Type.ValueType {
+	case pcommon.ValueTypeEmpty:
+		return ""
+	case pcommon.ValueTypeStr:
+		return fmt.Sprintf(`"%s-val-2"`, a.FullName)
+	case pcommon.ValueTypeInt:
+		return strconv.Itoa(len(a.FullName) + 1)
+	case pcommon.ValueTypeDouble:
+		return fmt.Sprintf("%f", 1.1+float64(len(a.FullName)))
+	case pcommon.ValueTypeBool:
+		return strconv.FormatBool(len(a.FullName)%2 == 1)
+	case pcommon.ValueTypeMap:
+		return fmt.Sprintf(`map[string]any{"key3": "%s-val3", "key4": "%s-val4"}`, a.FullName, a.FullName)
+	case pcommon.ValueTypeSlice:
+		return fmt.Sprintf(`[]any{"%s-item3", "%s-item4"}`, a.FullName, a.FullName)
+	case pcommon.ValueTypeBytes:
+		return fmt.Sprintf(`[]byte("%s-val-2")`, a.FullName)
+	}
+	return ""
+}
+
 type Signal struct {
 	// Enabled defines whether the signal is enabled by default.
 	Enabled bool `mapstructure:"enabled"`

cmd/mdatagen/internal/sampleconnector/documentation.md

@@ -60,6 +60,21 @@ Monotonic cumulative sum int metric with string input_type enabled by default.
 | slice_attr | Attribute with a slice value. | Any Slice | Recommended |
 | map_attr | Attribute with a map value. | Any Map | Recommended |
 
+### reaggregate.metric
+
+Metric for testing spacial reaggregation
+
+| Unit | Metric Type | Value Type | Stability |
+| ---- | ----------- | ---------- | --------- |
+| 1 | Gauge | Double | Beta |
+
+#### Attributes
+
+| Name | Description | Values | Requirement Level |
+| ---- | ----------- | ------ | -------- |
+| string_attr | Attribute with any string value. | Any Str | Recommended |
+| boolean_attr | Attribute with a boolean value. | Any Bool | Recommended |
+
 ## Optional Metrics
 
 The following metrics are not emitted by default. Each of them can be enabled by applying the following configuration: