pkg/ottl: add context inference for expressions (#40194)

#### Description

Adds support for context inference when parsing OTTL value expressions.
This is not used anywhere yet, but I intend to use it in a new processor
that partitions batches of data based on OTTL expressions.

#### Link to tracking issue

Fixes #39158

#### Testing

Added/updated unit tests based on existing ones for ParseConditions.

#### Documentation

None

---------

Co-authored-by: Edmo Vamerlatti Costa <[email protected]>
Co-authored-by: Evan Bradley <[email protected]>

Author: 3 people

.chloggen/ottl-inference-valuexpr.yaml

@@ -0,0 +1,27 @@
+# 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. filelogreceiver)
+component: pkg/ottl
+
+# A brief description of the change.  Surround your text with quotes ("") if it needs to start with a backtick (`).
+note: Add context inference support for OTTL value expressions
+
+# Mandatory: One or more tracking issues related to the change. You can use the PR number here if no issue exists.
+issues: [39158]
+
+# (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:
+
+# If your change doesn't affect end users or the exported elements of any package,
+# you should instead start your pull request title with [chore] or use the "Skip Changelog" label.
+# 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]

pkg/ottl/context_inferrer.go

@@ -33,8 +33,11 @@ type contextInferrer interface {
 	inferFromStatements(statements []string) (string, error)
 	// inferFromConditions returns the OTTL context inferred from the given conditions.
 	inferFromConditions(conditions []string) (string, error)
-	// infer returns the OTTL context inferred from the given statements and conditions.
-	infer(statements []string, conditions []string) (string, error)
+	// inferFromValueExpressions returns the OTTL context inferred from the given value expressions.
+	inferFromValueExpressions(expressions []string) (string, error)
+	// infer returns the OTTL context inferred from the given statements, conditions,
+	// and value expressions.
+	infer(statements, conditions, valueExpressions []string) (string, error)
 }
 
 type priorityContextInferrer struct {
@@ -88,15 +91,19 @@ func withContextInferrerPriorities(priorities []string) priorityContextInferrerO
 }
 
 func (s *priorityContextInferrer) inferFromConditions(conditions []string) (inferredContext string, err error) {
-	return s.infer(nil, conditions)
+	return s.infer(nil, conditions, nil)
 }
 
 func (s *priorityContextInferrer) inferFromStatements(statements []string) (inferredContext string, err error) {
-	return s.infer(statements, nil)
+	return s.infer(statements, nil, nil)
 }
 
-func (s *priorityContextInferrer) infer(statements []string, conditions []string) (inferredContext string, err error) {
-	var statementsHints, conditionsHints []priorityContextInferrerHints
+func (s *priorityContextInferrer) inferFromValueExpressions(expressions []string) (inferredContext string, err error) {
+	return s.infer(nil, nil, expressions)
+}
+
+func (s *priorityContextInferrer) infer(statements, conditions, valueExprs []string) (inferredContext string, err error) {
+	var statementsHints, conditionsHints, valueExprsHints []priorityContextInferrerHints
 	if len(statements) > 0 {
 		statementsHints, err = s.getStatementsHints(statements)
 		if err != nil {
@@ -109,23 +116,30 @@ func (s *priorityContextInferrer) infer(statements []string, conditions []string
 			return "", err
 		}
 	}
+	if len(valueExprs) > 0 {
+		valueExprsHints, err = s.getValueExpressionsHints(valueExprs)
+		if err != nil {
+			return "", err
+		}
+	}
 	if s.telemetrySettings.Logger.Core().Enabled(zap.DebugLevel) {
-		s.telemetrySettings.Logger.Debug("Inferring context from statements and conditions",
+		s.telemetrySettings.Logger.Debug("Inferring OTTL context",
 			zap.Strings("candidates", maps.Keys(s.contextCandidate)),
 			zap.Any("priority", s.contextPriority),
 			zap.Strings("statements", statements),
 			zap.Strings("conditions", conditions),
+			zap.Strings("value_expressions", valueExprs),
 		)
 	}
-	return s.inferFromHints(append(statementsHints, conditionsHints...))
+	return s.inferFromHints(slices.Concat(statementsHints, conditionsHints, valueExprsHints))
 }
 
 func (s *priorityContextInferrer) inferFromHints(hints []priorityContextInferrerHints) (inferredContext string, err error) {
 	defer func() {
 		if inferredContext != "" {
-			s.telemetrySettings.Logger.Debug(fmt.Sprintf(`Inferred context: "%s"`, inferredContext))
+			s.telemetrySettings.Logger.Debug(fmt.Sprintf(`Inferred OTTL context: "%s"`, inferredContext))
 		} else {
-			s.telemetrySettings.Logger.Debug("Unable to infer context from statements", zap.Error(err))
+			s.telemetrySettings.Logger.Debug("Unable to infer OTTL context", zap.Error(err))
 		}
 	}()
 
@@ -284,6 +298,24 @@ func (s *priorityContextInferrer) getStatementsHints(statements []string) ([]pri
 	return hints, nil
 }
 
+// getValueExpressionsHints extracts all path, function (converter) names, and enumSymbol
+// from the given value expressions. These values are used by the context inferrer as hints to
+// select a context in which the function/enum are supported.
+func (s *priorityContextInferrer) getValueExpressionsHints(exprs []string) ([]priorityContextInferrerHints, error) {
+	hints := make([]priorityContextInferrerHints, 0, len(exprs))
+	for _, expr := range exprs {
+		parsed, err := parseValueExpression(expr)
+		if err != nil {
+			return nil, err
+		}
+
+		visitor := newGrammarContextInferrerVisitor()
+		parsed.accept(&visitor)
+		hints = append(hints, visitor)
+	}
+	return hints, nil
+}
+
 // priorityContextInferrerHints is a grammarVisitor implementation that collects
 // all path, function names (converter.Function and editor.Function), and enumSymbol.
 type priorityContextInferrerHints struct {

pkg/ottl/context_inferrer_test.go

@@ -520,11 +520,12 @@ func Test_NewPriorityContextInferrer_InferConditions_DefaultContextsOrder(t *tes
 
 func Test_NewPriorityContextInferrer_Infer(t *testing.T) {
 	tests := []struct {
-		name       string
-		candidates map[string]*priorityContextInferrerCandidate
-		statements []string
-		conditions []string
-		expected   string
+		name        string
+		candidates  map[string]*priorityContextInferrerCandidate
+		statements  []string
+		conditions  []string
+		expressions []string
+		expected    string
 	}{
 		{
 			name: "with statements",
@@ -547,6 +548,18 @@ func Test_NewPriorityContextInferrer_Infer(t *testing.T) {
 			},
 			expected: "metric",
 		},
+		{
+			name: "with expressions",
+			candidates: map[string]*priorityContextInferrerCandidate{
+				"metric":   defaultDummyPriorityContextInferrerCandidate,
+				"resource": defaultDummyPriorityContextInferrerCandidate,
+			},
+			expressions: []string{
+				`resource.attributes["foo"]`,
+				`metric.name`,
+			},
+			expected: "metric",
+		},
 		{
 			name: "with statements and conditions",
 			candidates: map[string]*priorityContextInferrerCandidate{
@@ -560,6 +573,22 @@ func Test_NewPriorityContextInferrer_Infer(t *testing.T) {
 			},
 			expected: "metric",
 		},
+		{
+			name: "with statements, conditions, and value expressions",
+			candidates: map[string]*priorityContextInferrerCandidate{
+				"metric":   defaultDummyPriorityContextInferrerCandidate,
+				"resource": defaultDummyPriorityContextInferrerCandidate,
+			},
+			statements: []string{`set(resource.attributes["foo"], "bar")`},
+			expressions: []string{
+				`resource.attributes["foo"]`,
+			},
+			conditions: []string{
+				`IsMatch(metric.name, "^bar.*")`,
+				`IsMatch(metric.name, "^foo.*")`,
+			},
+			expected: "metric",
+		},
 	}
 
 	for _, tt := range tests {
@@ -568,7 +597,7 @@ func Test_NewPriorityContextInferrer_Infer(t *testing.T) {
 				componenttest.NewNopTelemetrySettings(),
 				tt.candidates,
 			)
-			inferredContext, err := inferrer.infer(tt.statements, tt.conditions)
+			inferredContext, err := inferrer.infer(tt.statements, tt.conditions, tt.expressions)
 			require.NoError(t, err)
 			assert.Equal(t, tt.expected, inferredContext)
 		})

pkg/ottl/parser.go

@@ -259,6 +259,20 @@ func (p *Parser[K]) prependContextToConditionPaths(context string, condition str
 	})
 }
 
+// prependContextToValueExpressionPaths changes the given OTTL value expression adding the context name prefix
+// to all context-less paths. No modifications are performed for paths which [Path.Context]
+// value matches any WithPathContextNames value.
+// The context argument must be valid WithPathContextNames value, otherwise an error is returned.
+func (p *Parser[K]) prependContextToValueExpressionPaths(context string, expr string) (string, error) {
+	return p.prependContextToPaths(context, expr, func(ottl string) ([]path, error) {
+		parsed, err := parseValueExpression(ottl)
+		if err != nil {
+			return nil, err
+		}
+		return getValuePaths(parsed), nil
+	})
+}
+
 var (
 	parser                = sync.OnceValue(newParser[parsedStatement])
 	conditionParser       = sync.OnceValue(newParser[booleanExpression])
@@ -493,14 +507,38 @@ func (c *ConditionSequence[K]) Eval(ctx context.Context, tCtx K) (bool, error) {
 // a mathematical expression.
 // This allows other components using this library to extract data from the context of the incoming signal using OTTL.
 type ValueExpression[K any] struct {
-	getter Getter[K]
+	getter   Getter[K]
+	origText string
 }
 
 // Eval evaluates the given expression and returns the value the expression resolves to.
 func (e *ValueExpression[K]) Eval(ctx context.Context, tCtx K) (any, error) {
 	return e.getter.Get(ctx, tCtx)
 }
 
+// ParseValueExpressions parses string expressions into a ValueExpression slice ready for execution.
+// Returns a slice of ValueExpression and a nil error on successful parsing.
+// If parsing fails, returns nil and an error containing each error per failed condition.
+func (p *Parser[K]) ParseValueExpressions(expressions []string) ([]*ValueExpression[K], error) {
+	parsedValueExpressions := make([]*ValueExpression[K], 0, len(expressions))
+	var parseErrs []error
+
+	for _, expression := range expressions {
+		ps, err := p.ParseValueExpression(expression)
+		if err != nil {
+			parseErrs = append(parseErrs, fmt.Errorf("unable to parse OTTL value expression %q: %w", expression, err))
+			continue
+		}
+		parsedValueExpressions = append(parsedValueExpressions, ps)
+	}
+
+	if len(parseErrs) > 0 {
+		return nil, errors.Join(parseErrs...)
+	}
+
+	return parsedValueExpressions, nil
+}
+
 // ParseValueExpression parses an expression string into a ValueExpression. The ValueExpression's Eval
 // method can then be used to extract the value from the context of the incoming signal.
 func (p *Parser[K]) ParseValueExpression(raw string) (*ValueExpression[K], error) {
@@ -514,6 +552,7 @@ func (p *Parser[K]) ParseValueExpression(raw string) (*ValueExpression[K], error
 	}
 
 	return &ValueExpression[K]{
+		origText: raw,
 		getter: &StandardGetSetter[K]{
 			Getter: func(ctx context.Context, tCtx K) (any, error) {
 				val, err := getter.Get(ctx, tCtx)