[CLEAN] Synthetic Benchmark PR #103102 - feat(tracemetrics): Parse metric from aggregations

[tomerqodo]

User description

Benchmark PR getsentry#103102

Type: Clean (correct implementation)

Original PR Title: feat(tracemetrics): Parse metric from aggregations
Original PR Description: This parses the metric from aggregations instead of relying on the top level metric name/type. The top level ones still take precedence for better compatibility. It has the limitation that it only allows a single metric across all the aggregations.
Original PR URL: getsentry#103102

---

PR Type

Enhancement

---

Description

• Parse metric information from aggregation functions instead of relying solely on top-level metric configuration

• Extract metric name, type, and unit from function arguments and store in new ResolvedMetricAggregate class

• Refactor metric filter generation into reusable utility functions for cleaner code organization

• Add support for multiple aggregations with the same metric while preventing cross-metric aggregations

---

Diagram Walkthrough

flowchart LR
  A["Aggregation Functions"] -->|extract metric info| B["ResolvedMetricAggregate"]
  C["Top-level Config"] -->|fallback| B
  B -->|generate filters| D["Metric Filters"]
  D -->|apply to query| E["Query Execution"]

Loading

File Walkthrough

	Relevant files
Enhancement	columns.py Add metric extraction from aggregation arguments                  src/sentry/search/eap/columns.py Add cast import for type casting Import MetricType from types module Create new ResolvedMetricAggregate dataclass extending ResolvedAggregate with metric metadata fields Update TraceMetricAggregateDefinition.resolve() to extract and store metric name, type, and unit from function arguments Return ResolvedMetricAggregate instead of ResolvedAggregate with populated metric fields +24/-6    resolver.py Add column-aware query resolution with dataset conditions src/sentry/search/eap/resolver.py Import and_trace_item_filters utility function Refactor resolve_query() to use and_trace_item_filters() for cleaner filter combination logic Add new resolve_query_with_columns() method that accepts selected columns and equations Implement resolve_dataset_conditions() to extract dataset-specific conditions from selected columns Call config.extra_conditions() with column and equation information for metric-aware filtering +35/-22  config.py Implement metric parsing from aggregations with fallback  src/sentry/search/eap/trace_metrics/config.py Move MetricType definition to types module Create new Metric dataclass to represent metric with name, type, and unit Refactor extra_conditions() signature to accept selected columns and equations Add _extra_conditions_from_metric() to handle top-level config metric filtering Add _extra_conditions_from_columns() to parse metrics from aggregation functions Extract get_metric_filter() as standalone function for reusability Add validation to prevent aggregating multiple different metrics in single query +118/-37 types.py Update config interface and centralize type definitions    src/sentry/search/eap/types.py Update SearchResolverConfig.extra_conditions() signature to accept selected columns and equations parameters Move MetricType type alias definition from trace_metrics config to types module +9/-1
Refactoring	rpc_utils.py Extract filter combination utility function                            src/sentry/search/eap/rpc_utils.py Create new utility module with and_trace_item_filters() function Consolidate filter combination logic to handle multiple TraceItemFilter objects Return single filter, combined filter, or None based on input count +14/-0    rpc_dataset_common.py Integrate column-aware query resolution and consolidate utilities src/sentry/snuba/rpc_dataset_common.py Import and_trace_item_filters from rpc_utils module Update get_table_rpc_request() to call resolve_query_with_columns() with selected columns and equations Move variable declarations in get_timeseries_query() before filter resolution for proper metric extraction Remove manual extra_conditions calls and let resolve_query_with_columns() handle dataset conditions Remove duplicate and_trace_item_filters() function definition +18/-29  trace_metrics.py Remove redundant extra conditions extraction                          src/sentry/snuba/trace_metrics.py Remove manual extra_conditions extraction calls from run_table_query() and run_timeseries_query() Let resolve_query_with_columns() handle metric condition injection automatically +0/-6
Tests	test_organization_events_trace_metrics.py Update tests for metric unit format and aggregation parsing tests/snuba/api/endpoints/test_organization_events_trace_metrics.py Update test cases to use - instead of none for missing metric unit parameter Add three new test cases for metric extraction from aggregation functions Test single metric aggregation with embedded metric info Test multiple aggregations with same metric name Test error handling for multiple different metrics in single query +74/-11

---

[qodo-code-review]

PR Compliance Guide 🔍

Below is a summary of compliance checks for this PR:

Security Compliance
🟢	No security concerns identified No security vulnerabilities detected by AI analysis. Human verification advised for critical code.
Ticket Compliance
⚪	🎫 No ticket provided Create ticket/issue
Codebase Duplication Compliance
⚪	Codebase context is not defined Follow the guide to enable codebase context checks.
Custom Compliance
🟢	Generic: Meaningful Naming and Self-Documenting Code Objective: Ensure all identifiers clearly express their purpose and intent, making code self-documenting Status: Passed Learn more about managing compliance generic rules or creating your own custom rules
	Generic: Secure Error Handling Objective: To prevent the leakage of sensitive system information through error messages while providing sufficient detail for internal debugging. Status: Passed Learn more about managing compliance generic rules or creating your own custom rules
	Generic: Secure Logging Practices Objective: To ensure logs are useful for debugging and auditing without exposing sensitive information like PII, PHI, or cardholder data. Status: Passed Learn more about managing compliance generic rules or creating your own custom rules
⚪	Generic: Comprehensive Audit Trails Objective: To create a detailed and reliable record of critical system actions for security analysis and compliance. Status: No audit logs: The new query resolution paths and condition injections (e.g., resolve_query_with_columns and resolve_dataset_conditions) introduce critical behavior changes without any added auditing or logging of these actions. Referred Code def resolve_query_with_columns( self, querystring: str | None, selected_columns: list[str] | None, equations: list[str] | None, ) -> tuple[ TraceItemFilter | None, AggregationFilter | None, list[VirtualColumnDefinition | None], ]: where, having, contexts = self.resolve_query(querystring) # Some datasets like trace metrics require we inject additional # conditions in the top level. dataset_conditions = self.resolve_dataset_conditions(selected_columns, equations) where = and_trace_item_filters(where, dataset_conditions) return where, having, contexts Learn more about managing compliance generic rules or creating your own custom rules
	Generic: Robust Error Handling and Edge Case Management Objective: Ensure comprehensive error handling that provides meaningful context and graceful degradation Status: Input validation: Metric parsing trusts argument positions and only validates emptiness/type loosely, potentially leading to runtime issues if fewer than expected arguments are supplied or wrong types are passed. Referred Code metric_name = None metric_type = None metric_unit = None if all( isinstance(resolved_argument, str) and resolved_argument != "" for resolved_argument in resolved_arguments[1:] ): # a metric was passed metric_name = cast(str, resolved_arguments[1]) metric_type = cast(MetricType, resolved_arguments[2]) metric_unit = None if resolved_arguments[3] == "-" else cast(str, resolved_arguments[3]) elif all(resolved_argument == "" for resolved_argument in resolved_arguments[1:]): # no metrics were specified, assume we query all metrics pass else: raise InvalidSearchQuery( f"Trace metric aggregates expect the full metric to be specified, got name:{resolved_arguments[1]} type:{resolved_arguments[2]} unit:{resolved_arguments[3]}" ) Learn more about managing compliance generic rules or creating your own custom rules
	Generic: Security-First Input Validation and Data Handling Objective: Ensure all data inputs are validated, sanitized, and handled securely to prevent vulnerabilities Status: Set handling risks: Using a set of Metric dataclass instances (selected_metrics: set[Metric]) relies on dataclass hash/eq defaults which may not behave as intended for uniqueness and could mis-detect multiple metrics. Referred Code selected_metrics: set[Metric] = set() if selected_columns: stripped_columns = [column.strip() for column in selected_columns] for column in stripped_columns: match = fields.is_function(column) if not match: continue resolved_function, _ = search_resolver.resolve_function(column) if not isinstance(resolved_function, ResolvedMetricAggregate): continue if not resolved_function.metric_name or not resolved_function.metric_type: continue metric = Metric( metric_name=resolved_function.metric_name, metric_type=resolved_function.metric_type, metric_unit=resolved_function.metric_unit, ) ... (clipped 13 lines) Learn more about managing compliance generic rules or creating your own custom rules

Compliance status legend

🟢 - Fully Compliant
🟡 - Partial Compliant
🔴 - Not Compliant
⚪ - Requires Further Human Verification
🏷️ - Compliance label

[qodo-code-review]

PR Code Suggestions ✨

Explore these optional code suggestions:

Category	Suggestion	Impact
Possible issue	Improve metric argument parsing logic Refine the metric argument parsing logic to be more flexible. Handle partially specified metrics and add validation for the metric_type to prevent potential errors. src/sentry/search/eap/columns.py [403-417] -if all( - isinstance(resolved_argument, str) and resolved_argument != "" - for resolved_argument in resolved_arguments[1:] -): - # a metric was passed - metric_name = cast(str, resolved_arguments[1]) - metric_type = cast(MetricType, resolved_arguments[2]) - metric_unit = None if resolved_arguments[3] == "-" else cast(str, resolved_arguments[3]) -elif all(resolved_argument == "" for resolved_argument in resolved_arguments[1:]): - # no metrics were specified, assume we query all metrics - pass -else: - raise InvalidSearchQuery( - f"Trace metric aggregates expect the full metric to be specified, got name:{resolved_arguments[1]} type:{resolved_arguments[2]} unit:{resolved_arguments[3]}" - ) +metric_name = cast(str, resolved_arguments[1]) if resolved_arguments[1] else None +metric_type_str = cast(str, resolved_arguments[2]) if resolved_arguments[2] else None +metric_unit = cast(str, resolved_arguments[3]) if resolved_arguments[3] not in ("-", "") else None +metric_type: MetricType | None = None +if metric_type_str: + if metric_type_str not in get_args(MetricType): + raise InvalidSearchQuery(f"Invalid metric type: {metric_type_str}") + metric_type = cast(MetricType, metric_type_str) + +if metric_name and not metric_type: + raise InvalidSearchQuery("metric type is required when a metric name is provided") + Apply / Chat Suggestion importance[1-10]: 7 __ Why: The suggestion correctly identifies that the current logic is too strict and proposes a more flexible approach for parsing metric arguments, including validation for metric_type. This improves robustness and user experience by allowing partially specified metrics.	Medium
More