Pre-check wether transactions have gen_ai spans when we construct transactions

to lessen the extraction performance impact

Author: andreiborza

packages/core/src/tracing/sentrySpan.ts

@@ -392,8 +392,12 @@ export class SentrySpan implements Span {
     // remove internal root span attributes we don't need to send.
     /* eslint-disable @typescript-eslint/no-dynamic-delete */
     delete this._attributes[SEMANTIC_ATTRIBUTE_SENTRY_CUSTOM_SPAN_NAME];
+    let hasGenAiSpans = false;
     spans.forEach(span => {
       delete span.data[SEMANTIC_ATTRIBUTE_SENTRY_CUSTOM_SPAN_NAME];
+      if (span.op?.startsWith('gen_ai.')) {
+        hasGenAiSpans = true;
+      }
     });
     // eslint-enabled-next-line @typescript-eslint/no-dynamic-delete
 
@@ -415,6 +419,7 @@ export class SentrySpan implements Span {
         capturedSpanScope,
         capturedSpanIsolationScope,
         dynamicSamplingContext: getDynamicSamplingContextFromSpan(this),
+        hasGenAiSpans,
       },
       request: normalizedRequest,
       ...(source && {

packages/core/src/tracing/spans/extractGenAiSpans.ts

@@ -15,7 +15,12 @@ import { spanJsonToSerializedStreamedSpan } from './spanJsonToStreamedSpan';
  * item does not include the extracted spans.
  */
 export function extractGenAiSpansFromEvent(event: Event, client: Client): SpanContainerItem | undefined {
-  if (event.type !== 'transaction' || !event.spans?.length || hasSpanStreamingEnabled(client)) {
+  if (
+    event.type !== 'transaction' ||
+    !event.spans?.length ||
+    !event.sdkProcessingMetadata?.hasGenAiSpans ||
+    hasSpanStreamingEnabled(client)
+  ) {
     return undefined;
   }
 

packages/core/test/lib/tracing/sentrySpan.test.ts

@@ -234,6 +234,7 @@ describe('SentrySpan', () => {
             trace_id: expect.stringMatching(/^[a-f0-9]{32}$/),
             transaction: 'test',
           },
+          hasGenAiSpans: false,
         },
         spans: [],
         start_timestamp: 1,

packages/core/test/lib/tracing/spans/extractGenAiSpans.test.ts

@@ -14,7 +14,7 @@ function makeSpanJSON(overrides: Partial<SpanJSON> = {}): SpanJSON {
   };
 }
 
-function makeTransactionEvent(spans: SpanJSON[]): Event {
+function makeTransactionEvent(spans: SpanJSON[], hasGenAiSpans = false): Event {
   return {
     type: 'transaction',
     transaction: 'GET /api/chat',
@@ -26,6 +26,9 @@ function makeTransactionEvent(spans: SpanJSON[]): Event {
         trace_id: '00112233445566778899aabbccddeeff',
       },
     },
+    sdkProcessingMetadata: {
+      ...(hasGenAiSpans && { hasGenAiSpans: true }),
+    },
     spans,
   };
 }
@@ -54,12 +57,9 @@ describe('extractGenAiSpansFromEvent', () => {
       timestamp: 1002,
     });
 
-    const event = makeTransactionEvent([genAiSpan, httpSpan]);
-    const client = makeClient();
-
-    const result = extractGenAiSpansFromEvent(event, client);
+    const event = makeTransactionEvent([genAiSpan, httpSpan], true);
+    const result = extractGenAiSpansFromEvent(event, makeClient());
 
-    // gen_ai spans should be in the container item
     expect(result).toBeDefined();
     const [headers, payload] = result!;
     expect(headers.type).toBe('span');
@@ -69,7 +69,6 @@ describe('extractGenAiSpansFromEvent', () => {
     expect(payload.items[0]!.span_id).toBe('genai001');
     expect(payload.items[0]!.name).toBe('chat gpt-4');
 
-    // gen_ai spans should be removed from the event
     expect(event.spans).toHaveLength(1);
     expect(event.spans![0]!.span_id).toBe('http001');
   });
@@ -80,54 +79,47 @@ describe('extractGenAiSpansFromEvent', () => {
     const agentSpan = makeSpanJSON({ span_id: 'agent001', op: 'gen_ai.invoke_agent', description: 'agent' });
     const dbSpan = makeSpanJSON({ span_id: 'db001', op: 'db.query', description: 'SELECT *' });
 
-    const event = makeTransactionEvent([chatSpan, embeddingsSpan, dbSpan, agentSpan]);
-    const client = makeClient();
-
-    const result = extractGenAiSpansFromEvent(event, client);
+    const event = makeTransactionEvent([chatSpan, embeddingsSpan, dbSpan, agentSpan], true);
+    const result = extractGenAiSpansFromEvent(event, makeClient());
 
     expect(result).toBeDefined();
     expect(result![0].item_count).toBe(3);
     expect(result![1].items).toHaveLength(3);
     expect(result![1].items.map(s => s.span_id)).toEqual(['chat001', 'embed001', 'agent001']);
 
-    // Only the db span should remain
     expect(event.spans).toHaveLength(1);
     expect(event.spans![0]!.span_id).toBe('db001');
   });
 
-  it('returns undefined when there are no gen_ai spans', () => {
-    const httpSpan = makeSpanJSON({ op: 'http.client' });
-    const dbSpan = makeSpanJSON({ op: 'db.query' });
+  it('returns undefined when hasGenAiSpans flag is not set', () => {
+    const event = makeTransactionEvent([makeSpanJSON({ op: 'gen_ai.chat' })], false);
 
-    const event = makeTransactionEvent([httpSpan, dbSpan]);
-    const client = makeClient();
+    expect(extractGenAiSpansFromEvent(event, makeClient())).toBeUndefined();
+    expect(event.spans).toHaveLength(1);
+  });
 
-    const result = extractGenAiSpansFromEvent(event, client);
+  it('returns undefined when there are no gen_ai spans', () => {
+    const event = makeTransactionEvent([makeSpanJSON({ op: 'http.client' }), makeSpanJSON({ op: 'db.query' })], true);
 
-    expect(result).toBeUndefined();
+    expect(extractGenAiSpansFromEvent(event, makeClient())).toBeUndefined();
     expect(event.spans).toHaveLength(2);
   });
 
   it('returns undefined when event has no spans', () => {
     const event = makeTransactionEvent([]);
-    const client = makeClient();
-
-    expect(extractGenAiSpansFromEvent(event, client)).toBeUndefined();
+    expect(extractGenAiSpansFromEvent(event, makeClient())).toBeUndefined();
   });
 
   it('returns undefined when event is not a transaction', () => {
     const event: Event = { type: undefined, spans: [makeSpanJSON({ op: 'gen_ai.chat' })] };
-    const client = makeClient();
-
-    expect(extractGenAiSpansFromEvent(event, client)).toBeUndefined();
+    expect(extractGenAiSpansFromEvent(event, makeClient())).toBeUndefined();
   });
 
   it('returns undefined when span streaming is enabled', () => {
-    const event = makeTransactionEvent([makeSpanJSON({ op: 'gen_ai.chat' })]);
+    const event = makeTransactionEvent([makeSpanJSON({ op: 'gen_ai.chat' })], true);
     const client = makeClient({ traceLifecycle: 'stream' });
 
     expect(extractGenAiSpansFromEvent(event, client)).toBeUndefined();
-    // Spans should not be modified
     expect(event.spans).toHaveLength(1);
   });
 
@@ -142,14 +134,10 @@ describe('extractGenAiSpansFromEvent', () => {
       op: 'http.client',
     });
 
-    const event = makeTransactionEvent([httpSpan, genAiSpan]);
-    const client = makeClient();
-
-    const result = extractGenAiSpansFromEvent(event, client);
+    const event = makeTransactionEvent([httpSpan, genAiSpan], true);
+    const result = extractGenAiSpansFromEvent(event, makeClient());
 
-    // The v2 span should still reference the v1 parent
     expect(result![1].items[0]!.parent_span_id).toBe('http001');
-    // The v1 parent should remain in the transaction
     expect(event.spans).toHaveLength(1);
     expect(event.spans![0]!.span_id).toBe('http001');
   });

packages/opentelemetry/src/spanExporter.ts

@@ -203,8 +203,11 @@ export class SentrySpanExporter {
       // We'll recursively add all the child spans to this array
       const spans = transactionEvent.spans || [];
 
+      let hasGenAiSpans = false;
       for (const child of root.children) {
-        createAndFinishSpanForOtelSpan(child, spans, sentSpans);
+        if (createAndFinishSpanForOtelSpan(child, spans, sentSpans)) {
+          hasGenAiSpans = true;
+        }
       }
 
       // spans.sort() mutates the array, but we do not use this anymore after this point
@@ -214,6 +217,13 @@ export class SentrySpanExporter {
           ? spans.sort((a, b) => a.start_timestamp - b.start_timestamp).slice(0, MAX_SPAN_COUNT)
           : spans;
 
+      if (hasGenAiSpans) {
+        transactionEvent.sdkProcessingMetadata = {
+          ...transactionEvent.sdkProcessingMetadata,
+          hasGenAiSpans: true,
+        };
+      }
+
       const measurements = timedEventsToMeasurements(span.events);
       if (measurements) {
         transactionEvent.measurements = measurements;
@@ -330,7 +340,10 @@ export function createTransactionForOtelSpan(span: ReadableSpan): TransactionEve
   return transactionEvent;
 }
 
-function createAndFinishSpanForOtelSpan(node: SpanNode, spans: SpanJSON[], sentSpans: Set<ReadableSpan>): void {
+/**
+ * Returns `true` if this span or any descendant is a gen_ai span.
+ */
+function createAndFinishSpanForOtelSpan(node: SpanNode, spans: SpanJSON[], sentSpans: Set<ReadableSpan>): boolean {
   const span = node.span;
 
   if (span) {
@@ -341,10 +354,13 @@ function createAndFinishSpanForOtelSpan(node: SpanNode, spans: SpanJSON[], sentS
 
   // If this span should be dropped, we still want to create spans for the children of this
   if (shouldDrop) {
+    let hasGenAiSpans = false;
     node.children.forEach(child => {
-      createAndFinishSpanForOtelSpan(child, spans, sentSpans);
+      if (createAndFinishSpanForOtelSpan(child, spans, sentSpans)) {
+        hasGenAiSpans = true;
+      }
     });
-    return;
+    return hasGenAiSpans;
   }
 
   const span_id = span.spanContext().spanId;
@@ -381,9 +397,13 @@ function createAndFinishSpanForOtelSpan(node: SpanNode, spans: SpanJSON[], sentS
 
   spans.push(spanJSON);
 
+  let hasGenAiSpans = !!op?.startsWith('gen_ai.');
   node.children.forEach(child => {
-    createAndFinishSpanForOtelSpan(child, spans, sentSpans);
+    if (createAndFinishSpanForOtelSpan(child, spans, sentSpans)) {
+      hasGenAiSpans = true;
+    }
   });
+  return hasGenAiSpans;
 }
 
 function getSpanData(span: ReadableSpan): {