feat(Replay): Add URLs Visited to EAP Replay Query

src/sentry/replays/endpoints/organization_replay_details.py

@@ -4,7 +4,17 @@
 from drf_spectacular.utils import extend_schema
 from rest_framework.request import Request
 from rest_framework.response import Response
-from snuba_sdk import Column, Condition, Entity, Function, Granularity, Op, Query
+from snuba_sdk import (
+    Column,
+    Condition,
+    Direction,
+    Entity,
+    Function,
+    Granularity,
+    Op,
+    OrderBy,
+    Query,
+)
 
 from sentry import features
 from sentry.api.api_owners import ApiOwner
@@ -24,6 +34,68 @@
 from sentry.replays.validators import ReplayValidator
 
 
+def _query_replay_urls_eap(
+    replay_id: str,
+    project_ids: list[int],
+    start: datetime,
+    end: datetime,
+    organization_id: int,
+) -> list[str]:
+    """Query URLs for a replay from EAP breadcrumb events."""
+    replay_id_no_dashes = replay_id.replace("-", "")
+
+    first_seen_agg = Function("min", parameters=[Column("sentry.timestamp")], alias="first_seen")
+
+    select = [
+        Column("to"),
+        first_seen_agg,
+    ]
+
+    snuba_query = Query(
+        match=Entity("replays"),
+        select=select,
+        where=[
+            Condition(Column("replay_id"), Op.EQ, replay_id_no_dashes),
+            Condition(Column("category"), Op.EQ, "navigation"),
+        ],
+        groupby=[Column("to")],
+        orderby=[OrderBy(first_seen_agg, Direction.ASC)],
+    )
+
+    settings = Settings(
+        attribute_types={
+            "replay_id": str,
+            "category": str,
+            "sentry.timestamp": float,
+            "to": str,
+        },
+        default_limit=1000,
+        default_offset=0,
+    )
+
+    request_meta = RequestMeta(
+        cogs_category="replays",
+        debug=False,
+        start_datetime=start,
+        end_datetime=end,
+        organization_id=organization_id,
+        project_ids=project_ids,
+        referrer="replays.query.urls",
+        request_id=str(uuid.uuid4().hex),
+        trace_item_type="replay",
+    )
+
+    result = eap_read.query(snuba_query, settings, request_meta, [])
+
+    urls: list[str] = []
+    for row in result.get("data", []):
+        url = row.get("to")
+        if url and isinstance(url, str):
+            urls.append(url)
+
+    return urls
+
+
 def _normalize_eap_response(data: list[dict]) -> list[dict]:
     """Normalize EAP response data for frontend compatibility.
 
@@ -202,6 +274,16 @@ def get(self, request: Request, organization: Organization, replay_id: str) -> R
                 organization_id=organization.id,
                 request_user_id=request.user.id,
             )["data"]
+
+            if snuba_response:
+                urls = _query_replay_urls_eap(
+                    replay_id=replay_id,
+                    project_ids=project_ids,
+                    start=filter_params["start"],
+                    end=filter_params["end"],
+                    organization_id=organization.id,
+                )
+                snuba_response[0]["urls_sorted"] = urls
         else:
             snuba_response = query_replay_instance(
                 project_id=project_ids,

tests/sentry/replays/endpoints/test_query_replay_instance_eap.py

@@ -4,6 +4,7 @@
 
 from sentry.replays.endpoints.organization_replay_details import (
     _normalize_eap_response,
+    _query_replay_urls_eap,
     query_replay_instance_eap,
 )
 from sentry.testutils.cases import ReplayBreadcrumbType, ReplayEAPTestCase, TestCase
@@ -47,6 +48,33 @@ def test_eap_replay_query(self) -> None:
                 breadcrumb_type=ReplayBreadcrumbType.CLICK,
                 timestamp=now,
             ),
+            self.create_eap_replay_breadcrumb(
+                project=self.project,
+                replay_id=replay_id1,
+                segment_id=0,
+                breadcrumb_type=ReplayBreadcrumbType.CLICK,
+                timestamp=now - datetime.timedelta(seconds=30),
+                category="navigation",
+                to="https://example.com/page1",
+            ),
+            self.create_eap_replay_breadcrumb(
+                project=self.project,
+                replay_id=replay_id1,
+                segment_id=0,
+                breadcrumb_type=ReplayBreadcrumbType.CLICK,
+                timestamp=now - datetime.timedelta(seconds=20),
+                category="navigation",
+                to="https://example.com/page2",
+            ),
+            self.create_eap_replay_breadcrumb(
+                project=self.project,
+                replay_id=replay_id1,
+                segment_id=0,
+                breadcrumb_type=ReplayBreadcrumbType.CLICK,
+                timestamp=now - datetime.timedelta(seconds=10),
+                category="navigation",
+                to="https://example.com/page3",
+            ),
         ]
 
         replay2_breadcrumbs = [
@@ -138,6 +166,30 @@ def test_eap_replay_query(self) -> None:
         assert replay2_data["count_dead_clicks"] == 3, "1 DEAD_CLICK + 2 RAGE_CLICK = 3 dead"
         assert replay2_data["count_rage_clicks"] == 2, "2 RAGE_CLICK = 2 rage"
 
+        # Test URL query for replay1
+        urls = _query_replay_urls_eap(
+            replay_id=replay_id1,
+            project_ids=project_ids,
+            start=start,
+            end=end,
+            organization_id=organization_id,
+        )
+        assert len(urls) == 3, f"Expected 3 URLs, got {len(urls)}"
+        assert urls == [
+            "https://example.com/page1",
+            "https://example.com/page2",
+            "https://example.com/page3",
+        ], f"URLs should be sorted by timestamp ascending, got {urls}"
+
+        urls2 = _query_replay_urls_eap(
+            replay_id=replay_id2,
+            project_ids=project_ids,
+            start=start,
+            end=end,
+            organization_id=organization_id,
+        )
+        assert len(urls2) == 0, f"Expected 0 URLs for replay2, got {len(urls2)}"
+
     def test_normalize_eap_response(self) -> None:
         """Test that EAP response data is correctly normalized.