feat(discover): add `cache_hit_rate` function into discover (#69415)

Add cache_hit_rate function into discover for the cache module to use.

---------

Co-authored-by: Edward Gou <edward.gou@sentry.io>

src/sentry/search/events/constants.py

@@ -191,6 +191,8 @@ HTTP_SERVER_ERROR_STATUS = {
     "511",
 }
 
+CACHE_HIT_STATUS = {"true", "false"}
+
 CONFIGURABLE_AGGREGATES = {
     "apdex()": "apdex({threshold}) as apdex",
     "user_misery()": "user_misery({threshold}) as user_misery",

src/sentry/search/events/datasets/spans_metrics.py

@@ -356,6 +356,15 @@ class SpansMetricsDatasetConfig(DatasetConfig):
                     ),
                     default_result_type="percentage",
                 ),
+                fields.MetricsFunction(
+                    "cache_hit_rate",
+                    snql_distribution=lambda args, alias: function_aliases.resolve_division(
+                        self._resolve_cache_hit_count(args),
+                        self._resolve_cache_hit_and_miss_count(args),
+                        alias,
+                    ),
+                    default_result_type="percentage",
+                ),
                 # TODO: Deprecated, use `http_response_rate(5)` instead
                 fields.MetricsFunction(
                     "http_error_rate",
@@ -685,6 +694,56 @@ class SpansMetricsDatasetConfig(DatasetConfig):
             alias,
         )
 
+    def _resolve_cache_hit_count(
+        self,
+        _: Mapping[str, str | Column | SelectType | int | float],
+        alias: str | None = None,
+    ) -> SelectType:
+
+        return self._resolve_count_if(
+            Function(
+                "equals",
+                [
+                    Column("metric_id"),
+                    self.resolve_metric("span.self_time"),
+                ],
+            ),
+            Function(
+                "equals",
+                [
+                    self.builder.column("cache.hit"),
+                    self.builder.resolve_tag_value("true"),
+                ],
+            ),
+            alias,
+        )
+
+    def _resolve_cache_hit_and_miss_count(
+        self,
+        _: Mapping[str, str | Column | SelectType | int | float],
+        alias: str | None = None,
+    ) -> SelectType:
+
+        statuses = [self.builder.resolve_tag_value(status) for status in constants.CACHE_HIT_STATUS]
+
+        return self._resolve_count_if(
+            Function(
+                "equals",
+                [
+                    Column("metric_id"),
+                    self.resolve_metric("span.self_time"),
+                ],
+            ),
+            Function(
+                "in",
+                [
+                    self.builder.column("cache.hit"),
+                    list(status for status in statuses if status is not None),
+                ],
+            ),
+            alias,
+        )
+
     def _resolve_http_response_count(
         self,
         args: Mapping[str, str | Column | SelectType | int | float],

tests/snuba/api/endpoints/test_organization_events_span_metrics.py

@@ -1031,6 +1031,35 @@ class OrganizationEventsMetricsEnhancedPerformanceEndpointTest(MetricsEnhancedPe
         assert data[0]["device.class"] == "Unknown"
         assert meta["fields"]["device.class"] == "string"
 
+    def test_cache_hit_rate(self):
+        self.store_span_metric(
+            1,
+            internal_metric=constants.SELF_TIME_LIGHT,
+            timestamp=self.min_ago,
+            tags={"cache.hit": "true"},
+        )
+        self.store_span_metric(
+            1,
+            internal_metric=constants.SELF_TIME_LIGHT,
+            timestamp=self.min_ago,
+            tags={"cache.hit": "false"},
+        )
+        response = self.do_request(
+            {
+                "field": ["cache_hit_rate()"],
+                "query": "",
+                "project": self.project.id,
+                "dataset": "spansMetrics",
+            }
+        )
+        assert response.status_code == 200, response.content
+        data = response.data["data"]
+        meta = response.data["meta"]
+        assert len(data) == 1
+        assert data[0]["cache_hit_rate()"] == 0.5
+        assert meta["dataset"] == "spansMetrics"
+        assert meta["fields"]["cache_hit_rate()"] == "percentage"
+
     def test_http_response_rate(self):
         self.store_span_metric(
             1,