feat(replay): add has_viewed to details and index (#68391)

Relates to #64924. Splitting
#67972 into smaller PRs.

---------

Co-authored-by: Colton Allen <[email protected]>

src/sentry/apidocs/examples/replay_examples.py

@@ -19,6 +19,7 @@
     "environment": "production",
     "error_ids": ["7e07485f-12f9-416b-8b14-26260799b51f"],
     "finished_at": "2022-07-07T14:15:33.201019",
+    "has_viewed": True,
     "id": "7e07485f-12f9-416b-8b14-26260799b51f",
     "is_archived": None,
     "os": {"name": "iOS", "version": "16.2"},

src/sentry/replays/endpoints/organization_replay_details.py

@@ -73,6 +73,7 @@ def get(self, request: Request, organization: Organization, replay_id: str) -> R
             start=filter_params["start"],
             end=filter_params["end"],
             organization=organization,
+            request_user_id=request.user.id,
         )
 
         response = process_raw_response(

src/sentry/replays/endpoints/project_replay_details.py

@@ -57,6 +57,7 @@ def get(self, request: Request, project: Project, replay_id: str) -> Response:
             start=filter_params["start"],
             end=filter_params["end"],
             organization=project.organization,
+            request_user_id=request.user.id,
         )
 
         response = process_raw_response(

src/sentry/replays/lib/new_query/parsers.py

@@ -2,6 +2,7 @@
 
 Functions in this module coerce external types to internal types.  Else they die.
 """
+
 import uuid
 
 from sentry.replays.lib.new_query.errors import CouldNotParseValue

src/sentry/replays/post_process.py

@@ -73,6 +73,7 @@ class ReplayDetailsResponse(TypedDict, total=False):
     info_ids: list[str] | None
     count_warnings: int | None
     count_infos: int | None
+    has_viewed: bool
 
 
 def process_raw_response(
@@ -181,6 +182,8 @@ def generate_normalized_output(
         ret_item["info_ids"] = item.pop("info_ids", None)
         ret_item["count_infos"] = item.pop("count_infos", None)
         ret_item["count_warnings"] = item.pop("count_warnings", None)
+        # Returns a UInt8 of either 0 or 1. We coerce to a bool.
+        ret_item["has_viewed"] = bool(item.get("has_viewed", 0))
         yield ret_item
 
 

src/sentry/replays/query.py

@@ -70,6 +70,7 @@ def query_replays_collection_raw(
         project_ids=project_ids,
         period_start=start,
         period_stop=end,
+        request_user_id=actor.id if actor else None,
     )
 
 
@@ -79,6 +80,7 @@ def query_replay_instance(
     start: datetime,
     end: datetime,
     organization: Organization | None = None,
+    request_user_id: int | None = None,
 ):
     """Query aggregated replay instance."""
     if isinstance(project_id, list):
@@ -93,6 +95,7 @@ def query_replay_instance(
             project_ids=project_ids,
             period_start=start,
             period_end=end,
+            request_user_id=request_user_id,
         ),
         tenant_id={"organization_id": organization.id} if organization else {},
         referrer="replays.query.details_query",
@@ -556,6 +559,7 @@ def _empty_uuids_lambda():
     "info_ids": ["info_ids"],
     "count_warnings": ["count_warnings"],
     "count_infos": ["count_infos"],
+    "has_viewed": ["has_viewed"],
 }
 
 
@@ -754,9 +758,16 @@ def collect_aliases(fields: list[str]) -> list[str]:
     return list(result)
 
 
-def select_from_fields(fields: list[str]) -> list[Column | Function]:
+def select_from_fields(fields: list[str], user_id: int | None) -> list[Column | Function]:
     """Return a list of columns to select."""
-    return [QUERY_ALIAS_COLUMN_MAP[alias] for alias in collect_aliases(fields)]
+    selection = []
+    for alias in collect_aliases(fields):
+        if alias == "has_viewed":
+            selection.append(compute_has_viewed(user_id))
+        else:
+            selection.append(QUERY_ALIAS_COLUMN_MAP[alias])
+
+    return selection
 
 
 def _extract_children(expression: ParenExpression) -> Generator[SearchFilter, None, None]:
@@ -765,3 +776,23 @@ def _extract_children(expression: ParenExpression) -> Generator[SearchFilter, No
             yield child
         elif isinstance(child, ParenExpression):
             yield from _extract_children(child)
+
+
+def compute_has_viewed(viewed_by_id: int | None) -> Function:
+    if viewed_by_id is None:
+        # Return the literal "false" if no user-id was specified.
+        return Function("equals", parameters=[1, 2])
+
+    return Function(
+        "greater",
+        parameters=[
+            Function(
+                "sum",
+                parameters=[
+                    Function("equals", parameters=[Column("viewed_by_id"), viewed_by_id]),
+                ],
+            ),
+            0,
+        ],
+        alias="has_viewed",
+    )

src/sentry/replays/testutils.py

@@ -119,6 +119,7 @@ def mock_expected_response(
         "count_errors": kwargs.pop("count_errors", 0),
         "count_warnings": kwargs.pop("count_warnings", 0),
         "count_infos": kwargs.pop("count_infos", 0),
+        "has_viewed": kwargs.pop("has_viewed", False),
     }
 
 
@@ -252,6 +253,31 @@ def mock_replay_click(
     }
 
 
+def mock_replay_viewed(
+    timestamp: float,
+    project_id: str,
+    replay_id: str,
+    viewed_by_id: int,
+    retention_days: int = 30,
+) -> dict[str, Any]:
+    return {
+        "type": "replay_event",
+        "start_time": int(timestamp),
+        "replay_id": replay_id,
+        "project_id": project_id,
+        "retention_days": retention_days,
+        "payload": list(
+            json.dumps(
+                {
+                    "type": "replay_viewed",
+                    "timestamp": timestamp,
+                    "viewed_by_id": viewed_by_id,
+                }
+            ).encode()
+        ),
+    }
+
+
 def mock_segment_init(
     timestamp: datetime.datetime,
     href: str = "http://localhost/",

src/sentry/replays/usecases/query/__init__.py

@@ -151,6 +151,7 @@ def query_using_optimized_search(
     project_ids: list[int],
     period_start: datetime,
     period_stop: datetime,
+    request_user_id: int | None = None,
 ):
     tenant_id = _make_tenant_id(organization)
 
@@ -212,6 +213,7 @@ def query_using_optimized_search(
             project_ids=project_ids,
             period_start=period_start,
             period_end=period_stop,
+            request_user_id=request_user_id,
         ),
         tenant_id,
         referrer="replays.query.browse_query",
@@ -285,15 +287,16 @@ def make_full_aggregation_query(
     project_ids: list[int],
     period_start: datetime,
     period_end: datetime,
+    request_user_id: int | None,
 ) -> Query:
     """Return a query to fetch every replay in the set."""
-    from sentry.replays.query import QUERY_ALIAS_COLUMN_MAP, select_from_fields
+    from sentry.replays.query import QUERY_ALIAS_COLUMN_MAP, compute_has_viewed, select_from_fields
 
     def _select_from_fields() -> list[Column | Function]:
         if fields:
-            return select_from_fields(list(set(fields)))
+            return select_from_fields(list(set(fields)), user_id=request_user_id)
         else:
-            return list(QUERY_ALIAS_COLUMN_MAP.values())
+            return list(QUERY_ALIAS_COLUMN_MAP.values()) + [compute_has_viewed(request_user_id)]
 
     return Query(
         match=Entity("replays"),

src/sentry/replays/usecases/query/conditions/__init__.py

@@ -11,6 +11,7 @@
     "SumOfClickScalar",
     "SumOfClickSelectorComposite",
     "SumOfDeadClickSelectorComposite",
+    "SumOfIntegerIdScalar",
     "SumOfIPv4Scalar",
     "SumOfRageClickSelectorComposite",
     "SumOfStringArray",
@@ -22,7 +23,13 @@
 
 
 from .activity import AggregateActivityScalar
-from .aggregate import SumOfIPv4Scalar, SumOfStringArray, SumOfStringScalar, SumOfUUIDArray
+from .aggregate import (
+    SumOfIntegerIdScalar,
+    SumOfIPv4Scalar,
+    SumOfStringArray,
+    SumOfStringScalar,
+    SumOfUUIDArray,
+)
 from .duration import SimpleAggregateDurationScalar
 from .error_ids import ErrorIdsArray
 from .selector import (

src/sentry/replays/usecases/query/conditions/aggregate.py

@@ -20,6 +20,7 @@
 aggregation of the row-wise condition visitor is 0 or not 0.  This works because we're looking
 for presence in the set; not for conformity across all rows.
 """
+
 from __future__ import annotations
 
 from uuid import UUID
@@ -29,6 +30,7 @@
 
 from sentry.replays.lib.new_query.conditions import (
     GenericBase,
+    IntegerScalar,
     IPv4Scalar,
     StringArray,
     StringScalar,
@@ -38,6 +40,24 @@
 from sentry.replays.lib.new_query.utils import contains, does_not_contain
 
 
+class SumOfIntegerIdScalar(GenericBase):
+    @staticmethod
+    def visit_eq(expression: Expression, value: int) -> Condition:
+        return contains(IntegerScalar.visit_eq(expression, value))
+
+    @staticmethod
+    def visit_neq(expression: Expression, value: int) -> Condition:
+        return does_not_contain(IntegerScalar.visit_eq(expression, value))
+
+    @staticmethod
+    def visit_in(expression: Expression, value: list[int]) -> Condition:
+        return contains(IntegerScalar.visit_in(expression, value))
+
+    @staticmethod
+    def visit_not_in(expression: Expression, value: list[int]) -> Condition:
+        return does_not_contain(IntegerScalar.visit_in(expression, value))
+
+
 class SumOfIPv4Scalar(GenericBase):
     @staticmethod
     def visit_eq(expression: Expression, value: str) -> Condition:

src/sentry/replays/usecases/query/configs/aggregate.py

@@ -10,13 +10,15 @@
 must parse to the data type of its source even if its later transformed into another type.  This
 acts as a validation step as must as a type coercion step.
 """
+
 from __future__ import annotations
 
 from sentry.replays.lib.new_query.conditions import IntegerScalar, UUIDScalar
 from sentry.replays.lib.new_query.fields import (
     ColumnField,
     CountField,
     FieldProtocol,
+    IntegerColumnField,
     StringColumnField,
     SumField,
     SumLengthField,
@@ -31,6 +33,7 @@
     SumOfClickScalar,
     SumOfClickSelectorComposite,
     SumOfDeadClickSelectorComposite,
+    SumOfIntegerIdScalar,
     SumOfIPv4Scalar,
     SumOfRageClickSelectorComposite,
     SumOfStringArray,
@@ -86,12 +89,12 @@ def array_string_field(column_name: str) -> StringColumnField:
     "click.textContent": click_field("click_text"),
     "click.title": click_field("click_title"),
     "count_dead_clicks": sum_field("click_is_dead"),
-    "count_infos": sum_field("count_info_events"),
-    "count_warnings": sum_field("count_warning_events"),
     "count_errors": sum_field("count_error_events"),
+    "count_infos": sum_field("count_info_events"),
     "count_rage_clicks": sum_field("click_is_rage"),
     "count_segments": count_field("segment_id"),
     "count_urls": sum_field("count_urls"),
+    "count_warnings": sum_field("count_warning_events"),
     "dead.selector": ComputedField(parse_selector, SumOfDeadClickSelectorComposite),
     "device.brand": string_field("device_brand"),
     "device.family": string_field("device_family"),
@@ -101,11 +104,10 @@ def array_string_field(column_name: str) -> StringColumnField:
     "duration": ComputedField(parse_int, SimpleAggregateDurationScalar),
     "environment": string_field("environment"),
     "error_ids": ComputedField(parse_uuid, SumOfErrorIdScalar),
-    "warning_ids": UUIDColumnField("warning_id", parse_uuid, SumOfUUIDScalar),
-    "info_ids": ComputedField(parse_uuid, SumOfInfoIdScalar),
     # Backwards Compat: We pass a simple string to the UUID column. Older versions of ClickHouse
     # do not understand the UUID type.
     "id": ColumnField("replay_id", parse_uuid, UUIDScalar),
+    "info_ids": ComputedField(parse_uuid, SumOfInfoIdScalar),
     "os.name": string_field("os_name"),
     "os.version": string_field("os_version"),
     "platform": string_field("platform"),
@@ -120,6 +122,8 @@ def array_string_field(column_name: str) -> StringColumnField:
     "user.id": string_field("user_id"),
     "user.ip_address": StringColumnField("ip_address_v4", parse_str, SumOfIPv4Scalar),
     "user.username": string_field("user_name"),
+    "viewed_by_id": IntegerColumnField("viewed_by_id", parse_int, SumOfIntegerIdScalar),
+    "warning_ids": UUIDColumnField("warning_id", parse_uuid, SumOfUUIDScalar),
 }
 
 
@@ -136,8 +140,8 @@ def array_string_field(column_name: str) -> StringColumnField:
 # Fields which have multiple names that represent the same search operation are defined here.
 # QQ:JFERG: why dont we have these on the scalar search
 search_config["error_id"] = search_config["error_ids"]
-search_config["warning_id"] = search_config["warning_ids"]
 search_config["info_id"] = search_config["info_ids"]
+search_config["warning_id"] = search_config["warning_ids"]
 
 
 search_config["release"] = search_config["releases"]