Replays Self-Serve Bulk Delete System

[ShashankFC]

Test 5

Summary by CodeRabbit

Release Notes

• New Features

  • Added AI-powered feedback summaries in the feedback list.
  • Introduced table widget visualization for dashboards with customizable columns.
  • Added owner assignment support for workflow detectors.
  • Enhanced replay summaries with error context integration.

• Bug Fixes

  • Improved PR comment formatting with better issue title and environment display.
  • Refined feedback summary generation with stricter length and format constraints.
  • Disabled autofix automation tuning by default for improved stability.

---

Replicated from ai-code-review-evaluation/sentry-coderabbit#5

[ShashankFC]

@cubic-dev-ai review this pull request

[cubic-dev-ai]

@cubic-dev-ai review this pull request

@ShashankFC I have started the AI code review. It will take a few minutes to complete.

[cubic-dev-ai]

21 issues found across 106 files

Note: This PR contains a large number of files. cubic only reviews up to 75 files per PR, so some files may not have been reviewed.

Prompt for AI agents (all issues)

Check if these issues are valid — if so, understand the root cause of each and fix them.


<file name="src/sentry/replays/usecases/delete.py">

<violation number="1" location="src/sentry/replays/usecases/delete.py:88">
P3: Typo: "segements" should be "segments".</violation>
</file>

<file name="static/app/views/dashboards/widgets/tableWidget/tableWidgetVisualization.tsx">

<violation number="1" location="static/app/views/dashboards/widgets/tableWidget/tableWidgetVisualization.tsx:94">
P2: Incomplete defensive coding: `Object.keys(tableData?.meta.fields)` will throw `TypeError` if `tableData` is nullish, because `Object.keys(undefined)` crashes. The optional chaining short-circuits to `undefined` but there's no fallback. Compare with `tableData?.data ?? []` on line 107, which properly falls back. Either remove the unnecessary optional chaining (since `tableData` is required) or add a proper fallback.</violation>
</file>

<file name="src/sentry/preprod/api/endpoints/organization_preprod_artifact_assemble.py">

<violation number="1" location="src/sentry/preprod/api/endpoints/organization_preprod_artifact_assemble.py:81">
P2: Analytics is recorded before the feature flag check, which means events will be tracked even for organizations that don't have the feature enabled (and receive a 404). Move the `analytics.record()` call after the `features.has()` check to avoid polluting analytics data with rejected requests.</violation>
</file>

<file name="static/app/components/scrollCarousel.tsx">

<violation number="1" location="static/app/components/scrollCarousel.tsx:204">
P2: Using CSS `transparent` here will cause a visible dark band artifact in the gradient. `transparent` is `rgba(0,0,0,0)` (transparent **black**), so the gradient interpolates through dark color values on the way to `p.theme.background`. The `LeftMask` correctly avoids this by using `Color(p.theme.background).alpha(0).rgb().string()` — the `RightMask` should do the same for a smooth, artifact-free gradient.</violation>
</file>

<file name="static/app/views/dashboards/widgetCard/chart.tsx">

<violation number="1" location="static/app/views/dashboards/widgetCard/chart.tsx:165">
P1: The `TableWidgetVisualization` is rendered with hardcoded empty `columns` and `tableData` instead of the actual query results available in scope (`result.data`, `result.meta`, `fields`). When the `use-table-widget-visualization` feature flag is enabled, users will see a blank table. Even if this is a scaffold, passing `columns={[]}` (an empty array rather than `undefined`) also defeats the component's own fallback column derivation logic. Consider wiring through the real data, or at minimum omit the `columns` prop so the component can self-derive columns from `tableData.meta`.</violation>
</file>

<file name="src/sentry/issues/endpoints/browser_reporting_collector.py">

<violation number="1" location="src/sentry/issues/endpoints/browser_reporting_collector.py:52">
P1: Bug: Truthiness check `self.initial_data.get("age")` will be falsy when `age=0`, which is a valid value (e.g., a report just generated). This silently bypasses the mutual exclusivity validation, allowing both `age` and `timestamp` to coexist. Use an identity check against `None` or `in` instead.</violation>

<violation number="2" location="src/sentry/issues/endpoints/browser_reporting_collector.py:58">
P1: Bug: Same truthiness issue — `self.initial_data.get("timestamp")` will be falsy when `timestamp=0` (allowed by `min_value=0`). This bypasses the mutual exclusivity check.</violation>
</file>

<file name="static/app/views/performance/newTraceDetails/traceWaterfall.tsx">

<violation number="1" location="static/app/views/performance/newTraceDetails/traceWaterfall.tsx:712">
P2: Potential layout regression: the old `FlexBox` had `height: 100%` which is not carried over to the `Flex` component. If any ancestor relies on `height: 100%` to size this container (rather than flexbox), the waterfall may collapse or not fill available vertical space. Consider adding a `style={{ height: '100%' }}` or `h="100%"` prop if the `Flex` component supports it.</violation>
</file>

<file name="static/app/views/feedback/feedbackListPage.tsx">

<violation number="1" location="static/app/views/feedback/feedbackListPage.tsx:125">
P2: `SummaryListContainer` is missing `overflow: hidden` (or `min-height: 0`). As a flex-column container placed in a grid cell, it inherits `min-height: auto` by default, which can prevent it from properly constraining the inner `Container` (a `FluidHeight` component). This can cause the feedback list to overflow its grid area instead of scrolling within the container.</violation>
</file>

<file name="static/app/views/performance/newTraceDetails/traceTabsAndVitals.tsx">

<violation number="1" location="static/app/views/performance/newTraceDetails/traceTabsAndVitals.tsx:27">
P2: Style regression: the second `<Flex>` in `Placeholder` is missing `align="center"` and `gap={space(1)}` that the original `FlexBox` had. The first `FlexBox` replacement correctly includes these props — this one should too.</violation>

<violation number="2" location="static/app/views/performance/newTraceDetails/traceTabsAndVitals.tsx:93">
P2: Style regression: the main `Container` replacement is missing `align="center"` and `gap={space(1)}` that the original `Container` (which extended `FlexBox`) had. Compare with the `Placeholder` component's outer `Flex`, which correctly preserves all three properties.</violation>
</file>

<file name="devservices/config.yml">

<violation number="1" location="devservices/config.yml:171">
P2: New subscription results consumers are missing from `full` and `ingest-all` modes. Both modes appear to be comprehensive configurations that include all consumers, yet the four new consumers (`eap-spans-subscription-results`, `subscription-results-eap-items`, `metrics-subscription-results`, `generic-metrics-subscription-results`) added here are only included in the `tracing` mode. They should likely be added to `full` and `ingest-all` as well for consistency.</violation>
</file>

<file name="static/app/components/feedback/list/useFeedbackSummary.tsx">

<violation number="1" location="static/app/components/feedback/list/useFeedbackSummary.tsx:37">
P2: `Boolean(normalizedDateRange)` is always `true` because `normalizeDateTimeParams` returns a plain object, which is inherently truthy in JavaScript. This guard has no effect. If the intent is to verify that meaningful date parameters exist, check for specific properties instead (e.g., that `start`/`end` or `statsPeriod` is present).</violation>
</file>

<file name="static/app/views/performance/newTraceDetails/traceDrawer/details/span/eapSections/attributes.tsx">

<violation number="1" location="static/app/views/performance/newTraceDetails/traceDrawer/details/span/eapSections/attributes.tsx:67">
P1: Bug: `HIDDEN_ATTRIBUTES` are only filtered when a search query is present. When the search box is empty (default view), the early return on line 65 (`return sorted;`) skips the hidden-attributes filter entirely, so `is_segment`, `project_id`, and `received` will still be displayed. The filter should be applied unconditionally.</violation>
</file>

<file name="src/sentry/integrations/source_code_management/commit_context.py">

<violation number="1" location="src/sentry/integrations/source_code_management/commit_context.py:582">
P2: `_truncate_title` can return a string up to `max_length + 3` characters due to the appended `"..."`. If the intent of `ISSUE_TITLE_MAX_LENGTH = 50` is to cap the final output at 50 characters, the slice should account for the ellipsis (e.g., `title[:max_length - 3]`).</violation>
</file>

<file name="static/app/components/feedback/feedbackSummary.tsx">

<violation number="1" location="static/app/components/feedback/feedbackSummary.tsx:11">
P2: Missing null/empty guard on `summary` before rendering. If the API returns successfully but with `summary: null` (or an empty string), the component will render the summary container with a header but no content. Add a check for `!summary` alongside the other early-return conditions.</violation>
</file>

<file name="static/app/views/explore/hooks/useTraceItemAttributeKeys.tsx">

<violation number="1" location="static/app/views/explore/hooks/useTraceItemAttributeKeys.tsx:37">
P2: Overly broad dependency: `useMemo` depends on `selection` but only uses `selection.datetime`. This causes unnecessary recomputations when unrelated parts of `selection` (e.g., `environments`) change.</violation>
</file>

<file name="src/sentry/hybridcloud/tasks/deliver_webhooks.py">

<violation number="1" location="src/sentry/hybridcloud/tasks/deliver_webhooks.py:239">
P1: `processing_deadline_duration=120` (2 min) is shorter than the task's internal `BATCH_SCHEDULE_OFFSET` deadline of 3 minutes. The taskworker will consider this task timed out before the internal delivery loop naturally breaks. This should be at least 180 seconds, or ideally match `drain_mailbox`'s 300 seconds to provide buffer for the stale-message cleanup that precedes the loop.</violation>
</file>

<file name="static/app/components/events/eventAttachments.tsx">

<violation number="1" location="static/app/components/events/eventAttachments.tsx:143">
P2: Replacing `FlexCenter` with `<Flex align="center">` drops `overflow: hidden` from the container. In a CSS grid with `1fr` column, this means long attachment names will overflow instead of truncating with ellipsis. The flex container needs `overflow: hidden` or `min-width: 0` so the child `Name`'s `text-overflow: ellipsis` can take effect.</violation>
</file>

<file name="src/sentry/replays/endpoints/project_replay_summarize_breadcrumbs.py">

<violation number="1" location="src/sentry/replays/endpoints/project_replay_summarize_breadcrumbs.py:118">
P1: Bug: `zip(error_ids, events.values())` will misalign error IDs with event data when nodestore entries are missing. The dict returned by `get_multi` is keyed by `node_id`, and missing entries shrink the dict, causing subsequent IDs to pair with wrong data. Instead, iterate over `(error_id, node_id)` pairs and look up each node_id in the dict.</violation>

<violation number="2" location="src/sentry/replays/endpoints/project_replay_summarize_breadcrumbs.py:155">
P1: Potential timestamp unit mismatch: error event timestamps from nodestore are in Unix seconds while RRWeb replay segment event timestamps are in milliseconds. This comparison will always evaluate to `True` (seconds << milliseconds), causing all errors to be emitted before any breadcrumb events rather than being interleaved chronologically. Consider converting one format to match the other (e.g., `error_timestamp * 1000` or `event_timestamp / 1000`).</violation>
</file>

_{Reply with feedback, questions, or to request a fix. Tag @cubic-dev-ai to re-run a review.}

[cubic-dev-ai]

P1: The TableWidgetVisualization is rendered with hardcoded empty columns and tableData instead of the actual query results available in scope (result.data, result.meta, fields). When the use-table-widget-visualization feature flag is enabled, users will see a blank table. Even if this is a scaffold, passing columns={[]} (an empty array rather than undefined) also defeats the component's own fallback column derivation logic. Consider wiring through the real data, or at minimum omit the columns prop so the component can self-derive columns from tableData.meta.

Prompt for AI agents

Check if this issue is valid — if so, understand the root cause and fix it. At static/app/views/dashboards/widgetCard/chart.tsx, line 165:

<comment>The `TableWidgetVisualization` is rendered with hardcoded empty `columns` and `tableData` instead of the actual query results available in scope (`result.data`, `result.meta`, `fields`). When the `use-table-widget-visualization` feature flag is enabled, users will see a blank table. Even if this is a scaffold, passing `columns={[]}` (an empty array rather than `undefined`) also defeats the component's own fallback column derivation logic. Consider wiring through the real data, or at minimum omit the `columns` prop so the component can self-derive columns from `tableData.meta`.</comment>

<file context>
@@ -162,23 +161,36 @@ class WidgetCardChart extends Component<WidgetCardChartProps> {
-            minColumnWidth={minTableColumnWidth}
-          />
+          {organization.features.includes('use-table-widget-visualization') ? (
+            <TableWidgetVisualization
+              columns={[]}
+              tableData={{
</file context>

Suggested change

	<TableWidgetVisualization
	columns={[]}
	tableData={{
	data: [],
	meta: {
	fields: {},
	units: {},
	},
	}}
	/>
	<TableWidgetVisualization
	tableData={{
	data: result.data,
	meta: {
	fields: result.meta?.fields ?? {},
	units: result.meta?.units ?? {},
	},
	}}
	/>

[cubic-dev-ai]

P1: Bug: Same truthiness issue — self.initial_data.get("timestamp") will be falsy when timestamp=0 (allowed by min_value=0). This bypasses the mutual exclusivity check.

Prompt for AI agents

Check if this issue is valid — if so, understand the root cause and fix it. At src/sentry/issues/endpoints/browser_reporting_collector.py, line 58:

<comment>Bug: Same truthiness issue — `self.initial_data.get("timestamp")` will be falsy when `timestamp=0` (allowed by `min_value=0`). This bypasses the mutual exclusivity check.</comment>

<file context>
@@ -17,30 +18,46 @@
+
+    def validate_age(self, value: int) -> int:
+        """Validate that age is present, but not timestamp."""
+        if self.initial_data.get("timestamp"):
+            raise serializers.ValidationError("If age is present, timestamp must be absent")
+        return value
</file context>

Suggested change

	if self.initial_data.get("timestamp"):
	if "timestamp" in self.initial_data:

[cubic-dev-ai]

P1: Bug: Truthiness check self.initial_data.get("age") will be falsy when age=0, which is a valid value (e.g., a report just generated). This silently bypasses the mutual exclusivity validation, allowing both age and timestamp to coexist. Use an identity check against None or in instead.

Prompt for AI agents

Check if this issue is valid — if so, understand the root cause and fix it. At src/sentry/issues/endpoints/browser_reporting_collector.py, line 52:

<comment>Bug: Truthiness check `self.initial_data.get("age")` will be falsy when `age=0`, which is a valid value (e.g., a report just generated). This silently bypasses the mutual exclusivity validation, allowing both `age` and `timestamp` to coexist. Use an identity check against `None` or `in` instead.</comment>

<file context>
@@ -17,30 +18,46 @@
+
+    def validate_timestamp(self, value: int) -> int:
+        """Validate that age is absent, but timestamp is present."""
+        if self.initial_data.get("age"):
+            raise serializers.ValidationError("If timestamp is present, age must be absent")
+        return value
</file context>

Suggested change

	if self.initial_data.get("age"):
	if "age" in self.initial_data:

[cubic-dev-ai]

P1: Bug: HIDDEN_ATTRIBUTES are only filtered when a search query is present. When the search box is empty (default view), the early return on line 65 (return sorted;) skips the hidden-attributes filter entirely, so is_segment, project_id, and received will still be displayed. The filter should be applied unconditionally.

Prompt for AI agents

Check if this issue is valid — if so, understand the root cause and fix it. At static/app/views/performance/newTraceDetails/traceDrawer/details/span/eapSections/attributes.tsx, line 67:

<comment>Bug: `HIDDEN_ATTRIBUTES` are only filtered when a search query is present. When the search box is empty (default view), the early return on line 65 (`return sorted;`) skips the hidden-attributes filter entirely, so `is_segment`, `project_id`, and `received` will still be displayed. The filter should be applied unconditionally.</comment>

<file context>
@@ -60,8 +62,10 @@ export function Attributes({
-      attribute.name.toLowerCase().trim().includes(searchQuery.toLowerCase().trim())
+    return sorted.filter(
+      attribute =>
+        !HIDDEN_ATTRIBUTES.includes(attribute.name) &&
+        attribute.name.toLowerCase().trim().includes(searchQuery.toLowerCase().trim())
     );
</file context>

[cubic-dev-ai]

P1: processing_deadline_duration=120 (2 min) is shorter than the task's internal BATCH_SCHEDULE_OFFSET deadline of 3 minutes. The taskworker will consider this task timed out before the internal delivery loop naturally breaks. This should be at least 180 seconds, or ideally match drain_mailbox's 300 seconds to provide buffer for the stale-message cleanup that precedes the loop.

Prompt for AI agents

Check if this issue is valid — if so, understand the root cause and fix it. At src/sentry/hybridcloud/tasks/deliver_webhooks.py, line 239:

<comment>`processing_deadline_duration=120` (2 min) is shorter than the task's internal `BATCH_SCHEDULE_OFFSET` deadline of 3 minutes. The taskworker will consider this task timed out before the internal delivery loop naturally breaks. This should be at least 180 seconds, or ideally match `drain_mailbox`'s 300 seconds to provide buffer for the stale-message cleanup that precedes the loop.</comment>

<file context>
@@ -234,6 +236,7 @@ def drain_mailbox(payload_id: int) -> None:
     silo_mode=SiloMode.CONTROL,
     taskworker_config=TaskworkerConfig(
         namespace=hybridcloud_control_tasks,
+        processing_deadline_duration=120,
     ),
 )
</file context>

Suggested change

	processing_deadline_duration=120,
	processing_deadline_duration=300,

[cubic-dev-ai]

P2: _truncate_title can return a string up to max_length + 3 characters due to the appended "...". If the intent of ISSUE_TITLE_MAX_LENGTH = 50 is to cap the final output at 50 characters, the slice should account for the ellipsis (e.g., title[:max_length - 3]).

Prompt for AI agents

Check if this issue is valid — if so, understand the root cause and fix it. At src/sentry/integrations/source_code_management/commit_context.py, line 582:

<comment>`_truncate_title` can return a string up to `max_length + 3` characters due to the appended `"..."`. If the intent of `ISSUE_TITLE_MAX_LENGTH = 50` is to cap the final output at 50 characters, the slice should account for the ellipsis (e.g., `title[:max_length - 3]`).</comment>

<file context>
@@ -570,6 +574,37 @@ def get_top_5_issues_by_count(
+        """Truncate title if it's too long and add ellipsis."""
+        if len(title) <= max_length:
+            return title
+        return title[:max_length].rstrip() + "..."
+
+    def get_environment_info(self, issue: Group) -> str:
</file context>

Suggested change

	return title[:max_length].rstrip() + "..."
	return title[:max_length - 3].rstrip() + "..."

[cubic-dev-ai]

P2: Missing null/empty guard on summary before rendering. If the API returns successfully but with summary: null (or an empty string), the component will render the summary container with a header but no content. Add a check for !summary alongside the other early-return conditions.

Prompt for AI agents

Check if this issue is valid — if so, understand the root cause and fix it. At static/app/components/feedback/feedbackSummary.tsx, line 11:

<comment>Missing null/empty guard on `summary` before rendering. If the API returns successfully but with `summary: null` (or an empty string), the component will render the summary container with a header but no content. Add a check for `!summary` alongside the other early-return conditions.</comment>

<file context>
@@ -0,0 +1,64 @@
+import useOrganization from 'sentry/utils/useOrganization';
+
+export default function FeedbackSummary() {
+  const {isError, isPending, summary, tooFewFeedbacks} = useFeedbackSummary();
+
+  const organization = useOrganization();
</file context>

[cubic-dev-ai]

P2: Overly broad dependency: useMemo depends on selection but only uses selection.datetime. This causes unnecessary recomputations when unrelated parts of selection (e.g., environments) change.

Prompt for AI agents

Check if this issue is valid — if so, understand the root cause and fix it. At static/app/views/explore/hooks/useTraceItemAttributeKeys.tsx, line 37:

<comment>Overly broad dependency: `useMemo` depends on `selection` but only uses `selection.datetime`. This causes unnecessary recomputations when unrelated parts of `selection` (e.g., `environments`) change.</comment>

<file context>
@@ -36,76 +21,43 @@ export function useTraceItemAttributeKeys({
+      datetime: selection.datetime,
+      projectIds,
+    });
+  }, [selection, traceItemType, type, projectIds]);
+
+  const queryKey = useMemo(
</file context>

Suggested change

	}, [selection, traceItemType, type, projectIds]);
	}, [selection.datetime, traceItemType, type, projectIds]);

[cubic-dev-ai]

P2: Replacing FlexCenter with <Flex align="center"> drops overflow: hidden from the container. In a CSS grid with 1fr column, this means long attachment names will overflow instead of truncating with ellipsis. The flex container needs overflow: hidden or min-width: 0 so the child Name's text-overflow: ellipsis can take effect.

Prompt for AI agents

Check if this issue is valid — if so, understand the root cause and fix it. At static/app/components/events/eventAttachments.tsx, line 143:

<comment>Replacing `FlexCenter` with `<Flex align="center">` drops `overflow: hidden` from the container. In a CSS grid with `1fr` column, this means long attachment names will overflow instead of truncating with ellipsis. The flex container needs `overflow: hidden` or `min-width: 0` so the child `Name`'s `text-overflow: ellipsis` can take effect.</comment>

<file context>
@@ -139,9 +140,10 @@ function EventAttachmentsContent({
           {attachments.map(attachment => (
             <Fragment key={attachment.id}>
-              <FlexCenter>
+              <Flex align="center">
                 <Name>{attachment.name}</Name>
-              </FlexCenter>
</file context>

Suggested change

	<Flex align="center">
	<Flex align="center" style={{overflow: 'hidden'}}>

[cubic-dev-ai]

P3: Typo: "segements" should be "segments".

Prompt for AI agents

Check if this issue is valid — if so, understand the root cause and fix it. At src/sentry/replays/usecases/delete.py, line 88:

<comment>Typo: "segements" should be "segments".</comment>

<file context>
@@ -84,6 +84,11 @@ def _delete_if_exists(filename: str) -> None:
 
 def _make_recording_filenames(project_id: int, row: MatchedRow) -> list[str]:
+    # Null segment_ids can cause this to fail. If no segments were ingested then we can skip
+    # deleting the segements.
+    if row["max_segment_id"] is None:
+        return []
</file context>

Suggested change

	# deleting the segements.
	# deleting the segments.