ref(avatar) refactor avatar primitives#107609
Closed
JonasBa wants to merge 216 commits intojb/avatar/reorgfrom
Closed
ref(avatar) refactor avatar primitives#107609JonasBa wants to merge 216 commits intojb/avatar/reorgfrom
JonasBa wants to merge 216 commits intojb/avatar/reorgfrom
Conversation
This PR removes the "Metrics" option from cross-event querying, as we're removing this for initial EA launch. Ticket: EXP-739
…rator and operand (#107512) Before: <img width="873" height="334" alt="Screenshot 2026-02-03 at 1 33 02 PM" src="https://github.com/user-attachments/assets/25191b0e-07a2-4b1b-906a-96a32114e59b" /> After: <img width="886" height="446" alt="Screenshot 2026-02-03 at 1 32 28 PM" src="https://github.com/user-attachments/assets/6ea00c13-3324-4ead-929b-f21a45c8561a" /> --------- Co-authored-by: Abdullah Khan <abdullahkhan@PG9Y57YDXQ.local>
## Description Moving the CacheAccess class to `caches/cache_access.py` this will become the basis for interacting with the Workflow / Detector caches.
…sioning (#106941) Closes https://linear.app/getsentry/issue/BIL-1928/add-formatting-for-admin-provisioning ## Summary Adds unit formatting for emerge data categories (SIZE_ANALYSIS, INSTALLABLE_BUILD) in the admin provisioning UI: - Add `isEmergeCategory()` and `getCategoryUnitSuffix()` helper functions in `dataCategory.tsx` - Fix `shortenedUnitName` values: "upload" → "build", "distribution" → "install" - Update provision subscription modal to use new helper for all category unit suffixes - Add emerge category handling in gift events modal with proper unit labels ## Test plan - [x] Unit tests added for `isEmergeCategory()` and `getCategoryUnitSuffix()` - [ ] Verify admin provisioning modal shows "(in builds)" suffix for SIZE_ANALYSIS - [ ] Verify admin provisioning modal shows "(in installs)" suffix for INSTALLABLE_BUILD - [ ] Verify gift events modal shows correct unit in label/help for emerge categories
…rder (#107455) This PR addresses the `TypeError: Dict key must be str` occurring in the Amazon SQS data forwarder when attempting to serialize event payloads containing `datetime.datetime` objects in breadcrumbs. The root cause was identified as `orjson.dumps()` failing to serialize these datetime objects when only the `orjson.OPT_UTC_Z` option was provided. To resolve this, the `orjson.OPT_NON_STR_KEYS` flag has been added to the `orjson.dumps()` calls within the `AmazonSQSForwarder`. This change aligns the SQS forwarder's serialization logic with other parts of the codebase that handle complex data structures, ensuring proper serialization of datetime objects and preventing the `TypeError`. Closes SENTRY-5HY1 <!-- Sentry employees and contractors can delete or ignore the following. --> ### Legal Boilerplate Look, I get it. The entity doing business as "Sentry" was incorporated in the State of Delaware in 2015 as Functional Software, Inc. and is gonna need some rights from me in order to utilize my contributions in this here PR. So here's the deal: I retain all rights, title and interest in and to my contributions, and by keeping this boilerplate intact I confirm that Sentry can use, modify, copy, and redistribute my contributions, under Sentry's choice of terms. --------- Co-authored-by: sentry[bot] <39604003+sentry[bot]@users.noreply.github.com> Co-authored-by: getsantry[bot] <66042841+getsantry[bot]@users.noreply.github.com> Co-authored-by: Tony Xiao <txiao@sentry.io>
This adds trace metric alerts for the old (existing) alert flow. Some notable changes: - We have slightly different dropdowns for trace metric alert, specifically the trace metric name MUST be picked. - We'll show loading state until the list of metric names resolves so a metric name can be picked to fulfuill the above condition
Remove ids from the create workflow example body for triggers, add the logic type options for action filters, and set a correct logic type example.
Tours inject a div element to the DOM, which causes invalid HTML structure in something like the sidebar, which becomes ul > div > li as opposed to ul > li The diff is unfortunately quite large due to the indentation change, but the components outside of the sidebar have been wrapped with a div, resembling previous functionality, while the sidebar links have received the props directly to preserve their intended DOM structure. Fix DE-772
…107465) ## PR Details + Skips automation for issues in the triage signals flow (>= 10 events path) if first_seen is more than 2 weeks ago + This prevents a flood of automation runs on historical issues when an org enables the new pricing + Summaries are still generated for old issues (< 10 events path is unaffected) - only automation is gated + **Known edge case: If an issue is created today but takes more than 14 days (2 weeks) to reach 10 events, it will also be skipped. I assume this is rare and acceptable for now.** --------- Co-authored-by: Rohan Agarwal <47861399+roaga@users.noreply.github.com>
- Gonna look into fixing the search separately - But even for convenience having both of these things on the same attribute is a pain - debug -> whether debug was enabled - debug_msg -> the message that was passed
If we run into this edge case where we can't find the sentry app installation, we've decided we want to delete the action because users can't see it or update it since the app has been uninstalled. See [this doc](https://www.notion.so/sentry/Invalid-Sentry-App-Actions-2f88b10e4b5d8046ac20faae3cfeb7c5) for more context on that decision. I will need to re-run [the migration](#107208) because I missed that we were dual writing actions with the installation uuid (fixed in #107436) and we may as well delete the orphaned actions in the same step rather that run a separate migration to do so. It will not be run until #107460 is merged.
… opening in explore (#107503) ~- When clicking `Open in Explore` from the Score breakdown chart in prebuilt Web Vitals dashboard, transforms the `performance_score(...)` yaxes to use equation format (ie `equation|performance_score(...)`).~ - Updates performance score breakdown chart to use `equation|` format. - This is necessary because the `performance_score` function is not a selectable yaxis aggregate in Explore (to avoid cluttering the dropdown). `performance_score` is however valid in equations. - Updates `isPerformanceScoreBreakdownChart` and `transformPerformanceScoreBreakdownSeries` to also handle `equation|` format - Also removes the 3 yaxes limit from `Open in Explore`. - This is necessary because some charts such as the score breakdown chart have more than 3 series
…ashboards to allow creating dashboard filters on multiple datasets (#107434) When navigating using dashboard links, we currently only apply a single global dashboard filter on a single dataset (the widget dataset). This is an issue because some dashboards use multiple datasets for a filterable field like `transaction`. Updates `linkedDashboards` to allow an `additionalDatasets` attribute to specify what datasets to also apply the global filter to.
…07030) Create a data condition to check if the open period for the issue is younger / older than X amount of time.
#107546) - Sending in as an object, causes snake -> came case convertions for keys, which we don't want for our usecase Co-authored-by: Abdullah Khan <abdullahkhan@PG9Y57YDXQ.local>
Replaces object-based display type definitions with an explicit array to make dropdown ordering safer and more maintainable. Object key ordering in JS is technically specified (ES2015+) but easy to break accidentally during refactoring, and it's annoying to inject new values into. Using an array makes the ordering intent explicit and safer for inserting new display types at specific positions. Also it removed a bunch of awkward TS ignores.
- We can't timeout the request midrun easily, so lets process it regardless so there's at least a payoff for making the user wait those extra seconds - Add number of iterations to the log message & as a metric
…ata` (#107447) Preparation for upcoming work on categorical bar charts. Categorical bars are a chart but they do _not_ use time series data, they use table data! So, `isChartDisplayType` isn't a helpful name anymore. Renames the `isChartDisplayType` helper function to `usesTimeSeriesData` to better reflect its actual purpose: determining whether a display type uses time-series data (events-stats endpoint) and yAxis state, versus table-style data (events endpoint) and fields state. This makes the conditionals a little easier to follow when making changes. Also added a few comments and extracted some conditionals to explain _why_ certain things are happening.
Fix an incorrect cast, resolving an apparent cross-type equality check.
- make the header larger - Remove unused props, remove spreading modal props into the form - rename input to "Team slug" to match team settings. Team name no longer really exists. before <img width="617" height="322" alt="image" src="https://github.com/user-attachments/assets/dc413340-d7ed-4fb8-8525-7b4071a1e726" /> after <img width="613" height="299" alt="image" src="https://github.com/user-attachments/assets/fb081668-f82b-4d8d-962e-e312440bfc49" />
- There's an option `performance.traces.transaction_query_timebuffer_days` which is how much of an offset we do to the ts passed to the trace endpoint, I wanna see how often we're querying more days of data than needed
Closes EME-764
JonasBa
force-pushed
the
jb/avatar/refactor
branch
from
bbde8d0 to
2e648ea
Compare
JonasBa
force-pushed
the
jb/avatar/refactor
branch
from
0a4f454 to
331d80f
Compare
JonasBa
force-pushed
the
jb/avatar/refactor
branch
from
40eb67e to
f597fbd
Compare
…107460) Update the sentry app installation deletion receiver to disable `Action`s using either the sentry app id or the sentry app installation uuid. Previously the sentry app installation deletion receiver was failing to update `Actions` that use the sentry app id (which is any single written `Action` or any `Action` that was dual written from a metric alert). <!-- BUGBOT_STATUS --><sup><a href="https://cursor.com/dashboard?tab=bugbot">Cursor Bugbot</a> reviewed your changes and found no issues for commit <u>63ffa9a</u></sup><!-- /BUGBOT_STATUS -->
JonasBa
force-pushed
the
jb/avatar/refactor
branch
from
f597fbd to
e8a0c03
Compare
ref(baseAvatar) inline avatar address comments streamline hook call and reset ref(avatar) resolve error handling avatar: add new primitives avatar: add new primitives avatar: improve types ref(avatar) wip test(avatar): rewrite LetterAvatar tests to focus on public API Rewrote tests to focus solely on the component's public behavior: - Initials rendering for various name formats - Fallback to question mark when name is empty or whitespace - Proper handling of multibyte characters, emails, and edge cases Removed tests for internal implementation details like styling and refs. test(avatar): add ImageAvatar tests for public API Added comprehensive tests for ImageAvatar component covering: - Image rendering when src is valid - Fallback to LetterAvatar when src is missing or image fails to load - Proper handling of src changes and error state reset - Props passing to both image and fallback components All tests focus on observable behavior rather than implementation details. test(avatar): enhance Gravatar tests to cover public API Enhanced Gravatar tests to cover: - SHA-256 hashing of gravatarId and URL generation - Remote size and d=404 parameters in URL - Fallback to LetterAvatar for empty/whitespace gravatarId - Props passing to both gravatar image and fallback - Hash regeneration when gravatarId changes - Switching between valid and empty gravatarId All tests focus on observable behavior and public API. fix(avatar): fix ImageAvatar fallback not triggering Fixed issues that prevented the LetterAvatar fallback from displaying when image load fails: 1. Properly destructure `name` from props before spreading to prevent it from being passed to the Image component 2. Remove incorrect `alt` prop being passed from Avatar component to ImageAvatar (ImageAvatarProps explicitly omits 'alt') These changes ensure the onError handler from mergeProps is properly attached without prop conflicts, allowing the error state to trigger and display the LetterAvatar fallback when an image fails to load. fix avatar correct usage correct usage correct usage use fixtures revert entire html interface update avatar chooser
JonasBa
force-pushed
the
jb/avatar/refactor
branch
from
e8a0c03 to
142877c
Compare
…roker (#107729) ## Summary Tests produce real Kafka messages via the taskworker pipeline, causing stale tasks to accumulate in the taskbroker's SQLite queue. Running 20 tests produced **266 tasks** (relay config invalidations, Slack notifications, code owners updates, spike projections). Over multiple test sessions, these accumulate into thousands. ### Root Cause Three factors combine to produce real Kafka messages during tests: 1. **`simulate_on_commit`** (autouse fixture) fires `on_commit` callbacks during tests 2. **Django signal handlers** queue tasks via `task.delay()` on model `post_save` 3. **`TASKWORKER_ALWAYS_EAGER=False`** sends those tasks to real Kafka at `127.0.0.1:9092` ### Fix Patch `TaskNamespace.send_task` as a no-op at session start via `pytest_sessionstart`. This is surgical — it blocks only the Kafka production step while preserving: - `_signal_send` hooks (`BurstTaskRunner`, `stale_database_reads`) - Serialization validation (`create_activation()` still runs) - `TaskRunner` (uses `ALWAYS_EAGER`, bypasses `send_task` entirely) - `BurstTaskRunner` (captures at `_signal_send`, before `send_task`) The 4 tests in `test_registry.py` that directly test `send_task` behavior use a `real_send_task` fixture to restore the original method. ### Verification Ran `tests/getsentry/tasks/test_quota_exceeded_notification.py` (20 tests) before and after, with taskbroker running but no taskworker draining the queue: | Metric | Before Fix | After Fix | |---|---|---| | Tests passing | 20/20 | 20/20 | | Baseline tasks (devserver scheduler) | 7 | 8 | | Tasks after test run | **273** (7 + 266 from tests) | **8** (0 from tests) | | Test-originated tasks | **266** | **0** | Before-fix breakdown (266 test-originated tasks from 20 tests): - 151 `invalidate_project_config` - 45 `update_code_owners_schema` - 45 `new_organization_notify` (Slack) - 25 `run_spike_projection` Additional regression tests all pass: - `tests/sentry/taskworker/test_registry.py` — 15/15 (including 4 with `real_send_task` fixture) - `tests/sentry/taskworker/` (full suite) — 177/177 - `tests/getsentry/consumers/test_outcomes_consumer.py` — 80/80 (`BurstTaskRunner` works correctly) ## Test plan - [x] Verified 266 tasks accumulate in taskbroker from 20 tests (before fix) - [x] Verified 0 test-originated tasks after fix - [x] All 177 taskworker tests pass - [x] BurstTaskRunner tests pass (80/80) - [x] Pre-commit hooks pass
We care about Rule also, and this will allow us to maintain a traffic and endpoint-based burndown tracker.
github-actions
bot
added
the
Scope: Backend
Contributor
|
🚨 Warning: This pull request contains Frontend and Backend changes! It's discouraged to make changes to Sentry's Frontend and Backend in a single pull request. The Frontend and Backend are not atomically deployed. If the changes are interdependent of each other, they must be separated into two pull requests and be made forward or backwards compatible, such that the Backend or Frontend can be safely deployed independently. Have questions? Please ask in the |
The product is Root Cause Analysis now, not Autofix anymore | Where | Before | After | | --- | --- | --- | | Across all the settings, projects, repos lists "Autofix" link renamed | <img width="628" height="167" alt="SCR-20260206-lkdw" src="https://github.com/user-attachments/assets/3c6e3b91-4208-4129-b440-a151529f73e0" /> | <img width="627" height="174" alt="SCR-20260206-lkeu" src="https://github.com/user-attachments/assets/4917f722-922e-4489-ab72-28b2f8fdd355" /> | Main seer settings, "Enable Autofix PR Creation by Default" renamed | <img width="564" height="256" alt="SCR-20260206-lkpx" src="https://github.com/user-attachments/assets/5081b1d3-ac19-4cfd-9c0e-03abd6963c8d" /> | <img width="556" height="257" alt="SCR-20260206-lkqu" src="https://github.com/user-attachments/assets/265b85f0-fad9-4355-b08d-e04f8c31f2ec" /> | Project List headers | <img width="915" height="192" alt="SCR-20260206-llfs" src="https://github.com/user-attachments/assets/8beceec3-2d77-4d60-9544-bef5982f5b71" /> | <img width="916" height="194" alt="SCR-20260206-lldz" src="https://github.com/user-attachments/assets/5cee355e-1156-4fdf-aff0-72439641eb4f" /> | Repo list header order | <img width="925" height="127" alt="SCR-20260206-llra" src="https://github.com/user-attachments/assets/af221517-b8a3-4aaa-aa7c-7e46e6aa6186" /> | <img width="910" height="120" alt="SCR-20260206-llrz" src="https://github.com/user-attachments/assets/932b986c-ac61-46b1-9463-784722a2821e" />
Updated test expectations to match the new avatar behavior where initials are generated from display names instead of slugs. Changes: - Team avatar: Expected initials "TS" → "TN" (from "Team Name" not "team-slug") - Organization avatar: Expected initials "OS" → "ON" (from "Organization Name" not "org-slug") - Changed test ID from "default-avatar" → "letter_avatar-avatar" to match new Avatar component behavior This is the correct behavior - avatars should use human-readable names for initials rather than technical slugs.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
20 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
todo: