* feat: fix Turborepo caching for app-store CLI build
- Fix incorrect output paths from app-store/ to packages/app-store/
- Fix incorrect input paths to match actual file locations
- Remove cache: false to enable caching
- Add all 11 generated files to outputs array
- Add NEXT_PUBLIC_IS_E2E env var for proper cache invalidation
- Add specific input patterns to avoid cache invalidation from generated files
- Improves local dev experience with cache hits vs ~3.1s cache misses
Performance improvement: builds go from ~3.1 seconds (cache miss) to
instant cache hits when inputs haven't changed.
Co-Authored-By: keith@cal.com <keithwillcode@gmail.com>
* Use new turbo command so it uses cache
* Made the output more flexible
* Added back a .tsx output
* fixed file path
* Removed global IS_E2E var that Devin added
---------
Co-authored-by: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* PBAC on team members page + org members page to use specific permissions with fallbacks
* Implement checks on edit sheet fetch to check for permisisons
* WIP permission fixes
* fix tests
* WIP
* fix edit mode perms
* fix remove and list view permissions
* fix add
* feat: add pbac permissions for impersonation (#23035)
* registery + migration + i18n
* UI for can impersonate
* impersonation backend
* remove permission from db query
* add missing prisma import
* add resend invitation and edit mode checks for delete/changeRole
* fix canChangeRole
* canChangeRole gate
* fix permission logic
* fix depends on
* push migration
* fix migration
* fix scoped advanced permissions
* use listMembers instead of invite proxy
* remove isOrgAdmin check to rely on pbac
* wait for session to load before navigating
* use event-types testId to ensure session loaded
* use profile page instead of eventTypes
* use correct roles
* fix wait for session
* correct orgs private isFallBackRoles
* set a timeout on waiting for session
* remove adminOrOwner check since its in fallpack
---------
Co-authored-by: Volnei Munhoz <volnei.munhoz@gmail.com>
Co-authored-by: Eunjae Lee <hey@eunjae.dev>
- Change workflow name from 'Atoms E2E Tests' to 'E2E Atoms'
- Update job name to match new convention
- Update step name for consistency
- Aligns with naming pattern used by other E2E workflows
Co-authored-by: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* fix: prevent non-E2E labels from triggering workflow runs
- Add conditional logic to all jobs to skip when non-ready-for-e2e labels are added
- Only allow ready-for-e2e label additions to trigger workflows
- Preserve all other trigger types (opened, synchronize, reopened)
- Fixes issue where labels like 'Improvements' cause unnecessary CI runs
Resolves the annoyance where adding labels like 'Improvements' to PRs
triggers full workflow runs including all checks, when only the
ready-for-e2e label should trigger E2E tests and associated builds.
Co-Authored-By: keith@cal.com <keithwillcode@gmail.com>
* refactor: consolidate workflow conditional logic into environment variable
- Replace repeated conditional logic with single SKIP_WORKFLOW env variable
- Simplify all job conditions from complex expression to env.SKIP_WORKFLOW != 'true'
- Maintain exact same functional behavior while improving maintainability
- Reduces code duplication across 20+ jobs in the workflow
This addresses feedback to consolidate the repeated conditional logic:
github.event.action != 'labeled' || github.event.label.name == 'ready-for-e2e'
The refactored approach is cleaner and easier to maintain.
Co-Authored-By: keith@cal.com <keithwillcode@gmail.com>
* fix: update required job logic to handle skipped jobs correctly
- Remove 'skipped' from failure conditions in required job
- Add SKIP_WORKFLOW condition to required job logic
- Prevents false failures when E2E jobs legitimately skip
- Maintains proper failure detection for actual job failures
This fixes the CI failure where the required job was treating
legitimately skipped E2E jobs as failures when no ready-for-e2e
label is present.
Co-Authored-By: keith@cal.com <keithwillcode@gmail.com>
---------
Co-authored-by: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* Fix vitest on prod
* Also extract out verify-email for the purpose of stopping test bundling
* Address NIT
* Fix lingo.dev imports
---------
Co-authored-by: Alex van Andel <me@alexvanandel.com>
* fix(navigation): make child links clickable on tab widths
* fix(navigation): highlight only active child route & remove underline animation
* fix(navigation): onclick single open tooltip and Tailwind color consistency for theme support
* Update NavigationItem.tsx
* Update Navigation.tsx
---------
Co-authored-by: Devanshu Sharma <devanshusharma658@gmail.com>
Co-authored-by: Anik Dhabal Babu <81948346+anikdhabal@users.noreply.github.com>
* refactor: separate task creation from processing to eliminate compilation overhead
- Create TaskProcessor class to handle task processing logic
- Move processQueue and cleanup methods to TaskProcessor
- Remove task handler imports from InternalTasker creation path
- Eliminate 2.5-3s compilation overhead by only loading task handlers during processing
- Maintain existing tasker.create() API for all callers
- Use composition pattern for clean separation of concerns
Co-Authored-By: keith@cal.com <keithwillcode@gmail.com>
refactor: completely separate TaskProcessor from InternalTasker
- Remove processQueue() and cleanup() methods from Tasker interface
- Eliminate TaskProcessor dependency from InternalTasker class
- Update API endpoints (cron.ts, cleanup.ts) to use TaskProcessor directly
- Update README documentation to show correct TaskProcessor usage
- Complete separation ensures task creation no longer imports task handlers
- Eliminates 2.5-3s compilation overhead while preserving all functionality
Co-Authored-By: keith@cal.com <keithwillcode@gmail.com>
* Moved the cleanup method back to internal tasker
---------
Co-authored-by: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* feat: allow organization owners access to shared billing portal
- Modify stripepayment portal endpoint to accept teamId parameter
- Use team's subscription ID to get customer ID from Stripe subscription
- Validate user permissions before allowing team billing access
- Update platform and regular billing views to pass teamId
- Maintain backward compatibility with individual user billing
Co-Authored-By: joe@cal.com <j.auyeung419@gmail.com>
* Add `getTeamByIdIfUserIsAdmin` to `TeamRepository`
* Add `getSubscriptionFromId` function
* Generate portal URL via the team's subscription not the requesting
user's customerId
* Pass teamId to `/portal` endpoint
* Undo Devin change to platform billing
* Address coderabbit comments
---------
Co-authored-by: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* chore: remove .yarn/ci-cache from git and ignore it
* Remove additional install-state.gz cache files
* Small fixes to the .gitignore
---------
Co-authored-by: Matthew Cengia <mattcen@mattcen.com>
Co-authored-by: Keith Williams <keithwillcode@gmail.com>
- Remove packages/features/index.ts barrel file that only exported ee namespace
- No imports were found using this barrel file - all imports use specific module paths
- This improves performance by eliminating unnecessary exports from ee folder
- Type checking passes confirming no breaking changes
Co-authored-by: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
emrysal