* enforce License.IsSeatCountEnforced if set
If a license sets `IsSeatCountEnforced`, enforce the user limit therein
as a hard cap.
Fixes: https://mattermost.atlassian.net/browse/CLD-9260
* remove duplicate tests
* Improve user limit error messages and display
- Add separate error messages for licensed vs unlicensed servers
- Licensed servers: "Server exceeds maximum licensed users. ERROR_LICENSED_USERS_LIMITS"
- Unlicensed servers: "Server exceeds safe user limit. ERROR_SAFETY_LIMITS_EXCEEDED"
- Remove redundant "Contact administrator" text from activation errors shown to admins
- Fix system console to display actual server error messages instead of generic "Failed to activate user"
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
* Add license nil check and test coverage
- Add license != nil check in GetServerLimits to prevent panic
- Add test case to verify graceful handling of license being set to nil
- Ensures fallback to hard-coded limits when license becomes nil
Co-authored-by: lieut-data <lieut-data@users.noreply.github.com>
* Fix user limits tests to expect license-specific error IDs
Update test expectations to use the new license-specific error IDs:
- app.user.update_active.license_user_limit.exceeded for licensed server user activation
- api.user.create_user.license_user_limits.exceeded for licensed server user creation
Also update frontend to show actual server error messages instead of generic ones in system console.
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
* Remove redundant license nil test
The test couldn't meaningfully verify nil license behavior since it relied on
hard-coded constants that can't be modified in the test.
Co-authored-by: lieut-data <lieut-data@users.noreply.github.com>
* Fix whitespace issue in limits_test.go
Remove unnecessary trailing newline to pass style checks.
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
* updated i18n
* s/ERROR_LICENSED_USERS_LIMITS/ERROR_LICENSED_USERS_LIMIT_EXCEEDED/, expand warning log
* Add 5% grace period for licensed user limits
- Add calculateGraceLimit() function with 5% or +1 minimum grace
- Apply grace period only to licensed servers with seat count enforcement
- Handle zero user licenses by returning zero grace limit
- Add comprehensive test coverage for grace period scenarios
- Unlicensed servers maintain existing hard-coded limits without grace
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
* Fix TestCreateUserOrGuestSeatCountEnforcement to account for 5% grace period
The test was failing because it expected user creation to fail at exactly
the license limit, but the implementation now includes a 5% grace period
before enforcement kicks in.
Changes:
- Update test cases to create users up to the grace limit (6 for a 5-user license)
- Add comments explaining the grace period calculation
- Both regular user and guest user creation tests now properly validate
enforcement at the grace limit rather than the base license limit
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
* Fix TestUpdateActiveWithUserLimits to account for 5% grace period
Update test expectations to match the new grace period behavior:
- At base limit (100) but below grace limit (105): should succeed
- At grace limit (105): should fail
- Above grace limit (106): should fail
This aligns the tests with the license enforcement implementation
that includes a 5% grace period above the licensed user count.
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
---------
Co-authored-by: Claude <noreply@anthropic.com>
Co-authored-by: claude[bot] <209825114+claude[bot]@users.noreply.github.com>
Co-authored-by: lieut-data <lieut-data@users.noreply.github.com>
Co-authored-by: Mattermost Build <build@mattermost.com>
* Fix join private channel not showing on all needed scenarios
* Fix for other two instances of the logic
* Fix test
---------
Co-authored-by: Mattermost Build <build@mattermost.com>
* Adds support for GMs in shared channels
* Fix linter
* Remove creatorID from slack call
---------
Co-authored-by: Miguel de la Cruz <miguel@ctrlz.es>
* [MM-63041] Convert many inputs to the Input component, replace clientError with more correct client-side validation that conforms to the input
* Fix line length
* PR feedback
---------
Co-authored-by: Mattermost Build <build@mattermost.com>
* Upgrade Go to 1.24.3
Updates the following files:
- server/.go-version: 1.23.9 → 1.24.3
- server/build/Dockerfile.buildenv: golang:1.23.9-bullseye → golang:1.24.3-bullseye
- server/go.mod: go 1.23.0 → go 1.24.3, toolchain go1.23.9 → go1.24.3
- server/public/go.mod: go 1.23.0 → go 1.24.3, toolchain go1.23.9 → go1.24.3
Also fixes non-constant format string errors introduced by Go 1.24.3's stricter format string checking:
- Added response() helper function in slashcommands/util.go for simple string responses
- Removed unused responsef() function from slashcommands/util.go
- Replaced responsef() with response() for translated strings that don't need formatting
- Fixed fmt.Errorf and fmt.Fprintf calls to use proper format verbs instead of string concatenation
- Updated marketplace buildURL to handle format strings conditionally
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
* Update generated mocks for Go 1.24.3
Regenerated mocks using mockery v2.53.4 to ensure compatibility with Go 1.24.3.
This addresses mock generation failures that occurred with the Go upgrade.
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
* Update to bookworm and fix non-existent sha
Signed-off-by: Stavros Foteinopoulos <stafot@gmail.com>
* fix non-constant format string
---------
Signed-off-by: Stavros Foteinopoulos <stafot@gmail.com>
Co-authored-by: Claude <noreply@anthropic.com>
Co-authored-by: Stavros Foteinopoulos <stafot@gmail.com>
* Remove aria-live from SuggestionBox
aria-live isn't the idiomatic way to make the autocomplete accessible to
screen readers. Instead, we should've used aria-activedescendant which
was done for the at-mention autocomplete in a previous ticket, but that
didn't apply to other types of autocompletes. That lead to the
at-mention autocomplete being too noisy (as there were two different
things telling the user what the results were), and it meant that other
types of autocompletes didn't function.
This needs a couple direct followups:
1. The E2E tests need to be updated since they test for aria-live.
2. The Suggestion items for other types of autocompletes need IDs for
the aria-activedescendant to work.
* Consistently set IDs for all SuggestionList items
Instead of leaving it up to the individual Suggestion components,
this'll ensure that a11y support works for all of them going forward as
long as they properly forward other props to the underlying li element.
I would've preferred if each instance of SuggestionList had unique IDs,
but they currently all use the ID suggestionList, and I didn't want to
update that ID across 55 different Cypress tests. There's should only be
a single SuggestionList visible at a time, so the current situation is
fine enough.
* Remove textboxId from AtMentionProvider
* Change suggestion list to scroll using IDs instead of findDOMNode and refs
* Add an aria-label to SuggestionList
* Make all Suggestion components use the option role
* Add number of results to SuggestionList readout
* Change SuggestionBox to only set aria-expanded with results
* Add tests for ARIA of SuggestionBox
* Address feedback
