TEST REQUESTFEEDBACK 143 UI Case

TEST-REQUESTFEEDBACK-143 - Loading, empty, failure and canonical retry feedback

Linked specification for the matching authoritative QBrain TEST record. Persistence is reported by receipt.json. Current implementation evidence is in implementation-receipt.json; the original documentation-only snapshot is retained under archive/design-20261027/.

Field	Specification
Objective	Verify Loading, empty, failure and canonical retry feedback.
QBrain reference/scope	TEST-REQUESTFEEDBACK-143; Word Pairs / PROD-WORDPAIRS; workspace <home>/github/WordGame/.mcpServer/wordpairs; layer layer-1
Source revision	Original designs at 435be05abac5310172cfb9a9290c9d7ea0d1cbde; exact dirty prose/decision hashes in sources.json. New UI-DEC decisions derive from the operator instruction to fill gaps.
Source links	decisions.md
Frame/element identifiers	GAP-FLOW-REQUESTFEEDBACK, GAP-TRANS-REQUESTFEEDBACK-01, GAP-TRANS-REQUESTFEEDBACK-02, GAP-TRANS-REQUESTFEEDBACK-03, GAP-TRANS-REQUESTFEEDBACK-04; exact SVG anchors are listed in coverage.md
FR/TR → specific AC references	FR-REQUESTFEEDBACK-030#UI-FEEDBACK-01; FR-REQUESTFEEDBACK-030#UI-FEEDBACK-02; FR-REQUESTFEEDBACK-030#UI-FEEDBACK-03; FR-REQUESTFEEDBACK-030#UI-FEEDBACK-04; FR-REQUESTFEEDBACK-030#UI-FEEDBACK-05
Priority	high
Platform/viewport	Phone Chromium/Android and Safari/iPhone where applicable; project matrix320x568,390x844,844x390;1280px at400%zoom and200%text enlargement. OS-native installation is a separate physical-device dependency.
Preconditions/roles	An isolated fixture controls pending reads, empty results, denied reads and mutation outcomes for Alice.
Test data	Draft Space; command C-1 with fixed payload; outcomes pending, unavailable, committed-response-lost, conflict, forbidden.
Synchronization	Await the explicit observable conditions and source-defined deadlines below; no fixed sleeps. Harness timeouts do not create a product response-time SLA.
Cleanup	Close case-owned contexts and reset only isolated fixture accounts, rooms or content, including after failure. Preserve shared catalog and real profile data.
Automation readiness	Automated isolated browser integration and supporting application/SQL checks executed; operational boundaries remain separate.
Basis	Specified by cited designs/QBrain; new gap behavior is specified by UI-DEC-01..08. Synthetic fixture identities/data are test configurations.
Execution	Executed applicable UI-DEC variants in Chromium with real React and isolated HTTP/application/domain fixtures; real disposable PostgreSQL covers persistence/audit. See execution receipt.

Ordered actions and observable results

Step	Action / condition	Observable expected result
1	Hold a read pending, then resolve empty and unavailable in separate examples.	Accessible loading feedback changes to distinct empty/allowed-action or unavailable/Retry states without a false success.
2	Hold a form submission pending and attempt another submit; then fail it.	Busy feedback prevents duplicate submission; the safe draft remains and a user-safe error/valid Retry appears.
3	Retry a committed-response-lost C-1, then exercise conflict in another fixture.	The uncertain retry sends identical C-1/payload and displays one committed result; conflict refreshes state without silently replaying a changed command.
4	Deny authorization while an older read is pending, then complete that older read.	Privileged records/drafts clear and the older response cannot restore them; no raw errors or credentials are shown.

Scenario

Feature: Loading, empty, failure and canonical retry feedback
  Scenario: TEST-REQUESTFEEDBACK-143 design verification
    Given An isolated fixture controls pending reads, empty results, denied reads and mutation outcomes for Alice.
    And the fixture data is Draft Space; command C-1 with fixed payload; outcomes pending, unavailable, committed-response-lost, conflict, forbidden.
    When Hold a read pending, then resolve empty and unavailable in separate examples.
    Then Accessible loading feedback changes to distinct empty/allowed-action or unavailable/Retry states without a false success.
    When Hold a form submission pending and attempt another submit; then fail it.
    Then Busy feedback prevents duplicate submission; the safe draft remains and a user-safe error/valid Retry appears.
    When Retry a committed-response-lost C-1, then exercise conflict in another fixture.
    Then The uncertain retry sends identical C-1/payload and displays one committed result; conflict refreshes state without silently replaying a changed command.
    When Deny authorization while an older read is pending, then complete that older read.
    Then Privileged records/drafts clear and the older response cannot restore them; no raw errors or credentials are shown.

Boundaries

Specified gap behavior is implemented and verified within the execution receipt's isolated UI/application/SQL scope. The original design-only snapshot remains historical.

Criterion references identify intended verification. Non-UI transaction, distribution, release/deployment and native OS guarantees retain separate technical/operational tests; this document is not a claim that a compound criterion passed.

Implementation evidence

Implementation snapshot and tiered results, hashed source inventory, and raw execution logs bind this case to the current tested implementation. Browser journeys use actual pages and real application/domain services behind isolated HTTP/persistence/time ports. They do not establish physical screen-reader behavior, native phone install, real OIDC, deployed availability or real private-push2..25UI delivery. Reflow at400% is the documented320CSS-pixel equivalent, not a physical browser-zoom certification.

Final hostile-review regression: malformed200room GET and POST bodies containing a private sentinel render only controlled room errors; the sentinel/parser/network text is absent and explicit Refresh restores authoritative room access. Both paths are included in the final61-browser run.