[P0] Align public pagination regression assertions with interactive-lift

[alexeygrigorev]

Product outcome

Restore the public pagination regression contract so the scheduled/full Django gate reflects the accepted shared pagination markup on current main. This is a test-contract correction only: the rendered paginator already intentionally carries the accepted interactive-lift design-system token.

Evidence and ownership boundary

• On current origin/main face8e4808d65afbf0374d1ced7a88079950d663, content/tests/test_public_pagination.py:423-424 still counts the obsolete exact strings:
  • class="filter-pill pagination-number"
  • class="filter-pill pagination-step"
• templates/public/_pagination.html:46,52,62 emits the accepted classes with interactive-lift appended.
• The token was introduced by the accepted a5d08f1 (“Unify clickable card and control interactions”), not by a paginator behavior change.
• Scheduled run 33513872199 and the integrated v3 Django evidence reproduce the failure; the first failing assertion is line 423.
• The collection-hub/archive class correction remains in Restore face8e4 formatting, typing, and focused content gates #261, whose groomed scope explicitly limits its focused test changes to test_archive_row.py and test_collection_hub.py. Keep failed structured-component evidence schema-valid #280 is CI evidence orchestration and explicitly excludes product/test pagination changes. This issue therefore owns only the separate test_public_pagination.py stale expectations.
• Shared paginator behavior and its accepted design-system contract are defined by closed Create one shared public catalogue pagination primitive #178.

Scope

Change only content/tests/test_public_pagination.py:

1. Update the numeric-control count to match class="filter-pill pagination-number interactive-lift" and retain the expected count of 7.
2. Update the direction-control count to match class="filter-pill pagination-step interactive-lift" and retain the expected count of 2.
3. Keep every other pagination assertion unchanged, including control counts, labels, aria-current, arrows, page-window/ellipsis behavior, URLs, canonical/SEO/cache semantics, boundary behavior, empty/single-page behavior, and source/design-system guards.

Acceptance criteria

• The candidate changes no production template, CSS, Python application code, fixture data, browser harness, generated projection, or Restore face8e4 formatting, typing, and focused content gates #261/Keep failed structured-component evidence schema-valid #280-owned path.
• The two rendered-class assertions recognize the accepted interactive-lift token and still prove exactly 7 numeric controls and 2 direction controls on the 12-page middle state.
• No assertion is weakened by removing the control class, reducing the expected count, using a broad substring that can match unrelated markup, or changing the paginator to satisfy the test.
• content.tests.test_public_pagination passes on the frozen candidate, including all existing query grammar, status, canonical, cache, accessibility, URL-window, boundary, and source-contract cases.
• The engineer runs the focused module plus make format-check, make lint, and the applicable versioned verification checks; an independent tester recomputes the plan and verifies the same frozen candidate. Final full-Django/full-regression evidence must be green after the separately owned Reconcile non-reproducible public projection after face8e4 #253/Restore face8e4 formatting, typing, and focused content gates #261 and other release blockers are repaired; this issue never waives a red gate.
• Screenshots are not_applicable: only a regression test file changes and no rendered product input changes.

Non-goals

• No paginator implementation, template, CSS, accessibility, URL, SEO, copy, or visual change.
• No archive-row/collection-hub class work, normalization/type work, formatter repair, or structured-evidence workflow work from Restore face8e4 formatting, typing, and focused content gates #261/Keep failed structured-component evidence schema-valid #280.
• No test deletion, skip, xfail, quarantine, timeout/retry change, assertion weakening, or broad test refactor.
• No commit, push, merge, deployment, or workflow rerun as implementation evidence before engineer/tester/PM gates.

Dependencies and handoff

No implementation dependency. This is a small independent testing/frontend lane and may be implemented without waiting for #261 because the changed file does not overlap #261's allowed paths. Final release verification remains subject to the common green-base/order recorded in #253/#261 and the release coordinator.

The engineer works uncommitted in an isolated worktree, posts the versioned handoff, and freezes it. A separate tester performs the focused and graph-selected checks. PM accepts only from a passing tester envelope; the orchestrator then owns commit/merge/push under _docs/PROCESS.md.

[alexeygrigorev]

PM grooming complete — #300

Disposition: new narrow P0 testing/frontend issue; no duplicate existing contract.

The stale expectations are limited to content/tests/test_public_pagination.py:423-424. The accepted shared paginator in templates/public/_pagination.html now emits interactive-lift on both numeric and direction controls. The issue body freezes the two count-preserving assertion updates and explicitly prohibits production/template changes or weakened checks.

Engineer lane: frontend/testing, test-only, isolated and uncommitted. It may proceed independently of #261 because there is no path overlap. Independent tester and PM gates remain required; final release evidence must still be green on the repaired release base. Screenshots are not applicable.

Related ownership: #178 is the accepted shared paginator contract; #261 owns the separate archive-row/collection-hub and gate-recovery work; #280 owns CI evidence orchestration only.

[alexeygrigorev]

Software Engineer Report — #300

Disposition: implementation complete and frozen for independent testing. The candidate is
test-only and uncommitted. It updates the two stale exact class-count expectations to include the
accepted interactive-lift token while preserving counts and all other pagination assertions.

Frozen candidate

• Worktree: /home/alexey/git/dtc-website/.tmp/issue-300-engineer
• Base/head: face8e4808d65afbf0374d1ced7a88079950d663 / the same commit (dirty worktree)
• Source mode: worktree; no commit, push, merge, PR, deployment, or issue closure
• Changed path: content/tests/test_public_pagination.py only
• Candidate diff SHA-256 (path + NUL + binary diff + NUL):
  1595276c68a6fab2c94fcc8729449fbe1063143d31365b2c48bb422a2ae5b4ab
• Source-tree manifest: 9b145fef5bb3290d51ba7a0e58b434599bf871348e73a4bc858a72ba902aee08
• git diff --check: passed; worktree contains no unrelated tracked changes

Implementation

• Numeric-control assertion now matches the exact class="filter-pill pagination-number interactive-lift" string and still requires exactly 7 controls.
• Direction-control assertion now matches the exact class="filter-pill pagination-step interactive-lift" string and still requires exactly 2 controls.
• The shared template, CSS, pagination implementation, URLs, labels, accessibility/SEO/cache/boundary assertions, fixtures, browser harness, generated data, and Restore face8e4 formatting, typing, and focused content gates #261/Keep failed structured-component evidence schema-valid #280-owned paths are unchanged.

Versioned verification plan

• Plan: .tmp/verification/issue-300-engineer/verification-plan.json
• Plan SHA-256: 9145cefb84e9cb93ea7bc7eeb18956587a078fb591ffd31411b4def66573c326
• Graph SHA-256: 05ffeee99244c9becd5e98e40be08188f1c5e2a61204daece984c3bebd1bce6
• Policy version: 5
• Profile/reason: full / test_infrastructure; browser profile smoke
• Direct node: app.content; downstream nodes: django.accounts, django.api, django.content, django.content_sync, django.core
• Risk flags: test_infrastructure; render impact: false; screenshots: not_applicable
• Plan buckets: rerun selector, quality, Django, Playwright smoke, container, and evidence validation; reused none; skipped none; not_applicable compatibility, content invariants, and screenshots.

The plan includes the uncommitted worktree change and binds the exact base/head above. The
graph-selected full runner was not started because an unrelated foreign Playwright suite was
already occupying the exclusive shared browser slot; the independent tester must recompute and
run the plan on this frozen candidate when that slot is available.

Commands and results

Passing:

• DJANGO_SETTINGS_MODULE=website.settings.test uv run --frozen python manage.py test content.tests.test_public_pagination --noinput — 28 tests passed
• make test-ci — 607 passed
• make lint — passed
• uv run --frozen ruff format --check content/tests/test_public_pagination.py — passed
• git diff --check — passed
• make migrations-check — passed (No changes detected)
• make django-check — passed (OpenAPI, management parity, Django system check)
• make deployment-check — passed
• make database-portability-check — passed
• make security-check — passed (baseline, vulnerability, canary, artifact scan)

Known frozen-base failures, all outside this candidate:

• make format-check reports existing drift in content/podcast_routes.py,
  content/public_data.py, content/tests/test_public_projection_builder.py,
  core/tests/test_interactive_surfaces.py, events/tests/test_prepare_registration_sources.py,
  and playwright_tests/test_interactive_surfaces.py; content/tests/test_public_pagination.py
  is formatted.
• make terminology-check reports frozen projection digest mismatches in
  content/public_projection/podcasts.json, wiki.json, and wiki_search.json (Reconcile non-reproducible public projection after face8e4 #253).
• make typecheck reports four existing errors in
  content/event_speaker_bio_normalization.py.

These baseline failures were not changed or masked. The candidate is ready for the independent
tester to recompute the plan, run focused/full graph-selected checks, and confirm the two exact
count assertions. Screenshots are not applicable because no rendered product input changed.

No secrets, production data, or foreign processes were accessed or modified.

[alexeygrigorev]

Independent tester report — issue #300

Result

FAIL for release acceptance; issue-specific focused checks pass. The frozen candidate's
single changed test passes, but the required full Django gate exposes a stale shared-paginator
assertion in content/tests/test_collection_hub.py. The candidate itself was not modified.

Frozen candidate and plan

• Worktree: /home/alexey/git/dtc-website/.tmp/issue-300-engineer
• Base/head: face8e4808d65afbf0374d1ced7a88079950d663 / the same commit, dirty only at
  content/tests/test_public_pagination.py
• Engineer plan SHA-256: 9145cefb84e9cb93ea7bc7eeb18956587a078fb591ffd31411b4def66573c326
• Independently recomputed tester plan SHA-256:
  d25a0445a248cf456f6c643782e2b254024a2539db57be59e861549004ac08f6
• Both plans have graph SHA-256
  05ffeee99244c9becd5e98e40be08188f1c5e2a61204daece984c3bebd1bce6, source-tree manifest
  SHA-256 9b145fef5bb3290d51ba7a0e58b434599bf871348e73a4bc858a72ba902aee08,
  test_infrastructure risk, render.impact=false, and screenshots not_applicable.
• Candidate diff remains one test path; no product/template/URL/generated-data changes were
  observed. The engineer-reported candidate diff digest is
  1595276c68a6fab2c94fcc8729449fbe1063143d31365b2c48bb422a2ae5b4ab.

Passing evidence

• DJANGO_SETTINGS_MODULE=website.settings.test uv run --frozen python manage.py test content.tests.test_public_pagination --noinput — 28 tests passed.
• make test-ci — 607 passed.
• make lint — passed.
• make migrations-check django-check deployment-check database-portability-check — passed;
  no migration drift, management OpenAPI/parity current, Django checks clean, deployment checks
  clean, and database portability clean.
• make security-check — passed; baseline, vulnerability, canary, and artifact scan completed.
• make verification-container VERIFY_CONTAINER_OUTPUT=.tmp/verification/issue-300-tester/evidence/container-check.json — pass; artifact SHA-256
  b3954adbc0731524e6d993d4c23d25d208c97a1d256118e79be796b7bb1f8647.
• No browser or screenshots were run per the test-only/no-browser assignment and the plan's
  render.impact=false/screenshots=not_applicable disposition.

Blocking findings

1. make test-django-full fails in
   content.tests.test_collection_hub.CollectionHubPaginationTests.test_controls_are_one_labelled_landmark_with_one_current_page.
   The assertion still expects class="filter-pill pagination-number", while the accepted
   shared paginator emits class="filter-pill pagination-number interactive-lift". An isolated
   rerun of content.tests.test_collection_hub reproduces 1 failure / 24 tests. This is
   outside the frozen candidate's changed path but contradicts the [P0] Align public pagination regression assertions with interactive-lift #300 grooming statement that
   the stale expectations are limited to test_public_pagination.py:423-424; coordinate the
   collection-hub owner (Restore face8e4 formatting, typing, and focused content gates #261) or expand [P0] Align public pagination regression assertions with interactive-lift #300's test-only scope before acceptance.
2. make format-check fails on six pre-existing files, including
   content/podcast_routes.py, content/public_data.py, and
   content/tests/test_public_projection_builder.py; the [P0] Align public pagination regression assertions with interactive-lift #300 file is formatted.
3. make typecheck fails with four pre-existing errors in
   content/event_speaker_bio_normalization.py.
4. make terminology-check fails on the frozen projection digests for
   content/public_projection/podcasts.json, wiki.json, and wiki_search.json (Reconcile non-reproducible public projection after face8e4 #253).

The focused #300 contract is therefore verified, but the independent tester cannot issue a
release-accepting PASS while the required full Django/quality gates remain red. No code, issue
labels, commits, pushes, merges, deployments, or screenshots were changed or created.

[alexeygrigorev]

Product Manager Disposition — HOLD\n\nVerdict: HOLD — #300 is technically correct but not yet accepted for merge/release.\n\nThe independent tester confirms the frozen #300 candidate passes its own 28-test content.tests.test_public_pagination contract and changes only content/tests/test_public_pagination.py. The tester also found one unrelated full-Django failure in content/tests/test_collection_hub.py: its exact paginator class expectation still omits interactive-lift.\n\nOwnership is preserved. The current #261 candidate already fixes that collection-hub assertion in its explicitly allowed content/tests/test_collection_hub.py path (and keeps #261’s archive-row/normalization/prose scope). #300 will not expand into that file, duplicate the assertion change, or alter #261’s scope. The #261 candidate is still PM-held and not merged, so this resolves ownership but does not yet make the shared gate green.\n\n### Acceptance criteria for #300\n\n- Keep the frozen one-file diff and exact class strings: filter-pill pagination-number interactive-lift counted as 7 and filter-pill pagination-step interactive-lift counted as 2.\n- Preserve all other pagination assertions; no production/template/CSS/browser/generated-data changes or weakened/broad assertions.\n- Re-freeze #300 on the exact release candidate containing the accepted #261 collection-hub fix, independently recompute the plan, and rerun the required full Django/quality envelope.\n- Obtain a tester-final PASS with no required unresolved failures; #253 projection/quality failures and other shared release blockers must be resolved by their owning issues rather than waived here. Screenshots remain not applicable because #300 changes only a regression-test file.\n\nUntil those conditions are met, #300 remains open and unaccepted. No commit, merge, push, or deployment is authorized.

[alexeygrigorev]

Independent tester report — issue #300

Verdict

NOT PASS for release acceptance; PASS for the frozen #300 candidate scope. The candidate's
test-only pagination contract passes its focused checks and the candidate remains unchanged.
The full release envelope is not green because a separate shared-paginator assertion in
content/tests/test_collection_hub.py is stale, and the frozen base has unrelated quality
failures. No browser or screenshot run was performed, as requested; this is a test-only change
with no render impact.

Frozen identity

• Worktree: /home/alexey/git/dtc-website/.tmp/issue-300-engineer
• Base/HEAD: face8e4808d65afbf0374d1ced7a88079950d663 / same commit, dirty only at
  content/tests/test_public_pagination.py
• Candidate diff SHA-256 (path + NUL + binary diff + NUL):
  1595276c68a6fab2c94fcc8729449fbe1063143d31365b2c48bb422a2ae5b4ab
• Engineer plan SHA-256:
  9145cefb84e9cb93ea7bc7eeb18956587a078fb591ffd31411b4def66573c326
• git diff --check: passed; no candidate edits were made

Passing evidence

• DJANGO_SETTINGS_MODULE=website.settings.test uv run --frozen pytest content/tests/test_public_pagination.py -q: 28 passed, 119 subtests passed
• DTC_TEST_RUN_ID=issue-300-tester-ci-20260901 make test-ci: 607 passed
• make lint: passed
• uv run --frozen ruff format --check content/tests/test_public_pagination.py: passed
• make migrations-check django-check deployment-check database-portability-check: passed
• make security-check: passed (baseline, vulnerability, canary, and artifact scan)
• make verification-container VERIFY_CONTAINER_OUTPUT=.tmp/verification/issue-300-tester-subagent/evidence/container-check.json: passed; artifact SHA-256 b3954adbc0731524e6d993d4c23d25d208c97a1d256118e79be796b7bb1f8647

Blocking / baseline evidence

• Isolated content.tests.test_collection_hub: 24 tests, 1 failure at
  test_controls_are_one_labelled_landmark_with_one_current_page; it still expects
  class="filter-pill pagination-number" while the accepted shared paginator emits
  interactive-lift. This is outside the candidate's sole changed path and blocks a green
  full-Django release envelope.
• make format-check: failed on six unrelated frozen-base files; the [P0] Align public pagination regression assertions with interactive-lift #300 file is formatted.
• make typecheck: failed on four unrelated existing errors in
  content/event_speaker_bio_normalization.py.
• make terminology-check: failed on frozen projection digests for podcasts.json, wiki.json,
  and wiki_search.json (Reconcile non-reproducible public projection after face8e4 #253).

No browser, screenshot, code modification, commit, push, merge, deployment, or issue closure was
performed. The candidate remains uncommitted and ready for PM disposition only after the release
baseline blockers are resolved or explicitly routed through their owning issues.

[alexeygrigorev]

Software Engineer Report — #300 fresh reconstruction

Disposition

HOLD for release acceptance; PASS for the bounded candidate scope. The
candidate is the accepted one-file regression-test correction and remains
uncommitted. Its focused pagination contract passes. The required full release
baseline is still red on failures owned by other issues, so this report does
not authorize a commit, merge, push, deployment, or issue closure.

Frozen candidate and provenance

• Worktree: /home/alexey/git/dtc-website/.tmp/issue-300-reconstruct-20260902
• Base/HEAD: face8e4808d65afbf0374d1ced7a88079950d663 / the same commit
• Source mode: uncommitted worktree; no commit, push, merge, deployment, or
  remote state change
• Changed-path census: exactly M content/tests/test_public_pagination.py
• Candidate diff SHA-256 (path + NUL + binary diff + NUL):
  1595276c68a6fab2c94fcc8729449fbe1063143d31365b2c48bb422a2ae5b4ab
• git diff --check: passed
• Worktree source manifest: 3,359 entries,
  9b145fef5bb3290d51ba7a0e58b434599bf871348e73a4bc858a72ba902aee08
• Git object format: sha1; base tree: c8496fb2a37529eb7c63e1eeb9e35e29a2aadada
• Non-candidate diff: empty; the production application, template, CSS,
  fixtures, generated projections, browser harness, migrations, and all
  Restore face8e4 formatting, typing, and focused content gates #261/Keep failed structured-component evidence schema-valid #280-owned paths are unchanged.

Bounded implementation

The two stale exact class-count assertions now match the accepted shared
markup emitted by templates/public/_pagination.html:

• class="filter-pill pagination-number interactive-lift" — exactly 7
  controls on the 12-page middle state;
• class="filter-pill pagination-step interactive-lift" — exactly 2 controls.

All other pagination assertions remain unchanged, including labels,
aria-current, arrows, ellipses, URL windows, canonical/SEO/cache behavior,
boundaries, empty/single-page behavior, and source/design-system guards. The
template itself is byte-identical to face8e4, proving no production behavior
change.

Candidate-specific checks

• DJANGO_SETTINGS_MODULE=website.settings.test uv run --frozen python manage.py test content.tests.test_public_pagination --noinput — 28 passed
• DJANGO_SETTINGS_MODULE=website.settings.test uv run --frozen pytest content/tests/test_public_pagination.py -q — 28 passed, 119 subtests passed
• uv run --frozen ruff format --check content/tests/test_public_pagination.py — passed
• uv run --frozen ruff check content/tests/test_public_pagination.py — passed
• git diff --check — passed

Repository controls

Passing controls:

• make lint — passed
• make test-ci — 607 passed
• make migrations-check — passed (No changes detected)
• make django-check — passed (OpenAPI, management parity, system check)
• make deployment-check — passed
• make database-portability-check — passed
• make security-check — passed (baseline, vulnerability, canary, artifact scan)
• make verification-container VERIFY_CONTAINER_OUTPUT=.tmp/verification/issue-300-reconstruct-20260902/evidence/container-check.json — passed, 13 assertions; artifact SHA-256 b3954adbc0731524e6d993d4c23d25d208c97a1d256118e79be796b7bb1f8647

Known frozen-base failures, not changed or masked by this candidate:

• make format-check — six unrelated files: content/podcast_routes.py,
  content/public_data.py, content/tests/test_public_projection_builder.py,
  core/tests/test_interactive_surfaces.py,
  events/tests/test_prepare_registration_sources.py, and
  playwright_tests/test_interactive_surfaces.py.
• make typecheck — four existing errors in
  content/event_speaker_bio_normalization.py (lines 505, 544, 550, 564).
• make terminology-check — frozen projection digests differ for
  content/public_projection/podcasts.json, wiki.json, and
  wiki_search.json (Reconcile non-reproducible public projection after face8e4 #253).

Versioned verification envelope

• Plan: .tmp/verification/issue-300-reconstruct-20260902/verification-plan.json
• Plan file SHA-256: 0ed62cb0d16ac6eff702174969df5b4554605f9463e0c556ae5e2a641f394009
• Report: .tmp/verification/issue-300-reconstruct-20260902/verification-report.json
• Report file SHA-256: aa1a205abf3e0db7d49f7b604729d7c41957fdc016086fba8df5c60ed1db6979
• Report-bound plan digest: b6dffba674b50601895adc695824d25bfb4ca54d9a81e371abea51d96a97351c
• Graph SHA-256: 05ffeee99244c9becd5e98e40be08188f1c5e2a61204daece984c3bebd1bce6
• Policy version: 5; profile/reason: full / test_infrastructure
• Direct node: app.content; downstream nodes:
  django.accounts, django.api, django.content,
  django.content_sync, django.core
• Render impact: false; screenshots: not_applicable
• Plan dispositions: compatibility, content invariants, and screenshots
  not_applicable; selector, quality, Django, Playwright smoke, container,
  and evidence validation rerun.
• make verification-evidence-check ... and
  make verification-report-check ... — passed validation.

The versioned runner's final verdict is failure because the frozen base
still fails the shared release envelope:

• Django full: one stale exact class expectation in
  content/tests/test_collection_hub.py (the Restore face8e4 formatting, typing, and focused content gates #261-owned path), with the
  secondary Django multiprocessing traceback-pickling error; the candidate's
  focused module is green.
• Playwright smoke: 30 selected, 23 passed, 7 failed, and 4 errors, caused by
  existing event/podcast/projection fixtures and not by this test-only diff.
• Quality: stops on the Reconcile non-reproducible public projection after face8e4 #253 frozen projection terminology digests above; the
  standalone format/type failures are recorded above.

Dependencies and handoff

The candidate must be re-frozen and independently tested on the release base
after #261's accepted collection-hub repair and #253's accepted deterministic
projection/source handoff are available. As of this handoff, both issues are
open and PM-held; their prepared candidates are not merge authorization or
green-base evidence. Full Django, quality, and applicable browser evidence
must be green before #300 can receive PM acceptance.

No screenshots are required because only a regression-test file changed. No
secrets, credentials, cookies, registration data, production exports, or
foreign processes were accessed or modified.