Files
ducoterra ffa919b8bf fix(chat): keep in-flight answers alive across in-app view switches
Root cause (owner repro, verified in a real browser 2026-09-06): the
five navbar views (Chat, RAG, Sources, Tuning, History) were separate
HTML documents, so a navbar click was a REAL cross-document navigation
— the chat page unloaded, the in-flight SSE fetch was aborted, and the
phase-48 teardown (app/api/chat.py `finally`, "chat: turn cancelled")
stopped the model. Observed: send question -> click RAG mid-stream ->
click Chat -> the answer never finished: no `query_log` row, and on
return a dangling question with no brain record (the pre-token pagehide
partial persist skips because `acc` is empty).

Phase-48 LOCKED-DECISION REFINEMENT (owner-confirmed 2026-09-06,
flagged per AGENTS.md rule 3, not silently deviated): "real navigation
cancels the fetch" now means LEAVING THE APP — tab close,
external/other-document navigation, the Stop button. In-app navbar
switches are client-side view switches and no longer cancel.

Fix — Option A (SPA shell), chosen over B (Service Worker owns the
stream) and C (server-side turn registry + resume):
- frontend/index.html is the shell: ONE `<main id="main">` holds the
  five `<section class="view">` blocks; hidden views carry BOTH
  `hidden` and `inert` (WCAG — no focus/keyboard traversal). The
  shared header, the single `doc-modal-*` skeleton, and the
  `#app-version` footer each exist exactly once; the per-view copies
  from the four folded pages are dropped.
- New frontend/assets/router.js (vanilla module — no framework, no
  bundler, No-CDN rule intact): lazy-imports a view module on FIRST
  show only (mount-once, hide-forever — the chat view's in-flight SSE
  reader persists across switches; that persistence IS the fix);
  intercepts same-shell navbar links with preventDefault +
  history.pushState (never a document load); handles popstate; single
  writer of `.nav-link` active state (is-active + aria-current),
  document.title, and the per-view meta description (values carried
  over from the old pages' heads, brand-resolved at write time).
- Each folded page's JS becomes `export async function mount(root)` —
  root-scoped queries; `initSharedHeader()` dropped (the header boots
  once in the shell via the chat module; the admin flag comes from the
  same cached `fetchIsAdmin()` promise — zero extra requests).
- app/main.py: a small list-driven route factory serves the shell for
  /tuning.html, /sources.html, /git-sources.html, /history.html —
  registered AFTER the API routers and BEFORE the static catch-all
  (routes-first). The phase-33 caching middleware applies no-cache +
  `?v=` rewriting unchanged; app/core/caching.py needed NO change
  (the view paths did not change — pinned by the integration tests).
- The four old view .html files are DELETED (one source of truth);
  deep links to the old URLs keep working (the router picks the view
  from the pathname); `/?chat=<id>` is unaffected; the Containerfile
  bundles router.js (inlining the lazy view modules) and drops the
  folded page files.
- app/schemas.py: HistoryTurn.text cap 4000 -> 32000 — the shell
  keeps long saved answers in the chat, and the old cap (stricter than
  the 24_000-char total history budget) 422-rejected any second turn
  in such a chat (found by the phase-42 E2E suite on the shell).

Boundaries: login.html, shared.html, doc-edit.html, document.html
REMAIN separate documents (flow pages, not navbar tabs); a mid-stream
navigation to doc-edit/document.html still cancels per phase 48
(follow-up candidate, out of scope). The SSE API is unchanged. Real
departures still cancel the turn — phase 48 intact (pinned by
tests/e2e/test_stop_generation.py, unchanged, and by the new suite's
real-departure control).

Tests:
- Phase-20 suite REWRITTEN to the new semantics
  (tests/e2e/test_sources_midstream_bug.py): a navbar switch no longer
  cancels — the stream survives the switch and the FULL answer
  settles; the pagehide partial persist REMAINS for real departures
  (the partial's exact shape — first streamed chunk prefix, no done
  metadata — is still pinned there).
- NEW story suite tests/e2e/test_nav_switch_keeps_stream.py (mock
  LLM): the owner repro (send -> RAG mid-stream -> Chat: window
  sentinel survives = same document, FULL answer, exactly one brain
  turn in bor.chat.v1, exactly one settled query_log row, auto-saved
  row matches) + the same mid-stream switch against the other three
  views + the real-departure-still-cancels control + the no-switch
  baseline.
- tests/unit/test_frontend_router.py: source-level pins of the router
  invariants (click interceptor targets ONLY same-shell view paths,
  pushState-only switches, mount-once guard, hidden+inert pair,
  single-writer active state/title); shell-route integration tests
  (each folded path serves the shell with no-cache + `?v=` body; a
  non-view path still 404s); the file-reading unit pins re-pointed at
  the shell (the four view files are gone — the shell is the source
  of truth).

Verification (this commit): full suite green — 1565 unit+integration
tests, app/ coverage 99% (>90% floor); ruff + pyright clean; the
phase's E2E suites green in isolation (house protocol, AGENTS.md rule
9). Owner repro verified in a real browser against the real LLM
(dev server :8010, headful Chromium): "tell me about everquest" ->
RAG mid-stream -> Chat — the answer completed with one brain bubble
and no error banner, `query_log` gained exactly one settled row
(deflected=True: the dev KB holds no EverQuest docs — the settle, not
the topic, is the proof), zero "chat: turn cancelled" lines for that
turn; the control (real navigation to /shared.html mid-stream) still
cancelled (no settled row, the cancel line logged, the partial
persisted on return). Screenshots: .agents/screenshots/76_manual_*.

Phase 76 (76_spa_nav_shell) complete — moved to
.agents/phases/complete/.
2026-09-06 06:31:31 -04:00

184 lines
7.6 KiB
Python

"""Unit: phase-61 text pins — the retired homelab-era copy is GONE, the
locked neutral replacements are PRESENT (house pattern: read the files
as text, assert substrings — no browser, cf. test_frontend_brand.py).
The browser-visible layer is gated by the story suite
(``tests/e2e/test_stale_ui_copy.py``); these pins catch a silent
regression in the templates / the ``Settings`` defaults without it.
"""
from __future__ import annotations
import re
from pathlib import Path
FRONTEND = Path(__file__).resolve().parents[2] / "frontend"
ROOT = FRONTEND.parent
CONFIG_PY = ROOT / "app" / "config.py"
#: Every page in the app ships a footer (the phase-39 ``HTML_PAGES``
#: tuple, repeated here so this file stands alone).
HTML_PAGES = (
"index.html",
# Phase 76 (task 01): tuning.html is folded into the shell (deleted)
# — the shell (index.html) stands in for it here.
# Phase 76 (task 02): sources.html + git-sources.html are folded
# too (both deleted — the shell stands in for both views).
# Phase 76 (task 03): history.html is folded too (deleted — the
# shell stands in for the History view; all four folded view files
# are gone).
"document.html",
"login.html",
"shared.html",
"doc-edit.html",
)
# --- the locked replacement copy (owner-locked A1 / A2) ----------------
META = "Ask anything about your indexed documents — every answer cites the exact doc."
EMPTY_SUB = (
"I've read through your documents — ask me anything, and I'll point "
"you at the exact doc. You've got this."
)
PLACEHOLDER = "Ask me anything…"
PAGE_SUB = (
"Every file indexed from your configured sources — git repositories, "
"local directories, and uploaded archives. Press <strong>Sync "
"sources</strong> to pull the latest and re-import."
)
FOOTER = "Powered by self-hosted models"
CHIPS = [
"What documents are in the knowledge base?",
"Which source does each answer come from?",
"How do I add a new source?",
"Summarize the most recent document.",
]
# --- the retired homelab-era copy (must be nowhere) --------------------
OLD_META_FRAG = "the homelab and deployments"
OLD_SUB_FRAG = "the homelab and deployment notes"
OLD_PLACEHOLDER_FRAG = "Ask me about the homelab"
OLD_FOOTER = "Reese's self-hosted models"
OLD_GIT_EXAMPLE = "github.com/you/homelab"
OLD_CHIPS = [
"How is my Kubernetes cluster set up?",
"What's my backup strategy?",
"How do I deploy a new service?",
"What's currently running in the homelab?",
]
def _text(path: Path) -> str:
return path.read_text(encoding="utf-8")
def _norm(text: str) -> str:
"""Collapse whitespace runs — the templates wrap long lines, so the
locked copy is pinned against the normalized text."""
return re.sub(r"\s+", " ", text).strip()
def test_chat_page_old_copy_is_gone() -> None:
"""index.html: the old meta fragment, the old empty-state sub
fragment, the old composer placeholder and the old footer are all
retired. (The L176 code comment still says 'homelab' — A3: comments
are not user-visible and stay untouched, so only the exact old
user-visible strings are pinned.)"""
html = _text(FRONTEND / "index.html")
for frag in (OLD_META_FRAG, OLD_SUB_FRAG, OLD_PLACEHOLDER_FRAG, OLD_FOOTER):
assert frag not in html, f"retired chat copy still present: {frag!r}"
def test_sources_page_old_copy_is_gone() -> None:
"""The shell's RAG view (formerly sources.html, phase 76 task 02):
the ~/Homelab + ~/Deployments page-sub citations and the old footer
are retired (the whole shell — which carries the view — must be
clean)."""
html = _text(FRONTEND / "index.html")
for frag in ("~/Homelab", "~/Deployments", OLD_FOOTER):
assert frag not in html, f"retired sources copy still present: {frag!r}"
def test_git_sources_page_old_copy_is_gone() -> None:
"""The shell's Sources view (formerly git-sources.html, phase 76
task 02): the old example repo URL (A3) and the old footer are
retired (the whole shell — which carries the view — must be
clean)."""
html = _text(FRONTEND / "index.html")
for frag in (OLD_GIT_EXAMPLE, OLD_FOOTER):
assert frag not in html, f"retired git-sources copy still present: {frag!r}"
def test_chat_page_locked_copy_present_exactly_once() -> None:
"""The three locked (A1) chat strings, each exactly once in
index.html (a second copy could drift out of sync)."""
html = _norm(_text(FRONTEND / "index.html"))
assert html.count(META) == 1, "the locked meta description"
assert html.count(EMPTY_SUB) == 1, "the locked empty-state sub"
assert html.count(PLACEHOLDER) == 1, "the locked composer placeholder"
def test_sources_page_sub_is_the_locked_copy() -> None:
"""The RAG view's .page-sub (the string the TODO cited by name) reads
the locked (A1) copy, including the <strong> around Sync sources —
pinned inside the .page-sub element, not anywhere in the file.
Phase 76 (task 02): scoped to the RAG view — the shell carries one
.page-sub per view, and the earlier views' subs come first."""
html = _text(FRONTEND / "index.html")
i = html.find('id="view-rag"')
j = html.find('id="view-git-sources"', i)
assert -1 < i < j, "the RAG view section must be in the shell"
m = re.search(r'<p class="page-sub">(.*?)</p>', html[i:j], re.DOTALL)
assert m, "the RAG view must keep the .page-sub"
assert _norm(m.group(1)) == PAGE_SUB
def test_all_five_footers_are_the_locked_neutral_default() -> None:
"""Every standalone page carries the locked (A1) footer inside
exactly one ``class="footer-text"`` span (the stable hook phase 62's
BOR_FOOTER_TEXT env var drives). Phase 76 (task 02): the folded RAG
+ Sources files are gone — the shell carries its footer once.
Phase 76 (task 03): the History file is gone too — the shell's
single (chat) footer stands in for the History view (its per-page
footer copy, with the duplicate #app-version span, is dropped with
the move)."""
span = f'<span class="footer-text">{FOOTER}</span>'
for page in HTML_PAGES:
html = _text(FRONTEND / page)
assert html.count('class="footer-text"') == 1, (
f"{page}: exactly one .footer-text span"
)
assert html.count(span) == 1, f"{page}: the locked neutral footer"
def test_config_default_chips_are_the_locked_list_in_order() -> None:
"""The ``Settings.suggestions`` default (app/config.py) is exactly
the four locked (A2) chips, in order (the E2E conftest derives the
app-under-test env from this field — it can never drift)."""
config = _text(CONFIG_PY)
m = re.search(r"suggestions: list\[str\] = \[(.*?)\]", config, re.DOTALL)
assert m, "the suggestions default list must exist in app/config.py"
block = m.group(1)
positions = []
for chip in CHIPS:
assert f'"{chip}"' in block, f"locked chip missing from the default: {chip!r}"
positions.append(block.index(f'"{chip}"'))
assert positions == sorted(positions), "the chips must keep the locked order"
def test_old_chips_are_retired_not_relocated() -> None:
"""The four old default chips are retired, not relocated: they must
appear NOWHERE in the app source (app/ + frontend/) — config.py
included (the scan covers it)."""
for base in (ROOT / "app", FRONTEND):
for path in sorted(base.rglob("*")):
if not path.is_file() or "__pycache__" in path.parts:
continue
text = path.read_text(encoding="utf-8")
for chip in OLD_CHIPS:
assert chip not in text, (
f"retired chip {chip!r} relocated to {path.relative_to(ROOT)}"
)