Add test runner dashboard service (test.rose-ash.com)

Public Quart microservice that runs pytest against shared/tests/ and
shared/sexp/tests/, serving an HTMX-powered sexp-rendered dashboard
with pass/fail/running status, auto-refresh polling, and re-run button.
No database — results stored in memory.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

deploy.sh

@@ -2,7 +2,7 @@
 set -euo pipefail
 
 REGISTRY="registry.rose-ash.com:5000"
-APPS="blog market cart events federation account relations likes orders"
+APPS="blog market cart events federation account relations likes orders test"
 
 usage() {
   echo "Usage: deploy.sh [app ...]"

dev.sh

@@ -20,8 +20,8 @@ case "${1:-up}" in
     shift
     $COMPOSE logs -f "$@"
     ;;
-  test)
+  test-run)
-    # One-shot: all unit tests
+    # One-shot: all unit tests (headless, no dashboard)
     $COMPOSE run --rm test-unit python -m pytest \
       shared/ artdag/core/tests/ artdag/core/artdag/sexp/ \
       artdag/l1/tests/ artdag/l1/sexp_effects/ \

docker-compose.dev.yml

@@ -351,6 +351,42 @@ services:
       - ./account/__init__.py:/app/account/__init__.py:ro
       - ./account/models:/app/account/models:ro
 
+  test:
+    restart: unless-stopped
+    ports:
+      - "8011:8000"
+    environment:
+      <<: *dev-env
+    volumes:
+      - /root/rose-ash/_config/app-config.yaml:/app/config/app-config.yaml:ro
+      - ./shared:/app/shared
+      - ./test/app.py:/app/app.py
+      - ./test/sexp:/app/sexp
+      - ./test/bp:/app/bp
+      - ./test/services:/app/services
+      - ./test/runner.py:/app/runner.py
+      - ./test/path_setup.py:/app/path_setup.py
+      - ./test/entrypoint.sh:/usr/local/bin/entrypoint.sh
+      # sibling models
+      - ./blog/__init__.py:/app/blog/__init__.py:ro
+      - ./blog/models:/app/blog/models:ro
+      - ./market/__init__.py:/app/market/__init__.py:ro
+      - ./market/models:/app/market/models:ro
+      - ./cart/__init__.py:/app/cart/__init__.py:ro
+      - ./cart/models:/app/cart/models:ro
+      - ./events/__init__.py:/app/events/__init__.py:ro
+      - ./events/models:/app/events/models:ro
+      - ./federation/__init__.py:/app/federation/__init__.py:ro
+      - ./federation/models:/app/federation/models:ro
+      - ./account/__init__.py:/app/account/__init__.py:ro
+      - ./account/models:/app/account/models:ro
+      - ./relations/__init__.py:/app/relations/__init__.py:ro
+      - ./relations/models:/app/relations/models:ro
+      - ./likes/__init__.py:/app/likes/__init__.py:ro
+      - ./likes/models:/app/likes/models:ro
+      - ./orders/__init__.py:/app/orders/__init__.py:ro
+      - ./orders/models:/app/orders/models:ro
+
   test-unit:
     build:
       context: .

docker-compose.yml

@@ -35,6 +35,7 @@ x-app-env: &app-env
   APP_URL_ORDERS: https://orders.rose-ash.com
   APP_URL_RELATIONS: http://relations:8000
   APP_URL_LIKES: http://likes:8000
+  APP_URL_TEST: https://test.rose-ash.com
   APP_URL_ARTDAG: https://celery-artdag.rose-ash.com
   APP_URL_ARTDAG_L2: https://artdag.rose-ash.com
   INTERNAL_URL_BLOG: http://blog:8000
@@ -46,6 +47,7 @@ x-app-env: &app-env
   INTERNAL_URL_ORDERS: http://orders:8000
   INTERNAL_URL_RELATIONS: http://relations:8000
   INTERNAL_URL_LIKES: http://likes:8000
+  INTERNAL_URL_TEST: http://test:8000
   INTERNAL_URL_ARTDAG: http://l1-server:8100
   AP_DOMAIN: federation.rose-ash.com
   AP_DOMAIN_BLOG: blog.rose-ash.com
@@ -201,6 +203,17 @@ services:
       RUN_MIGRATIONS: "true"
       WORKERS: "1"
 
+  test:
+    <<: *app-common
+    image: registry.rose-ash.com:5000/test:latest
+    build:
+      context: .
+      dockerfile: test/Dockerfile
+    environment:
+      <<: *app-env
+      REDIS_URL: redis://redis:6379/9
+      WORKERS: "1"
+
   db:
     image: postgres:16
     environment:

test/Dockerfile

@@ -0,0 +1,60 @@
+# syntax=docker/dockerfile:1
+
+FROM python:3.11-slim AS base
+
+ENV PYTHONDONTWRITEBYTECODE=1 \
+    PYTHONUNBUFFERED=1 \
+    PYTHONPATH=/app \
+    PIP_NO_CACHE_DIR=1 \
+    APP_PORT=8000 \
+    APP_MODULE=app:app
+
+WORKDIR /app
+
+RUN apt-get update && apt-get install -y --no-install-recommends \
+      ca-certificates \
+    && rm -rf /var/lib/apt/lists/*
+
+COPY shared/requirements.txt ./requirements.txt
+RUN pip install -r requirements.txt && \
+    pip install pytest pytest-json-report
+
+# Shared code (including tests)
+COPY shared/ ./shared/
+
+# App code
+COPY test/ ./test-app-tmp/
+# Move service files into /app (flatten), but keep Dockerfile.* in place
+RUN cp -r test-app-tmp/app.py test-app-tmp/path_setup.py \
+         test-app-tmp/bp test-app-tmp/sexp test-app-tmp/services \
+         test-app-tmp/runner.py test-app-tmp/__init__.py ./ 2>/dev/null || true && \
+    rm -rf test-app-tmp
+
+# Sibling models for cross-domain SQLAlchemy imports
+COPY blog/__init__.py     ./blog/__init__.py
+COPY blog/models/         ./blog/models/
+COPY market/__init__.py   ./market/__init__.py
+COPY market/models/       ./market/models/
+COPY cart/__init__.py     ./cart/__init__.py
+COPY cart/models/         ./cart/models/
+COPY events/__init__.py   ./events/__init__.py
+COPY events/models/       ./events/models/
+COPY federation/__init__.py  ./federation/__init__.py
+COPY federation/models/      ./federation/models/
+COPY account/__init__.py  ./account/__init__.py
+COPY account/models/      ./account/models/
+COPY relations/__init__.py ./relations/__init__.py
+COPY relations/models/ ./relations/models/
+COPY likes/__init__.py ./likes/__init__.py
+COPY likes/models/ ./likes/models/
+COPY orders/__init__.py ./orders/__init__.py
+COPY orders/models/ ./orders/models/
+
+COPY test/entrypoint.sh /usr/local/bin/entrypoint.sh
+RUN chmod +x /usr/local/bin/entrypoint.sh
+
+RUN useradd -m -u 10001 appuser && chown -R appuser:appuser /app
+USER appuser
+
+EXPOSE ${APP_PORT}
+ENTRYPOINT ["/usr/local/bin/entrypoint.sh"]

test/app.py

@@ -0,0 +1,44 @@
+from __future__ import annotations
+import path_setup  # noqa: F401
+import sexp.sexp_components as sexp_components  # noqa: F401
+
+from shared.infrastructure.factory import create_base_app
+
+from bp import register_dashboard
+from services import register_domain_services
+
+
+async def test_context() -> dict:
+    """Test app context processor — minimal, no cross-service fragments."""
+    from shared.infrastructure.context import base_context
+
+    ctx = await base_context()
+    ctx["menu_items"] = []
+    ctx["cart_mini_html"] = ""
+    ctx["auth_menu_html"] = ""
+    ctx["nav_tree_html"] = ""
+    return ctx
+
+
+def create_app() -> "Quart":
+    app = create_base_app(
+        "test",
+        context_fn=test_context,
+        domain_services_fn=register_domain_services,
+    )
+
+    import sexp.sexp_components  # noqa: F401
+
+    app.register_blueprint(register_dashboard(url_prefix="/"))
+
+    # Run tests on startup
+    @app.before_serving
+    async def _run_tests_on_startup():
+        import runner
+        import asyncio
+        asyncio.create_task(runner.run_tests())
+
+    return app
+
+
+app = create_app()

test/bp/__init__.py

@@ -0,0 +1 @@
+from .dashboard.routes import register as register_dashboard

test/bp/dashboard/routes.py

@@ -0,0 +1,64 @@
+"""Test dashboard routes."""
+from __future__ import annotations
+
+import asyncio
+
+from quart import Blueprint, Response, make_response, request
+
+
+def register(url_prefix: str = "/") -> Blueprint:
+    bp = Blueprint("dashboard", __name__, url_prefix=url_prefix)
+
+    @bp.get("/")
+    async def index():
+        """Full page dashboard with last results."""
+        from shared.sexp.page import get_template_context
+        from shared.browser.app.csrf import generate_csrf_token
+        from sexp.sexp_components import render_dashboard_page
+        import runner
+
+        ctx = await get_template_context()
+        result = runner.get_results()
+        running = runner.is_running()
+        csrf = generate_csrf_token()
+
+        html = await render_dashboard_page(ctx, result, running, csrf)
+        return await make_response(html, 200)
+
+    @bp.post("/run")
+    async def run():
+        """Trigger a test run, redirect to /."""
+        import runner
+
+        if not runner.is_running():
+            asyncio.create_task(runner.run_tests())
+
+        # HX-Redirect for HTMX, regular redirect for non-HTMX
+        if request.headers.get("HX-Request"):
+            resp = Response("", status=200)
+            resp.headers["HX-Redirect"] = "/"
+            return resp
+
+        from quart import redirect as qredirect
+        return qredirect("/")
+
+    @bp.get("/results")
+    async def results():
+        """HTMX partial — poll target for results table."""
+        from shared.browser.app.csrf import generate_csrf_token
+        from sexp.sexp_components import render_results_partial
+        import runner
+
+        result = runner.get_results()
+        running = runner.is_running()
+        csrf = generate_csrf_token()
+
+        html = await render_results_partial(result, running, csrf)
+
+        resp = Response(html, status=200, content_type="text/html")
+        # If still running, tell HTMX to keep polling
+        if running:
+            resp.headers["HX-Trigger-After-Swap"] = "test-still-running"
+        return resp
+
+    return bp

test/entrypoint.sh

@@ -0,0 +1,24 @@
+#!/usr/bin/env bash
+set -euo pipefail
+
+# No database — skip DB wait and migrations
+
+# Clear Redis page cache on deploy
+if [[ -n "${REDIS_URL:-}" && "${REDIS_URL}" != "no" ]]; then
+  python3 -c "
+import redis, os
+r = redis.from_url(os.environ['REDIS_URL'])
+r.flushdb()
+" || echo "Redis flush failed (non-fatal), continuing..."
+fi
+
+# Start the app
+RELOAD_FLAG=""
+if [[ "${RELOAD:-}" == "true" ]]; then
+  RELOAD_FLAG="--reload"
+fi
+PYTHONUNBUFFERED=1 exec hypercorn "${APP_MODULE:-app:app}" \
+  --bind 0.0.0.0:${PORT:-8000} \
+  --workers ${WORKERS:-1} \
+  --keep-alive 75 \
+  ${RELOAD_FLAG}

test/path_setup.py

@@ -0,0 +1,9 @@
+import sys
+import os
+
+_app_dir = os.path.dirname(os.path.abspath(__file__))
+_project_root = os.path.dirname(_app_dir)
+
+for _p in (_project_root, _app_dir):
+    if _p not in sys.path:
+        sys.path.insert(0, _p)

test/runner.py

@@ -0,0 +1,135 @@
+"""Pytest subprocess runner + in-memory result storage."""
+from __future__ import annotations
+
+import asyncio
+import json
+import logging
+import time
+from pathlib import Path
+
+log = logging.getLogger(__name__)
+
+# In-memory state
+_last_result: dict | None = None
+_running: bool = False
+
+# Paths to test directories (relative to /app in Docker)
+_TEST_DIRS = [
+    "shared/tests/",
+    "shared/sexp/tests/",
+]
+
+_REPORT_PATH = "/tmp/test-report.json"
+
+
+async def run_tests() -> dict:
+    """Run pytest in subprocess, parse JSON report, store results."""
+    global _last_result, _running
+
+    if _running:
+        return {"status": "already_running"}
+
+    _running = True
+    started_at = time.time()
+
+    try:
+        cmd = [
+            "python3", "-m", "pytest",
+            *_TEST_DIRS,
+            "--json-report",
+            f"--json-report-file={_REPORT_PATH}",
+            "-q",
+            "--tb=short",
+        ]
+
+        proc = await asyncio.create_subprocess_exec(
+            *cmd,
+            stdout=asyncio.subprocess.PIPE,
+            stderr=asyncio.subprocess.STDOUT,
+            cwd="/app",
+        )
+        stdout, _ = await proc.communicate()
+        finished_at = time.time()
+
+        # Parse JSON report
+        report_path = Path(_REPORT_PATH)
+        if report_path.exists():
+            try:
+                report = json.loads(report_path.read_text())
+            except (json.JSONDecodeError, OSError):
+                report = {}
+        else:
+            report = {}
+
+        summary = report.get("summary", {})
+        tests_raw = report.get("tests", [])
+
+        tests = []
+        for t in tests_raw:
+            tests.append({
+                "nodeid": t.get("nodeid", ""),
+                "outcome": t.get("outcome", "unknown"),
+                "duration": round(t.get("duration", 0), 4),
+                "longrepr": (t.get("call", {}) or {}).get("longrepr", ""),
+            })
+
+        passed = summary.get("passed", 0)
+        failed = summary.get("failed", 0)
+        errors = summary.get("error", 0)
+        skipped = summary.get("skipped", 0)
+        total = summary.get("total", len(tests))
+
+        if failed > 0 or errors > 0:
+            status = "failed"
+        else:
+            status = "passed"
+
+        _last_result = {
+            "status": status,
+            "started_at": started_at,
+            "finished_at": finished_at,
+            "duration": round(finished_at - started_at, 2),
+            "passed": passed,
+            "failed": failed,
+            "errors": errors,
+            "skipped": skipped,
+            "total": total,
+            "tests": tests,
+            "stdout": (stdout or b"").decode("utf-8", errors="replace")[-5000:],
+        }
+
+        log.info(
+            "Test run complete: %s (%d passed, %d failed, %d errors, %.1fs)",
+            status, passed, failed, errors, _last_result["duration"],
+        )
+        return _last_result
+
+    except Exception:
+        log.exception("Test run failed")
+        finished_at = time.time()
+        _last_result = {
+            "status": "error",
+            "started_at": started_at,
+            "finished_at": finished_at,
+            "duration": round(finished_at - started_at, 2),
+            "passed": 0,
+            "failed": 0,
+            "errors": 1,
+            "skipped": 0,
+            "total": 0,
+            "tests": [],
+            "stdout": "",
+        }
+        return _last_result
+    finally:
+        _running = False
+
+
+def get_results() -> dict | None:
+    """Return last run results."""
+    return _last_result
+
+
+def is_running() -> bool:
+    """Check if tests are currently running."""
+    return _running

test/services/__init__.py

@@ -0,0 +1,6 @@
+"""Test app service registration."""
+from __future__ import annotations
+
+
+def register_domain_services() -> None:
+    """Register services for the test app (none needed)."""

test/sexp/dashboard.sexpr

@@ -0,0 +1,88 @@
+;; Test dashboard components
+
+(defcomp ~test-status-badge (&key status)
+  (span :class (str "inline-flex items-center rounded-full border px-3 py-1 text-sm font-medium "
+                (if (= status "running") "border-amber-300 bg-amber-50 text-amber-700 animate-pulse"
+                (if (= status "passed") "border-emerald-300 bg-emerald-50 text-emerald-700"
+                (if (= status "failed") "border-rose-300 bg-rose-50 text-rose-700"
+                "border-stone-300 bg-stone-50 text-stone-700"))))
+    status))
+
+(defcomp ~test-run-button (&key running csrf)
+  (form :method "POST" :action "/run" :class "inline"
+    (input :type "hidden" :name "csrf_token" :value csrf)
+    (button :type "submit"
+      :class (str "rounded bg-stone-800 px-4 py-2 text-sm font-medium text-white hover:bg-stone-700 "
+              "disabled:opacity-50 disabled:cursor-not-allowed transition-colors")
+      :disabled (if running "true" nil)
+      (if running "Running..." "Run Tests"))))
+
+(defcomp ~test-summary (&key status passed failed errors skipped total duration last-run running csrf)
+  (div :class "space-y-4"
+    (div :class "flex items-center justify-between flex-wrap gap-3"
+      (div :class "flex items-center gap-3"
+        (h2 :class "text-lg font-semibold text-stone-800" "Test Results")
+        (when status (~test-status-badge :status status)))
+      (~test-run-button :running running :csrf csrf))
+    (when status
+      (div :class "grid grid-cols-2 sm:grid-cols-3 md:grid-cols-6 gap-3"
+        (div :class "rounded border border-stone-200 bg-white p-3 text-center"
+          (div :class "text-2xl font-bold text-stone-800" total)
+          (div :class "text-xs text-stone-500" "Total"))
+        (div :class "rounded border border-emerald-200 bg-emerald-50 p-3 text-center"
+          (div :class "text-2xl font-bold text-emerald-700" passed)
+          (div :class "text-xs text-emerald-600" "Passed"))
+        (div :class "rounded border border-rose-200 bg-rose-50 p-3 text-center"
+          (div :class "text-2xl font-bold text-rose-700" failed)
+          (div :class "text-xs text-rose-600" "Failed"))
+        (div :class "rounded border border-orange-200 bg-orange-50 p-3 text-center"
+          (div :class "text-2xl font-bold text-orange-700" errors)
+          (div :class "text-xs text-orange-600" "Errors"))
+        (div :class "rounded border border-sky-200 bg-sky-50 p-3 text-center"
+          (div :class "text-2xl font-bold text-sky-700" skipped)
+          (div :class "text-xs text-sky-600" "Skipped"))
+        (div :class "rounded border border-stone-200 bg-white p-3 text-center"
+          (div :class "text-2xl font-bold text-stone-800" (str duration "s"))
+          (div :class "text-xs text-stone-500" "Duration")))
+      (div :class "text-xs text-stone-400" (str "Last run: " last-run)))))
+
+(defcomp ~test-row (&key nodeid outcome duration longrepr)
+  (tr :class (str "border-b border-stone-100 "
+              (if (= outcome "passed") "bg-white"
+              (if (= outcome "failed") "bg-rose-50"
+              (if (= outcome "skipped") "bg-sky-50"
+              "bg-orange-50"))))
+    (td :class "px-3 py-2 text-xs font-mono text-stone-700 max-w-0 truncate" :title nodeid nodeid)
+    (td :class "px-3 py-2 text-center"
+      (span :class (str "inline-flex items-center rounded-full border px-2 py-0.5 text-[11px] font-medium "
+                    (if (= outcome "passed") "border-emerald-300 bg-emerald-50 text-emerald-700"
+                    (if (= outcome "failed") "border-rose-300 bg-rose-50 text-rose-700"
+                    (if (= outcome "skipped") "border-sky-300 bg-sky-50 text-sky-700"
+                    "border-orange-300 bg-orange-50 text-orange-700"))))
+        outcome))
+    (td :class "px-3 py-2 text-right text-xs text-stone-500 tabular-nums" (str duration "s"))
+    (td :class "px-3 py-2 text-xs text-rose-600 font-mono max-w-xs truncate" :title longrepr
+      (when longrepr longrepr))))
+
+(defcomp ~test-results-table (&key rows-html has-failures)
+  (div :class "overflow-x-auto rounded border border-stone-200 bg-white"
+    (table :class "w-full text-left"
+      (thead
+        (tr :class "border-b border-stone-200 bg-stone-50"
+          (th :class "px-3 py-2 text-xs font-medium text-stone-600" "Test")
+          (th :class "px-3 py-2 text-xs font-medium text-stone-600 text-center w-24" "Status")
+          (th :class "px-3 py-2 text-xs font-medium text-stone-600 text-right w-20" "Time")
+          (th :class "px-3 py-2 text-xs font-medium text-stone-600 w-48" "Error")))
+      (tbody rows-html))))
+
+(defcomp ~test-running-indicator ()
+  (div :class "flex items-center justify-center py-12 text-stone-500"
+    (div :class "flex items-center gap-3"
+      (div :class "animate-spin h-6 w-6 border-2 border-stone-300 border-t-stone-600 rounded-full")
+      (span :class "text-sm" "Running tests..."))))
+
+(defcomp ~test-no-results ()
+  (div :class "flex items-center justify-center py-12 text-stone-400"
+    (div :class "text-center"
+      (div :class "text-4xl mb-2" "?")
+      (div :class "text-sm" "No test results yet. Click \"Run Tests\" to start."))))

test/sexp/sexp_components.py

@@ -0,0 +1,105 @@
+"""Test service s-expression page components."""
+from __future__ import annotations
+
+import os
+from datetime import datetime
+
+from shared.sexp.jinja_bridge import render, load_service_components
+from shared.sexp.helpers import root_header_html, full_page
+
+# Load test-specific .sexpr components at import time
+load_service_components(os.path.dirname(os.path.dirname(__file__)))
+
+
+def _format_time(ts: float | None) -> str:
+    """Format a unix timestamp for display."""
+    if not ts:
+        return "never"
+    return datetime.fromtimestamp(ts).strftime("%-d %b %Y, %H:%M:%S")
+
+
+def _test_rows_html(tests: list[dict]) -> str:
+    """Render all test result rows."""
+    parts = []
+    for t in tests:
+        parts.append(render(
+            "test-row",
+            nodeid=t["nodeid"],
+            outcome=t["outcome"],
+            duration=str(t["duration"]),
+            longrepr=t.get("longrepr", ""),
+        ))
+    return "".join(parts)
+
+
+def _results_partial_html(result: dict | None, running: bool, csrf: str) -> str:
+    """Render the results section (summary + table or running indicator)."""
+    if running and not result:
+        summary = render(
+            "test-summary",
+            status="running", passed="0", failed="0", errors="0",
+            skipped="0", total="0", duration="...",
+            last_run="in progress", running=True, csrf=csrf,
+        )
+        return summary + render("test-running-indicator")
+
+    if not result:
+        summary = render(
+            "test-summary",
+            status=None, passed="0", failed="0", errors="0",
+            skipped="0", total="0", duration="0",
+            last_run="never", running=running, csrf=csrf,
+        )
+        return summary + render("test-no-results")
+
+    status = "running" if running else result["status"]
+    summary = render(
+        "test-summary",
+        status=status,
+        passed=str(result["passed"]),
+        failed=str(result["failed"]),
+        errors=str(result["errors"]),
+        skipped=str(result.get("skipped", 0)),
+        total=str(result["total"]),
+        duration=str(result["duration"]),
+        last_run=_format_time(result["finished_at"]) if not running else "in progress",
+        running=running,
+        csrf=csrf,
+    )
+
+    if running:
+        return summary + render("test-running-indicator")
+
+    tests = result.get("tests", [])
+    if not tests:
+        return summary + render("test-no-results")
+
+    has_failures = result["failed"] > 0 or result["errors"] > 0
+    rows = _test_rows_html(tests)
+    table = render("test-results-table", rows_html=rows,
+                   has_failures=str(has_failures).lower())
+    return summary + table
+
+
+def _wrap_results_div(inner_html: str, running: bool) -> str:
+    """Wrap results in a div with HTMX polling when running."""
+    attrs = 'id="test-results" class="space-y-6 p-4"'
+    if running:
+        attrs += ' hx-get="/results" hx-trigger="every 2s" hx-swap="outerHTML"'
+    return f'<div {attrs}>{inner_html}</div>'
+
+
+async def render_dashboard_page(ctx: dict, result: dict | None,
+                                running: bool, csrf: str) -> str:
+    """Full page: test dashboard."""
+    hdr = root_header_html(ctx)
+    inner = _results_partial_html(result, running, csrf)
+    content = _wrap_results_div(inner, running)
+    return full_page(ctx, header_rows_html=hdr, content_html=content)
+
+
+async def render_results_partial(result: dict | None, running: bool,
+                                 csrf: str) -> str:
+    """HTMX partial: just the results section (wrapped in polling div)."""
+    inner = _results_partial_html(result, running, csrf)
+    return _wrap_results_div(inner, running)