feat(ralph): smart gates + DAG + dashboard live (W3)
Restructurare Ralph QC loop pe smart gate dispatcher tag-driven (în loc de 5 faze fixe), DAG dependsOn cu propagare blocked, retry guard 3-strike, rate limit detection, plus dashboard live cu polling 5s. Changes: - tools/ralph_prd_generator.py: parametru optional final_plan_path; când e furnizat, invocă Claude Opus pe final-plan.md pentru extragere user stories cu schema extinsă (tags, dependsOn, acceptanceCriteria 3-5). Backward compat păstrat — fără final_plan_path, fallback la heuristic-ul vechi. - tools/ralph/prd-template.json: schema W3 (tags[], dependsOn[], retries, failed, blocked, failureReason, requiresDesignReview). - tools/ralph/prompt.md: 4 faze (impl, base quality, smart gates, commit) + dispatcher pe story.tags. Tags vide → run-all-gates fallback (safe default). - tools/ralph_dag.py (nou): tag validation heuristic anti-silent-regression (force ui dacă diff atinge .vue/.tsx/.html/.css/.scss; force db pentru migrations sau .sql; force vercel dacă există vercel.json) + topological sort cu blocked propagation + atomic prd.json updates. - tools/ralph/ralph.sh: --max-turns 30, DAG-aware story selection, retry counter cu auto-fail la 3, rate limit detection (sleep 30min + 1 retry), CLI subcommands prin tools/ralph_dag.py helper. - dashboard/handlers/ralph.py (nou): /api/ralph/status + /<slug>/log + /prd + /stop. Defensive vs corrupt prd.json. Sandbox-ed PID kill. - dashboard/ralph.html (nou): live cards 3/2/1 col responsive, polling 5s, drawer pentru log/PRD viewer, status colors (--status-running/blocked/ failed/complete declarate inline), Lucide icons cu aria-labels. - dashboard/api.py: mount /api/ralph/* (GET status/log/prd, POST stop). - tests/: 72 teste noi (smart gates, DAG, retry, dashboard endpoint). Note arhitecturale: - Polling 5s ales peste SSE/WebSocket (suficient pentru iter Ralph 8-15min) - Tag validation rulează POST-iter pe diff git pentru anti-silent-regression - Rate limit retry: 1 dată per rulare, apoi mark failed=rate_limited Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
This commit is contained in:
305
dashboard/handlers/ralph.py
Normal file
305
dashboard/handlers/ralph.py
Normal file
@@ -0,0 +1,305 @@
|
||||
"""Ralph live dashboard endpoints (W3).
|
||||
|
||||
Endpoints:
|
||||
GET /api/ralph/status — toate proiectele Ralph (cards data)
|
||||
GET /api/ralph/<slug>/log — tail progress.txt (default 100 lines)
|
||||
GET /api/ralph/<slug>/prd — full prd.json content
|
||||
POST /api/ralph/<slug>/stop — SIGTERM la Ralph PID
|
||||
|
||||
Polling: 5s din ralph.html (suficient pentru iter 8-15min Ralph).
|
||||
NU SSE/WebSocket pentru MVP.
|
||||
|
||||
Citește status din `~/workspace/<slug>/scripts/ralph/`:
|
||||
- prd.json → stories (passes/failed/blocked/retries)
|
||||
- progress.txt → log human-readable
|
||||
- logs/iteration-*.log → mtime ultimului iter
|
||||
- .ralph.pid → PID activ (verificat cu os.kill 0)
|
||||
|
||||
Reuse path constants din `dashboard/constants.py` (WORKSPACE_DIR).
|
||||
"""
|
||||
import json
|
||||
import os
|
||||
import signal
|
||||
from datetime import datetime
|
||||
from pathlib import Path
|
||||
from urllib.parse import unquote
|
||||
|
||||
import constants
|
||||
|
||||
|
||||
# Path Ralph per proiect (mereu în scripts/ralph/)
|
||||
def _ralph_dir(project_dir: Path) -> Path:
|
||||
return project_dir / "scripts" / "ralph"
|
||||
|
||||
|
||||
# Estimare ETA simplistă: avg iter time × stories rămase
|
||||
DEFAULT_ITER_MINUTES = 12 # midpoint din intervalul 8-15min menționat în plan
|
||||
|
||||
|
||||
class RalphHandlers:
|
||||
"""Mixin pentru /api/ralph/* — Ralph live status + control."""
|
||||
|
||||
# ── helpers ────────────────────────────────────────────────
|
||||
def _ralph_validate_slug(self, slug: str):
|
||||
"""Validează slug-ul + returnează project_dir sau None."""
|
||||
if not slug or "/" in slug or ".." in slug:
|
||||
return None
|
||||
slug = unquote(slug)
|
||||
project_dir = constants.WORKSPACE_DIR / slug
|
||||
try:
|
||||
resolved = project_dir.resolve()
|
||||
workspace_resolved = constants.WORKSPACE_DIR.resolve()
|
||||
resolved.relative_to(workspace_resolved)
|
||||
except (ValueError, OSError):
|
||||
return None
|
||||
if not project_dir.exists() or not project_dir.is_dir():
|
||||
return None
|
||||
return project_dir
|
||||
|
||||
def _ralph_pid_alive(self, ralph_dir: Path):
|
||||
"""Întoarce (running: bool, pid: int|None)."""
|
||||
pid_file = ralph_dir / ".ralph.pid"
|
||||
if not pid_file.exists():
|
||||
return False, None
|
||||
try:
|
||||
pid = int(pid_file.read_text().strip())
|
||||
os.kill(pid, 0) # signal 0 = check existence
|
||||
return True, pid
|
||||
except (ValueError, ProcessLookupError, PermissionError, OSError):
|
||||
return False, None
|
||||
|
||||
def _ralph_eta_minutes(self, stories_remaining: int, last_iter_mtime: float | None) -> int | None:
|
||||
"""Estimează minute rămase — None dacă nu avem date."""
|
||||
if stories_remaining <= 0:
|
||||
return 0
|
||||
return stories_remaining * DEFAULT_ITER_MINUTES
|
||||
|
||||
def _ralph_summarize_project(self, project_dir: Path) -> dict | None:
|
||||
"""Construiește dict de status per proiect — None dacă nu e Ralph project."""
|
||||
ralph_dir = _ralph_dir(project_dir)
|
||||
prd_json = ralph_dir / "prd.json"
|
||||
if not prd_json.exists():
|
||||
return None
|
||||
|
||||
# Defensive parse — corupt prd.json nu trebuie să dărâme dashboard
|
||||
try:
|
||||
prd = json.loads(prd_json.read_text(encoding="utf-8"))
|
||||
except (json.JSONDecodeError, OSError):
|
||||
return {
|
||||
"slug": project_dir.name,
|
||||
"status": "error",
|
||||
"error": "prd.json invalid sau ilizibil",
|
||||
"running": False,
|
||||
"pid": None,
|
||||
"stories": [],
|
||||
"storiesTotal": 0,
|
||||
"storiesComplete": 0,
|
||||
"storiesFailed": 0,
|
||||
"storiesBlocked": 0,
|
||||
}
|
||||
|
||||
stories = prd.get("userStories", []) or []
|
||||
total = len(stories)
|
||||
complete = sum(1 for s in stories if s.get("passes"))
|
||||
failed = sum(1 for s in stories if s.get("failed"))
|
||||
blocked = sum(1 for s in stories if s.get("blocked"))
|
||||
remaining = total - complete - failed - blocked
|
||||
|
||||
running, pid = self._ralph_pid_alive(ralph_dir)
|
||||
|
||||
# Last iteration mtime (pentru "acum X")
|
||||
logs_dir = ralph_dir / "logs"
|
||||
last_iter_mtime = None
|
||||
last_iter_iso = None
|
||||
if logs_dir.exists():
|
||||
iter_logs = sorted(logs_dir.glob("iteration-*.log"), key=lambda f: f.stat().st_mtime, reverse=True)
|
||||
if iter_logs:
|
||||
last_iter_mtime = iter_logs[0].stat().st_mtime
|
||||
last_iter_iso = datetime.fromtimestamp(last_iter_mtime).isoformat()
|
||||
|
||||
# Status compus pentru UI cards
|
||||
if running:
|
||||
top_status = "running"
|
||||
elif failed > 0 and remaining == 0:
|
||||
top_status = "failed"
|
||||
elif complete == total and total > 0:
|
||||
top_status = "complete"
|
||||
elif blocked > 0 and running is False:
|
||||
top_status = "blocked"
|
||||
else:
|
||||
top_status = "idle"
|
||||
|
||||
# Current story (DAG-eligible cel mai mic priority)
|
||||
current_story = None
|
||||
if running:
|
||||
eligible = [
|
||||
s for s in stories
|
||||
if not s.get("passes") and not s.get("failed") and not s.get("blocked")
|
||||
]
|
||||
eligible.sort(key=lambda s: (s.get("priority", 999), s.get("id", "")))
|
||||
if eligible:
|
||||
current_story = {
|
||||
"id": eligible[0].get("id"),
|
||||
"title": eligible[0].get("title"),
|
||||
"tags": eligible[0].get("tags", []),
|
||||
"retries": eligible[0].get("retries", 0),
|
||||
}
|
||||
|
||||
return {
|
||||
"slug": project_dir.name,
|
||||
"status": top_status,
|
||||
"running": running,
|
||||
"pid": pid,
|
||||
"branchName": prd.get("branchName", ""),
|
||||
"storiesTotal": total,
|
||||
"storiesComplete": complete,
|
||||
"storiesFailed": failed,
|
||||
"storiesBlocked": blocked,
|
||||
"storiesRemaining": remaining,
|
||||
"currentStory": current_story,
|
||||
"lastIterAt": last_iter_iso,
|
||||
"etaMinutes": self._ralph_eta_minutes(remaining, last_iter_mtime),
|
||||
"stories": [
|
||||
{
|
||||
"id": s.get("id"),
|
||||
"title": s.get("title"),
|
||||
"passes": bool(s.get("passes")),
|
||||
"failed": bool(s.get("failed")),
|
||||
"blocked": bool(s.get("blocked")),
|
||||
"retries": int(s.get("retries", 0)),
|
||||
"tags": s.get("tags", []),
|
||||
"failureReason": s.get("failureReason", ""),
|
||||
}
|
||||
for s in stories
|
||||
],
|
||||
}
|
||||
|
||||
# ── /api/ralph/status (GET) ────────────────────────────────
|
||||
def handle_ralph_status(self):
|
||||
"""Întoarce status pentru toate proiectele Ralph din workspace."""
|
||||
try:
|
||||
projects = []
|
||||
if not constants.WORKSPACE_DIR.exists():
|
||||
self.send_json({"projects": [], "fetchedAt": datetime.now().isoformat()})
|
||||
return
|
||||
|
||||
for entry in sorted(constants.WORKSPACE_DIR.iterdir()):
|
||||
if not entry.is_dir() or entry.name.startswith("."):
|
||||
continue
|
||||
summary = self._ralph_summarize_project(entry)
|
||||
if summary is not None:
|
||||
projects.append(summary)
|
||||
|
||||
self.send_json({
|
||||
"projects": projects,
|
||||
"fetchedAt": datetime.now().isoformat(),
|
||||
"count": len(projects),
|
||||
})
|
||||
except Exception as exc:
|
||||
self.send_json({"error": str(exc)}, 500)
|
||||
|
||||
# ── /api/ralph/<slug>/log (GET) ────────────────────────────
|
||||
def handle_ralph_log(self, slug: str):
|
||||
"""Tail progress.txt pentru un slug. Default last 100 lines."""
|
||||
try:
|
||||
project_dir = self._ralph_validate_slug(slug)
|
||||
if not project_dir:
|
||||
self.send_json({"error": "Invalid project slug"}, 400)
|
||||
return
|
||||
|
||||
from urllib.parse import parse_qs, urlparse
|
||||
qs = parse_qs(urlparse(self.path).query)
|
||||
try:
|
||||
lines_n = min(int(qs.get("lines", ["100"])[0]), 1000)
|
||||
except ValueError:
|
||||
lines_n = 100
|
||||
|
||||
progress = _ralph_dir(project_dir) / "progress.txt"
|
||||
if not progress.exists():
|
||||
self.send_json({"slug": slug, "lines": [], "total": 0})
|
||||
return
|
||||
|
||||
try:
|
||||
content = progress.read_text(encoding="utf-8", errors="replace")
|
||||
except OSError as exc:
|
||||
self.send_json({"error": f"read failed: {exc}"}, 500)
|
||||
return
|
||||
|
||||
all_lines = content.splitlines()
|
||||
tail = all_lines[-lines_n:] if len(all_lines) > lines_n else all_lines
|
||||
self.send_json({
|
||||
"slug": slug,
|
||||
"lines": tail,
|
||||
"total": len(all_lines),
|
||||
})
|
||||
except Exception as exc:
|
||||
self.send_json({"error": str(exc)}, 500)
|
||||
|
||||
# ── /api/ralph/<slug>/prd (GET) ────────────────────────────
|
||||
def handle_ralph_prd(self, slug: str):
|
||||
"""Returnează full prd.json pentru un slug."""
|
||||
try:
|
||||
project_dir = self._ralph_validate_slug(slug)
|
||||
if not project_dir:
|
||||
self.send_json({"error": "Invalid project slug"}, 400)
|
||||
return
|
||||
|
||||
prd_json = _ralph_dir(project_dir) / "prd.json"
|
||||
if not prd_json.exists():
|
||||
self.send_json({"error": "prd.json not found"}, 404)
|
||||
return
|
||||
|
||||
try:
|
||||
data = json.loads(prd_json.read_text(encoding="utf-8"))
|
||||
except json.JSONDecodeError as exc:
|
||||
self.send_json({"error": f"prd.json invalid: {exc}"}, 500)
|
||||
return
|
||||
|
||||
self.send_json(data)
|
||||
except Exception as exc:
|
||||
self.send_json({"error": str(exc)}, 500)
|
||||
|
||||
# ── /api/ralph/<slug>/stop (POST) ──────────────────────────
|
||||
def handle_ralph_stop(self, slug: str):
|
||||
"""Trimite SIGTERM la Ralph PID. Verifică că PID-ul e în WORKSPACE_DIR."""
|
||||
try:
|
||||
project_dir = self._ralph_validate_slug(slug)
|
||||
if not project_dir:
|
||||
self.send_json({"success": False, "error": "Invalid project slug"}, 400)
|
||||
return
|
||||
|
||||
ralph_dir = _ralph_dir(project_dir)
|
||||
pid_file = ralph_dir / ".ralph.pid"
|
||||
if not pid_file.exists():
|
||||
self.send_json({"success": False, "error": "No PID file"}, 404)
|
||||
return
|
||||
|
||||
try:
|
||||
pid = int(pid_file.read_text().strip())
|
||||
except (ValueError, OSError) as exc:
|
||||
self.send_json({"success": False, "error": f"Invalid PID file: {exc}"}, 500)
|
||||
return
|
||||
|
||||
# Sandbox: verifică că procesul e în workspace (nu omoară random PID)
|
||||
try:
|
||||
proc_cwd = Path(f"/proc/{pid}/cwd").resolve()
|
||||
if not str(proc_cwd).startswith(str(constants.WORKSPACE_DIR)):
|
||||
self.send_json({"success": False, "error": "PID not in workspace"}, 403)
|
||||
return
|
||||
except (FileNotFoundError, PermissionError):
|
||||
# Procesul nu mai există — best-effort cleanup
|
||||
self.send_json({"success": True, "message": "Process already stopped"})
|
||||
return
|
||||
|
||||
try:
|
||||
os.killpg(os.getpgid(pid), signal.SIGTERM)
|
||||
except ProcessLookupError:
|
||||
self.send_json({"success": True, "message": "Process already stopped"})
|
||||
return
|
||||
except PermissionError:
|
||||
self.send_json({"success": False, "error": "Permission denied"}, 403)
|
||||
return
|
||||
|
||||
self.send_json({"success": True, "message": f"Ralph stopped (PID {pid})"})
|
||||
except Exception as exc:
|
||||
self.send_json({"success": False, "error": str(exc)}, 500)
|
||||
Reference in New Issue
Block a user