diff --git a/frontend_multi_user/Dockerfile b/frontend_multi_user/Dockerfile index 37812eb5..18966fad 100644 --- a/frontend_multi_user/Dockerfile +++ b/frontend_multi_user/Dockerfile @@ -37,5 +37,5 @@ ENV PLANEXE_FRONTEND_MULTIUSER_WORKERS=4 CMD gunicorn wsgi:app \ --bind 0.0.0.0:${PLANEXE_FRONTEND_MULTIUSER_PORT:-5000} \ --workers ${PLANEXE_FRONTEND_MULTIUSER_WORKERS} \ - --timeout 60 \ + --timeout 120 \ --chdir /app/frontend_multi_user/src diff --git a/frontend_multi_user/src/app.py b/frontend_multi_user/src/app.py index cd2661c7..b96fade4 100644 --- a/frontend_multi_user/src/app.py +++ b/frontend_multi_user/src/app.py @@ -531,11 +531,12 @@ def _ensure_stopped_state() -> None: one enum name does not poison the attempt for the other. """ for type_name in ("taskstate", "planstate"): - try: - with self.db.engine.begin() as conn: - conn.execute(text(f"ALTER TYPE {type_name} ADD VALUE IF NOT EXISTS 'stopped'")) - except Exception as exc: - logger.debug("ALTER TYPE %s: %s", type_name, exc) + for enum_value in ("stopped",): + try: + with self.db.engine.begin() as conn: + conn.execute(text(f"ALTER TYPE {type_name} ADD VALUE IF NOT EXISTS '{enum_value}'")) + except Exception as exc: + logger.debug("ALTER TYPE %s ADD VALUE %s: %s", type_name, enum_value, exc) def _ensure_last_progress_at_column() -> None: insp = inspect(self.db.engine) @@ -1061,7 +1062,8 @@ def index(): user_id = str(user.id) credits_balance = to_credit_decimal(user.credits_balance) credits_balance_display = format_credit_display(user.credits_balance) - can_create_plan = credits_balance >= Decimal("2") + min_credits = Decimal(os.environ.get("PLANEXE_MIN_CREDITS_TO_CREATE_PLAN", "2")) + can_create_plan = credits_balance >= min_credits if user_id: # Generate a nonce so the user can start a plan from the dashboard diff --git a/frontend_multi_user/src/plan_routes.py b/frontend_multi_user/src/plan_routes.py index 6744bb95..e39a61d5 100644 --- a/frontend_multi_user/src/plan_routes.py +++ b/frontend_multi_user/src/plan_routes.py @@ -755,8 +755,9 @@ def create_plan(): user = db.session.get(UserAccount, uuid.UUID(str(current_user.id))) if not user: return jsonify({"error": "User not found"}), 400 - if to_credit_decimal(user.credits_balance) < Decimal("2"): - return jsonify({"error": "Insufficient credits (minimum 2 required)"}), 402 + min_credits = Decimal(os.environ.get("PLANEXE_MIN_CREDITS_TO_CREATE_PLAN", "2")) + if to_credit_decimal(user.credits_balance) < min_credits: + return jsonify({"error": f"Insufficient credits (minimum {min_credits} required)"}), 402 first_key = ( UserApiKey.query .filter_by(user_id=user.id, revoked_at=None) @@ -949,6 +950,202 @@ def plan(): ) +def _validate_and_clean_import_zip(zip_data: bytes) -> dict: + """Validate and clean an uploaded zip for plan import. + + Returns dict with keys: + error: str or None — error message if invalid + cleaned_zip: bytes or None — cleaned zip data if valid + plan_raw: dict or None — parsed plan_raw.json if present + metadata: dict or None — parsed planexe_metadata.json if present + """ + # Verify it's a valid zip + try: + zf = zipfile.ZipFile(io.BytesIO(zip_data)) + except (zipfile.BadZipFile, Exception) as exc: + return {"error": f"Invalid zip file: {exc}", "cleaned_zip": None, "plan_raw": None, "metadata": None} + + # Build the set of allowed filenames from FilenameEnum + allowed_filenames: set[str] = set() + for member in FilenameEnum: + value = member.value + if "{}" in value: + # Template filenames like "expert_criticism_{}_raw.json" — skip exact match, + # we'll match these by suffix below + continue + allowed_filenames.add(value) + + # Template suffixes for filenames with {} placeholders + template_suffixes: list[tuple[str, str]] = [] + for member in FilenameEnum: + value = member.value + if "{}" in value: + # e.g. "expert_criticism_{}_raw.json" -> "_raw.json" after the placeholder + suffix = value.split("{}")[1] + prefix = value.split("{}")[0] + template_suffixes.append((prefix, suffix)) + + files_to_delete = {e.value for e in ExtraFilenameEnum} + unrecognized = [] + + for info in zf.infolist(): + # Skip directories + if info.is_dir(): + continue + name = info.filename + # Skip path traversal attempts + if ".." in name or name.startswith("/"): + continue + # Strip leading directory (e.g. "run_id/plan.txt" -> "plan.txt") + basename = name.split("/")[-1] if "/" in name else name + # Skip dotfiles (e.g. .DS_Store, ._resource_forks) + if basename.startswith("."): + continue + + if basename in files_to_delete: + continue + if basename in allowed_filenames: + continue + # Check template patterns + matched = False + for prefix, suffix in template_suffixes: + if basename.startswith(prefix) and basename.endswith(suffix): + matched = True + break + if not matched: + unrecognized.append(basename) + + # Rebuild the zip: keep only recognized FilenameEnum files, flatten paths to basenames + skip_files = files_to_delete | set(unrecognized) + out_buf = io.BytesIO() + kept_count = 0 + with zipfile.ZipFile(out_buf, "w", compression=zipfile.ZIP_DEFLATED) as out_zf: + for info in zf.infolist(): + if info.is_dir(): + continue + if info.external_attr >> 16 & 0o170000 == 0o120000: + continue + name = info.filename + if ".." in name or name.startswith("/"): + continue + basename = name.split("/")[-1] if "/" in name else name + if basename.startswith("."): + continue + if basename in skip_files: + continue + # Flatten: write with basename only so extractall puts files directly in run_id_dir + out_zf.writestr(basename, zf.read(info.filename)) + kept_count += 1 + + if unrecognized: + logger.info("Plan import: skipped %d unrecognized files", len(unrecognized)) + + if kept_count == 0: + return {"error": "Zip contains no recognized PlanExe files.", "cleaned_zip": None, "plan_raw": None, "metadata": None} + + # Extract plan_raw.json and planexe_metadata.json if present + plan_raw = None + metadata = None + for info in zf.infolist(): + basename = info.filename.split("/")[-1] if "/" in info.filename else info.filename + if basename == FilenameEnum.INITIAL_PLAN_RAW.value: + try: + plan_raw = json.loads(zf.read(info.filename)) + except (json.JSONDecodeError, Exception): + pass + elif basename == FilenameEnum.PLANEXE_METADATA.value: + try: + metadata = json.loads(zf.read(info.filename)) + except (json.JSONDecodeError, Exception): + pass + zf.close() + + return {"error": None, "cleaned_zip": out_buf.getvalue(), "plan_raw": plan_raw, "metadata": metadata} + + +@plan_routes_bp.route("/plan/resume-from-zip", methods=["GET"]) +@login_required +def plan_resume_from_zip(): + from src.app import _model_profile_options + return render_template("plan_resume_from_zip.html", model_profile_options=_model_profile_options()) + + +@plan_routes_bp.route("/plan/resume-from-zip/upload", methods=["POST"]) +@login_required +def plan_resume_from_zip_upload(): + """JSON API for zip upload. Called via fetch() from the resume-from-zip page.""" + if not current_user.is_admin: + user = _get_current_user_account() + min_credits = Decimal(os.environ.get("PLANEXE_MIN_CREDITS_TO_CREATE_PLAN", "2")) + if user and to_credit_decimal(user.credits_balance) < min_credits: + return jsonify({"error": f"Insufficient credits. At least {min_credits} credits required."}), 402 + + zip_file = request.files.get("zip_file") + if zip_file is None or not zip_file.filename: + return jsonify({"error": "No file selected."}), 400 + if not zip_file.filename.endswith(".zip"): + return jsonify({"error": "Please upload a .zip file."}), 400 + + zip_data = zip_file.read() + zip_size = len(zip_data) + max_zip_size = 10 * 1024 * 1024 # 10 MB + if zip_size > max_zip_size: + return jsonify({"error": f"Zip file too large ({zip_size / 1024 / 1024:.1f} MB). Maximum is {max_zip_size // 1024 // 1024} MB."}), 400 + + result = _validate_and_clean_import_zip(zip_data) + if result["error"]: + return jsonify({"error": result["error"]}), 400 + + try: + user_id = str(current_user.id) + raw_profile = request.form.get("model_profile") + selected_model_profile = normalize_model_profile(raw_profile).value + + # Use plan_raw.json prompt if available, otherwise fallback + plan_raw = result.get("plan_raw") + prompt = plan_raw.get("plan_prompt", "") if plan_raw else "" + if not prompt: + prompt = f"[Imported from {zip_file.filename}]" + + # Use pipeline_version from planexe_metadata.json if available + metadata = result.get("metadata") + snapshot_version = metadata.get("pipeline_version") if metadata else None + effective_version = snapshot_version if snapshot_version is not None else PIPELINE_VERSION + + parameters = { + "trigger_source": "frontend import", + "import_filename": zip_file.filename, + "pipeline_version": effective_version, + "model_profile": selected_model_profile, + "resume": True, + } + if plan_raw and plan_raw.get("pretty_date"): + try: + parsed = datetime.strptime(plan_raw["pretty_date"], "%Y-%b-%d") + parameters["start_date"] = parsed.strftime("%Y-%m-%d") + except (ValueError, TypeError): + pass # Skip unparseable dates + + plan = PlanItem(**{ + "prompt": prompt, + "state": PlanState.pending, + "user_id": user_id, + "parameters": parameters, + "run_zip_snapshot": result["cleaned_zip"], + }) + db.session.add(plan) + db.session.commit() + logger.info( + "Plan import: created plan %s from %r (%s bytes, cleaned %s bytes) for user %s", + plan.id, zip_file.filename, zip_size, len(result["cleaned_zip"]), user_id, + ) + return jsonify({"plan_id": str(plan.id)}), 200 + except Exception as exc: + db.session.rollback() + logger.error("Plan import failed for %r: %s", zip_file.filename, exc) + return jsonify({"error": "Import failed. Please try again."}), 500 + + @plan_routes_bp.route("/plan/stop", methods=["POST"]) @login_required def plan_stop(): diff --git a/frontend_multi_user/templates/plan_list.html b/frontend_multi_user/templates/plan_list.html index 43a0f043..155097c7 100644 --- a/frontend_multi_user/templates/plan_list.html +++ b/frontend_multi_user/templates/plan_list.html @@ -131,7 +131,10 @@ {% endblock %} {% block content %} -

Plans

+
+

Plans

+ Resume from Zip +

Technical queue view · newest first · click row to inspect

{% if plan_rows %} diff --git a/frontend_multi_user/templates/plan_resume_from_zip.html b/frontend_multi_user/templates/plan_resume_from_zip.html new file mode 100644 index 00000000..116e8951 --- /dev/null +++ b/frontend_multi_user/templates/plan_resume_from_zip.html @@ -0,0 +1,261 @@ +{% extends "base.html" %} +{% block title %}Resume from Zip - PlanExe{% endblock %} +{% block head %} + +{% endblock %} + +{% block content %} +
+

Resume from Zip

+

+ Upload a PlanExe zip to resume generation. Edit files inside the zip before uploading to steer the output — changes to early-stage files will propagate downstream. +

+ +
+ How it works +
    +
  1. Download a plan zip from an existing plan.
    2. +
  2. Edit any files you want to change (e.g. rewrite assumptions, adjust the project plan).
    3. +
  3. Upload the modified zip here. PlanExe will skip completed steps and regenerate everything downstream from your edits.
    4. +
+

+ Topological order matters. The pipeline runs in stages — each stage reads from earlier stages. If you edit an early file (like the plan prompt or assumptions), all later stages that depend on it will be regenerated. If you only edit a late file (like the executive summary), only the final report is affected. +

+

+ To force a stage to re-run, delete its output file from the zip. PlanExe will regenerate it using the files that remain. +

+
+ + + +
+
📦
+
Drop a PlanExe zip here, or click to browse
+
.zip only, max 10 MB
+
+
+ + +
+
+ + +{% endblock %} diff --git a/worker_plan_database/app.py b/worker_plan_database/app.py index e486f0a1..5e582b8c 100644 --- a/worker_plan_database/app.py +++ b/worker_plan_database/app.py @@ -360,12 +360,13 @@ def ensure_stopped_state() -> None: the TaskState → PlanState Python rename (proposal 74). Fresh databases created after that rename will have ``planstate``. We try both names. """ - with db.engine.begin() as conn: - for type_name in ("taskstate", "planstate"): + for type_name in ("taskstate", "planstate"): + for enum_value in ("stopped",): try: - conn.execute(text(f"ALTER TYPE {type_name} ADD VALUE IF NOT EXISTS 'stopped'")) + with db.engine.begin() as conn: + conn.execute(text(f"ALTER TYPE {type_name} ADD VALUE IF NOT EXISTS '{enum_value}'")) except Exception as exc: - logger.debug("ALTER TYPE %s: %s", type_name, exc) + logger.debug("ALTER TYPE %s ADD VALUE %s: %s", type_name, enum_value, exc) def worker_process_started() -> None: planexe_worker_id = os.environ.get("PLANEXE_WORKER_ID")