""" ALLI Codegen Subpackage: Built-in Python Libraries Preset Emitters. Keeps runtime/codegen parity for Python Libraries palette cards (`alli.sys.env`, `requests.get`, `statistics.mean`, `pandas.dataframe`, `numpy.array`). """ from typing import Optional, Tuple from lambdas.codegen.context import NodeCodegenCtx from language_core.codegen_contracts import format_literal, sanitize_identifier def emit_python_library_preset_node(nctx: NodeCodegenCtx) -> Tuple[bool, Optional[str]]: """Emit PEP 8 Python code for built-in Python Libraries preset cards.""" raw_type, node_id, safe_id, params, cg = nctx.raw_type, nctx.node_id, nctx.safe_id, nctx.params, nctx.cg lines, ovm = cg.code_body_lines, cg.output_var_map if raw_type == "alli.primitive.datetime": cg.discovered_modules.add("datetime") op = str(params.get("operation") or "strftime").lower() fmt = str(params.get("format_pattern") or "%Y-%m-%d %H:%M:%S") nv = f"dt_{safe_id}" if op in ("now", "utcnow", "isoformat", "today"): lines.append(f" {nv} = datetime.datetime.now(datetime.timezone.utc).{'date().' if op == 'today' else ''}isoformat()") elif op == "timedelta": days = int(params.get("days") or 1) lines.append(f" {nv} = (datetime.datetime.now(datetime.timezone.utc) + datetime.timedelta(days={days})).isoformat()") else: lines.append(f" {nv} = datetime.datetime.now(datetime.timezone.utc).strftime({format_literal(fmt)})") ovm.update({f"{node_id}:result": nv, f"{node_id}:iso": nv, f"{node_id}:val": nv, f"{node_id}:out": nv, node_id: nv}) return True, nv if raw_type in ("alli.sys.env", "sys.env_get", "sys.env", "sys.env_set"): from lambdas.codegen.wave1_core_emitters import emit_safe_sys_env return emit_safe_sys_env(nctx) if raw_type == "requests.get": cg.discovered_modules.add("requests") cg.discovered_requirements.add("requests") url_expr = nctx.resolve_input("url", format_literal(str(params.get("url") or "https://api.example.com/data"))) nv = f"res_{safe_id}" lines.append(f" {nv} = requests.get({url_expr}, timeout=5).json()") ovm.update({f"{node_id}:result": nv, f"{node_id}:response": nv, f"{node_id}:val": nv, f"{node_id}:out": nv, node_id: nv}) return True, nv if raw_type == "statistics.mean": cg.discovered_modules.add("statistics") func = str(params.get("func") or "mean").strip().lower() data_expr = nctx.resolve_input("data", format_literal(params.get("data") or [10, 20, 30])) nv = f"stat_{safe_id}" lines.append(f" {nv} = statistics.{func}({data_expr})") ovm.update({f"{node_id}:result": nv, f"{node_id}:val": nv, f"{node_id}:out": nv, node_id: nv}) return True, nv if raw_type == "pandas.dataframe": cg.discovered_modules.add("pandas") cg.discovered_requirements.add("pandas") nv = f"df_{safe_id}" func = str(params.get("func") or "read_csv") fp = format_literal(str(params.get("filepath") or "data.csv")) lines.append( f" {nv} = pandas.read_csv({fp})" if func == "read_csv" and not nctx.incoming else f" {nv} = pandas.DataFrame({nctx.resolve_input('data', '[]')})" ) ovm.update({f"{node_id}:result": nv, f"{node_id}:rows": nv, f"{node_id}:val": nv, f"{node_id}:out": nv, node_id: nv}) return True, nv if raw_type == "numpy.array": cg.discovered_modules.add("numpy") cg.discovered_requirements.add("numpy") nv = f"arr_{safe_id}" dtype = format_literal(str(params.get("dtype") or "float64")) if nctx.incoming or params.get("values") is not None: vals_expr = nctx.resolve_input("data", format_literal(params.get("values") or [0.0, 0.0])) lines.append(f" {nv} = numpy.array({vals_expr}, dtype={dtype})") else: lines.append(f" {nv} = numpy.zeros((2, 2), dtype={dtype})") ovm.update({f"{node_id}:result": nv, f"{node_id}:val": nv, f"{node_id}:out": nv, node_id: nv}) return True, nv return False, None