diff --git a/.github/workflows/ci.yml b/.github/workflows/ci.yml index fdb4ff48f..a019a2604 100644 --- a/.github/workflows/ci.yml +++ b/.github/workflows/ci.yml @@ -2,12 +2,18 @@ name: T3 CI on: workflow_dispatch: + push: + branches: [main] pull_request: branches: [main] types: [opened, synchronize, reopened] schedule: - cron: '0 0 * * *' +concurrency: + group: ${{ github.workflow }}-${{ github.ref }} + cancel-in-progress: true + jobs: lint: name: Lint (ruff) diff --git a/.vscode/launch.json b/.vscode/launch.json deleted file mode 100644 index 2436e2308..000000000 --- a/.vscode/launch.json +++ /dev/null @@ -1,17 +0,0 @@ -{ - // Use IntelliSense to learn about possible attributes. - // Hover to view descriptions of existing attributes. - // For more information, visit: https://go.microsoft.com/fwlink/?linkid=830387 - "version": "0.2.0", - "configurations": [ - { - "name": "Python: Current File", - "type": "python", - "request": "launch", - "program": "${file}", - "env": {"PYDEVD_WARN_EVALUATION_TIMEOUT": "500"}, - "console": "integratedTerminal", - "justMyCode": true - } - ] -} \ No newline at end of file diff --git a/.vscode/settings.json b/.vscode/settings.json deleted file mode 100644 index 9b388533a..000000000 --- a/.vscode/settings.json +++ /dev/null @@ -1,7 +0,0 @@ -{ - "python.testing.pytestArgs": [ - "tests" - ], - "python.testing.unittestEnabled": false, - "python.testing.pytestEnabled": true -} \ No newline at end of file diff --git a/CODE_OF_CONDUCT.md b/CODE_OF_CONDUCT.md index 1fb726ea8..abdcc4ff9 100644 --- a/CODE_OF_CONDUCT.md +++ b/CODE_OF_CONDUCT.md @@ -1,4 +1,4 @@ -# The RMG-ARC Tandem Tool (T3) Code of Conduct +# T3 Code of Conduct ## Our Pledge diff --git a/Makefile b/Makefile index 2901f4866..aa6b21063 100644 --- a/Makefile +++ b/Makefile @@ -10,7 +10,7 @@ test: pytest tests/ --cov -ra -vv test-main: - pytest tests/test_main.py -ra -vv + pytest tests/test_main.py --cov -ra -vv test-functional: - pytest tests/test_functional.py -ra -vv + pytest tests/test_functional.py --cov -ra -vv diff --git a/README.md b/README.md index a1c0b8f07..c039c7b1a 100644 --- a/README.md +++ b/README.md @@ -1,8 +1,3 @@ -

- T3 -

- -# The Tandem Tool (T3) for automated chemical kinetic model development [![CI](https://github.com/ReactionMechanismGenerator/T3/actions/workflows/ci.yml/badge.svg?branch=main)](https://github.com/ReactionMechanismGenerator/T3/actions/workflows/ci.yml) [![Docs](https://github.com/ReactionMechanismGenerator/T3/actions/workflows/gh-pages.yml/badge.svg)](https://reactionmechanismgenerator.github.io/T3/) @@ -11,6 +6,8 @@ [![MIT license](http://img.shields.io/badge/license-MIT-brightgreen.svg)](http://opensource.org/licenses/MIT) ![python](https://img.shields.io/badge/Python-3.12+-blue.svg) +# The Tandem Tool (T3) for automated chemical kinetic model development + **T3** automates the development of detailed chemical kinetic models. Given a set of initial species and conditions, it produces a validated model with high-fidelity thermochemistry and rate parameters by iteratively combining diff --git a/pyproject.toml b/pyproject.toml index 423163ae7..ed8ee7efd 100644 --- a/pyproject.toml +++ b/pyproject.toml @@ -38,6 +38,12 @@ ignore = [ "W291", # trailing whitespace ] +[tool.pytest.ini_options] +filterwarnings = [ + "ignore::DeprecationWarning", + "ignore::PendingDeprecationWarning", +] + [tool.ruff.lint.per-file-ignores] "__init__.py" = ["F401"] # unused imports in __init__.py are re-exports "t3/runners/rmg_incore_sa.py" = ["E701"] # subprocess script, compact one-liners are fine diff --git a/pytest.ini b/pytest.ini deleted file mode 100644 index df291533a..000000000 --- a/pytest.ini +++ /dev/null @@ -1,4 +0,0 @@ -[pytest] -filterwarnings = - ignore::DeprecationWarning - ignore::PendingDeprecationWarning diff --git a/t3/main.py b/t3/main.py index 4b681ae19..a964348df 100755 --- a/t3/main.py +++ b/t3/main.py @@ -582,6 +582,7 @@ def run_rmg(self, restart_rmg: bool = False): t3_project_name=self.project, rmg_execution_type=self.rmg['rmg_execution_type'], restart_rmg=restart_rmg, + walltime=self.t3['options']['max_RMG_walltime'], ) if rmg_exception_encountered: self.rmg_exceptions_counter += 1 @@ -1151,13 +1152,13 @@ def get_reaction_key(self, try: if label == t3_reaction.get_reaction_smiles_label(): return key - except Exception: + except (AttributeError, ValueError): pass elif label_type == 'Chemkin': try: if label == t3_reaction.to_chemkin(): return key - except Exception: + except (AttributeError, ValueError): pass return None diff --git a/t3/runners/rmg_runner.py b/t3/runners/rmg_runner.py index 4a4157da6..b48781671 100644 --- a/t3/runners/rmg_runner.py +++ b/t3/runners/rmg_runner.py @@ -4,9 +4,11 @@ """ import datetime +import logging import os import shlex import shutil +import subprocess import time from typing import TYPE_CHECKING, List, Optional, Tuple @@ -161,9 +163,27 @@ def rmg_job_converged(project_directory: str) -> Tuple[bool, Optional[str]]: return rmg_converged, error +_DEFAULT_RMG_TIMEOUT_S = 6 * 3600 # 6 hours + +logger = logging.getLogger(__name__) + + +def _parse_walltime_to_seconds(walltime: str) -> int: + """Parse a 'DD:HH:MM:SS' walltime string to total seconds. Returns 0 for '00:00:00:00'.""" + parts = walltime.split(':') + if len(parts) != 4: + return 0 + try: + days, hours, minutes, seconds = (int(p) for p in parts) + except ValueError: + return 0 + return days * 86400 + hours * 3600 + minutes * 60 + seconds + + def run_rmg_incore(rmg_input_file_path: str, verbose: Optional[int] = None, max_iterations: Optional[int] = None, + walltime: Optional[str] = None, ) -> bool: """ Run RMG incore under the rmg_env. @@ -172,10 +192,14 @@ def run_rmg_incore(rmg_input_file_path: str, rmg_input_file_path (str): The path to the RMG input file. max_iterations (int, optional): Max RMG iterations. verbose (int, optional): Level of verbosity. + walltime (str, optional): Max walltime in 'DD:HH:MM:SS' format. Defaults to 6 hours. Returns: bool: Whether an exception was raised. """ + timeout_s = _parse_walltime_to_seconds(walltime) if walltime else 0 + if timeout_s <= 0: + timeout_s = _DEFAULT_RMG_TIMEOUT_S project_directory = os.path.abspath(os.path.dirname(rmg_input_file_path)) verbose = f' -v {verbose}' if verbose is not None else '' max_iterations = f' -m {max_iterations}' if max_iterations is not None else '' @@ -192,8 +216,16 @@ def run_rmg_incore(rmg_input_file_path: str, echo "Micromamba/Mamba/Conda required" >&2 exit 1 fi' ''' - stdout, stderr = execute_command(shell_script, shell=True, no_fail=True, executable='/bin/bash') - stderr_text = ''.join(stderr) if isinstance(stderr, list) else (stderr or '') + try: + result = subprocess.run(shell_script, shell=True, executable='/bin/bash', + capture_output=True, text=True, timeout=timeout_s) + stderr_text = result.stderr or '' + except subprocess.TimeoutExpired: + logger.error(f'RMG incore timed out after {timeout_s}s') + return True + if result.returncode != 0: + logger.error(f'RMG incore exited with code {result.returncode}') + return True if 'RMG threw an exception and did not converge.' in stderr_text: return True return False @@ -271,6 +303,7 @@ def rmg_runner(rmg_input_file_path: str, t3_project_name: Optional[str] = None, rmg_execution_type: Optional[str] = None, restart_rmg: bool = False, + walltime: Optional[str] = None, ) -> bool: """ Run an RMG job as a subprocess under the rmg_env. @@ -286,6 +319,7 @@ def rmg_runner(rmg_input_file_path: str, t3_project_name (str, optional): The T3 project name, used for setting a job name on the server for the RMG run. rmg_execution_type (str, optional): The RMG execution type (incore or local). Also set via settings.py. restart_rmg (bool, optional): Whether to restart RMG from seed. + walltime (str, optional): Max walltime in 'DD:HH:MM:SS' format. Defaults to 6 hours. Returns: bool: Whether an exception was raised. @@ -299,6 +333,7 @@ def rmg_runner(rmg_input_file_path: str, rmg_exception_encountered = run_rmg_incore(rmg_input_file_path=rmg_input_file_path, verbose=verbose, max_iterations=max_iterations, + walltime=walltime, ) return rmg_exception_encountered elif rmg_execution_type == 'local': diff --git a/t3/schema.py b/t3/schema.py index 1dd2b6ee6..0810856d7 100644 --- a/t3/schema.py +++ b/t3/schema.py @@ -477,7 +477,7 @@ class Config: @classmethod def check_units(cls, value): """RMGOptions.units validator""" - if value.lower() is not None and value != 'si': + if value.lower() != 'si': raise ValueError(f'Currently RMG only supports SI units, got "{value}"') return value.lower() diff --git a/t3/simulate/cantera_base.py b/t3/simulate/cantera_base.py index 07cece2ec..96ad652d6 100644 --- a/t3/simulate/cantera_base.py +++ b/t3/simulate/cantera_base.py @@ -8,6 +8,7 @@ different reactor configuration. """ +import logging import cantera as ct import numpy as np from abc import abstractmethod @@ -331,6 +332,9 @@ def simulate(self): non_inert_mask = np.ones(self.num_ct_species, dtype=bool) non_inert_mask[self.inert_index_list] = False + if not np.all(np.isfinite(mass_frac_sa)): + logging.getLogger(__name__).warning( + 'NaN/inf detected in mass fraction SA matrix — check for numerical issues') kin_correction = mass_frac_sa[non_inert_mask, :self.num_ct_reactions].sum(axis=0) thermo_correction = mass_frac_sa[non_inert_mask, self.num_ct_reactions:].sum(axis=0) @@ -462,9 +466,17 @@ def get_idt_by_T(self): time, data_list, reaction_sensitivity_data, thermodynamic_sensitivity_data = condition_data T_data = data_list[0] - dTdt = np.diff(T_data.data) / np.diff(time.data) - idt_dict['idt_index'].append(int(np.argmax(dTdt))) - idt_dict['idt'].append(time.data[idt_dict['idt_index'][i]]) + dt = np.diff(time.data) + dt = np.where(dt == 0, np.finfo(float).tiny, dt) + dTdt = np.diff(T_data.data) / dt + valid = np.isfinite(dTdt) + if valid.any(): + masked = np.where(valid, dTdt, -np.inf) + idx = int(np.argmax(masked)) + else: + idx = 0 + idt_dict['idt_index'].append(idx) + idt_dict['idt'].append(time.data[min(idx + 1, len(time.data) - 1)]) return idt_dict diff --git a/t3/simulate/cantera_jsr.py b/t3/simulate/cantera_jsr.py index c7f0e34e8..ea2b590df 100644 --- a/t3/simulate/cantera_jsr.py +++ b/t3/simulate/cantera_jsr.py @@ -114,6 +114,8 @@ def reinitialize_simulation(self, T0=None, P0=None, X0=None, V0=None): self._jsr_exhaust = ct.Reservoir(self.model) self.cantera_reactor = ct.IdealGasReactor(self.model, energy='off', volume=VOLUME) + if residence_time <= 0: + raise ValueError(f'Invalid residence time: {residence_time}') self._jsr_mfc = ct.MassFlowController( upstream=self._jsr_inlet, downstream=self.cantera_reactor, diff --git a/t3/simulate/cantera_pfr.py b/t3/simulate/cantera_pfr.py index dcc1d1982..411daec2c 100644 --- a/t3/simulate/cantera_pfr.py +++ b/t3/simulate/cantera_pfr.py @@ -129,6 +129,7 @@ def _compute_distance(self): # Velocity: u = u_0 * mean_MW_0 / mean_MW (P, T, A all cancel) total_residence_time = self.conditions[idx].reaction_time.value_si u_0 = LENGTH / total_residence_time + mean_MW = np.where(mean_MW > 0, mean_MW, np.finfo(float).tiny) u = u_0 * mean_MW[0] / mean_MW # Axial distance [m] via cumulative integration @@ -203,6 +204,8 @@ def _simulate_chain(self): sim.advance_to_steady_state() # Residence time in this cell + if mass_flow_rate <= 0: + raise ValueError(f'Invalid mass flow rate: {mass_flow_rate}') cell_residence_time = reactor.mass / mass_flow_rate cumulative_time += cell_residence_time diff --git a/t3/simulate/cantera_pfr_t_profile.py b/t3/simulate/cantera_pfr_t_profile.py index 94435c40e..dfb0e5d04 100644 --- a/t3/simulate/cantera_pfr_t_profile.py +++ b/t3/simulate/cantera_pfr_t_profile.py @@ -107,6 +107,11 @@ T_HOT = 900.0 # K, isothermal plateau temperature T_OUTLET = 500.0 # K, outlet temperature +if not (0 < RAMP_UP_END < ISO_END < LENGTH): + raise ValueError( + f'Temperature profile constants must satisfy 0 < RAMP_UP_END ({RAMP_UP_END}) < ISO_END ({ISO_END}) < LENGTH ({LENGTH})' + ) + def temperature_profile(z): """ @@ -307,7 +312,11 @@ def simulate(self): # Local velocity at the new (T, rho) and the corresponding # time step to traverse one segment. rho = self.model.density + if rho * AREA <= 0: + raise ValueError(f'Invalid density ({rho}) or area ({AREA}) in PFR T-profile cell {n}') u = mass_flow_rate / (rho * AREA) + if u <= 0: + raise ValueError(f'Invalid velocity ({u}) in PFR T-profile cell {n}') dt = dz / u sim.advance(sim.time + dt) diff --git a/t3/simulate/rmg_constant_tp.py b/t3/simulate/rmg_constant_tp.py index 42b917951..8c196e831 100755 --- a/t3/simulate/rmg_constant_tp.py +++ b/t3/simulate/rmg_constant_tp.py @@ -5,6 +5,7 @@ import datetime import itertools +import logging import os import numpy as np from typing import List, Optional, TYPE_CHECKING @@ -279,6 +280,12 @@ def get_species_concentration_lists_from_ranged_params(self) -> List[List[dict]] species_lists.append(new_species_list) for species_list in species_lists: + total = sum(s['concentration'] for s in species_list + if isinstance(s['concentration'], (int, float))) + if total > 1.0 or total <= 0: + logging.getLogger(__name__).warning( + f'Species concentrations sum to {total:.4f}, expected (0, 1]. ' + f'Cantera will normalize mole fractions.') species_list.sort(key=lambda spc: spc['concentration'][0] if isinstance(spc['concentration'], (tuple, list)) else spc['concentration'], reverse=True) return species_lists diff --git a/t3/utils/libraries.py b/t3/utils/libraries.py index 66bd04b63..51e7ad63c 100644 --- a/t3/utils/libraries.py +++ b/t3/utils/libraries.py @@ -21,8 +21,8 @@ from t3.logger import Logger -_LOCK_TIMEOUT_S = 3600 # 1 hour wait for lock acquisition -_LOCK_STALE_S = 3600 # 1 hour before breaking a stale lock +_LOCK_TIMEOUT_S = 300 # 5 min wait for lock acquisition +_LOCK_STALE_S = 300 # 5 min before breaking a stale lock _LOCK_POLL_S = 10.0 # Poll every 10 seconds diff --git a/t3/utils/rmg_sa_parser.py b/t3/utils/rmg_sa_parser.py index 3a7da3553..24a9daa5b 100644 --- a/t3/utils/rmg_sa_parser.py +++ b/t3/utils/rmg_sa_parser.py @@ -93,7 +93,10 @@ def parse_rmg_sa_csv(csv_path: str) -> Tuple[List[float], Dict[str, Dict[int, Li value = float(row[col_idx]) if kind == 'kinetics': # param_raw is e.g. 'k10' -> extract integer 10 - rxn_idx = int(param_raw.lstrip('k')) + try: + rxn_idx = int(param_raw.lstrip('k')) + except ValueError: + continue kinetics.setdefault(obs_label, {}).setdefault(rxn_idx, []).append(value) else: spc_label = _strip_rmg_index(param_raw) diff --git a/t3/utils/rmg_shim.py b/t3/utils/rmg_shim.py index d95e0e79b..7e48ff2e5 100644 --- a/t3/utils/rmg_shim.py +++ b/t3/utils/rmg_shim.py @@ -5,6 +5,7 @@ in-process dependency on rmgpy itself. """ +import logging import os import tempfile from dataclasses import dataclass, field @@ -690,6 +691,9 @@ def write_atomic(path: str, content: str) -> None: f.write(content) os.replace(tmp_path, path) except Exception: - if os.path.exists(tmp_path): - os.unlink(tmp_path) + try: + if os.path.exists(tmp_path): + os.unlink(tmp_path) + except OSError: + logging.getLogger(__name__).warning(f'Failed to clean up temp file: {tmp_path}') raise diff --git a/t3/utils/writer.py b/t3/utils/writer.py index da2e312bf..06f351881 100644 --- a/t3/utils/writer.py +++ b/t3/utils/writer.py @@ -42,11 +42,11 @@ def write_rmg_input_file(rmg: dict, # database database = rmg['database'] # the following args type could be either str or list, detect str and format accordingly - if isinstance(database['kinetics_depositories'], str) and database['kinetics_depositories'][0] != "'": + if isinstance(database['kinetics_depositories'], str) and database['kinetics_depositories'] and database['kinetics_depositories'][0] != "'": database['kinetics_depositories'] = f"'{database['kinetics_depositories']}'" - if isinstance(database['kinetics_estimator'], str) and database['kinetics_estimator'][0] != "'": + if isinstance(database['kinetics_estimator'], str) and database['kinetics_estimator'] and database['kinetics_estimator'][0] != "'": database['kinetics_estimator'] = f"'{database['kinetics_estimator']}'" - if isinstance(database['kinetics_families'], str) and database['kinetics_families'][0] != "'": + if isinstance(database['kinetics_families'], str) and database['kinetics_families'] and database['kinetics_families'][0] != "'": database['kinetics_families'] = f"'{database['kinetics_families']}'" database_template = """database( thermoLibraries=${thermo_libraries}, @@ -374,25 +374,30 @@ def write_pdep_network_file(network_name: str, if 'reactants =' in line: parse_isomers = (False, False) if parse_tp: - if 'Tmin' in line: + if 'Tmin' in line and '(' in line: # Tmin = (300, 'K'), t_min = line.split('(')[1].split(',')[0] - elif 'Tmax' in line: + elif 'Tmax' in line and '(' in line: # Tmax = (2200, 'K'), t_max = line.split('(')[1].split(',')[0] - elif 'Pmin' in line: + elif 'Pmin' in line and '(' in line: # Pmin = (0.01, 'bar'), p_min = line.split('(')[1].split(',')[0] - elif 'Pmax' in line: + elif 'Pmax' in line and '(' in line: # Pmax = (100, 'bar'), p_max = line.split('(')[1].split(',')[0] - if all(parse_isomers) and "'," in line: + if all(parse_isomers) and "'," in line and "'" in line: # 'C=O(26)', - isomer_labels.append(line.split("'")[1]) + parts = line.split("'") + if len(parts) >= 2: + isomer_labels.append(parts[1]) if 'method = ' in line: # method = 'chemically-significant eigenvalues', splits = line.split("'") - new_lines.append(f"{splits[0]}'{METHOD_MAP[method]}'{splits[2]}") + if len(splits) >= 3: + new_lines.append(f"{splits[0]}'{METHOD_MAP[method]}'{splits[2]}") + else: + new_lines.append(line) elif 'rmgmode' in line: new_lines.append(line) if any(param is None for param in [t_min, t_max, p_min, p_max]): diff --git a/tests/test_main.py b/tests/test_main.py index 9a7ef3e64..9436fb340 100644 --- a/tests/test_main.py +++ b/tests/test_main.py @@ -1017,7 +1017,6 @@ def test_add_reaction(): set_paths=True, ) rmg_species, rmg_reactions = t3.load_species_and_reactions_from_yaml_file() - print(f"DEBUG: len(rmg_reactions) = {len(rmg_reactions)}") # Filter for valid/balanced reactions to use in test valid_reactions = [] @@ -1031,7 +1030,6 @@ def test_add_reaction(): except Exception: pass - print(f"DEBUG: len(valid_reactions) = {len(valid_reactions)}") assert len(valid_reactions) > 0, "No valid balanced reactions found in the test dataset!" # Use valid reactions for adding