Bugfix: MOPAC files with references in comments

  
* Fixed a bug that caused a crash when reading MOPAC files with references in the
  comments.
* Updated the MOPAC reader to the new approach for running MOPAC in the cases that
  it is needed: Z-matrices and mixed inputs that OpenBabel can't handle.

HISTORY.rst

@@ -1,6 +1,12 @@
 =======
 History
 =======
+2024.11.3 -- Bugfix: MOPAC files with references in comments
+  * Fixed a bug that caused a crash when reading MOPAC files with references in the
+    comments.
+  * Updated the MOPAC reader to the new approach for running MOPAC in the cases that
+    it is needed: Z-matrices and mixed inputs that OpenBabel can't handle.
+
 2024.8.23 -- Enhancements to directory handling
   * Changed the handling of paths to make them relative to the directory that the step
     is running in. In a loop, for instance, files are relative to the iteration

read_structure_step/formats/mop/find_mopac.py

@@ -1,7 +1,8 @@
-import seamm_util
 from pathlib import Path
 import os
 
+import seamm_util
+
 mopac_error_identifiers = []
 
 

read_structure_step/formats/mop/obabel.py

@@ -16,9 +16,8 @@
 
 from openbabel import openbabel
 from read_structure_step.formats.registries import register_reader
-import seamm
 from seamm_util import Q_
-from .find_mopac import find_mopac
+from .find_mopac import find_mopac  # noqa: F401
 
 if "OpenBabel_version" not in globals():
     OpenBabel_version = None
@@ -196,6 +195,7 @@ def load_mop(
     references=None,
     bibliography=None,
     save_data=True,
+    step=None,
     **kwargs,
 ):
     """Read a MOPAC input file.
@@ -405,10 +405,6 @@ def load_mop(
         except Exception:
             logger.info("**** falling back to MOPAC")
             # Try using a MOPAC output file instead. Works for e.g. mixed coordinates
-            mopac_exe = find_mopac()
-            if mopac_exe is None:
-                raise FileNotFoundError("The MOPAC executable could not be found")
-
             # Create an input file
             text = ["0SCF", "title", "description"]
             text.extend(raw_geometry_lines)
@@ -418,10 +414,7 @@ def load_mop(
 
             logger.debug(f"MOPAC input file:\n\n{files['mopac.dat']}\n")
 
-            local = seamm.ExecLocal()
-            result = local.run(
-                cmd=[mopac_exe, "mopac.dat"], files=files, return_files=["mopac.out"]
-            )
+            result = step.run_mopac(files=files, return_files=["mopac.out"])
 
             if result["mopac.out"]["data"] is None:
                 raise RuntimeError("MOPAC failed: " + result["mopac.out"]["exception"])
@@ -642,6 +635,7 @@ def load_mop(
                             keyword = metadata[keyword]
                             if value == "":
                                 print(f"Value for {keyword} missing in MOPAC .mop file")
+                                print("\n\t".join(description_lines))
                                 continue
                             if "reference" in keyword:
                                 description = keyword.split(".")[0]
@@ -672,7 +666,7 @@ def load_mop(
                                 ):
                                     stderr = float(stderr) * kcal2kJ
                                 system_properties.put(new_keyword, stderr)
-                            if (
+                            if "reference" not in keyword and (
                                 "heat capacity" in keyword
                                 or "enthalpy" in keyword
                                 or "entropy" in keyword

read_structure_step/read.py

@@ -21,6 +21,7 @@ def read(
     printer=None,
     references=None,
     bibliography=None,
+    step=None,
 ):
     """
     Calls the appropriate functions to parse the requested file.
@@ -70,6 +71,9 @@ def read(
     bibliography : dict
         The bibliography as a dictionary.
 
+    step : seamm.Node = None
+        The node in the flowchart, used for running e.g. MOPAC.
+
     Returns
     -------
     [Configuration]
@@ -124,6 +128,7 @@ def read(
         printer=printer,
         references=references,
         bibliography=bibliography,
+        step=step,
     )
 
     return configurations

read_structure_step/read_structure.py

@@ -11,9 +11,13 @@
 directory, and is used for all normal output from this step.
 """
 
+import configparser
+import importlib
 import logging
+import os
 from pathlib import PurePath, Path
 import pprint  # noqa: F401
+import shutil
 import tarfile
 import tempfile
 import textwrap
@@ -23,7 +27,7 @@
 from .read import read
 import seamm
 from seamm_util import ureg, Q_  # noqa: F401
-from seamm_util import getParser
+from seamm_util import Configuration, getParser
 import seamm_util.printing as printing
 from seamm_util.printing import FormattedText as __
 from .utils import guess_extension
@@ -259,6 +263,7 @@ def run(self):
                 printer=printer.important,
                 references=self.references,
                 bibliography=self._bibliography,
+                step=self,
             )
 
             # Finish the output
@@ -371,6 +376,7 @@ def read_tarfile(self, tarfile_path, P):
                         printer=printer.important,
                         references=self.references,
                         bibliography=self._bibliography,
+                        step=self,
                     )
 
                     tmp_path.unlink()
@@ -384,3 +390,75 @@ def read_tarfile(self, tarfile_path, P):
                 indent=4 * " ",
             )
         )
+
+    def run_mopac(self, files=None, return_files=["mopac.out"]):
+        """Run MOPAC to parse the input file."""
+
+        import mopac_step
+
+        # Access the options
+        seamm_options = self.global_options
+
+        executor = self.flowchart.executor
+
+        # Read configuration file for MOPAC if it exists
+        executor_type = executor.name
+        full_config = configparser.ConfigParser()
+        ini_dir = Path(seamm_options["root"]).expanduser()
+        path = ini_dir / "mopac.ini"
+        # If the config file doesn't exists, get the default
+        if not path.exists():
+            resources = importlib.resources.files("mopac_step") / "data"
+            ini_text = (resources / "mopac.ini").read_text()
+            txt_config = Configuration(path)
+            txt_config.from_string(ini_text)
+
+            # Work out the conda info needed
+            txt_config.set_value("local", "conda", os.environ["CONDA_EXE"])
+            txt_config.set_value("local", "conda-environment", "seamm-mopac")
+            txt_config.save()
+
+        full_config.read(ini_dir / "mopac.ini")
+
+        # Getting desperate! Look for an executable in the path
+        if executor_type not in full_config:
+            path = shutil.which("mopac")
+            if path is None:
+                raise RuntimeError(
+                    f"No section for '{executor_type}' in MOPAC ini file "
+                    f"({ini_dir / 'mopac.ini'}), nor in the defaults, nor "
+                    "in the path!"
+                )
+            else:
+                txt_config = Configuration(path)
+                txt_config.add_section(executor_type)
+                txt_config.set_value(executor_type, "installation", "local")
+                txt_config.set_value(executor_type, "code", str(path))
+                txt_config.save()
+                full_config.read(ini_dir / "mopac.ini")
+
+        config = dict(full_config.items(executor_type))
+
+        # Use the matching version of the seamm-mopac image by default.
+        config["version"] = mopac_step.__version__
+
+        env = {
+            "OMP_NUM_THREADS": "1",
+        }
+
+        result = executor.run(
+            cmd=["{code}", "mopac.dat", ">", "stdout.txt", "2>", "stderr.txt"],
+            config=config,
+            directory=self.directory,
+            files=files,
+            return_files=return_files,
+            in_situ=True,
+            shell=True,
+            env=env,
+        )
+
+        if not result:
+            self.logger.error("There was an error running MOPAC")
+            return None
+
+        return result