orto 1.8.2__tar.gz → 1.8.3__tar.gz

{orto-1.8.2 → orto-1.8.3}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: orto
-Version: 1.8.2
+Version: 1.8.3
 Summary: A package to make life easier when performing Orca calculations.
 Home-page: https://orto.kragskow.group
 Author: Jon Kragskow

orto-1.8.3/orto/__version__.py

@@ -0,0 +1 @@
+__version__ = '1.8.3'

{orto-1.8.2 → orto-1.8.3}/orto/cli.py

@@ -527,6 +527,14 @@ def gen_job_func(uargs):
         else:
             nbo_module = None
 
+        # Check structure in input file or xyz file
+        # unless skip_structure flag is set
+        if not uargs.skip_structure:
+            try:
+                inp.check_structure(oj.input_file)
+            except (DataNotFoundError, DataFormattingError) as e:
+                ut.red_exit(str(e))
+
         # Set SLURM error and output file names
         config['error'] = 'slurm.%j.e'
         config['output'] = 'slurm.%j.o'
@@ -1880,6 +1888,15 @@ def read_args(arg_list=None):
         )
     )
 
+    gen_job.add_argument(
+        '--skip_structure',
+        '-ss',
+        action='store_true',
+        help=(
+            'Disables checking of structure in input file or xyz file'
+        )
+    )
+
     gen_job.add_argument(
         '-om',
         '--orca_load',

{orto-1.8.2 → orto-1.8.3}/orto/extractor.py

@@ -2204,6 +2204,85 @@ class XYZInputExtractor(extto.LineExtractor):
         return _ext.data
 
 
+class StructureInputExtractor(extto.BetweenExtractor):
+    '''
+    Extracts structure from xyz file
+    '''
+
+    # Regex Start Pattern
+    START_PATTERN = rb'(?<=\*xyz )'
+
+    # Regex End Pattern
+    END_PATTERN = rb'(?=\*)'
+
+    @property
+    def data(self) -> tuple[list[str], NDArray]:
+        '''
+        list of atom labels and ndarray of coordinates (N, 3)
+        '''
+        return self._data
+
+    @staticmethod
+    def _process_block(block: str) -> tuple[list[str], NDArray]:
+        '''
+        Converts single block into data entries described in self.data
+
+        Parameters
+        ----------
+        block: str
+            String block extracted from file
+
+        Returns
+        -------
+        tuple[list[str], NDArray]
+        '''
+
+        # Get labels and coordinates
+        all_info = re.findall(
+            r'([A-Za-z]+\s*-?\d\.\d*\s+-?\d\.\d*\s+-?\d\.\d*)',
+            block
+        )
+
+        labels = re.findall(
+            r'([A-Za-z]+)',
+            block
+        )
+
+        coords = np.asarray([
+            [float(v) for v in val.split()[1:]]
+            for val in all_info
+        ])
+
+        if len(labels) != coords.shape[0]:
+            raise DataFormattingError(
+                'Number of atom labels does not match number of coordinates'
+            )
+
+        data = (labels, coords)
+
+        return data
+
+    @classmethod
+    def extract(cls, file_name: str | pathlib.Path) -> tuple[list[str], NDArray]: # noqa
+        '''
+        Convenience method which instantiates class, extracts blocks, and
+        returns processed datasets
+
+        Parameters
+        ----------
+        file_name: str | pathlib.Path
+            File to parse
+
+        Returns
+        -------
+        tuple[list[str], NDArray]
+            Each entry contains processed data, as defined in cls.data
+        '''
+        _ext = cls()
+        _ext(file_name, process=True)
+        return _ext.data
+
+
 class IntInputExtractor(XYZInputExtractor):
     '''
     Extracts *int line of an input file

{orto-1.8.2 → orto-1.8.3}/orto/input.py

@@ -1,6 +1,7 @@
 from pathlib import Path
 import re
 import numpy as np
+import numpy.linalg as la
 import copy
 import xyz_py as xyzp
 
@@ -304,6 +305,77 @@ def check_xyz(file_name: str | Path, skip_check) -> None:
     return
 
 
+def check_structure(file_name: str | Path, threshold: float = 0.1) -> None:
+    '''
+    Checks that no atoms of the structure are overlapping.\n
+
+    Parameters
+    ----------
+    file_name: str | Path
+        Orca input file as either name or Path object
+    threshold: float
+        Minimum allowed distance between atoms in Angstroms
+
+    Returns
+    -------
+    None
+
+    Raises
+    ------
+    DataFormattingError
+        If any atoms are overlapping
+    '''
+
+    # Get xyz file name and check it exists and is formatted correctly
+    try:
+        xyz_file = oe.XYZFileInputExtractor.extract(file_name)
+    except DataNotFoundError:
+        xyz_file = []
+
+    try:
+        xyzline = oe.XYZInputExtractor.extract(file_name)
+    except DataNotFoundError:
+        xyzline = []
+
+    if not len(xyz_file) and not len(xyzline):
+        ut.red_exit(
+            'Error: missing or incorrect *xyzfile or *xyz line in input'
+        )
+
+    if len(xyz_file):
+        # Load structure from xyz file
+        xyz_file = Path(xyz_file[0])
+        labels, coords = xyzp.load_xyz(xyz_file)
+    else:
+        # Extract structure from input file
+        labels, coords = oe.StructureInputExtractor.extract(file_name)[0]
+
+    # Compute pairwise differences using broadcasting
+    # Broadcast to shape (N, N, 3)
+    # from (N, 1, 3) and (1, N, 3)
+    diffs = coords[:, np.newaxis, :] - coords[np.newaxis, :, :]
+
+    # Compute Euclidean distances
+    distances = la.norm(diffs, axis=-1)  # shape (N, N)
+
+    # Add large value to diagonal to avoid zero self-distances
+    distances += np.diag([100]*len(labels))
+    # Find index of zero distances
+    zero_dist = np.where(
+        distances < threshold
+    )
+    # and report
+    if len(zero_dist[0]):
+        raise DataFormattingError(
+            'Error: Overlapping atoms detected in structure'
+            ' at atom numbers:\n' + ', '.join(
+                str(i+1) for i in zero_dist[0]
+            )
+        )
+
+    return
+
+
 def check_moinp_moread(file_name: str | Path) -> None:
     '''
     Checks if MORead and/or MOInp are present in the input file.\n

{orto-1.8.2 → orto-1.8.3}/orto.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: orto
-Version: 1.8.2
+Version: 1.8.3
 Summary: A package to make life easier when performing Orca calculations.
 Home-page: https://orto.kragskow.group
 Author: Jon Kragskow

{orto-1.8.2 → orto-1.8.3}/setup.py

@@ -8,7 +8,7 @@ Please see the `orto` documentation for more details.
 
 # DO NOT EDIT THIS NUMBER!
 # IT IS AUTOMATICALLY CHANGED BY python-semantic-release
-__version__ = '1.8.2'
+__version__ = '1.8.3'
 
 setuptools.setup(
     name='orto',

orto-1.8.2/orto/__version__.py

@@ -1 +0,0 @@
-__version__ = '1.8.2'