Merge pull request #13 from nlesc-nano/devel

Data-CAT 0.1.2

.travis.yml

@@ -0,0 +1,37 @@
+sudo: false
+language:
+    python: 3.7
+
+env:
+    global:
+        - COMMIT_AUTHOR_EMAIL: b.f.van.beek@vu.nl
+
+install:
+    # Install python
+    - wget https://repo.continuum.io/miniconda/Miniconda3-latest-Linux-x86_64.sh -O miniconda.sh
+    - bash miniconda.sh -b -p $HOME/miniconda
+    - export PATH="$HOME/miniconda/bin:$PATH"
+    - hash -r
+    - conda config --set always_yes yes --set changeps1 no --set auto_update_conda False
+    - conda update -q conda
+
+    # Useful for debugging any issues with conda
+    - conda info -a
+
+    # Install virtual enviroment
+    - conda create -n CAT python=3.7
+    - conda install -n CAT -c conda-forge h5py rdkit
+    - source activate CAT
+
+    # Install tests, data-CAT and nano-CAT
+    - pip install .[test]
+    - pip install cat@git+https://github.com/nlesc-nano/CAT@devel --upgrade
+
+script:
+    # Run the unitary tests excluding the expensive computations
+    - pytest -m "not (slow or long)" --cov=CAT tests
+    - coverage xml && coverage report -m
+
+branches:
+    only:
+        - master

CHANGELOG.rst

@@ -5,6 +5,15 @@ Change Log
 All notable changes to this project will be documented in this file.
 This project adheres to `Semantic Versioning <http://semver.org/>`_.
 
+0.1.2
+*****
+
+* Updated many ``__str__`` and ``__repr__`` methods.
+* Added the ``Database.__eq__`` method.
+* Moved context managers to ``dataCAT.context_managers``
+* Moved (and renamed) the ``DF()`` class to ``dataCAT.df_collection.DFCollection()``.
+* Added more tests.
+
 
 0.1.1
 *****

README.rst

@@ -3,7 +3,7 @@
 
 
 ##############
-Data-CAT 0.1.1
+Data-CAT 0.1.2
 ##############
 
 Data-CAT is a databasing framework for the Compound Attachment Tools package (CAT_).

dataCAT/__version__.py

@@ -1 +1 @@
-__version__ = '0.1.1'
+__version__ = '0.1.2'

dataCAT/context_managers.py

@@ -0,0 +1,249 @@
+"""
+dataCAT.context_managers
+========================
+
+A module which holds context managers for the :class:`.Database` class.
+
+Index
+-----
+.. currentmodule:: dataCAT.context_managers
+.. autosummary::
+    MetaManager
+    OpenYaml
+    OpenLig
+    OpenQD
+
+API
+---
+.. autoclass:: MetaManager
+    :members:
+.. autoclass:: OpenYaml
+.. autoclass:: OpenLig
+.. autoclass:: OpenQD
+
+"""
+
+from os import getcwd, sep
+from os.path import basename
+from typing import (Callable, Optional, Any)
+from contextlib import AbstractContextManager
+from dataclasses import dataclass
+
+import yaml
+import pandas as pd
+
+from scm.plams import Settings
+
+from .df_collection import get_df_collection
+
+__all__ = ['MetaManager', 'OpenYaml', 'OpenLig', 'OpenQD']
+
+
+@dataclass(frozen=True)
+class MetaManager:
+    """A wrapper for context managers.
+
+    Has a single important method, :meth:`MetaManager.open`,
+    which calls and returns the context manager stored in :attr:`MetaManager.manager`.
+
+    Note
+    ----
+    :attr:`MetaManager.filename` will be the first positional argument provided
+    to :attr:`MetaManager.manager`.
+
+    Paramaters
+    ----------
+    filename : str
+        The path+filename of a database component
+        See :attr:`MetaManager.filename`.
+
+    manager : |type|_ [|AbstractContextManager|_]
+        A type object of a context manager.
+        TThe first positional argument of the context manager should be the filename.
+        See :attr:`MetaManager.manager`.
+
+    Attributes
+    ----------
+    filename : str
+        The path+filename of a database component.
+
+    manager : |type|_ [|AbstractContextManager|_]
+        A type object of a context manager.
+        The first positional argument of the context manager should be the filename.
+
+    """
+
+    filename: str
+    manager: Callable[..., AbstractContextManager]
+
+    def __repr__(self) -> str:
+        filename = repr(f'...{sep}{basename(self.filename)}')
+        return f'{self.__class__.__name__}(filename={filename}, manager={repr(self.manager)})'
+
+    def __str__(self) -> str:
+        args = self.__class__.__name__, repr(self.filename), repr(self.manager)
+        return '{}(\n    filename = {},\n    manager  = {}\n)'.format(*args)
+
+    def open(self, *args: Any, **kwargs: Any) -> AbstractContextManager:
+        """Call and return :attr:`MetaManager.manager`."""
+        return self.manager(self.filename, *args, **kwargs)
+
+
+class OpenYaml(AbstractContextManager):
+    """Context manager for opening and closing job settings (:attr:`.Database.yaml`).
+
+    Parameters
+    ----------
+    filename : str
+        The path+filename to the database component.
+
+    write : bool
+        Whether or not the database file should be updated after closing this instance.
+
+    Attributes
+    ----------
+    filename : str
+        The path+filename to the database component.
+
+    write : bool
+        Whether or not the database file should be updated after closing this instance.
+
+    settings : |None|_ or |plams.Settings|_
+        An attribute for (temporary) storing the opened .yaml file
+        (:attr:`OpenYaml.filename`) as :class:`.Settings` instance.
+
+    """
+
+    def __init__(self, filename: Optional[str] = None,
+                 write: bool = True) -> None:
+        """Initialize the :class:`.OpenYaml` context manager."""
+        self.filename: str = filename or getcwd()
+        self.write: bool = write
+        self.settings = None
+
+    def __enter__(self) -> Settings:
+        """Open the :class:`.OpenYaml` context manager, importing :attr:`.settings`."""
+        with open(self.filename, 'r') as f:
+            self.settings = Settings(yaml.load(f, Loader=yaml.FullLoader))
+        return self.settings
+
+    def __exit__(self, exc_type, exc_value, traceback) -> None:
+        """Close the :class:`.OpenYaml` context manager, exporting :attr:`.settings`."""
+        if self.write:
+            yml_dict = self.settings.as_dict()
+
+            # A fix for Settings.as_dict() not functioning when containg a lists of Settings
+            for key in yml_dict:
+                for i, value in enumerate(yml_dict[key]):
+                    if isinstance(value, Settings):
+                        yml_dict[key][i] = value.as_dict()
+
+            # Write to the .yaml file
+            with open(self.filename, 'w') as f:
+                f.write(yaml.dump(yml_dict, default_flow_style=False, indent=4))
+        self.settings = None
+        assert self.settings is None
+
+
+class OpenLig(AbstractContextManager):
+    """Context manager for opening and closing the ligand database (:attr:`.Database.csv_lig`).
+
+    Parameters
+    ----------
+    filename : str
+        The path+filename to the database component.
+
+    write : bool
+        Whether or not the database file should be updated after closing this instance.
+
+    Attributes
+    ----------
+    filename : str
+        The path+filename to the database component.
+
+    write : bool
+        Whether or not the database file should be updated after closing this instance.
+
+    df : |None|_ or |pd.DataFrame|_
+        An attribute for (temporary) storing the opened .csv file
+        (see :attr:`OpenLig.filename`) as a :class:`.DataFrame` instance.
+
+    """
+
+    def __init__(self, filename: Optional[str] = None,
+                 write: bool = True) -> None:
+        """Initialize the :class:`.OpenLig` context manager."""
+        self.filename: str = filename or getcwd()
+        self.write: bool = write
+        self.df: Optional['DFCollection'] = None
+
+    def __enter__(self) -> 'DFCollection':
+        """Open the :class:`.OpenLig` context manager, importing :attr:`.df`."""
+        # Open the .csv file
+        dtype = {'hdf5 index': int, 'formula': str, 'settings': str, 'opt': bool}
+        self.df = df = get_df_collection(
+            pd.read_csv(self.filename, index_col=[0, 1], header=[0, 1], dtype=dtype)
+        )
+
+        # Fix the columns
+        idx_tups = [(i, '') if 'Unnamed' in j else (i, j) for i, j in df.columns]
+        df.columns = pd.MultiIndex.from_tuples(idx_tups, names=df.columns.names)
+        return df
+
+    def __exit__(self, exc_type, exc_value, traceback) -> None:
+        """Close the :class:`.OpenLig` context manager, exporting :attr:`.df`."""
+        if self.write:
+            self.df.to_csv(self.filename)
+        self.df = None
+
+
+class OpenQD(AbstractContextManager):
+    """Context manager for opening and closing the QD database (:attr:`.Database.csv_qd`).
+
+    Parameters
+    ----------
+    filename : str
+        The path+filename to the database component.
+
+    write : bool
+        Whether or not the database file should be updated after closing this instance.
+
+    Attributes
+    ----------
+    filename : str
+        The path+filename to the database component.
+
+    write : bool
+        Whether or not the database file should be updated after closing this instance.
+
+    df : |None|_ or |pd.DataFrame|_
+        An attribute for (temporary) storing the opened .csv file
+        (:attr:`OpenQD.filename`) as :class:`.DataFrame` instance.
+
+    """
+
+    def __init__(self, filename: Optional[str] = None,
+                 write: bool = True) -> None:
+        """Initialize the :class:`.OpenQD` context manager."""
+        self.filename: str = filename or getcwd()
+        self.write: bool = write
+        self.df: Optional['DFCollection'] = None
+
+    def __enter__(self) -> 'DFCollection':
+        """Open the :class:`.OpenQD` context manager, importing :attr:`.df`."""
+        # Open the .csv file
+        dtype = {'hdf5 index': int, 'settings': str, 'opt': bool}
+        self.df = df = get_df_collection(
+            pd.read_csv(self.filename, index_col=[0, 1, 2, 3], header=[0, 1], dtype=dtype)
+        )
+
+        # Fix the columns
+        idx_tups = [(i, '') if 'Unnamed' in j else (i, j) for i, j in df.columns]
+        df.columns = pd.MultiIndex.from_tuples(idx_tups, names=df.columns.names)
+        return df
+
+    def __exit__(self, exc_type, exc_value, traceback) -> None:
+        """Close the :class:`.OpenQD` context manager, exporting :attr:`.df`."""
+        if self.write:
+            self.df.to_csv(self.filename)
+        self.df = None

dataCAT/create_database.py

@@ -58,20 +58,18 @@ def _create_csv(path: str,
         The absolute path to the ligand or QD database.
 
     """
-    path = join(path, database + '_database.csv')
+    filename = join(path, database + '_database.csv')
 
     # Check if the database exists and has the proper keys; create it if it does not
-    if not isfile(path):
-        msg = f'{database}_database.csv not found in {path}, creating {database} database'
-        logger.info(msg)
-
+    if not isfile(filename):
         if database == 'ligand':
-            _create_csv_lig(path)
+            _create_csv_lig(filename)
         elif database == 'QD':
-            _create_csv_qd(path)
+            _create_csv_qd(filename)
         else:
             raise ValueError(f"'{database}' is not an accepated value for the 'database' argument")
-    return path
+        logger.info(f'{database}_database.csv not found in {path}, creating {database} database')
+    return filename
 
 
 def _create_csv_lig(filename: str) -> None:
@@ -186,13 +184,13 @@ def _create_yaml(path: str,
 
     """
     # Define arguments
-    path = join(path, name)
+    filename = join(path, name)
 
     # Create a new .yaml file if it does not yet exist
-    if not isfile(path):
-        with open(path, 'w') as f:
+    if not isfile(filename):
+        with open(filename, 'w') as f:
             f.write(yaml.dump({None: [None]}, default_flow_style=False, indent=4))
-    return path
+    return filename
 
 
 def _create_mongodb(host: str = 'localhost',