ChemAxonNMRPredictor

Complexity

Total Complexity

28

Size/Duplication

Total Lines	109
Duplicated Lines	0 %

Importance

Changes	1
Bugs	0	Features	1

9 Methods

Rating	Name	Duplication	Size	Complexity
B	transform()	0	11	5
A	max_atoms()	0	3	2
B	_transform_series()	0	23	4
A	feature_names()	0	3	1
A	index()	0	3	1
A	__init__()	0	4	1
A	_transform_mol()	0	5	1
F	parse_output()	0	27	9
A	element()	0	3	2

#! /usr/bin/env python
#
# Copyright (C) 2015-2016 Rich Lewis <rl403@cam.ac.uk>
# License: 3-clause BSD

"""
## skchem.descriptors.nmr

Module for NMR prediction.
"""
import tempfile
import subprocess
import re
import logging
import time

import numpy as np

The import numpy could not be resolved.

import pandas as pd

The import pandas could not be resolved.

from ..io import write_sdf
from ..utils import NamedProgressBar
from .. import core

LOGGER = logging.getLogger(__name__)

def mol_count(filename):

This function should have a docstring.

    return sum(1 for l in open(filename, 'rb') if l == b'##PEAKASSIGNMENTS=(XYMA)\r\n')


class ChemAxonNMRPredictor(object):

This class should have a docstring.

    def __init__(self, element='C', max_atoms='auto'):
        self.element = element
        self._detect_max_atoms = False
        self.max_atoms = max_atoms

    @property
    def element(self):

This method should have a docstring.

        return self._element

    @element.setter
    def element(self, val):

This method should have a docstring.

        val = val.upper()
        if val in ('C', 'H'):
            self._element = val

The attribute _element was defined outside __init__.

        else:
            raise ValueError('ChemAxon can only predict 1H or 13C shifts.')

    @property
    def max_atoms(self):

This method should have a docstring.

        return self._max_atoms

    @max_atoms.setter
    def max_atoms(self, arg):

This method should have a docstring.

        if arg == 'auto':
            self._detect_max_atoms = True
            self._max_atoms = -1

The attribute _max_atoms was defined outside __init__.

        else:
            self._max_atoms = arg

The attribute _max_atoms was defined outside __init__.

    @property
    def index(self):

This method should have a docstring.

        return pd.RangeIndex(self.max_atoms, name='atom_idx')

    @property
    def feature_names(self):

This method should have a docstring.

        return self.index

    def transform(self, obj):

This method should have a docstring.

        if isinstance(obj, core.Mol):
            return self._transform_mol(obj)
        elif isinstance(obj, pd.Series):
            return self._transform_series(obj)
        elif isinstance(obj, pd.DataFrame):
            return self._transform_series(obj.structure)
        elif isinstance(obj, (tuple, list)):
            return self._transform_series(obj)
        else:
            raise NotImplementedError

    def _transform_mol(self, mol):
        # make into series then use self._transform_mol
        ser = pd.Series([mol], name=mol.name)
        res = self._transform_series(ser)
        return res.iloc[0]

    def _transform_series(self, ser):

        if self._detect_max_atoms:
            self.max_atoms = ser.ms.atoms.apply(len).max()

        with tempfile.NamedTemporaryFile(suffix='.sdf') as infile, tempfile.NamedTemporaryFile() as outfile:

This line is too long as per the coding-style (108/100).

            write_sdf(ser, infile.name)
            args = ['cxcalc', infile.name, '-o', outfile.name] + [self.element.lower() + 'nmr']

            LOGGER.info('Running: ' + ' '.join(args))

            p = subprocess.Popen(args, stdout=subprocess.DEVNULL, stderr=subprocess.DEVNULL)

The name p does not conform to the variable naming conventions ([a-z_][a-z0-9_]{2,30}$).

The Module subprocess does not seem to have a member named DEVNULL.

            bar = NamedProgressBar(name=self.__class__.__name__, max_value=len(ser))

Black listed name "bar"

            while p.poll() is None:
                bar.update(mol_count(outfile.name))

The Instance of NamedProgressBar does not seem to have a member named update.

                time.sleep(1)
            bar.update(len(ser))

The Instance of NamedProgressBar does not seem to have a member named update.

            p.wait()

            res = self.parse_output(outfile.name, n_mols=len(ser))
            return pd.DataFrame(res, index=ser.index, columns=self.index)


    def parse_output(self, outfile, n_mols=None):

        """ Read the NMR output file. """

        if n_mols is None:
            n_mols = mol_count(outfile)

        res = np.repeat(np.nan, n_mols * self.max_atoms).reshape(n_mols, self.max_atoms)
        regex = re.compile(b'\((-?\d+.\d+),\d+,[A-Z],<([0-9\,]+)>\)\r\n')

A suspicious escape sequence \( was found. Did you maybe forget to add an r prefix?

A suspicious escape sequence \d was found. Did you maybe forget to add an r prefix?

A suspicious escape sequence \, was found. Did you maybe forget to add an r prefix?

A suspicious escape sequence \) was found. Did you maybe forget to add an r prefix?

        mol_idx = 0

        with open(outfile, 'rb') as f:

The name f does not conform to the variable naming conventions ([a-z_][a-z0-9_]{2,30}$).

            # loop through the file - inner loop will also advance the pointer
            for l in f:

The name l does not conform to the variable naming conventions ([a-z_][a-z0-9_]{2,30}$).

                if l == b'##PEAKASSIGNMENTS=(XYMA)\r\n':
                    for row in f:
                        if row == b'##END=\r\n':
                            break
                        else:
                            LOGGER.debug('Row to parse: %s', row)
                            shift, idxs = regex.match(row).groups()
                            shift, idxs = float(shift), [int(idx) for idx in idxs.split(b',')]
                            for atom_idx in idxs:
                                res[mol_idx, atom_idx] = shift
                    mol_idx += 1
        return res