Source code for matador.query.query

# coding: utf-8
# Distributed under the terms of the MIT License.

""" This file implements all queries to the database, including parsing
user inputs, displaying results and calling other functionality.

"""


import sys
import random
from os import devnull
from itertools import combinations
from traceback import print_exc

import pymongo as pm
import numpy as np
from bson.json_util import dumps
from bson.objectid import ObjectId

from matador.utils.print_utils import print_warning, print_success, print_notify
from matador.utils.chem_utils import get_periodic_table
from matador.utils.chem_utils import parse_element_string, get_stoich_from_formula
from matador.utils.cursor_utils import display_results, filter_cursor_by_chempots
from matador.db import make_connection_to_collection
from matador.config import load_custom_settings


[docs]class DBQuery: """ Class that implements queries to MongoDB structure database. Attributes: cursor (list of dict or :obj:`pymongo.Cursor`): list or cursor of structures matching query. args (dict): contains all keyword arguments used to construct the query (see matador query --help) for list. query_dict (dict): dictionary passed to database for query calc_dict (dict): if performing a matching query (e.g. self.args['subcmd'] = 'hull'), this dictionary contains the parameters used to match to other structures repo (pymongo.collection.Collection): the pymongo collection that is being queried. top (int): number of structures to print/export set by self.args.get('top') (DEFAULT: 10). cursor_min_limit (int): if a query returns more structures than this, do not implicitly convert to a list. """ # below this number of documents, # all queries will return a list rather than a pymongo Cursor cursor_min_limit = 1000 def __init__( self, client=False, collections=False, subcmd='query', debug=False, quiet=False, mongo_settings=None, **kwargs ): """ Parse arguments from matador or API call before calling query. Keyword arguments: client (pm.MongoClient): the MongoClient to connect to. collections (dict of pm.collections.Collection): dictionary of pymongo Collections. subcmd (str): either 'query' or 'hull', 'voltage', 'hulldiff'. These will decide whether calcuation accuracies are matched in the final results. """ # read args and set housekeeping self.args = kwargs self.debug = debug if self.args.get('subcmd') is None: self.args['subcmd'] = subcmd if self.args.get('testing') is None: self.args['testing'] = False if self.args.get('as_crystal') is None: self.args['as_crystal'] = False if subcmd in ['hull', 'hulldiff', 'voltage'] and self.args.get('composition') is None: raise RuntimeError('{} requires composition query'.format(subcmd)) self._create_hull = (self.args.get('subcmd') in ['hull', 'hulldiff', 'voltage'] or self.args.get('hull_cutoff') is not None) # public attributes self.cursor = EmptyCursor() self.query_dict = None self.calc_dict = None self.repo = None # private attributes to be set later self._empty_query = None self._gs_enthalpy = None self._non_elemental = None self._chempots = None self._num_to_display = None if debug: print(self.args) if quiet: f = open(devnull, 'w') sys.stdout = f # if testing keyword is used, all database operations are ignored if not self.args.get('testing'): # connect to db or use passed client if client: self._client = client self._db = client.crystals if collections is not False: _collections = collections if (not collections or not client): # use passed settings or load from config file if mongo_settings: self.mongo_settings = mongo_settings else: self.mongo_settings = load_custom_settings( config_fname=self.args.get('config'), debug=self.args.get('debug') ) result = make_connection_to_collection( self.args.get('db'), mongo_settings=self.mongo_settings ) # ideally this would be rewritten to use a context manager to ensure # that connections are _always_ cleaned up self._client, self._db, _collections = result if len(_collections) > 1: raise NotImplementedError("Querying multiple collections is no longer supported.") else: for collection in _collections: self._collection = _collections[collection] break # define some periodic table macros self._periodic_table = get_periodic_table() # set default top value to 10 if self.args.get('summary') or self.args.get('subcmd') in ['swaps', 'polish']: self.top = None else: self.top = self.args.get('top') if self.args.get('top') is not None else 10 # create the dictionary to pass to MongoDB self._construct_query() if not self.args.get('testing'): if self.args.get('id') is not None and (self._create_hull or self.args.get('calc_match')): # if we've requested and ID and hull/calc_match, do the ID query self.perform_id_query() self.perform_query() if self._create_hull and self.args.get('id') is None: # if we're making a normal hull, find the sets of calculations to use self.perform_hull_query() if not self._create_hull: # only filter for uniqueness if not eventually making a hull if self.args.get('uniq'): from matador.utils.cursor_utils import filter_unique_structures print_notify('Filtering for unique structures...') if isinstance(self.cursor, pm.cursor.Cursor): raise RuntimeError("Unable to filter pymongo cursor for uniqueness directly.") if self.args.get('top') is not None: top = self.args['top'] else: top = len(self.cursor) self.cursor = filter_unique_structures( self.cursor[:top], debug=self.args.get('debug'), sim_tol=self.args.get('uniq'), energy_tol=1e20 ) if self.args.get('available_values') is not None: print('Querying available values...') self._query_available_values(self.args.get('available_values'), self.cursor) # if no client was passed, then we need to close the one we made if not client and not self.args.get('testing'): self._client.close() if quiet: f.close() sys.stdout = sys.__stdout__ def _construct_query(self): """ Set up query dict and perform query depending on command-line / API arguments. Sets self.query_dict. """ self.cursor = EmptyCursor() # initalize query_dict to '$and' all queries self.query_dict = dict() self.query_dict['$and'] = [] self._empty_query = True # benchmark enthalpy to display (set by calc_match) self._gs_enthalpy = 0.0 # operate on one structure and related others if self.args.get('id') is not None: if not self._create_hull and not self.args.get('calc_match'): self.query_dict['$and'].append(self._query_id()) self._empty_query = False # create alias for formula for backwards-compatibility self.args['stoichiometry'] = self.args.get('formula') if self.args.get('stoichiometry') is not None: self.query_dict['$and'].append(self._query_stoichiometry()) self._empty_query = False if self.args.get('composition') is not None: self.query_dict['$and'].append(self._query_composition()) self._empty_query = False if self.args.get('num_species') is not None: self.query_dict['$and'].append(self._query_num_species()) self._empty_query = False if self.args.get('space_group') is not None: self.query_dict['$and'].append(self._query_space_group()) self._empty_query = False if self.args.get('num_fu') is not None: self.query_dict['$and'].append(self._query_num_fu()) self._empty_query = False if self.args.get('tags') is not None: self.query_dict['$and'].append(self._query_tags()) self._empty_query = False if self.args.get('doi') is not None: self.query_dict['$and'].append(self._query_doi()) self._empty_query = False if self.args.get('icsd') is not None: self.query_dict['$and'].append(self._query_icsd()) self._empty_query = False if self.args.get('field') is not None: try: for ind, field in enumerate(self.args.get('field')): _filter = self.args.get('filter')[ind] try: for i, value in enumerate(_filter): _filter[i] = float(value) filter_type = 'float' except ValueError: filter_type = 'string' if filter_type == 'float': self.query_dict['$and'].append(self._query_float_range( field, _filter)) else: self.query_dict['$and'].append(self._query_string( field, _filter)) except Exception: raise RuntimeError( "Unexpected field/filter format. Both must be " "provided as lists, even if only one field is being filtered." ) self._empty_query = False if self.args.get('cutoff') is not None: self.query_dict['$and'].append(self._query_float_range( 'cut_off_energy', self.args.get('cutoff'))) self._empty_query = False if self.args.get('geom_force_tol') is not None: self.query_dict['$and'].append(self._query_float_range( 'geom_force_tol', self.args.get('geom_force_tol'))) self._empty_query = False if self.args.get('grid_scale') is not None: self.query_dict['$and'].append(self._query_float_range( 'grid_scale', self.args.get('grid_scale'))) self._empty_query = False if self.args.get('fine_grid_scale') is not None: self.query_dict['$and'].append(self._query_float_range( 'fine_grid_scale', self.args.get('fine_grid_scale'))) self._empty_query = False if self.args.get('src_str') is not None: self.query_dict['$and'].append(self._query_source()) self._empty_query = False if self.args.get('root_src') is not None: self.query_dict['$and'].append(self._query_root_source()) self._empty_query = False if self.args.get('pressure') is not None: self.query_dict['$and'].append(self._query_float_range( 'pressure', self.args.get('pressure') or 0.0, tolerance=self.args.get('pressure_tolerance') or 0.5)) self._empty_query = False elif self.args['subcmd'] in ['hull', 'hulldiff', 'voltage']: self.query_dict['$and'].append(self._query_float_range( 'pressure', 0.0, tolerance=self.args.get('pressure_tolerance') or 0.5)) if self.args.get('encapsulated') is True: self.query_dict['$and'].append(self._query_encap()) self._empty_query = False if self.args.get('cnt_radius') is not None: self.query_dict['$and'].append(self._query_float_range( 'cnt_radius', self.args.get('cnt_radius'), tolerance=0.01)) self._empty_query = False if self.args.get('cnt_vector') is not None: self.query_dict['$and'].append(self._query_cnt_vector()) self._empty_query = False if self.args.get('sedc') is not None: self.query_dict['$and'].append(self._query_sedc()) self._empty_query = False if self.args.get('xc_functional') is not None: self.query_dict['$and'].append(self._query_xc_functional()) self._empty_query = False if self.args.get('mp_spacing') is not None: self.query_dict['$and'].append(self._query_float_range( 'kpoints_mp_spacing', self.args.get('mp_spacing'), tolerance=self.args.get('kpoint_tolerance') or 0.01)) self._empty_query = False if self.args.get('spin') is not None: tmp_dict = self._query_spin() if tmp_dict: self.query_dict['$and'].append(tmp_dict) self._empty_query = False if not self.args.get('ignore_warnings'): self.query_dict['$and'].append(self._query_quality()) if self.args.get('time') is not None: self.query_dict['$and'].append(self._query_time(self.args.get('since') or False)) self._empty_query = False def _perform_empty_query(self, as_list=False): """ No parameters were asked for, so just return a cursor that contains the entire collection currently stored as :attr:`repo`. Returns: list or :obj:`pymongo.cursor.Cursor`: the results of the query. """ num_documents = self.repo.count_documents({}) cursor = self.repo.find().sort('enthalpy_per_atom', pm.ASCENDING) if self.debug: print('Empty query, showing all...') if num_documents < self.cursor_min_limit or as_list: return list(cursor), num_documents return cursor, num_documents
[docs] def perform_query(self): """ Find results that match the query_dict inside the MongoDB database. """ # if no query submitted, find all if self._empty_query and self.args.get('id') is None: self.repo = self._collection self.cursor, cursor_count = self._perform_empty_query() # if no special query has been made already, begin executing the query if not self._empty_query: self.repo = self._collection if self.debug: print('Query dict:') print(dumps(self.query_dict, indent=1)) # execute query self.cursor, cursor_count = self._find_and_sort(self.query_dict) if self._non_elemental: self.cursor = filter_cursor_by_chempots(self._chempots, self.cursor) print('{} results found for query in {}.'.format(cursor_count, self.repo.name)) self._num_to_display = cursor_count if self.args.get('subcmd') != 'swaps' and not self._create_hull: self._set_filter_display_results(cursor_count) # if a summary has been requested, cursor must be converted to list if self.args.get('summary'): self.cursor = list(self.cursor) if self.args.get('subcmd') != 'swaps' and not self._create_hull: if self._num_to_display >= 1 or self._num_to_display is None: if self._num_to_display == cursor_count: display_results(self.cursor, **self.args) else: display_results(self.cursor[:self._num_to_display], **self.args) if isinstance(self.cursor, pm.cursor.Cursor): self.cursor.rewind()
def _set_filter_display_results(self, cursor_count): """ Filter and display the results based on the command line parameters. """ # by default, show the top structures only # if delta_E requested, count how many exist below that energy if self.args.get('delta_E') is not None: if isinstance(self.cursor, pm.cursor.Cursor) and len(self.cursor.distinct('stoichiometry')) > 1: print('Multiple stoichiometries in cursor, unable to filter by energy with --delta_E.') else: self.cursor = list(self.cursor) gs_enthalpy = self.cursor[0]['enthalpy_per_atom'] for ind, doc in enumerate(self.cursor[1:]): if abs(doc['enthalpy_per_atom'] - gs_enthalpy) > self.args.get('delta_E'): self._num_to_display = ind + 1 break elif self.top == -1 or self.top is None or cursor_count <= self.top: self._num_to_display = cursor_count self.top = cursor_count elif cursor_count > self.top: self._num_to_display = self.top def _find_and_sort(self, query_filter=None, as_list=False, **kwargs): """ Query `self.repo` using Pymongo arguments/kwargs. Sorts based on enthalpy_per_atom and optionally returns list of Crystals. Keyword arguments: query_filter (dict): the query to use. If None, perform a blank query. as_list (bool): whether to return a list of a pm.cursor.Cursor object. Returns: list/pm.cursor.Cursor: the results of the query. int: the number of results in the query. """ from matador.crystal import Crystal if query_filter is None: query_filter = {} count = self.repo.count_documents(query_filter, **kwargs) cursor = self.repo.find(query_filter, **kwargs).sort('enthalpy_per_atom', pm.ASCENDING) if self.args.get('as_crystal'): return [Crystal(doc) for doc in cursor], count if count < self.cursor_min_limit or as_list: return list(cursor), count return cursor, count
[docs] def perform_hull_query(self): """ Perform the multiple queries necessary to find possible calculation sets to create a convex hull from. Raises: SystemExit: if no structures are found for hull. """ if self._collection is not None: self.repo = self._collection print('Creating hull from structures in query results.') if self.args.get('biggest'): print('\nFinding biggest calculation set for hull...\n') else: print('\nFinding the best calculation set for hull...') test_cursors = [] test_cursor_count = [] text_ids = [] calc_dicts = [] cutoff = [] num_sample = 2 num_rand_sample = 5 if self.args.get('biggest') else 3 if isinstance(self.cursor, pm.cursor.Cursor): count = self.cursor.count() else: count = len(self.cursor) if count <= 0: raise SystemExit('No structures found for hull.') # generate some random indices to match to, make sure they are in order # so can be accessed without cursor rewinds sampling_indices = list(range(num_sample)) + sorted(random.sample(range(2, count), num_rand_sample)) for ind in sampling_indices: doc = self.cursor[ind] text_ids.append(doc['text_id']) try: self.query_dict = self._query_calc(doc) cutoff.append(doc['cut_off_energy']) calc_dicts.append(dict()) calc_dicts[-1]['$and'] = list(self.query_dict['$and']) self.query_dict['$and'].append(self._query_composition()) if not self.args.get('ignore_warnings'): self.query_dict['$and'].append(self._query_quality()) probe_cursor, probe_count = self._find_and_sort(self.query_dict) if self._non_elemental: probe_cursor = filter_cursor_by_chempots(self._chempots, probe_cursor) probe_count = len(probe_cursor) test_cursors.append(probe_cursor) test_cursor_count.append(probe_count) print("{:^24}: matched {} structures." .format(' '.join(doc['text_id']), probe_count), end='\t-> ') print('{spin}{sedc}{functional} {cutoff} eV, {geom_force_tol} eV/A, {kpoints} 1/A.' .format(spin="S-" if doc.get('spin_polarized') else '', sedc="+" + doc.get('sedc') + "+" if doc.get('sedc') else "", functional=doc["xc_functional"], cutoff=doc["cut_off_energy"], geom_force_tol=doc.get('geom_force_tol', 'xxx'), kpoints=doc.get('kpoints_mp_spacing', 'xxx'))) if test_cursor_count[-1] == count: print('Matched all structures...') break if test_cursor_count[-1] > 2 * int(count / 3): print('Matched at least 2/3 of total number, composing hull...') break except Exception: print_exc() print_warning('Error with {}'.format(' '.join(doc['text_id']))) if self.args.get('biggest'): choice = np.argmax(np.asarray(test_cursor_count)) else: # by default, find highest cutoff hull as first proxy for quality choice = np.argmax(np.asarray(cutoff)) text_id = text_ids[choice] self.cursor = test_cursors[choice] self.calc_dict = calc_dicts[choice] if not test_cursor_count[choice]: raise RuntimeError('No structures found that match chemical potentials.') print_success('Composing hull from set containing {}'.format(' '.join(text_id)))
[docs] def perform_id_query(self): """ Query the `text_id` field for the ID provided in the args for a calc_match or hull/voltage query. Use the results of the text_id query to match to other entries that have the same calculation parameters. Sets self.query_dict and self.calc_dict. Raises: RuntimeError: if no structures are found. """ self.cursor = [] query_dict = dict() query_dict['$and'] = [] query_dict['$and'].append(self._query_id()) if not self.args.get('ignore_warnings'): query_dict['$and'].append(self._query_quality()) self.repo = self._collection self.cursor = list(self._find_and_sort(query_dict)) if not self.cursor: raise RuntimeError('Could not find a match with {} try widening your search.'.format(self.args.get('id'))) if len(self.cursor) >= 1: display_results(list(self.cursor)[:self.top], **self.args) if len(self.cursor) > 1: print_warning('Matched multiple structures with same text_id. The first one will be used.') # save special copy of calc_dict for hulls self.calc_dict = dict() self.calc_dict['$and'] = [] # to avoid deep recursion, and since this is always called first # don't append, just set self.query_dict = self._query_calc(self.cursor[0]) if self.args.get('composition'): self.args['intersection'] = True self.query_dict['$and'].append(self._query_composition()) self.calc_dict['$and'] = list(self.query_dict['$and'])
[docs] def query_stoichiometry(self, **kwargs): """ Alias for private function of the same name. """ return self._query_stoichiometry(**kwargs)
[docs] def query_composition(self, **kwargs): """ Alias for private function of the same name. """ return self._query_composition(**kwargs)
[docs] def query_tags(self, **kwargs): """ Alias for private function of the same name. """ return self._query_tags(**kwargs)
[docs] def query_quality(self, **kwargs): """ Alias for private function of the same name. """ return self._query_quality(**kwargs)
@staticmethod def _query_float_range(field, values, tolerance=None): """ Query all entries with field between float value range, or with float value. Parameters: field (str): the field to query. values (float/list of float): either single value, or list of 2 floats. Keyword arguments: tolerance (float): tolerance to add and subtract if single value is provided. Returns: dict: the constructed query. """ query_dict = dict() query_dict[field] = dict() if not isinstance(values, list): values = [values] if len(values) == 2: if values[0] > values[1]: tmp = values[0] values[0] = values[1] values[1] = tmp query_dict[field]['$gte'] = values[0] query_dict[field]['$lte'] = values[1] else: if tolerance is None: query_dict[field]['$eq'] = values[0] else: query_dict[field]['$gte'] = round(values[0] - tolerance, 8) query_dict[field]['$lte'] = round(values[0] + tolerance, 8) return query_dict @staticmethod def _query_string(field, values): """ Query all entries for an exact string match on field. Parameters: field (str): the field to query. values (list or str): strings to query with $or joins. Returns: dict: the constructed query. """ query_dict = dict() if not isinstance(values, list): values = [values] if len(values) > 1: query_dict['$or'] = [] for value in values: query_dict['$or'].append({field: value}) else: query_dict[field] = values[0] return query_dict def _query_stoichiometry(self, custom_stoich=None, partial_formula=None): """ Query DB for particular stoichiometry. """ # alias stoichiometry if custom_stoich is None: stoich = self.args.get('stoichiometry') if isinstance(stoich, str): stoich = [stoich] else: stoich = custom_stoich if partial_formula is None: partial_formula = self.args.get('partial_formula') if ':' in stoich[0]: raise RuntimeError('Formula cannot contain ":", you probably meant to query composition.') stoich = get_stoich_from_formula(stoich[0], sort=False) query_dict = dict() query_dict['$and'] = [] for ind, _ in enumerate(stoich): elem = stoich[ind][0] fraction = int(stoich[ind][1]) if '[' in elem or ']' in elem: types_dict = dict() types_dict['$or'] = list() elem = elem.strip('[').strip(']') if elem in self._periodic_table: for group_elem in self._periodic_table[elem]: types_dict['$or'].append(dict()) types_dict['$or'][-1]['stoichiometry'] = dict() types_dict['$or'][-1]['stoichiometry']['$in'] = [[group_elem, fraction]] query_dict['$and'].append(types_dict) elif ',' in elem: for group_elem in elem.split(','): types_dict['$or'].append(dict()) types_dict['$or'][-1]['stoichiometry'] = dict() types_dict['$or'][-1]['stoichiometry']['$in'] = [[group_elem, fraction]] query_dict['$and'].append(types_dict) else: stoich_dict = dict() stoich_dict['stoichiometry'] = dict() stoich_dict['stoichiometry']['$in'] = [[elem, fraction]] query_dict['$and'].append(stoich_dict) if not partial_formula: size_dict = dict() size_dict['stoichiometry'] = dict() size_dict['stoichiometry']['$size'] = len(stoich) query_dict['$and'].append(size_dict) return query_dict @staticmethod def _query_ratio(ratios): """ Query DB for ratio of two elements. Ratios must be integers. Parameters: ratios (list): e.g. ratios = [['MoS', 2], ['LiS', 1]] """ query_dict = dict() for pair in ratios: query_dict['ratios.' + pair[0]] = int(pair[1]) return query_dict def _query_composition(self, custom_elem=None, partial_formula=None, elem_field='elems'): """ Query DB for all structures containing all the elements taken as input. Passing this function a number is a deprecated feature, replaced by query_num_species. Keyword arguments: custom_elem (str): use to query custom string, rather than CLI args partial_formula (bool): remove stoich size from query if True elem_field (str): which field to query for elems, either `atom_types` or `elems` Returns: dict: dictionary containing database query. """ if custom_elem is None: if isinstance(self.args.get('composition'), str): elements = [self.args.get('composition')] else: elements = self.args.get('composition') else: elements = custom_elem if partial_formula is None: partial_formula = self.args.get('partial_formula') self._non_elemental = False if ':' in elements[0]: self._non_elemental = True self.args['intersection'] = True self._chempots = elements[0].split(':') elements = [parse_element_string(elem) for elem in self._chempots] elements = list(dict.fromkeys([char for elem in elements for char in elem if char.isalpha()])) # if there's only one string, try split it by caps if not self._non_elemental: for char in elements[0]: if char.isdigit(): raise SystemExit('Composition cannot contain a number.') elements = parse_element_string(elements[0]) or_preference = False for _, elem in enumerate(elements): if '{' in elem or '}' in elem: or_preference = True elements_tmp = [element for ind, element in enumerate(elements) if element not in elements[:ind]] if len(elements_tmp) < len(elements): print('Ignoring duplicate element...') elements = elements_tmp if self.args.get('intersection'): if or_preference: raise RuntimeError('Intersection not implemented for overlapping sets, e.g. {}') query_dict = dict() query_dict['$or'] = [] size = len(elements) # iterate over all combinations, limited by num species if self.args.get('num_species'): max_num = self.args.get('num_species') min_num = max_num else: max_num = 8 min_num = 1 if len(elements) > max_num: print('Limiting query to up to {} elements per structure...'.format(max_num)) for rlen in range(min_num, max_num+1): for combi in combinations(elements, r=rlen): list_combi = list(combi) types_dict = dict() types_dict['$and'] = list() types_dict['$and'].append(dict()) types_dict['$and'][-1]['stoichiometry'] = dict() types_dict['$and'][-1]['stoichiometry']['$size'] = len(list_combi) for elem in list_combi: types_dict['$and'].append(dict()) types_dict['$and'][-1][elem_field] = dict() types_dict['$and'][-1][elem_field]['$in'] = [elem] query_dict['$or'].append(types_dict) else: # expand group macros query_dict = dict() query_dict['$and'] = [] size = len(elements) if or_preference: element_slots = [] for elem in elements: if '[' in elem or '{' in elem: elem = elem.strip('{').strip('}').strip('[').strip(']') if elem in self._periodic_table: element_slots.append(self._periodic_table[elem]) elif ',' in elem: element_slots.append(elem.split(',')) else: element_slots.append([elem]) else: element_slots.append([elem]) from itertools import product slots = [list(config) for config in product(*element_slots)] types_dict = dict() types_dict['$or'] = list() for slot in slots: if len({elem for elem in slot}) == len(slot): types_dict['$or'].append(dict()) types_dict['$or'][-1]['$and'] = [] for elem in slot: types_dict['$or'][-1]['$and'].append(dict()) types_dict['$or'][-1]['$and'][-1][elem_field] = dict() types_dict['$or'][-1]['$and'][-1][elem_field]['$in'] = [elem] query_dict['$and'].append(types_dict) else: for elem in elements: if '[' in elem or ']' in elem: types_dict = dict() types_dict['$or'] = list() elem = elem.strip('[').strip(']') if elem in self._periodic_table: for group_elem in self._periodic_table[elem]: types_dict['$or'].append(dict()) types_dict['$or'][-1][elem_field] = dict() types_dict['$or'][-1][elem_field]['$in'] = [group_elem] elif ',' in elem: for group_elem in elem.split(','): types_dict['$or'].append(dict()) types_dict['$or'][-1][elem_field] = dict() types_dict['$or'][-1][elem_field]['$in'] = [group_elem] else: types_dict = dict() types_dict[elem_field] = dict() types_dict[elem_field]['$in'] = [elem] query_dict['$and'].append(types_dict) if not partial_formula and not self.args.get('intersection'): size_dict = dict() size_dict['stoichiometry'] = dict() size_dict['stoichiometry']['$size'] = size query_dict['$and'].append(size_dict) return query_dict def _query_num_species(self): """ Query database for all structures with a given number of elements, e.g. binaries, ternaries etc. """ num = self.args.get('num_species') if not isinstance(num, list): num = num elif isinstance(num, list): num = num[0] else: sys.exit('--num_species takes a single integer or list containing a single integer') query_dict = dict() query_dict['stoichiometry'] = dict() query_dict['stoichiometry']['$size'] = num return query_dict def _query_space_group(self): """ Query DB for all structures with given space group. """ query_dict = dict() if not isinstance(self.args.get('space_group'), list): spg = [self.args.get('space_group')] else: spg = self.args.get('space_group') query_dict['space_group'] = str(spg[0]) return query_dict def _query_num_fu(self): """ Query DB for all structures with more than a given number of formula units in the simulation. """ query_dict = dict() num = self.args.get('num_fu') if isinstance(num, list): num = num[0] query_dict['num_fu'] = dict() query_dict['num_fu']['$gte'] = num return query_dict def _query_tags(self): """ Find all structures matching given tags. """ query_dict = dict() query_dict['$and'] = [] for tag in self.args.get('tags'): temp_dict = dict() temp_dict['tags'] = dict() temp_dict['tags']['$in'] = [tag] query_dict['$and'].append(temp_dict) return query_dict def _query_doi(self): """ Find all structures matching given DOI, in format xxxx/xxxx. """ doi = self.args.get('doi') if not isinstance(doi, list): doi = [doi] query_dict = dict() query_dict['doi'] = dict() query_dict['doi']['$in'] = doi return query_dict def _query_id(self): """ Find all structures matching given tags. """ if isinstance(self.args.get('id'), str): self.args['id'] = self.args['id'].strip().split(' ') query_dict = dict() query_dict['text_id'] = self.args.get('id') return query_dict def _query_icsd(self): """ Find all structures matching given ICSD CollCode. """ if not isinstance(self.args.get('icsd'), list): icsd = [self.args.get('icsd')] else: icsd = self.args.get('icsd') query_dict = dict() if isinstance(icsd[0], bool): query_dict['icsd'] = dict() query_dict['icsd']['$exists'] = icsd[0] elif icsd[0] == 0: query_dict['icsd'] = dict() query_dict['icsd']['$exists'] = True else: query_dict['$or'] = [{'icsd': {'$eq': str(icsd[0])}}, {'icsd': {'$eq': icsd[0]}}] return query_dict def _query_source(self): """ Find all structures with source string from args. """ import re src_str = self.args.get('src_str') if not isinstance(src_str, list): src_str = [src_str] query_dict = dict() query_dict['source'] = dict() query_dict['source']['$in'] = [re.compile(src) for src in src_str] return query_dict def _query_root_source(self): """ Find all structures with root source string from args. """ root_src = self.args.get('root_src') if not isinstance(root_src, list): root_src = [root_src] query_dict = dict() for src in root_src: query_dict['$or'] = [] query_dict['$or'].append(dict()) query_dict['$or'][-1]['root_source'] = src return query_dict @staticmethod def _query_available_values(field, cursor): """ Query the values stored under a particular field and print the information. Parameters: field (str): the field to query. cursor (list): the cursor to query. """ supported_fields = [ 'doi', 'tags', 'root_source', 'cnt_vector', 'castep_version', 'cut_off_energy' ] number_containing_field = sum([1 for doc in cursor if field in doc]) if field in supported_fields and number_containing_field != 0: value_degeneracy = dict() for doc in cursor: if doc.get(field) is not None: values = doc.get(field) if isinstance(values, list): for value in values: if value in value_degeneracy: value_degeneracy[value] += 1 else: value_degeneracy[value] = 1 else: if values in value_degeneracy: value_degeneracy[values] += 1 else: value_degeneracy[values] = 1 print('Set of values under key {}:'.format(field)) for value in sorted(value_degeneracy, key=value_degeneracy.get): print('{:<10} -> {:<10}'.format(value_degeneracy[value], value)) else: print('Field {} unsupported for finding all possible values, must be one of {}' .format(field, supported_fields)) print('{}/{} contain field {}'.format(number_containing_field, len(cursor), field)) @staticmethod def _query_quality(): """ Find all structures with non-zero or non-existent (e.g. OQMD) quality. """ query_dict = dict() query_dict['$or'] = [] query_dict['$or'].append(dict()) query_dict['$or'][-1]['quality'] = dict() query_dict['$or'][-1]['quality']['$gt'] = 0 query_dict['$or'].append(dict()) query_dict['$or'][-1]['quality'] = dict() query_dict['$or'][-1]['quality']['$exists'] = False return query_dict @staticmethod def _query_encap(): """ Query only CNT encapsulated structures. """ query_dict = dict() query_dict['encapsulated'] = dict() query_dict['encapsulated']['$exists'] = True return query_dict def _query_cnt_vector(self): """ Query structures within a nanotube of given chiral vector. """ query_dict = dict() if not isinstance(self.args.get('cnt_vector'), list) or len(self.args.get('cnt_vector')) != 2: raise SystemExit('CNT vector query needs to be of form [n, m]') chiral_vec = self.args.get('cnt_vector') query_dict['cnt_chiral'] = dict() query_dict['cnt_chiral']['$eq'] = chiral_vec return query_dict def _query_sedc(self): """ Query all calculations using given SEDC scheme. Use --sedc null to query for no dispersion correction. """ query_dict = dict() if self.args.get('sedc') != 'null': query_dict['sedc_scheme'] = self.args.get('sedc') else: query_dict['sedc_scheme'] = dict() query_dict['sedc_scheme']['$exists'] = False return query_dict def _query_xc_functional(self, xc_functional=None): """ Query all calculations with specified xc-functional. Keyword arguments: xc_functional (str): CASTEP string for xc-functional to override CLI. """ query_dict = dict() if xc_functional is None: if isinstance(self.args.get('xc_functional'), list): xc_functional = self.args.get('xc_functional')[0] else: xc_functional = self.args.get('xc_functional') if xc_functional is not None: query_dict['xc_functional'] = xc_functional return query_dict def _query_spin(self): """ Query all calculations with spin polarisation, i.e. --spin n!=0, or non-spin-polarization, i.e. --spin 0. """ query_dict = dict() if isinstance(self.args.get('spin'), list): spin = self.args.get('spin')[0] else: spin = self.args.get('spin') if spin == 'any': query_dict = dict() elif int(spin) == 0: query_dict['spin_polarized'] = dict() query_dict['spin_polarized']['$ne'] = True elif int(spin) > 0: query_dict['spin_polarized'] = True return query_dict def _query_calc(self, doc): """ Find all structures with matching accuracy to specified structure. """ self._gs_enthalpy = doc['enthalpy_per_atom'] query_dict = {} query_dict['$and'] = [] query_dict['$and'].append(self._query_xc_functional(xc_functional=doc.get('xc_functional'))) query_dict['$and'].append(self._query_float_range( 'pressure', doc.get('pressure', 0.0), tolerance=self.args.get('pressure_tolerance') or 0.5)) if self.args.get('time') is not None: query_dict['$and'].append(self._query_time()) if 'spin_polarized' in doc and doc['spin_polarized']: if self.args.get('spin') != 'any': temp_dict = dict() temp_dict['spin_polarized'] = doc['spin_polarized'] query_dict['$and'].append(temp_dict) else: if self.args.get('spin') != 'any': temp_dict = dict() temp_dict['spin_polarized'] = dict() temp_dict['spin_polarized']['$ne'] = True query_dict['$and'].append(temp_dict) if doc.get('grid_scale', 1.75) == 1.75: temp_dict = dict() temp_dict['$or'] = [] temp_dict['$or'].append({'grid_scale': {'$exists': False}}) temp_dict['$or'].append({'grid_scale': {'$eq': 1.75}}) query_dict['$and'].append(temp_dict) else: temp_dict = dict() temp_dict['$or'] = [] temp_dict['$or'].append({'grid_scale': {'$eq': doc.get('grid_scale')}}) query_dict['$and'].append(temp_dict) if doc.get('fine_grid_scale', 1.75) == 1.75: temp_dict = dict() temp_dict['$or'] = [] temp_dict['$or'].append({'fine_grid_scale': {'$exists': False}}) temp_dict['$or'].append({'fine_grid_scale': {'$eq': 1.75}}) query_dict['$and'].append(temp_dict) else: temp_dict = dict() temp_dict['$or'] = [] temp_dict['$or'].append({'fine_grid_scale': {'$eq': doc.get('fine_grid_scale')}}) query_dict['$and'].append(temp_dict) if 'geom_force_tol' in doc and doc['geom_force_tol'] != 0.05: temp_dict = dict() temp_dict['geom_force_tol'] = doc['geom_force_tol'] query_dict['$and'].append(temp_dict) else: temp_dict = dict() temp_dict['$or'] = dict() temp_dict['$or'] = [] temp_dict['$or'].append({'geom_force_tol': {'$exists': False}}) temp_dict['$or'].append({'geom_force_tol': {'$eq': 0.05}}) query_dict['$and'].append(temp_dict) if 'sedc_scheme' in doc: temp_dict = dict() temp_dict['sedc_scheme'] = doc['sedc_scheme'] query_dict['$and'].append(temp_dict) else: temp_dict = dict() temp_dict['sedc_scheme'] = dict() temp_dict['sedc_scheme']['$exists'] = False query_dict['$and'].append(temp_dict) db = self.args.get('db') if isinstance(db, list): db = db[0] if self.args.get('loose') or (db is not None and 'oqmd' in db): return query_dict query_dict['$and'].append(self._query_float_range( 'kpoints_mp_spacing', doc.get('kpoints_mp_spacing'), tolerance=self.args.get('kpoint_tolerance') or 0.01)) query_dict['$and'].append(dict()) query_dict['$and'][-1]['cut_off_energy'] = doc['cut_off_energy'] if 'species_pot' in doc: for species in doc['species_pot']: temp_dict = dict() temp_dict['$or'] = [] temp_dict['$or'].append(dict()) temp_dict['$or'][-1]['species_pot.' + species] = dict() temp_dict['$or'][-1]['species_pot.' + species]['$exists'] = False temp_dict['$or'].append(dict()) temp_dict['$or'][-1]['species_pot.' + species] = doc['species_pot'][species] query_dict['$and'].append(temp_dict) if self.debug: print('Calc match dict:') print(dumps(query_dict, indent=2)) return query_dict def _query_time(self, since=False): """ Only include structures added before or after (depending on since) the date given in args['time']. Keyword arguments: since (bool): query before or after this time. """ from datetime import datetime, timedelta from time import mktime query_dict = dict() time_period = timedelta(days=int(self.args.get('time'))) time = (datetime.today() - time_period).timetuple() elapsed = str(hex(int(mktime(time))))[2:] cutoff_id = ObjectId(elapsed + '0000000000000000') query_dict['_id'] = dict() if since: query_dict['_id']['$gte'] = cutoff_id else: query_dict['_id']['$lte'] = cutoff_id return query_dict
[docs]class EmptyCursor: """ Empty cursor class for failures. """
[docs] @staticmethod def count_documents(*args, **kwargs): """ Dummy function always returns 0. """ return 0