Coverage for /builds/kinetik161/ase/ase/symbols.py: 96.08%

1import collections.abc 

2import numbers 

3import warnings 

4from typing import Dict, Iterator, List, Sequence, Set, Union 

5 

6import numpy as np 

7 

8from ase.data import atomic_numbers, chemical_symbols 

9from ase.formula import Formula 

10 

11Integers = Union[Sequence[int], np.ndarray] 

12 

13 

14def string2symbols(s: str) -> List[str]: 

15 """Convert string to list of chemical symbols.""" 

16 return list(Formula(s)) 

17 

18 

19def symbols2numbers(symbols) -> List[int]: 

20 if isinstance(symbols, str): 

21 symbols = string2symbols(symbols) 

22 numbers = [] 

23 for s in symbols: 

24 if isinstance(s, str): 

25 numbers.append(atomic_numbers[s]) 

26 else: 

27 numbers.append(int(s)) 

28 return numbers 

29 

30 

31class Symbols(collections.abc.Sequence): 

32 """A sequence of chemical symbols. 

33 

34 ``atoms.symbols`` is a :class:`ase.symbols.Symbols` object. This 

35 object works like an editable view of ``atoms.numbers``, except 

36 its elements are manipulated as strings. 

37 

38 Examples: 

39 

40 >>> from ase.build import molecule 

41 >>> atoms = molecule('CH3CH2OH') 

42 >>> atoms.symbols 

43 Symbols('C2OH6') 

44 >>> atoms.symbols[:3] 

45 Symbols('C2O') 

46 >>> atoms.symbols == 'H' # doctest: +ELLIPSIS 

47 array([False, False, False, True, True, True, True, True, True]...) 

48 >>> atoms.symbols[-3:] = 'Pu' 

49 >>> atoms.symbols 

50 Symbols('C2OH3Pu3') 

51 >>> atoms.symbols[3:6] = 'Mo2U' 

52 >>> atoms.symbols 

53 Symbols('C2OMo2UPu3') 

54 >>> atoms.symbols.formula 

55 Formula('C2OMo2UPu3') 

56 

57 The :class:`ase.formula.Formula` object is useful for extended 

58 formatting options and analysis. 

59 

60 """ 

61 

62 def __init__(self, numbers) -> None: 

63 self.numbers = np.asarray(numbers, int) 

64 

65 @classmethod 

66 def fromsymbols(cls, symbols) -> 'Symbols': 

67 numbers = symbols2numbers(symbols) 

68 return cls(np.array(numbers)) 

69 

70 @property 

71 def formula(self) -> Formula: 

72 """Formula object.""" 

73 string = Formula.from_list(self).format('reduce') 

74 return Formula(string) 

75 

76 def __getitem__(self, key) -> Union['Symbols', str]: 

77 num = self.numbers[key] 

78 if isinstance(num, numbers.Integral): 

79 return chemical_symbols[num] 

80 return Symbols(num) 

81 

82 def __iter__(self) -> Iterator[str]: 

83 for num in self.numbers: 

84 yield chemical_symbols[num] 

85 

86 def __setitem__(self, key, value) -> None: 

87 numbers = symbols2numbers(value) 

88 if len(numbers) == 1: 

89 self.numbers[key] = numbers[0] 

90 else: 

91 self.numbers[key] = numbers 

92 

93 def __len__(self) -> int: 

94 return len(self.numbers) 

95 

96 def __str__(self) -> str: 

97 return self.get_chemical_formula('reduce') 

98 

99 def __repr__(self) -> str: 

100 return f'Symbols(\'{self}\')' 

101 

102 def __eq__(self, obj) -> bool: 

103 if not hasattr(obj, '__len__'): 

104 return False 

105 

106 try: 

107 symbols = Symbols.fromsymbols(obj) 

108 except Exception: 

109 # Typically this would happen if obj cannot be converged to 

110 # atomic numbers. 

111 return False 

112 return self.numbers == symbols.numbers 

113 

114 def get_chemical_formula( 

115 self, 

116 mode: str = 'hill', 

117 empirical: bool = False, 

118 ) -> str: 

119 """Get chemical formula. 

120 

121 See documentation of ase.atoms.Atoms.get_chemical_formula().""" 

122 # XXX Delegate the work to the Formula object! 

123 if mode in ('reduce', 'all') and empirical: 

124 warnings.warn("Empirical chemical formula not available " 

125 "for mode '{}'".format(mode)) 

126 

127 if len(self) == 0: 

128 return '' 

129 

130 numbers = self.numbers 

131 

132 if mode == 'reduce': 

133 n = len(numbers) 

134 changes = np.concatenate(([0], np.arange(1, n)[numbers[1:] != 

135 numbers[:-1]])) 

136 symbols = [chemical_symbols[e] for e in numbers[changes]] 

137 counts = np.append(changes[1:], n) - changes 

138 

139 tokens = [] 

140 for s, c in zip(symbols, counts): 

141 tokens.append(s) 

142 if c > 1: 

143 tokens.append(str(c)) 

144 formula = ''.join(tokens) 

145 elif mode == 'all': 

146 formula = ''.join([chemical_symbols[n] for n in numbers]) 

147 else: 

148 symbols = [chemical_symbols[Z] for Z in numbers] 

149 f = Formula('', _tree=[(symbols, 1)]) 

150 if empirical: 

151 f, _ = f.reduce() 

152 if mode in {'hill', 'metal'}: 

153 formula = f.format(mode) 

154 else: 

155 raise ValueError( 

156 "Use mode = 'all', 'reduce', 'hill' or 'metal'.") 

157 

158 return formula 

159 

160 def search(self, symbols) -> Integers: 

161 """Return the indices of elements with given symbol or symbols.""" 

162 numbers = set(symbols2numbers(symbols)) 

163 indices = [i for i, number in enumerate(self.numbers) 

164 if number in numbers] 

165 return np.array(indices, int) 

166 

167 def species(self) -> Set[str]: 

168 """Return unique symbols as a set.""" 

169 return set(self) 

170 

171 def indices(self) -> Dict[str, Integers]: 

172 """Return dictionary mapping each unique symbol to indices. 

173 

174 >>> from ase.build import molecule 

175 >>> atoms = molecule('CH3CH2OH') 

176 >>> atoms.symbols.indices() 

177 {'C': array([0, 1]), 'O': array([2]), 'H': array([3, 4, 5, 6, 7, 8])} 

178 

179 """ 

180 dct: Dict[str, List[int]] = {} 

181 for i, symbol in enumerate(self): 

182 dct.setdefault(symbol, []).append(i) 

183 return {key: np.array(value, int) for key, value in dct.items()} 

184 

185 def species_indices(self) -> Sequence[int]: 

186 """Return the indices of each atom within their individual species. 

187 

188 >>> from ase import Atoms 

189 >>> atoms = Atoms('CH3CH2OH') 

190 >>> atoms.symbols.species_indices() 

191 [0, 0, 1, 2, 1, 3, 4, 0, 5] 

192 

193 ^ ^ ^ ^ ^ ^ ^ ^ ^ 

194 C H H H C H H O H 

195 

196 """ 

197 

198 counts: Dict[str, int] = {} 

199 result = [] 

200 for i, n in enumerate(self.numbers): 

201 counts[n] = counts.get(n, -1) + 1 

202 result.append(counts[n]) 

203 

204 return result