Coverage for scripts/scratch.py: 4%

1import timeit 

2from src.overturetoosm.resources import places_tags 

3from src.overturetoosm.places import process_place 

4 

5 

6sample = { 

7 "id": "123", 

8 "version": 1, 

9 "update_time": "2022-01-01T00:00:00Z", 

10 "sources": [ 

11 { 

12 "property": "property1", 

13 "dataset": "dataset1", 

14 "record_id": "record1", 

15 "confidence": 0.8, 

16 } 

17 ], 

18 "names": { 

19 "primary": "Primary Name", 

20 "common": "Common Name", 

21 "rules": "Rules Name", 

22 }, 

23 "brand": { 

24 "wikidata": "Q123", 

25 "names": { 

26 "primary": "Brand Name", 

27 "common": "Common Name", 

28 "rules": "Rules Name", 

29 }, 

30 }, 

31 "categories": { 

32 "main": "notary_public", 

33 "alternate": ["alternate_category1", "alternate_category2"], 

34 }, 

35 "confidence": 0.8, 

36 "websites": ["https://example.com"], 

37 "socials": ["www.facebook.com/example"], 

38 "phones": ["+1234567890"], 

39 "addresses": [ 

40 { 

41 "freeform": "123 Main St", 

42 "locality": "City", 

43 "postcode": "12345", 

44 "region": "State", 

45 "country": "Country", 

46 } 

47 ], 

48} 

49 

50after = { 

51 "names.primary": "Primary Name", 

52 "brand.wikidata": "Q123", 

53 "brand.names.primary": "Brand Name", 

54 "confidence": 0.8, 

55 "websites": "https://example.com", 

56 "socials": ["www.facebook.com/example"], 

57 "phones": "+1234567890", 

58 "freeform": "123 Main St", 

59 "locality": "City", 

60 "postcode": "12345", 

61 "region": "State", 

62 "country": "Country", 

63 "source": "dataset1 via overturetoosm", 

64 "office": "lawyer", 

65 "lawyer": "notary", 

66} 

67 

68 

69mapping = { 

70 "names.primary": "name", 

71 "brand.wikidata": "brand:wikidata", 

72 "brand.names.primary": "brand", 

73 "phones": "phone", 

74 "freeform": "addr:street_address", 

75 "locality": "addr:city", 

76 "postcode": "addr:postcode", 

77 "region": "addr:state", 

78 "country": "addr:country", 

79 "websites": "website", 

80} 

81 

82 

83def flatten(dictionary: dict, parent_key: str = "", separator: str = ".") -> dict: 

84 """Flatten a nested dictionary into a single-level dictionary.""" 

85 items = [] 

86 for key, value in dictionary.items(): 

87 new_key = parent_key + separator + key if parent_key else key 

88 if isinstance(value, dict): 

89 items.extend(flatten(value, new_key, separator=separator).items()) 

90 elif not new_key.endswith(("rules", "common")): 

91 items.append((new_key, value)) 

92 return dict(items) 

93 

94 

95def get_first_item(props: dict, keys: list[str]): 

96 for key in keys: 

97 props[key] = props.get(key, [])[0] 

98 return props 

99 

100 

101def handle(props: dict) -> dict: 

102 flattened = flatten(props) 

103 x = flattened | flatten( 

104 flattened.get("addresses", [])[0], parent_key="addr", separator=":" 

105 ) 

106 x["source"] = ( 

107 ", ".join(source["dataset"] for source in x.get("sources", [])) 

108 + " via overturetoosm" 

109 ) 

110 x = get_first_item(x, ["websites", "phones"]) 

111 x = x | places_tags.get(x.get("categories.main", ""), {}) 

112 for social in x.get("socials", []): 

113 if "facebook" in social: 

114 x["contact:facebook"] = social 

115 elif "twitter" in social: 

116 x["contact:twitter"] = social 

117 remove = [ 

118 "update_time", 

119 "id", 

120 "version", 

121 "addresses", 

122 "sources", 

123 "categories.main", 

124 "categories.alternate", 

125 "socials", 

126 "confidence", 

127 ] 

128 return {mapping.get(k, k): v for k, v in x.items() if k not in remove} 

129 

130 

131if 0: 

132 a = timeit.timeit( 

133 "handle(sample)", setup="from __main__ import sample, handle", number=100000 

134 ) 

135 b = timeit.timeit( 

136 "process_props(sample)", 

137 setup="from __main__ import sample, process_props", 

138 number=100000, 

139 ) 

140 print("here:", a) 

141 print("finished:", b) 

142else: 

143 print(sample) 

144 print(process_place(sample))