elid 0.4.24

Embedding Locality IDentifier - encode embeddings into sortable string IDs for vector search without vector stores, plus fast string similarity algorithms
Documentation 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159

"""
Basic usage examples for ELID Python bindings.

Run with: python basic_usage.py

First, install the package:
  maturin develop
"""

import elid

print("=== ELID Python Example ===\n")

# Example 1: Levenshtein Distance
print("1. Levenshtein Distance:")
distance = elid.levenshtein("kitten", "sitting")
print(f"   Distance between 'kitten' and 'sitting': {distance}")

similarity = elid.normalized_levenshtein("kitten", "sitting")
print(f"   Normalized similarity: {similarity:.2f}\n")

# Example 2: Jaro-Winkler Similarity
print("2. Jaro-Winkler Similarity (best for names):")
names = [
    ("Martha", "Marhta"),
    ("John Smith", "Jon Smith"),
    ("DIXON", "DICKSON"),
]

for name1, name2 in names:
    jaro_sim = elid.jaro(name1, name2)
    jw_sim = elid.jaro_winkler(name1, name2)
    print(f"   '{name1}' vs '{name2}':")
    print(f"     Jaro: {jaro_sim:.3f}, Jaro-Winkler: {jw_sim:.3f}")
print()

# Example 3: Finding best match
print("3. Finding Best Match:")
candidates = ["apple", "application", "apply", "apricot", "banana"]
query = "app"

result = elid.find_best_match(query, candidates)
print(f"   Query: '{query}'")
print(
    f"   Best match: '{candidates[result['index']]}' (score: {result['score']:.3f})\n"
)

# Example 4: Finding all matches above threshold
print("4. Finding All Matches Above Threshold:")
matches = elid.find_matches_above_threshold(query, candidates, 0.5)
print(f"   Query: '{query}', Threshold: 0.5")
print("   Matches:")
for match in matches:
    print(f"     - '{candidates[match['index']]}' (score: {match['score']:.3f})")
print()

# Example 5: Product Search Simulation
print("5. Product Search Simulation:")
products = [
    "iPhone 14 Pro Max",
    "iPhone 14 Pro",
    "iPhone 14",
    "iPhone 13 Pro",
    "Samsung Galaxy S23",
    "Google Pixel 7",
]

searches = ["iphone 14 pro", "galaxy", "pixel"]

for search in searches:
    # Score all products
    scored = [
        {
            "index": i,
            "product": product,
            "score": elid.best_match(search.lower(), product.lower()),
        }
        for i, product in enumerate(products)
    ]

    # Sort by score descending
    scored.sort(key=lambda x: x["score"], reverse=True)

    print(f"   Search: '{search}'")
    print("   Top 3 matches:")
    for match in scored[:3]:
        print(f"     - {match['product']} (score: {match['score']:.3f})")
    print()

# Example 6: Using SimilarityOpts
print("6. Using SimilarityOpts (case-insensitive, trimmed):")
opts = elid.SimilarityOpts(case_sensitive=False, trim_whitespace=True)

pairs = [
    ("  HELLO  ", "hello"),
    ("World", "WORLD"),
    ("\tTest\n", "test"),
]

for str1, str2 in pairs:
    dist = elid.levenshtein_with_opts(str1, str2, opts)
    print(f"   '{str1}' vs '{str2}': distance = {dist}")
print()

# Example 7: Hamming Distance
print("7. Hamming Distance (equal-length strings):")
hamming_pairs = [
    ("ACGTACGT", "ACGTACCT"),
    ("hello", "hallo"),
]

for seq1, seq2 in hamming_pairs:
    dist = elid.hamming(seq1, seq2)
    if dist is not None:
        print(f"   '{seq1}' vs '{seq2}': {dist} differences")
print()

# Example 8: OSA Distance (with transpositions)
print("8. OSA Distance (handles transpositions):")
osa_pairs = [
    ("ca", "ac"),
    ("abcd", "acbd"),
]

for str1, str2 in osa_pairs:
    osa = elid.osa_distance(str1, str2)
    lev = elid.levenshtein(str1, str2)
    print(f"   '{str1}' vs '{str2}':")
    print(f"     OSA: {osa}, Levenshtein: {lev}")
print()

# Example 9: Spell Checker
print("9. Simple Spell Checker:")
dictionary = ["receive", "believe", "achieve", "receive", "ceiling", "deceive"]
misspellings = ["recieve", "beleive", "achive", "cieling"]

for misspelling in misspellings:
    result = elid.find_best_match(misspelling, dictionary)
    suggestion = dictionary[result["index"]]
    print(f"   '{misspelling}' → '{suggestion}' (confidence: {result['score']:.2%})")
print()

# Example 10: Name Deduplication
print("10. Name Deduplication:")
names = [
    ("John Smith", "Jon Smith"),
    ("Robert Johnson", "Bob Johnson"),
    ("Mary Williams", "Marie Williams"),
    ("James Brown", "Jane Brown"),
]

threshold = 0.85
for name1, name2 in names:
    similarity = elid.jaro_winkler(name1, name2)
    is_duplicate = similarity > threshold
    status = "✓ Likely duplicate" if is_duplicate else "✗ Different people"
    print(f"   '{name1}' vs '{name2}': {similarity:.3f} - {status}")

print("\n=== End of Examples ===")