56 lines
1.3 KiB
Markdown
56 lines
1.3 KiB
Markdown
|
```swift title="Swift"
|
||
|
|
import Foundation
|
||
|
|
import Kreuzberg
|
||
|
|
import RustBridge
|
||
|
|
|
||
|
|
struct VectorRecord {
|
||
|
|
let id: String
|
||
|
|
let content: String
|
||
|
|
let embedding: [Float]
|
||
|
|
let metadata: [String: String]
|
||
|
|
}
|
||
|
|
|
||
|
|
let configJson = """
|
||
|
|
{
|
||
|
|
"chunking": {
|
||
|
|
"max_characters": 512,
|
||
|
|
"overlap": 50,
|
||
|
|
"embedding": {
|
||
|
|
"model": {"preset": {"name": "balanced"}},
|
||
|
|
"batch_size": 32,
|
||
|
|
"normalize": true
|
||
|
|
}
|
||
|
|
}
|
||
|
|
}
|
||
|
|
"""
|
||
|
|
|
||
|
|
let documentId = "doc_001"
|
||
|
|
let config = try extractionConfigFromJson(configJson)
|
||
|
|
let result = try extractFileSync("document.pdf", nil, config)
|
||
|
|
|
||
|
|
var records: [VectorRecord] = []
|
||
|
|
|
||
|
|
if let chunks = result.chunks() {
|
||
|
|
for (index, chunk) in chunks.enumerated() {
|
||
|
|
guard let embedding = chunk.embedding() else { continue }
|
||
|
|
|
||
|
|
let content = chunk.content().toString()
|
||
|
|
let vector = embedding.map { $0 }
|
||
|
|
|
||
|
|
var metadata: [String: String] = [:]
|
||
|
|
metadata["document_id"] = documentId
|
||
|
|
metadata["chunk_index"] = String(index)
|
||
|
|
metadata["content_length"] = String(content.count)
|
||
|
|
|
||
|
|
records.append(VectorRecord(
|
||
|
|
id: "\(documentId)_chunk_\(index)",
|
||
|
|
content: content,
|
||
|
|
embedding: vector,
|
||
|
|
metadata: metadata
|
||
|
|
))
|
||
|
|
}
|
||
|
|
}
|
||
|
|
|
||
|
|
print("Generated \(records.count) vector records")
|
||
|
|
```
|