Files
fil/docs/snippets/zig/metadata/vector_database_integration.md

48 lines
1.4 KiB
Markdown
Raw Permalink Normal View History

2026-06-01 23:40:55 +02:00
```zig title="Zig"
const std = @import("std");
const kreuzberg = @import("kreuzberg");
pub fn main() !void {
var gpa = std.heap.GeneralPurposeAllocator(.{}){};
defer _ = gpa.deinit();
const allocator = gpa.allocator();
const config_json =
\\{
\\ "chunking": {
\\ "max_characters": 512,
\\ "overlap": 50,
\\ "embedding": {
\\ "model": "balanced",
\\ "normalize": true
\\ }
\\ }
\\}
;
const result_json = try kreuzberg.extract_file_sync("document.pdf", null, config_json);
defer std.heap.c_allocator.free(result_json);
var parsed = try std.json.parseFromSlice(std.json.Value, allocator, result_json, .{});
defer parsed.deinit();
const root = parsed.value;
if (root != .object) return;
const stdout = std.io.getStdOut().writer();
const chunks_val = root.object.get("chunks") orelse return;
if (chunks_val != .array) return;
for (chunks_val.array.items, 0..) |chunk, index| {
if (chunk != .object) continue;
const embedding_val = chunk.object.get("embedding") orelse continue;
if (embedding_val != .array) continue;
try stdout.print("Chunk {d}: {d} dimensions\n", .{ index, embedding_val.array.items.len });
// Store embedding_val.array.items in vector database
}
}
```