File size: 2,052 Bytes
cdfaab8
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
cf386b1
cdfaab8
 
 
 
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
from __future__ import annotations
from pinecone import Pinecone, ServerlessSpec
from typing import Any
import time

class PineconeVectorStore:
    def __init__(self, api_key: str, index_name: str, dimension: int, cloud: str, region: str):
        self.pc = Pinecone(api_key=api_key)
        self.index_name = index_name
        self.dimension = dimension
        self.cloud = cloud
        self.region = region

        self._ensure_index()
        self.index = self.pc.Index(index_name)

    def _ensure_index(self):
        existing = {idx["name"] for idx in self.pc.list_indexes()}
        if self.index_name not in existing:
            self.pc.create_index(
                name=self.index_name,
                dimension=self.dimension,
                metric="cosine",
                spec=ServerlessSpec(cloud=self.cloud, region=self.region),
            )
            for _ in range(30):
                desc = self.pc.describe_index(self.index_name)
                if desc.status.get("ready"):
                    break
                time.sleep(1)

    def upsert(self, vectors: list[dict], namespace: str = "resumes"):
        if not vectors:
            return
        self.index.upsert(vectors=vectors, namespace=namespace)

    def query(self, vector: list[float], top_k: int = 8, namespace: str = "resumes",
              filter: dict | None = None) -> list[dict[str, Any]]:
        res = self.index.query(
            vector=vector,
            top_k=top_k,
            include_metadata=True,
            include_values=False,
            namespace=namespace,
            filter=filter
        )
        matches = res.get("matches", []) if isinstance(res, dict) else res.matches
        out = []
        for m in matches:
            if isinstance(m, dict):
                out.append(m)
            else:
                out.append({"id": m.id, "score": m.score, "metadata": m.metadata})
        return out

    def delete_namespace(self, namespace: str = "resumes"):
        self.index.delete(delete_all=True, namespace=namespace)