highperfocused/open-webui
1
0
Fork 0
mirror of https://github.com/open-webui/open-webui.git synced 2025-03-17 21:32:42 +01:00
Code Issues Packages Projects Releases Wiki Activity

chore: format

Browse Source
This commit is contained in:
Timothy Jaeryang Baek 2025-03-05 19:17:41 -08:00
parent aaaebfabbe
commit d4fca9dabf
1 changed files with 81 additions and 102 deletions

183
backend/open_webui/retrieval/vector/dbs/elasticsearch.py

@ -1,30 +1,28 @@
from elasticsearch import Elasticsearch, BadRequestError from elasticsearch import Elasticsearch, BadRequestError
from typing import Optional from typing import Optional
import ssl import ssl
from elasticsearch.helpers import bulk,scan from elasticsearch.helpers import bulk, scan
from open_webui.retrieval.vector.main import VectorItem, SearchResult, GetResult from open_webui.retrieval.vector.main import VectorItem, SearchResult, GetResult
from open_webui.config import ( from open_webui.config import (
ELASTICSEARCH_URL, ELASTICSEARCH_URL,
ELASTICSEARCH_CA_CERTS, ELASTICSEARCH_CA_CERTS,
ELASTICSEARCH_API_KEY, ELASTICSEARCH_API_KEY,
ELASTICSEARCH_USERNAME, ELASTICSEARCH_USERNAME,
ELASTICSEARCH_PASSWORD, ELASTICSEARCH_PASSWORD,
ELASTICSEARCH_CLOUD_ID, ELASTICSEARCH_CLOUD_ID,
ELASTICSEARCH_INDEX_PREFIX, ELASTICSEARCH_INDEX_PREFIX,
SSL_ASSERT_FINGERPRINT, SSL_ASSERT_FINGERPRINT,
) )
class ElasticsearchClient: class ElasticsearchClient:
""" """
Important: Important:
in order to reduce the number of indexes and since the embedding vector length is fixed, we avoid creating in order to reduce the number of indexes and since the embedding vector length is fixed, we avoid creating
an index for each file but store it as a text field, while seperating to different index an index for each file but store it as a text field, while seperating to different index
baesd on the embedding length. baesd on the embedding length.
""" """
def __init__(self): def __init__(self):
self.index_prefix = ELASTICSEARCH_INDEX_PREFIX self.index_prefix = ELASTICSEARCH_INDEX_PREFIX
self.client = Elasticsearch( self.client = Elasticsearch(
@ -32,15 +30,19 @@ class ElasticsearchClient:
ca_certs=ELASTICSEARCH_CA_CERTS, ca_certs=ELASTICSEARCH_CA_CERTS,
api_key=ELASTICSEARCH_API_KEY, api_key=ELASTICSEARCH_API_KEY,
cloud_id=ELASTICSEARCH_CLOUD_ID, cloud_id=ELASTICSEARCH_CLOUD_ID,
basic_auth=(ELASTICSEARCH_USERNAME,ELASTICSEARCH_PASSWORD) if ELASTICSEARCH_USERNAME and ELASTICSEARCH_PASSWORD else None, basic_auth=(
ssl_assert_fingerprint=SSL_ASSERT_FINGERPRINT (ELASTICSEARCH_USERNAME, ELASTICSEARCH_PASSWORD)
if ELASTICSEARCH_USERNAME and ELASTICSEARCH_PASSWORD
else None
),
ssl_assert_fingerprint=SSL_ASSERT_FINGERPRINT,
) )
#Status: works
def _get_index_name(self,dimension:int)->str: # Status: works
def _get_index_name(self, dimension: int) -> str:
return f"{self.index_prefix}_d{str(dimension)}" return f"{self.index_prefix}_d{str(dimension)}"
#Status: works # Status: works
def _scan_result_to_get_result(self, result) -> GetResult: def _scan_result_to_get_result(self, result) -> GetResult:
if not result: if not result:
return None return None
@ -55,7 +57,7 @@ class ElasticsearchClient:
return GetResult(ids=[ids], documents=[documents], metadatas=[metadatas]) return GetResult(ids=[ids], documents=[documents], metadatas=[metadatas])
#Status: works # Status: works
def _result_to_get_result(self, result) -> GetResult: def _result_to_get_result(self, result) -> GetResult:
if not result["hits"]["hits"]: if not result["hits"]["hits"]:
return None return None
@ -70,7 +72,7 @@ class ElasticsearchClient:
return GetResult(ids=[ids], documents=[documents], metadatas=[metadatas]) return GetResult(ids=[ids], documents=[documents], metadatas=[metadatas])
#Status: works # Status: works
def _result_to_search_result(self, result) -> SearchResult: def _result_to_search_result(self, result) -> SearchResult:
ids = [] ids = []
distances = [] distances = []
@ -84,19 +86,21 @@ class ElasticsearchClient:
metadatas.append(hit["_source"].get("metadata")) metadatas.append(hit["_source"].get("metadata"))
return SearchResult( return SearchResult(
ids=[ids], distances=[distances], documents=[documents], metadatas=[metadatas] ids=[ids],
distances=[distances],
documents=[documents],
metadatas=[metadatas],
) )
#Status: works
# Status: works
def _create_index(self, dimension: int): def _create_index(self, dimension: int):
body = { body = {
"mappings": { "mappings": {
"dynamic_templates": [ "dynamic_templates": [
{ {
"strings": { "strings": {
"match_mapping_type": "string", "match_mapping_type": "string",
"mapping": { "mapping": {"type": "keyword"},
"type": "keyword"
}
} }
} }
], ],
@ -111,68 +115,52 @@ class ElasticsearchClient:
}, },
"text": {"type": "text"}, "text": {"type": "text"},
"metadata": {"type": "object"}, "metadata": {"type": "object"},
} },
} }
} }
self.client.indices.create(index=self._get_index_name(dimension), body=body) self.client.indices.create(index=self._get_index_name(dimension), body=body)
#Status: works
# Status: works
def _create_batches(self, items: list[VectorItem], batch_size=100): def _create_batches(self, items: list[VectorItem], batch_size=100):
for i in range(0, len(items), batch_size): for i in range(0, len(items), batch_size):
yield items[i : min(i + batch_size,len(items))] yield items[i : min(i + batch_size, len(items))]
#Status: works # Status: works
def has_collection(self,collection_name) -> bool: def has_collection(self, collection_name) -> bool:
query_body = {"query": {"bool": {"filter": []}}} query_body = {"query": {"bool": {"filter": []}}}
query_body["query"]["bool"]["filter"].append({"term": {"collection": collection_name}}) query_body["query"]["bool"]["filter"].append(
{"term": {"collection": collection_name}}
)
try: try:
result = self.client.count( result = self.client.count(index=f"{self.index_prefix}*", body=query_body)
index=f"{self.index_prefix}*",
body=query_body return result.body["count"] > 0
)
return result.body["count"]>0
except Exception as e: except Exception as e:
return None return None
def delete_collection(self, collection_name: str): def delete_collection(self, collection_name: str):
query = { query = {"query": {"term": {"collection": collection_name}}}
"query": {
"term": {"collection": collection_name}
}
}
self.client.delete_by_query(index=f"{self.index_prefix}*", body=query) self.client.delete_by_query(index=f"{self.index_prefix}*", body=query)
#Status: works
# Status: works
def search( def search(
self, collection_name: str, vectors: list[list[float]], limit: int self, collection_name: str, vectors: list[list[float]], limit: int
) -> Optional[SearchResult]: ) -> Optional[SearchResult]:
query = { query = {
"size": limit, "size": limit,
"_source": [ "_source": ["text", "metadata"],
"text",
"metadata"
],
"query": { "query": {
"script_score": { "script_score": {
"query": { "query": {
"bool": { "bool": {"filter": [{"term": {"collection": collection_name}}]}
"filter": [
{
"term": {
"collection": collection_name
}
}
]
}
}, },
"script": { "script": {
"source": "cosineSimilarity(params.vector, 'vector') + 1.0", "source": "cosineSimilarity(params.vector, 'vector') + 1.0",
"params": { "params": {
"vector": vectors[0] "vector": vectors[0]
}, # Assuming single query vector }, # Assuming single query vector
}, },
} }
}, },
@ -183,7 +171,8 @@ class ElasticsearchClient:
) )
return self._result_to_search_result(result) return self._result_to_search_result(result)
#Status: only tested halfwat
# Status: only tested halfwat
def query( def query(
self, collection_name: str, filter: dict, limit: Optional[int] = None self, collection_name: str, filter: dict, limit: Optional[int] = None
) -> Optional[GetResult]: ) -> Optional[GetResult]:
@ -197,7 +186,9 @@ class ElasticsearchClient:
for field, value in filter.items(): for field, value in filter.items():
query_body["query"]["bool"]["filter"].append({"term": {field: value}}) query_body["query"]["bool"]["filter"].append({"term": {field: value}})
query_body["query"]["bool"]["filter"].append({"term": {"collection": collection_name}}) query_body["query"]["bool"]["filter"].append(
{"term": {"collection": collection_name}}
)
size = limit if limit else 10 size = limit if limit else 10
try: try:
@ -206,59 +197,53 @@ class ElasticsearchClient:
body=query_body, body=query_body,
size=size, size=size,
) )
return self._result_to_get_result(result) return self._result_to_get_result(result)
except Exception as e: except Exception as e:
return None return None
#Status: works
def _has_index(self,dimension:int):
return self.client.indices.exists(index=self._get_index_name(dimension=dimension))
# Status: works
def _has_index(self, dimension: int):
return self.client.indices.exists(
index=self._get_index_name(dimension=dimension)
)
def get_or_create_index(self, dimension: int): def get_or_create_index(self, dimension: int):
if not self._has_index(dimension=dimension): if not self._has_index(dimension=dimension):
self._create_index(dimension=dimension) self._create_index(dimension=dimension)
#Status: works
# Status: works
def get(self, collection_name: str) -> Optional[GetResult]: def get(self, collection_name: str) -> Optional[GetResult]:
# Get all the items in the collection. # Get all the items in the collection.
query = { query = {
"query": { "query": {"bool": {"filter": [{"term": {"collection": collection_name}}]}},
"bool": { "_source": ["text", "metadata"],
"filter": [ }
{
"term": {
"collection": collection_name
}
}
]
}
}, "_source": ["text", "metadata"]}
results = list(scan(self.client, index=f"{self.index_prefix}*", query=query)) results = list(scan(self.client, index=f"{self.index_prefix}*", query=query))
return self._scan_result_to_get_result(results) return self._scan_result_to_get_result(results)
#Status: works # Status: works
def insert(self, collection_name: str, items: list[VectorItem]): def insert(self, collection_name: str, items: list[VectorItem]):
if not self._has_index(dimension=len(items[0]["vector"])): if not self._has_index(dimension=len(items[0]["vector"])):
self._create_index(dimension=len(items[0]["vector"])) self._create_index(dimension=len(items[0]["vector"]))
for batch in self._create_batches(items): for batch in self._create_batches(items):
actions = [ actions = [
{ {
"_index":self._get_index_name(dimension=len(items[0]["vector"])), "_index": self._get_index_name(dimension=len(items[0]["vector"])),
"_id": item["id"], "_id": item["id"],
"_source": { "_source": {
"collection": collection_name, "collection": collection_name,
"vector": item["vector"], "vector": item["vector"],
"text": item["text"], "text": item["text"],
"metadata": item["metadata"], "metadata": item["metadata"],
}, },
} }
for item in batch for item in batch
] ]
bulk(self.client,actions) bulk(self.client, actions)
# Upsert documents using the update API with doc_as_upsert=True. # Upsert documents using the update API with doc_as_upsert=True.
def upsert(self, collection_name: str, items: list[VectorItem]): def upsert(self, collection_name: str, items: list[VectorItem]):
@ -280,8 +265,7 @@ class ElasticsearchClient:
} }
for item in batch for item in batch
] ]
bulk(self.client,actions) bulk(self.client, actions)
# Delete specific documents from a collection by filtering on both collection and document IDs. # Delete specific documents from a collection by filtering on both collection and document IDs.
def delete( def delete(
@ -292,21 +276,16 @@ class ElasticsearchClient:
): ):
query = { query = {
"query": { "query": {"bool": {"filter": [{"term": {"collection": collection_name}}]}}
"bool": {
"filter": [
{"term": {"collection": collection_name}}
]
}
}
} }
#logic based on chromaDB # logic based on chromaDB
if ids: if ids:
query["query"]["bool"]["filter"].append({"terms": {"_id": ids}}) query["query"]["bool"]["filter"].append({"terms": {"_id": ids}})
elif filter: elif filter:
for field, value in filter.items(): for field, value in filter.items():
query["query"]["bool"]["filter"].append({"term": {f"metadata.{field}": value}}) query["query"]["bool"]["filter"].append(
{"term": {f"metadata.{field}": value}}
)
self.client.delete_by_query(index=f"{self.index_prefix}*", body=query) self.client.delete_by_query(index=f"{self.index_prefix}*", body=query)