core,langchain,community[patch]: allow langsmith 0.2 (#28598)

libs/community/langchain_community/embeddings/fastembed.py

@@ -1,6 +1,6 @@
 import importlib
 import importlib.metadata
-from typing import Any, Dict, List, Literal, Optional
+from typing import Any, Dict, List, Literal, Optional, cast
 
 import numpy as np
 from langchain_core.embeddings import Embeddings
@@ -117,7 +117,7 @@ def embed_documents(self, texts: List[str]) -> List[List[float]]:
             embeddings = self.model.embed(
                 texts, batch_size=self.batch_size, parallel=self.parallel
             )
-        return [e.tolist() for e in embeddings]
+        return [cast(List[float], e.tolist()) for e in embeddings]
 
     def embed_query(self, text: str) -> List[float]:
         """Generate query embeddings using FastEmbed.
@@ -133,4 +133,4 @@ def embed_query(self, text: str) -> List[float]:
                 text, batch_size=self.batch_size, parallel=self.parallel
             )
         )
-        return query_embeddings.tolist()
+        return cast(List[float], query_embeddings.tolist())

libs/community/langchain_community/embeddings/laser.py

@@ -1,4 +1,4 @@
-from typing import Any, Dict, List, Optional
+from typing import Any, Dict, List, Optional, cast
 
 import numpy as np
 from langchain_core.embeddings import Embeddings
@@ -73,7 +73,7 @@ def embed_documents(self, texts: List[str]) -> List[List[float]]:
         embeddings: np.ndarray
         embeddings = self._encoder_pipeline.encode_sentences(texts)
 
-        return embeddings.tolist()
+        return cast(List[List[float]], embeddings.tolist())
 
     def embed_query(self, text: str) -> List[float]:
         """Generate single query text embeddings using LASER.
@@ -86,4 +86,4 @@ def embed_query(self, text: str) -> List[float]:
         """
         query_embeddings: np.ndarray
         query_embeddings = self._encoder_pipeline.encode_sentences([text])
-        return query_embeddings.tolist()[0]
+        return cast(List[List[float]], query_embeddings.tolist())[0]

libs/community/langchain_community/utilities/redis.py

@@ -2,7 +2,7 @@
 
 import logging
 import re
-from typing import TYPE_CHECKING, Any, List, Optional, Pattern
+from typing import TYPE_CHECKING, Any, List, Optional, Pattern, cast
 from urllib.parse import urlparse
 
 import numpy as np
@@ -18,7 +18,7 @@ def _array_to_buffer(array: List[float], dtype: Any = np.float32) -> bytes:
 
 
 def _buffer_to_array(buffer: bytes, dtype: Any = np.float32) -> List[float]:
-    return np.frombuffer(buffer, dtype=dtype).tolist()
+    return cast(List[float], np.frombuffer(buffer, dtype=dtype).tolist())
 
 
 class TokenEscaper:

libs/community/langchain_community/vectorstores/semadb.py

@@ -1,4 +1,4 @@
-from typing import Any, Iterable, List, Optional, Tuple
+from typing import Any, Iterable, List, Optional, Tuple, cast
 from uuid import uuid4
 
 import numpy as np
@@ -111,7 +111,7 @@ def add_texts(
             embed_matrix = embed_matrix / np.linalg.norm(
                 embed_matrix, axis=1, keepdims=True
             )
-            embeddings = embed_matrix.tolist()
+            embeddings = cast(List[List[float]], embed_matrix.tolist())
         # Create points
         ids: List[str] = []
         points = []
@@ -186,7 +186,7 @@ def _search_points(self, embedding: List[float], k: int = 4) -> List[dict]:
         if self.distance_strategy == DistanceStrategy.COSINE:
             vec = np.array(embedding)
             vec = vec / np.linalg.norm(vec)
-            embedding = vec.tolist()
+            embedding = cast(List[float], vec.tolist())
         # Perform search request
         payload = {
             "vector": embedding,

libs/community/langchain_community/vectorstores/usearch.py

@@ -1,6 +1,6 @@
 from __future__ import annotations
 
-from typing import Any, Dict, Iterable, List, Optional, Tuple, Union
+from typing import Any, Dict, Iterable, List, Optional, Tuple, Union, cast
 
 import numpy as np
 from langchain_core.documents import Document
@@ -75,7 +75,7 @@ def add_texts(
         self.index.add(np.array(ids), np.array(embeddings))
         self.docstore.add(dict(zip(ids, documents)))
         self.ids.extend(ids)
-        return ids.tolist()
+        return cast(List[str], ids.tolist())
 
     def similarity_search_with_score(
         self,
@@ -171,4 +171,4 @@ def from_texts(
         usearch = guard_import("usearch.index")
         index = usearch.Index(ndim=len(embeddings[0]), metric=metric)
         index.add(np.array(ids), np.array(embeddings))
-        return cls(embedding, index, docstore, ids.tolist())
+        return cls(embedding, index, docstore, cast(List[str], ids.tolist()))

libs/community/pyproject.toml

@@ -39,7 +39,7 @@ aiohttp = "^3.8.3"
 tenacity = ">=8.1.0,!=8.4.0,<10"
 dataclasses-json = ">= 0.5.7, < 0.7"
 pydantic-settings = "^2.4.0"
-langsmith = "^0.1.125"
+langsmith = ">=0.1.125,<0.3"
 httpx-sse = "^0.4.0"
 [[tool.poetry.dependencies.numpy]]
 version = ">=1.22.4,<2"

libs/community/tests/integration_tests/vectorstores/test_singlestoredb.py

@@ -3,7 +3,7 @@
 import math
 import os
 import tempfile
-from typing import List
+from typing import List, cast
 
 import numpy as np
 import pytest
@@ -60,13 +60,13 @@ class RandomEmbeddings(Embeddings):
     """Fake embeddings with random vectors. For testing purposes."""
 
     def embed_documents(self, texts: List[str]) -> List[List[float]]:
-        return [np.random.rand(100).tolist() for _ in texts]
+        return [cast(list[float], np.random.rand(100).tolist()) for _ in texts]
 
     def embed_query(self, text: str) -> List[float]:
-        return np.random.rand(100).tolist()
+        return cast(list[float], np.random.rand(100).tolist())
 
     def embed_image(self, uris: List[str]) -> List[List[float]]:
-        return [np.random.rand(100).tolist() for _ in uris]
+        return [cast(list[float], np.random.rand(100).tolist()) for _ in uris]
 
 
 class IncrementalEmbeddings(Embeddings):

libs/community/tests/unit_tests/vectorstores/test_utils.py

@@ -1,5 +1,7 @@
 """Test vector store utility functions."""
 
+from typing import cast
+
 import numpy as np
 from langchain_core.documents import Document
 
@@ -53,7 +55,7 @@ def test_maximal_marginal_relevance() -> None:
 def test_maximal_marginal_relevance_query_dim() -> None:
     query_embedding = np.random.random(size=5)
     query_embedding_2d = query_embedding.reshape((1, 5))
-    embedding_list = np.random.random(size=(4, 5)).tolist()
+    embedding_list = cast(list[list[float]], np.random.random(size=(4, 5)).tolist())
     first = maximal_marginal_relevance(query_embedding, embedding_list)
     second = maximal_marginal_relevance(query_embedding_2d, embedding_list)
     assert first == second

libs/core/langchain_core/tracers/context.py

@@ -6,6 +6,7 @@
 from typing import (
     TYPE_CHECKING,
     Any,
+    Literal,
     Optional,
     Union,
     cast,
@@ -141,7 +142,7 @@ def _get_trace_callbacks(
     return cb
 
 
-def _tracing_v2_is_enabled() -> bool:
+def _tracing_v2_is_enabled() -> Union[bool, Literal["local"]]:
     if tracing_v2_callback_var.get() is not None:
         return True
     return ls_utils.tracing_is_enabled()