Merge branch 'master' into patch-6

.github/workflows/_release.yml

@@ -95,16 +95,31 @@ jobs:
           PKG_NAME: ${{ needs.build.outputs.pkg-name }}
           VERSION: ${{ needs.build.outputs.version }}
         run: |
-          REGEX="^$PKG_NAME==\\d+\\.\\d+\\.\\d+\$"
-          echo $REGEX
-          PREV_TAG=$(git tag --sort=-creatordate | grep -P $REGEX || true | head -1)
+          PREV_TAG="${curr_version%.*}.$(( ${curr_version##*.} - 1 ))"; [[ "${curr_version##*.}" -eq 0 ]] && PREV_TAG=""
+
+          # backup case if releasing e.g. 0.3.0, looks up last release
+          # note if last release (chronologically) was e.g. 0.1.47 it will get 
+          # that instead of the last 0.2 release
+          if [ -z "$PREV_TAG" ]; then
+            REGEX="^$PKG_NAME==\\d+\\.\\d+\\.\\d+\$"
+            echo $REGEX
+            PREV_TAG=$(git tag --sort=-creatordate | grep -P $REGEX || true | head -1)
+          fi
+
+          # confirm prev-tag actually exists in git repo with git tag
+          if [ -z git tag -l "$PREV_TAG" ]; then
+            echo "Previous tag $PREV_TAG not found in git repo"
+            exit 1
+          fi
+
+
           TAG="${PKG_NAME}==${VERSION}"
           if [ "$TAG" == "$PREV_TAG" ]; then
             echo "No new version to release"
             exit 1
           fi
           echo tag="$TAG" >> $GITHUB_OUTPUT
-          echo prev-tag="$PREV_TAG" >> $GITHUB_OUTPUT
+          echo prev-tag="$PKG_NAME==$last_version" >> $GITHUB_OUTPUT
       - name: Generate release body
         id: generate-release-body
         working-directory: langchain

docs/docs/how_to/document_loader_csv.ipynb

@@ -157,7 +157,7 @@
     "    temp_file_path = temp_file.name\n",
     "\n",
     "loader = CSVLoader(file_path=temp_file_path)\n",
-    "loader.load()\n",
+    "data = loader.load()\n",
     "for record in data[:2]:\n",
     "    print(record)"
    ]

docs/src/theme/FeatureTables.js

@@ -300,6 +300,12 @@ const FEATURE_TABLES = {
                 package: "langchain-google_vertexai",
                 apiLink: "https://python.langchain.com/api_reference/google_vertexai/llms/langchain_google_vertexai.llms.VertexAI.html"
             },
+            {
+                name: "NVIDIA",
+                link: "NVIDIA",
+                package: "langchain-nvidia",
+                apiLink: "https://python.langchain.com/api_reference/nvidia_ai_endpoints/llm/langchain_nvidia_ai_endpoints.llm.NVIDIA.html"
+            },
         ],
     },
     text_embedding: {
@@ -387,6 +393,12 @@ const FEATURE_TABLES = {
                 package: "langchain-ibm",
                 apiLink: "https://python.langchain.com/api_reference/ibm/embeddings/langchain_ibm.embeddings.WatsonxEmbeddings.html"
             },
+            {
+                name: "NVIDIA",
+                link: "NVIDIAEmbeddings",
+                package: "langchain-nvidia",
+                apiLink: "https://python.langchain.com/api_reference/nvidia_ai_endpoints/embeddings/langchain_nvidia_ai_endpoints.embeddings.NVIDIAEmbeddings.html"
+            },
         ]
     },
     document_retrievers: {

docs/vercel.json

@@ -87,7 +87,7 @@
       "destination": "/docs/integrations/providers/:path*"
     },
     {
-      "source": "/docs/troubleshooting/errors/:path((?:GRAPH_RECURSION_LIMIT|INVALID_CONCURRENT_GRAPH_UPDATE|INVALID_GRAPH_NODE_RETURN_VALUE|MULTIPLE_SUBGRAPHS)/?)*",
+      "source": "/docs/troubleshooting/errors/:path((?:GRAPH_RECURSION_LIMIT|INVALID_CONCURRENT_GRAPH_UPDATE|INVALID_GRAPH_NODE_RETURN_VALUE|MULTIPLE_SUBGRAPHS|INVALID_CHAT_HISTORY)/?)*",
       "destination": "https://langchain-ai.github.io/langgraph/troubleshooting/errors/:path*"
     },
     {

libs/community/langchain_community/agent_toolkits/powerbi/toolkit.py

@@ -51,7 +51,7 @@ class PowerBIToolkit(BaseToolkit):
         examples: Optional. The examples for the prompt. Default is None.
         max_iterations: Optional. The maximum iterations to run. Default is 5.
         callback_manager: Optional. The callback manager. Default is None.
-        output_token_limit: Optional. The output token limit. Default is None.
+        output_token_limit: The output token limit. Default is 4000.
         tiktoken_model_name: Optional. The TikToken model name. Default is None.
     """
 
@@ -60,7 +60,7 @@ class PowerBIToolkit(BaseToolkit):
     examples: Optional[str] = None
     max_iterations: int = 5
     callback_manager: Optional[BaseCallbackManager] = None
-    output_token_limit: Optional[int] = None
+    output_token_limit: int = 4000
     tiktoken_model_name: Optional[str] = None
 
     model_config = ConfigDict(

libs/community/langchain_community/chat_models/baidu_qianfan_endpoint.py

@@ -68,9 +68,16 @@ def convert_message_to_dict(message: BaseMessage) -> dict:
         message_dict = {"role": "assistant", "content": message.content}
         if "function_call" in message.additional_kwargs:
             message_dict["function_call"] = message.additional_kwargs["function_call"]
-            # If function call only, content is None not empty string
-            if message_dict["content"] == "":
-                message_dict["content"] = None
+        elif len(message.tool_calls) != 0:
+            tool_call = message.tool_calls[0]
+            message_dict["function_call"] = {
+                "name": tool_call["name"],
+                "args": tool_call["args"],
+            }
+
+        # If function call only, content is None not empty string
+        if "function_call" in message_dict and message_dict["content"] == "":
+            message_dict["content"] = None
     elif isinstance(message, (FunctionMessage, ToolMessage)):
         message_dict = {
             "role": "function",

libs/core/langchain_core/output_parsers/list.py

@@ -1,9 +1,11 @@
 from __future__ import annotations
 
+import csv
 import re
 from abc import abstractmethod
 from collections import deque
 from collections.abc import AsyncIterator, Iterator
+from io import StringIO
 from typing import Optional as Optional
 from typing import TypeVar, Union
 
@@ -162,7 +164,14 @@ def parse(self, text: str) -> list[str]:
         Returns:
             A list of strings.
         """
-        return [part.strip() for part in text.split(",")]
+        try:
+            reader = csv.reader(
+                StringIO(text), quotechar='"', delimiter=",", skipinitialspace=True
+            )
+            return [item for sublist in reader for item in sublist]
+        except csv.Error:
+            # keep old logic for backup
+            return [part.strip() for part in text.split(",")]
 
     @property
     def _type(self) -> str:

libs/core/tests/unit_tests/output_parsers/test_list_parser.py

@@ -64,6 +64,25 @@ def test_multiple_items() -> None:
     assert list(parser.transform(iter([text]))) == [[a] for a in expected]
 
 
+def test_multiple_items_with_comma() -> None:
+    """Test that a string with multiple comma-separated items with 1 item containing a
+    comma is parsed to a list."""
+    parser = CommaSeparatedListOutputParser()
+    text = '"foo, foo2",bar,baz'
+    expected = ["foo, foo2", "bar", "baz"]
+
+    assert parser.parse(text) == expected
+    assert add(parser.transform(t for t in text)) == expected
+    assert list(parser.transform(t for t in text)) == [[a] for a in expected]
+    assert list(parser.transform(t for t in text.splitlines(keepends=True))) == [
+        [a] for a in expected
+    ]
+    assert list(
+        parser.transform(" " + t if i > 0 else t for i, t in enumerate(text.split(" ")))
+    ) == [[a] for a in expected]
+    assert list(parser.transform(iter([text]))) == [[a] for a in expected]
+
+
 def test_numbered_list() -> None:
     parser = NumberedListOutputParser()
     text1 = (