langchain-dev-utils 1.4.0__tar.gz → 1.4.2__tar.gz

{langchain_dev_utils-1.4.0 → langchain_dev_utils-1.4.2}/.vscode/settings.json

@@ -3,5 +3,6 @@
   "python.testing.unittestEnabled": false,
   "python.testing.pytestEnabled": true,
   "search.useIgnoreFiles": true,
-  "python.analysis.typeCheckingMode": "off"
+  "python.analysis.typeCheckingMode": "off",
+  "python.languageServer": "None"
 }

langchain_dev_utils-1.4.2/AGENTS.md

@@ -0,0 +1,173 @@
+# Agentic Coding Guidelines for langchain-dev-utils
+
+This file provides guidelines for AI agents working in this repository.
+
+## Project Overview
+
+A Python utility library for LangChain and LangGraph development. Uses hatchling build system, uv for package management, ruff for linting, and pytest for testing.
+
+---
+
+## Build/Lint/Test Commands
+
+### Package Management (uv)
+```bash
+# Install dependencies
+uv sync
+
+# Install with optional dependencies
+uv sync --extra standard
+
+# Install dev dependencies
+uv sync --group dev
+
+# Install test dependencies
+uv sync --group tests
+```
+
+### Linting & Formatting (ruff)
+```bash
+# Check linting
+uv run ruff check .
+
+# Check specific file
+uv run ruff check src/langchain_dev_utils/path/to/file.py
+
+# Fix auto-fixable issues
+uv run ruff check --fix .
+
+# Format code
+uv run ruff format .
+
+# Format specific file
+uv run ruff format src/langchain_dev_utils/path/to/file.py
+```
+
+### Testing (pytest)
+```bash
+# Run all tests
+uv run pytest
+
+# Run with verbose output
+uv run pytest -v
+
+# Run specific test file
+uv run pytest tests/test_agent.py
+
+# Run specific test function
+uv run pytest tests/test_agent.py::test_prebuilt_agent
+
+# Run specific test class
+uv run pytest tests/test_chat_models.py::TestImageProcessing
+
+# Run with asyncio support (automatically configured)
+uv run pytest -s
+
+# Run tests matching pattern
+uv run pytest -k "test_load"
+```
+
+### Build
+```bash
+# Build package
+uv build
+
+# Build wheel only
+uv build --wheel
+```
+
+---
+
+## Code Style Guidelines
+
+### Imports
+- Use absolute imports for external packages
+- Use relative imports within the package (e.g., `from ..chat_models import ...`)
+- Group imports: stdlib → third-party → local
+- Ruff handles import sorting automatically
+
+### Type Hints
+- Use type hints for all function parameters and return types
+- Use `Optional[Type]` or `Type | None` for nullable types (both acceptable)
+- Use `Any` sparingly and only when necessary
+- Use `Sequence`, `Mapping` for generic collections
+- Use generics with TypeVars where appropriate
+
+### Naming Conventions
+- `snake_case` for functions, methods, variables
+- `PascalCase` for classes
+- `UPPER_CASE` for constants
+- `_leading_underscore` for private/internal functions
+- Leading double underscore for name mangling when needed
+
+### Docstrings
+- Use Google-style docstrings
+- Include Args, Returns, Raises sections for public functions
+- Include Examples section for complex functions
+- Keep docstrings under 100 characters per line when possible
+
+### Code Structure
+- Maximum line length: 88 characters (ruff enforces, E501 ignored)
+- Use trailing commas in multi-line structures
+- Two blank lines between top-level functions/classes
+- One blank line between methods
+
+### Error Handling
+- Use specific exception types, not bare `except:`
+- Provide descriptive error messages with f-strings
+- Use `raise ValueError(msg)` pattern with descriptive messages
+- Avoid bare `raise` statements
+
+### Async Code
+- Use `pytest.mark.asyncio` for async test functions
+- Use `async`/`await` consistently
+- Prefer `asyncio` primitives from standard library
+
+### Ruff Configuration
+- Enabled rules: E, F, I, PGH003, T201
+- Import sorting (I) is enforced
+- No print statements in production code (T201)
+
+---
+
+## Testing Guidelines
+
+- Tests live in `tests/` directory
+- Test files named `test_*.py`
+- Test functions named `test_*`
+- Use pytest fixtures for setup/teardown
+- Use `pytest.mark.asyncio` for async tests
+- Mock external API calls in unit tests
+- Integration tests use actual APIs (marked implicitly by test names)
+- Tests use `langchain-tests` for standard integration test suites
+
+---
+
+## Project Structure
+
+```
+langchain-dev-utils/
+├── src/langchain_dev_utils/      # Source code
+│   ├── agents/                   # Agent utilities
+│   ├── chat_models/              # Chat model utilities
+│   ├── embeddings/               # Embedding utilities
+│   ├── graph/                    # Graph utilities
+│   ├── message_convert/          # Message conversion
+│   ├── pipeline/                 # Pipeline utilities (deprecated since v1.4.0, will be removed in v1.5.0)
+│   └── tool_calling/             # Tool calling utilities
+├── tests/                        # Test files
+├── docs/                         # Documentation
+├── pyproject.toml                # Project configuration
+└── uv.lock                       # Dependency lock file
+```
+
+---
+
+## Dependencies
+
+- Core: langchain, langchain-core, langgraph
+- Optional: jinja2, json-repair, langchain-openai
+- Dev: ruff, dashscope, langchain-model-profiles
+- Test: python-dotenv, langchain-tests, langchain-deepseek, langchain-qwq, langchain-ollama, langchain-community
+
+Python version: >=3.11

{langchain_dev_utils-1.4.0 → langchain_dev_utils-1.4.2}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: langchain-dev-utils
-Version: 1.4.0
+Version: 1.4.2
 Summary: A practical utility library for LangChain and LangGraph development
 Project-URL: Source Code, https://github.com/TBice123123/langchain-dev-utils
 Project-URL: repository, https://github.com/TBice123123/langchain-dev-utils
@@ -12,6 +12,7 @@ Requires-Dist: langchain-core>=1.2.5
 Requires-Dist: langchain>=1.2.0
 Requires-Dist: langgraph>=1.0.0
 Provides-Extra: standard
+Requires-Dist: jinja2>=3.1.6; extra == 'standard'
 Requires-Dist: json-repair>=0.53.1; extra == 'standard'
 Requires-Dist: langchain-openai; extra == 'standard'
 Description-Content-Type: text/markdown

{langchain_dev_utils-1.4.0 → langchain_dev_utils-1.4.2}/pyproject.toml

@@ -1,11 +1,15 @@
 [project]
 name = "langchain-dev-utils"
-version = "1.4.0"
+version = "1.4.2"
 description = "A practical utility library for LangChain and LangGraph development"
 readme = "README.md"
 authors = [{ name = "tiebingice", email = "tiebingice123@outlook.com" }]
 requires-python = ">=3.11"
-dependencies = ["langchain>=1.2.0", "langchain-core>=1.2.5", "langgraph>=1.0.0"]
+dependencies = [
+    "langchain>=1.2.0",
+    "langchain-core>=1.2.5",
+    "langgraph>=1.0.0",
+]
 
 [project.urls]
 "Source Code" = "https://github.com/TBice123123/langchain-dev-utils"
@@ -14,7 +18,11 @@ documentation = "https://tbice123123.github.io/langchain-dev-utils"
 
 
 [project.optional-dependencies]
-standard = ["json-repair>=0.53.1", "langchain-openai"]
+standard = [
+    "jinja2>=3.1.6",
+    "json-repair>=0.53.1",
+    "langchain-openai",
+]
 
 [build-system]
 requires = ["hatchling"]
@@ -30,7 +38,11 @@ python_files = ["test_*.py"]
 python_functions = ["test_*"]
 
 [dependency-groups]
-dev = ["langchain-model-profiles>=0.0.5", "ruff>=0.14.5"]
+dev = [
+    "dashscope>=1.25.9",
+    "langchain-model-profiles>=0.0.5",
+    "ruff>=0.14.5",
+]
 docs = [
     "jupyter>=1.1.1",
     "mkdocs-material>=9.7.0",

langchain_dev_utils-1.4.2/src/langchain_dev_utils/__init__.py

@@ -0,0 +1 @@
+__version__ = "1.4.2"

{langchain_dev_utils-1.4.0 → langchain_dev_utils-1.4.2}/src/langchain_dev_utils/_utils.py

@@ -1,11 +1,48 @@
 from importlib import util
 from typing import Literal, Optional, cast
 
+from langchain_core.tools import BaseTool
 from langgraph.graph import StateGraph
 from langgraph.graph.state import StateNode
 from pydantic import BaseModel
 
 
+def _duplicate_tools(tools: list[BaseTool]) -> list[BaseTool]:
+    """Duplicate tools with the same name.
+
+    Args:
+        tools (list[BaseTool]): The list of tools.
+
+    Returns:
+        list[BaseTool]: The duplicated tools.
+    """
+    tool_name_set = set()
+    duplicated_tools = []
+    for tool_obj in tools:
+        if tool_obj.name not in tool_name_set:
+            duplicated_tools.append(tool_obj)
+            tool_name_set.add(tool_obj.name)
+    return duplicated_tools
+
+
+def _merge_tools(tools: list[BaseTool]) -> list[BaseTool]:
+    """Merge tools with the same name.
+
+    Args:
+        tools (list[BaseTool]): The list of tools.
+
+    Returns:
+        list[BaseTool]: The merged tools.
+    """
+    tool_name_set = set()
+    merged_tools = []
+    for tool_obj in tools:
+        if tool_obj.name not in tool_name_set:
+            merged_tools.append(tool_obj)
+            tool_name_set.add(tool_obj.name)
+    return merged_tools
+
+
 def _transform_node_to_tuple(
     node: StateNode | tuple[str, StateNode],
 ) -> tuple[str, StateNode]:
@@ -23,13 +60,15 @@ def _transform_node_to_tuple(
 
 
 def _check_pkg_install(
-    pkg: Literal["langchain_openai", "json_repair"],
+    pkg: Literal["langchain_openai", "json_repair", "jinja2"],
 ) -> None:
     if not util.find_spec(pkg):
         if pkg == "langchain_openai":
             msg = "Please install langchain_dev_utils[standard],when use 'openai-compatible'"
-        else:
+        elif pkg == "json_repair":
             msg = "Please install langchain_dev_utils[standard] to use ToolCallRepairMiddleware."
+        else:
+            msg = "Please install langchain_dev_utils[standard] to use FormatPromptMiddleware."
         raise ImportError(msg)
 
 

{langchain_dev_utils-1.4.0 → langchain_dev_utils-1.4.2}/src/langchain_dev_utils/agents/middleware/__init__.py

@@ -1,10 +1,10 @@
-from .format_prompt import format_prompt
+from .format_prompt import FormatPromptMiddleware, format_prompt
 from .handoffs import HandoffAgentMiddleware
 from .model_fallback import ModelFallbackMiddleware
 from .model_router import ModelRouterMiddleware
 from .plan import PlanMiddleware
 from .summarization import SummarizationMiddleware
-from .tool_call_repair import ToolCallRepairMiddleware
+from .tool_call_repair import ToolCallRepairMiddleware, tool_call_repair
 from .tool_emulator import LLMToolEmulator
 from .tool_selection import LLMToolSelectorMiddleware
 
@@ -16,6 +16,8 @@ __all__ = [
     "LLMToolEmulator",
     "ModelRouterMiddleware",
     "ToolCallRepairMiddleware",
-    "format_prompt",
+    "FormatPromptMiddleware",
     "HandoffAgentMiddleware",
+    "tool_call_repair",
+    "format_prompt",
 ]

langchain_dev_utils-1.4.2/src/langchain_dev_utils/agents/middleware/format_prompt.py

@@ -0,0 +1,156 @@
+from typing import Awaitable, Callable, Literal
+
+from langchain.agents.middleware import ModelRequest
+from langchain.agents.middleware.types import (
+    AgentMiddleware,
+    ModelCallResult,
+    ModelResponse,
+)
+from langchain_core.messages import SystemMessage
+from langchain_core.prompts.string import get_template_variables
+
+from langchain_dev_utils._utils import _check_pkg_install
+
+
+class FormatPromptMiddleware(AgentMiddleware):
+    """Format the system prompt with variables from state and context.
+
+    This middleware function extracts template variables from the system prompt
+    and populates them with values from the agent's state and runtime context.
+    Variables are first resolved from the state, then from the context if not found.
+
+    Args:
+        template_format: The format of the template. Defaults to "f-string".
+
+    Example:
+
+        # Use format_prompt middleware instance rather than FormatPromptMiddleware class (Recommended)
+
+        ```python
+        from langchain_dev_utils.agents.middleware import format_prompt
+        from langchain.agents import create_agent
+        from langchain_core.messages import HumanMessage
+        from dataclasses import dataclass
+
+        @dataclass
+        class Context:
+            name: str
+            user: str
+
+        agent = create_agent(
+            model=model,
+            tools=tools,
+            system_prompt="You are a helpful assistant. Your name is {name}. Your user is {user}.",
+            middleware=[format_prompt],
+            context_schema=Context,
+        )
+        agent.invoke(
+            {
+                "messages": [HumanMessage(content="Hello")],
+            },
+            context=Context(name="assistant", user="Tom"),
+        )
+        ```
+
+        # Use FormatPromptMiddleware class(Use when template_format is jinja2)
+
+        ```python
+        from langchain_dev_utils.agents.middleware import FormatPromptMiddleware
+        from langchain.agents import create_agent
+        from langchain_core.messages import HumanMessage
+        from dataclasses import dataclass
+
+        @dataclass
+        class Context:
+            name: str
+            user: str
+
+        agent = create_agent(
+            model=model,
+            tools=tools,
+            system_prompt="You are a helpful assistant. Your name is {{ name }}. Your user is {{ user }}.",
+            middleware=[FormatPromptMiddleware(template_format="jinja2")],
+            context_schema=Context,
+        )
+        agent.invoke(
+            {
+                "messages": [HumanMessage(content="Hello")],
+            },
+            context=Context(name="assistant", user="Tom"),
+        )
+        ```
+    """
+
+    def __init__(
+        self,
+        *,
+        template_format: Literal["f-string", "jinja2"] = "f-string",
+    ) -> None:
+        super().__init__()
+
+        self.template_format = template_format
+
+        if template_format == "jinja2":
+            _check_pkg_install("jinja2")
+
+    def _format_prompt(self, request: ModelRequest) -> str:
+        """Add the plan system prompt to the system message."""
+        system_msg = request.system_message
+        if system_msg is None:
+            raise ValueError(
+                "system_message must be provided,while use format_prompt in middleware."
+            )
+
+        system_prompt = "\n".join(
+            [content.get("text", "") for content in system_msg.content_blocks]
+        )
+        variables = get_template_variables(system_prompt, self.template_format)
+
+        format_params = {}
+
+        state = request.state
+        for key in variables:
+            if var := state.get(key, None):
+                format_params[key] = var
+
+        other_var_keys = set(variables) - set(format_params.keys())
+
+        if other_var_keys:
+            context = request.runtime.context
+            if context is not None:
+                for key in other_var_keys:
+                    if var := getattr(context, key, None):
+                        format_params[key] = var
+
+        if self.template_format == "jinja2":
+            from jinja2 import Template
+
+            template = Template(system_prompt)
+            return template.render(**format_params)
+        else:
+            return system_prompt.format(**format_params)
+
+    def wrap_model_call(
+        self,
+        request: ModelRequest,
+        handler: Callable[[ModelRequest], ModelResponse],
+    ) -> ModelCallResult:
+        """Update the system prompt with variables from state and context."""
+        prompt = self._format_prompt(request)
+        request = request.override(system_message=SystemMessage(content=prompt))
+        return handler(request)
+
+    async def awrap_model_call(
+        self,
+        request: ModelRequest,
+        handler: Callable[[ModelRequest], Awaitable[ModelResponse]],
+    ) -> ModelCallResult:
+        """Update the system prompt with variables from state and context."""
+        prompt = self._format_prompt(request)
+        override_request = request.override(
+            system_message=SystemMessage(content=prompt)
+        )
+        return await handler(override_request)
+
+
+format_prompt = FormatPromptMiddleware()

{langchain_dev_utils-1.4.0 → langchain_dev_utils-1.4.2}/src/langchain_dev_utils/agents/middleware/handoffs.py

@@ -1,4 +1,4 @@
-from typing import Any, Awaitable, Callable, Literal, cast
+from typing import Awaitable, Callable, Literal, cast
 
 from langchain.agents import AgentState
 from langchain.agents.middleware import AgentMiddleware, ModelRequest, ModelResponse
@@ -9,6 +9,7 @@ from langchain_core.messages import SystemMessage, ToolMessage
 from langgraph.types import Command
 from typing_extensions import NotRequired, Optional, TypedDict
 
+from langchain_dev_utils._utils import _duplicate_tools, _merge_tools
 from langchain_dev_utils.chat_models import load_chat_model
 
 
@@ -19,7 +20,7 @@ class MultiAgentState(AgentState):
 class AgentConfig(TypedDict):
     model: NotRequired[str | BaseChatModel]
     prompt: str | SystemMessage
-    tools: NotRequired[list[BaseTool | dict[str, Any]]]
+    tools: NotRequired[list[BaseTool]]
     default: NotRequired[bool]
     handoffs: list[str] | Literal["all"]
 
@@ -176,7 +177,16 @@ class HandoffAgentMiddleware(AgentMiddleware):
             agents_config,
             handoffs_tools,
         )
-        self.tools = handoffs_tools
+
+        all_tools = [
+            *handoffs_tools,
+        ]
+
+        for agent_name in agents_config:
+            tools = agents_config.get(agent_name, {}).get("tools", [])
+            all_tools.extend(tools)
+
+        self.tools = _merge_tools(all_tools)
 
     def _get_override_request(self, request: ModelRequest) -> ModelRequest:
         active_agent_name = request.state.get("active_agent", self.default_agent_name)
@@ -184,15 +194,14 @@ class HandoffAgentMiddleware(AgentMiddleware):
         _config = self.agents_config[active_agent_name]
 
         params = {}
-        if _config.get("model"):
-            model = _config.get("model")
+        if model := _config.get("model"):
             if isinstance(model, str):
                 model = load_chat_model(model)
             params["model"] = model
-        if _config.get("prompt"):
-            params["system_prompt"] = _config.get("prompt")
-        if _config.get("tools"):
-            params["tools"] = _config.get("tools")
+        if prompt := _config.get("prompt"):
+            params["system_prompt"] = prompt
+        if tools := _config.get("tools"):
+            params["tools"] = _duplicate_tools(tools)
 
         if params:
             return request.override(**params)

{langchain_dev_utils-1.4.0 → langchain_dev_utils-1.4.2}/src/langchain_dev_utils/agents/middleware/model_router.py

@@ -10,6 +10,7 @@ from langgraph.runtime import Runtime
 from pydantic import BaseModel, Field
 from typing_extensions import TypedDict
 
+from langchain_dev_utils._utils import _duplicate_tools, _merge_tools
 from langchain_dev_utils.chat_models import load_chat_model
 from langchain_dev_utils.message_convert import format_sequence
 
@@ -120,6 +121,13 @@ class ModelRouterMiddleware(AgentMiddleware):
 
         self.router_prompt = router_prompt
 
+        all_tools = []
+        for model in model_list:
+            if tools := model.get("tools"):
+                all_tools.extend(tools)
+
+        self.tools = _merge_tools(all_tools)
+
     def _select_model(self, messages: list[AnyMessage]):
         response = cast(
             SelectModel,
@@ -174,7 +182,7 @@ class ModelRouterMiddleware(AgentMiddleware):
                     model = load_chat_model(select_model_name)
             override_kwargs["model"] = model
             if model_values["tools"] is not None:
-                override_kwargs["tools"] = model_values["tools"]
+                override_kwargs["tools"] = _duplicate_tools(model_values["tools"])
             if model_values["system_prompt"] is not None:
                 override_kwargs["system_message"] = SystemMessage(
                     content=model_values["system_prompt"]

{langchain_dev_utils-1.4.0 → langchain_dev_utils-1.4.2}/src/langchain_dev_utils/agents/middleware/tool_call_repair.py

@@ -94,3 +94,6 @@ class ToolCallRepairMiddleware(AgentMiddleware):
             result=results,
             structured_response=response.structured_response,
         )
+
+
+tool_call_repair = ToolCallRepairMiddleware()

{langchain_dev_utils-1.4.0 → langchain_dev_utils-1.4.2}/src/langchain_dev_utils/agents/wrap.py

@@ -17,7 +17,7 @@ def _process_input(request: str, runtime: ToolRuntime) -> str:
 def _process_output(
     request: str, response: dict[str, Any], runtime: ToolRuntime
 ) -> Any:
-    return response["messages"][-1].content
+    return response["messages"][-1].text
 
 
 def get_subagent_name(runtime: ToolRuntime) -> str:

{langchain_dev_utils-1.4.0 → langchain_dev_utils-1.4.2}/src/langchain_dev_utils/chat_models/adapters/openai_compatible.py

@@ -261,6 +261,8 @@ class _BaseChatOpenAICompatible(BaseChatOpenAI):
                 payload_messages.append(_convert_message_to_dict(m))
 
         payload["messages"] = payload_messages
+        if "tools" in payload and len(payload["tools"]) == 0:
+            payload.pop("tools")
         return payload
 
     @model_validator(mode="after")
@@ -302,6 +304,8 @@ class _BaseChatOpenAICompatible(BaseChatOpenAI):
                 ModelProfile,
                 _get_profile_by_provider_and_model(self._provider, self.model_name),
             )
+        if "json_schema" in self.supported_response_format:
+            self.profile.update({"structured_output": True})
         return self
 
     def _create_chat_result(
@@ -347,9 +351,9 @@ class _BaseChatOpenAICompatible(BaseChatOpenAI):
             if isinstance(model_extra, dict) and (
                 reasoning := model_extra.get("reasoning")
             ):
-                rtn.generations[0].message.additional_kwargs["reasoning_content"] = (
-                    reasoning
-                )
+                rtn.generations[0].message.additional_kwargs[
+                    "reasoning_content"
+                ] = reasoning
 
         return rtn
 

{langchain_dev_utils-1.4.0 → langchain_dev_utils-1.4.2}/tests/test_embedding.py

@@ -5,8 +5,8 @@ from langchain_tests.integration_tests.embeddings import EmbeddingsIntegrationTe
 
 from langchain_dev_utils.embeddings.adapters import create_openai_compatible_embedding
 
-SiliconFlowEmbeddings = create_openai_compatible_embedding(
-    "siliconflow", embedding_model_cls_name="SiliconFlowEmbeddings"
+ZAIEmbeddings = create_openai_compatible_embedding(
+    "zai", embedding_model_cls_name="ZAIEmbeddings"
 )
 
 
@@ -14,9 +14,9 @@ class TestStandard(EmbeddingsIntegrationTests):
     @property
     def embeddings_class(self) -> type[Embeddings]:
         """Embeddings class."""
-        return cast("type[Embeddings]", SiliconFlowEmbeddings)
+        return cast("type[Embeddings]", ZAIEmbeddings)
 
     @property
     def embedding_model_params(self) -> dict[str, Any]:
         """Embeddings model parameters."""
-        return {"model": "BAAI/bge-m3"}
+        return {"model": "embedding-3"}

{langchain_dev_utils-1.4.0 → langchain_dev_utils-1.4.2}/tests/test_handoffs_middleware.py

@@ -6,9 +6,7 @@ from langgraph.checkpoint.memory import InMemorySaver
 from langgraph.types import Command
 
 from langchain_dev_utils.agents import create_agent
-from langchain_dev_utils.agents.middleware import (
-    HandoffAgentMiddleware,
-)
+from langchain_dev_utils.agents.middleware import HandoffAgentMiddleware
 from langchain_dev_utils.agents.middleware.handoffs import AgentConfig
 from langchain_dev_utils.chat_models import load_chat_model
 
@@ -91,10 +89,6 @@ def test_handoffs_middleware():
                 handoffs_tool_overrides=handoffs_tool_map,
             )
         ],
-        tools=[
-            get_current_time,
-            run_code,
-        ],
         checkpointer=InMemorySaver(),
     )
 

{langchain_dev_utils-1.4.0 → langchain_dev_utils-1.4.2}/tests/test_load_embbeding.py

@@ -7,7 +7,6 @@ from langchain_dev_utils.embeddings import load_embeddings
 @pytest.fixture(
     params=[
         "dashscope:text-embedding-v4",
-        "siliconflow:BAAI/bge-m3",
         "ollama:bge-m3:latest",
     ]
 )

{langchain_dev_utils-1.4.0 → langchain_dev_utils-1.4.2}/tests/utils/register.py

@@ -27,10 +27,6 @@ def register_all_model_providers():
 def register_all_embeddings_providers():
     batch_register_embeddings_provider(
         [
-            {
-                "provider_name": "siliconflow",
-                "embeddings_model": "openai-compatible",
-            },
             {"provider_name": "dashscope", "embeddings_model": DashScopeEmbeddings},
         ]
     )

{langchain_dev_utils-1.4.0 → langchain_dev_utils-1.4.2}/uv.lock

@@ -483,6 +483,80 @@ wheels = [
     { url = "https://mirrors.ustc.edu.cn/pypi/packages/60/97/891a0971e1e4a8c5d2b20bbe0e524dc04548d2307fee33cdeba148fd4fc7/comm-0.2.3-py3-none-any.whl", hash = "sha256:c615d91d75f7f04f095b30d1c1711babd43bdc6419c1be9886a85f2f4e489417", size = 7294, upload-time = "2025-07-25T14:02:02.896Z" },
 ]
 
+[[package]]
+name = "cryptography"
+version = "46.0.4"
+source = { registry = "https://pypi.mirrors.ustc.edu.cn/simple/" }
+dependencies = [
+    { name = "cffi", marker = "platform_python_implementation != 'PyPy'" },
+]
+sdist = { url = "https://mirrors.ustc.edu.cn/pypi/packages/78/19/f748958276519adf6a0c1e79e7b8860b4830dda55ccdf29f2719b5fc499c/cryptography-46.0.4.tar.gz", hash = "sha256:bfd019f60f8abc2ed1b9be4ddc21cfef059c841d86d710bb69909a688cbb8f59", size = 749301, upload-time = "2026-01-28T00:24:37.379Z" }
+wheels = [
+    { url = "https://mirrors.ustc.edu.cn/pypi/packages/8d/99/157aae7949a5f30d51fcb1a9851e8ebd5c74bf99b5285d8bb4b8b9ee641e/cryptography-46.0.4-cp311-abi3-macosx_10_9_universal2.whl", hash = "sha256:281526e865ed4166009e235afadf3a4c4cba6056f99336a99efba65336fd5485", size = 7173686, upload-time = "2026-01-28T00:23:07.515Z" },
+    { url = "https://mirrors.ustc.edu.cn/pypi/packages/87/91/874b8910903159043b5c6a123b7e79c4559ddd1896e38967567942635778/cryptography-46.0.4-cp311-abi3-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:5f14fba5bf6f4390d7ff8f086c566454bff0411f6d8aa7af79c88b6f9267aecc", size = 4275871, upload-time = "2026-01-28T00:23:09.439Z" },
+    { url = "https://mirrors.ustc.edu.cn/pypi/packages/c0/35/690e809be77896111f5b195ede56e4b4ed0435b428c2f2b6d35046fbb5e8/cryptography-46.0.4-cp311-abi3-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:47bcd19517e6389132f76e2d5303ded6cf3f78903da2158a671be8de024f4cd0", size = 4423124, upload-time = "2026-01-28T00:23:11.529Z" },
+    { url = "https://mirrors.ustc.edu.cn/pypi/packages/1a/5b/a26407d4f79d61ca4bebaa9213feafdd8806dc69d3d290ce24996d3cfe43/cryptography-46.0.4-cp311-abi3-manylinux_2_28_aarch64.whl", hash = "sha256:01df4f50f314fbe7009f54046e908d1754f19d0c6d3070df1e6268c5a4af09fa", size = 4277090, upload-time = "2026-01-28T00:23:13.123Z" },
+    { url = "https://mirrors.ustc.edu.cn/pypi/packages/0c/d8/4bb7aec442a9049827aa34cee1aa83803e528fa55da9a9d45d01d1bb933e/cryptography-46.0.4-cp311-abi3-manylinux_2_28_ppc64le.whl", hash = "sha256:5aa3e463596b0087b3da0dbe2b2487e9fc261d25da85754e30e3b40637d61f81", size = 4947652, upload-time = "2026-01-28T00:23:14.554Z" },
+    { url = "https://mirrors.ustc.edu.cn/pypi/packages/2b/08/f83e2e0814248b844265802d081f2fac2f1cbe6cd258e72ba14ff006823a/cryptography-46.0.4-cp311-abi3-manylinux_2_28_x86_64.whl", hash = "sha256:0a9ad24359fee86f131836a9ac3bffc9329e956624a2d379b613f8f8abaf5255", size = 4455157, upload-time = "2026-01-28T00:23:16.443Z" },
+    { url = "https://mirrors.ustc.edu.cn/pypi/packages/0a/05/19d849cf4096448779d2dcc9bb27d097457dac36f7273ffa875a93b5884c/cryptography-46.0.4-cp311-abi3-manylinux_2_31_armv7l.whl", hash = "sha256:dc1272e25ef673efe72f2096e92ae39dea1a1a450dd44918b15351f72c5a168e", size = 3981078, upload-time = "2026-01-28T00:23:17.838Z" },
+    { url = "https://mirrors.ustc.edu.cn/pypi/packages/e6/89/f7bac81d66ba7cde867a743ea5b37537b32b5c633c473002b26a226f703f/cryptography-46.0.4-cp311-abi3-manylinux_2_34_aarch64.whl", hash = "sha256:de0f5f4ec8711ebc555f54735d4c673fc34b65c44283895f1a08c2b49d2fd99c", size = 4276213, upload-time = "2026-01-28T00:23:19.257Z" },
+    { url = "https://mirrors.ustc.edu.cn/pypi/packages/da/9f/7133e41f24edd827020ad21b068736e792bc68eecf66d93c924ad4719fb3/cryptography-46.0.4-cp311-abi3-manylinux_2_34_ppc64le.whl", hash = "sha256:eeeb2e33d8dbcccc34d64651f00a98cb41b2dc69cef866771a5717e6734dfa32", size = 4912190, upload-time = "2026-01-28T00:23:21.244Z" },
+    { url = "https://mirrors.ustc.edu.cn/pypi/packages/a6/f7/6d43cbaddf6f65b24816e4af187d211f0bc536a29961f69faedc48501d8e/cryptography-46.0.4-cp311-abi3-manylinux_2_34_x86_64.whl", hash = "sha256:3d425eacbc9aceafd2cb429e42f4e5d5633c6f873f5e567077043ef1b9bbf616", size = 4454641, upload-time = "2026-01-28T00:23:22.866Z" },
+    { url = "https://mirrors.ustc.edu.cn/pypi/packages/9e/4f/ebd0473ad656a0ac912a16bd07db0f5d85184924e14fc88feecae2492834/cryptography-46.0.4-cp311-abi3-musllinux_1_2_aarch64.whl", hash = "sha256:91627ebf691d1ea3976a031b61fb7bac1ccd745afa03602275dda443e11c8de0", size = 4405159, upload-time = "2026-01-28T00:23:25.278Z" },
+    { url = "https://mirrors.ustc.edu.cn/pypi/packages/d1/f7/7923886f32dc47e27adeff8246e976d77258fd2aa3efdd1754e4e323bf49/cryptography-46.0.4-cp311-abi3-musllinux_1_2_x86_64.whl", hash = "sha256:2d08bc22efd73e8854b0b7caff402d735b354862f1145d7be3b9c0f740fef6a0", size = 4666059, upload-time = "2026-01-28T00:23:26.766Z" },
+    { url = "https://mirrors.ustc.edu.cn/pypi/packages/eb/a7/0fca0fd3591dffc297278a61813d7f661a14243dd60f499a7a5b48acb52a/cryptography-46.0.4-cp311-abi3-win32.whl", hash = "sha256:82a62483daf20b8134f6e92898da70d04d0ef9a75829d732ea1018678185f4f5", size = 3026378, upload-time = "2026-01-28T00:23:28.317Z" },
+    { url = "https://mirrors.ustc.edu.cn/pypi/packages/2d/12/652c84b6f9873f0909374864a57b003686c642ea48c84d6c7e2c515e6da5/cryptography-46.0.4-cp311-abi3-win_amd64.whl", hash = "sha256:6225d3ebe26a55dbc8ead5ad1265c0403552a63336499564675b29eb3184c09b", size = 3478614, upload-time = "2026-01-28T00:23:30.275Z" },
+    { url = "https://mirrors.ustc.edu.cn/pypi/packages/b9/27/542b029f293a5cce59349d799d4d8484b3b1654a7b9a0585c266e974a488/cryptography-46.0.4-cp314-cp314t-macosx_10_9_universal2.whl", hash = "sha256:485e2b65d25ec0d901bca7bcae0f53b00133bf3173916d8e421f6fddde103908", size = 7116417, upload-time = "2026-01-28T00:23:31.958Z" },
+    { url = "https://mirrors.ustc.edu.cn/pypi/packages/f8/f5/559c25b77f40b6bf828eabaf988efb8b0e17b573545edb503368ca0a2a03/cryptography-46.0.4-cp314-cp314t-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:078e5f06bd2fa5aea5a324f2a09f914b1484f1d0c2a4d6a8a28c74e72f65f2da", size = 4264508, upload-time = "2026-01-28T00:23:34.264Z" },
+    { url = "https://mirrors.ustc.edu.cn/pypi/packages/49/a1/551fa162d33074b660dc35c9bc3616fefa21a0e8c1edd27b92559902e408/cryptography-46.0.4-cp314-cp314t-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:dce1e4f068f03008da7fa51cc7abc6ddc5e5de3e3d1550334eaf8393982a5829", size = 4409080, upload-time = "2026-01-28T00:23:35.793Z" },
+    { url = "https://mirrors.ustc.edu.cn/pypi/packages/b0/6a/4d8d129a755f5d6df1bbee69ea2f35ebfa954fa1847690d1db2e8bca46a5/cryptography-46.0.4-cp314-cp314t-manylinux_2_28_aarch64.whl", hash = "sha256:2067461c80271f422ee7bdbe79b9b4be54a5162e90345f86a23445a0cf3fd8a2", size = 4270039, upload-time = "2026-01-28T00:23:37.263Z" },
+    { url = "https://mirrors.ustc.edu.cn/pypi/packages/4c/f5/ed3fcddd0a5e39321e595e144615399e47e7c153a1fb8c4862aec3151ff9/cryptography-46.0.4-cp314-cp314t-manylinux_2_28_ppc64le.whl", hash = "sha256:c92010b58a51196a5f41c3795190203ac52edfd5dc3ff99149b4659eba9d2085", size = 4926748, upload-time = "2026-01-28T00:23:38.884Z" },
+    { url = "https://mirrors.ustc.edu.cn/pypi/packages/43/ae/9f03d5f0c0c00e85ecb34f06d3b79599f20630e4db91b8a6e56e8f83d410/cryptography-46.0.4-cp314-cp314t-manylinux_2_28_x86_64.whl", hash = "sha256:829c2b12bbc5428ab02d6b7f7e9bbfd53e33efd6672d21341f2177470171ad8b", size = 4442307, upload-time = "2026-01-28T00:23:40.56Z" },
+    { url = "https://mirrors.ustc.edu.cn/pypi/packages/8b/22/e0f9f2dae8040695103369cf2283ef9ac8abe4d51f68710bec2afd232609/cryptography-46.0.4-cp314-cp314t-manylinux_2_31_armv7l.whl", hash = "sha256:62217ba44bf81b30abaeda1488686a04a702a261e26f87db51ff61d9d3510abd", size = 3959253, upload-time = "2026-01-28T00:23:42.827Z" },
+    { url = "https://mirrors.ustc.edu.cn/pypi/packages/01/5b/6a43fcccc51dae4d101ac7d378a8724d1ba3de628a24e11bf2f4f43cba4d/cryptography-46.0.4-cp314-cp314t-manylinux_2_34_aarch64.whl", hash = "sha256:9c2da296c8d3415b93e6053f5a728649a87a48ce084a9aaf51d6e46c87c7f2d2", size = 4269372, upload-time = "2026-01-28T00:23:44.655Z" },
+    { url = "https://mirrors.ustc.edu.cn/pypi/packages/17/b7/0f6b8c1dd0779df2b526e78978ff00462355e31c0a6f6cff8a3e99889c90/cryptography-46.0.4-cp314-cp314t-manylinux_2_34_ppc64le.whl", hash = "sha256:9b34d8ba84454641a6bf4d6762d15847ecbd85c1316c0a7984e6e4e9f748ec2e", size = 4891908, upload-time = "2026-01-28T00:23:46.48Z" },
+    { url = "https://mirrors.ustc.edu.cn/pypi/packages/83/17/259409b8349aa10535358807a472c6a695cf84f106022268d31cea2b6c97/cryptography-46.0.4-cp314-cp314t-manylinux_2_34_x86_64.whl", hash = "sha256:df4a817fa7138dd0c96c8c8c20f04b8aaa1fac3bbf610913dcad8ea82e1bfd3f", size = 4441254, upload-time = "2026-01-28T00:23:48.403Z" },
+    { url = "https://mirrors.ustc.edu.cn/pypi/packages/9c/fe/e4a1b0c989b00cee5ffa0764401767e2d1cf59f45530963b894129fd5dce/cryptography-46.0.4-cp314-cp314t-musllinux_1_2_aarch64.whl", hash = "sha256:b1de0ebf7587f28f9190b9cb526e901bf448c9e6a99655d2b07fff60e8212a82", size = 4396520, upload-time = "2026-01-28T00:23:50.26Z" },
+    { url = "https://mirrors.ustc.edu.cn/pypi/packages/b3/81/ba8fd9657d27076eb40d6a2f941b23429a3c3d2f56f5a921d6b936a27bc9/cryptography-46.0.4-cp314-cp314t-musllinux_1_2_x86_64.whl", hash = "sha256:9b4d17bc7bd7cdd98e3af40b441feaea4c68225e2eb2341026c84511ad246c0c", size = 4651479, upload-time = "2026-01-28T00:23:51.674Z" },
+    { url = "https://mirrors.ustc.edu.cn/pypi/packages/00/03/0de4ed43c71c31e4fe954edd50b9d28d658fef56555eba7641696370a8e2/cryptography-46.0.4-cp314-cp314t-win32.whl", hash = "sha256:c411f16275b0dea722d76544a61d6421e2cc829ad76eec79280dbdc9ddf50061", size = 3001986, upload-time = "2026-01-28T00:23:53.485Z" },
+    { url = "https://mirrors.ustc.edu.cn/pypi/packages/5c/70/81830b59df7682917d7a10f833c4dab2a5574cd664e86d18139f2b421329/cryptography-46.0.4-cp314-cp314t-win_amd64.whl", hash = "sha256:728fedc529efc1439eb6107b677f7f7558adab4553ef8669f0d02d42d7b959a7", size = 3468288, upload-time = "2026-01-28T00:23:55.09Z" },
+    { url = "https://mirrors.ustc.edu.cn/pypi/packages/56/f7/f648fdbb61d0d45902d3f374217451385edc7e7768d1b03ff1d0e5ffc17b/cryptography-46.0.4-cp38-abi3-macosx_10_9_universal2.whl", hash = "sha256:a9556ba711f7c23f77b151d5798f3ac44a13455cc68db7697a1096e6d0563cab", size = 7169583, upload-time = "2026-01-28T00:23:56.558Z" },
+    { url = "https://mirrors.ustc.edu.cn/pypi/packages/d8/cc/8f3224cbb2a928de7298d6ed4790f5ebc48114e02bdc9559196bfb12435d/cryptography-46.0.4-cp38-abi3-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:8bf75b0259e87fa70bddc0b8b4078b76e7fd512fd9afae6c1193bcf440a4dbef", size = 4275419, upload-time = "2026-01-28T00:23:58.364Z" },
+    { url = "https://mirrors.ustc.edu.cn/pypi/packages/17/43/4a18faa7a872d00e4264855134ba82d23546c850a70ff209e04ee200e76f/cryptography-46.0.4-cp38-abi3-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:3c268a3490df22270955966ba236d6bc4a8f9b6e4ffddb78aac535f1a5ea471d", size = 4419058, upload-time = "2026-01-28T00:23:59.867Z" },
+    { url = "https://mirrors.ustc.edu.cn/pypi/packages/ee/64/6651969409821d791ba12346a124f55e1b76f66a819254ae840a965d4b9c/cryptography-46.0.4-cp38-abi3-manylinux_2_28_aarch64.whl", hash = "sha256:812815182f6a0c1d49a37893a303b44eaac827d7f0d582cecfc81b6427f22973", size = 4278151, upload-time = "2026-01-28T00:24:01.731Z" },
+    { url = "https://mirrors.ustc.edu.cn/pypi/packages/20/0b/a7fce65ee08c3c02f7a8310cc090a732344066b990ac63a9dfd0a655d321/cryptography-46.0.4-cp38-abi3-manylinux_2_28_ppc64le.whl", hash = "sha256:a90e43e3ef65e6dcf969dfe3bb40cbf5aef0d523dff95bfa24256be172a845f4", size = 4939441, upload-time = "2026-01-28T00:24:03.175Z" },
+    { url = "https://mirrors.ustc.edu.cn/pypi/packages/db/a7/20c5701e2cd3e1dfd7a19d2290c522a5f435dd30957d431dcb531d0f1413/cryptography-46.0.4-cp38-abi3-manylinux_2_28_x86_64.whl", hash = "sha256:a05177ff6296644ef2876fce50518dffb5bcdf903c85250974fc8bc85d54c0af", size = 4451617, upload-time = "2026-01-28T00:24:05.403Z" },
+    { url = "https://mirrors.ustc.edu.cn/pypi/packages/00/dc/3e16030ea9aa47b63af6524c354933b4fb0e352257c792c4deeb0edae367/cryptography-46.0.4-cp38-abi3-manylinux_2_31_armv7l.whl", hash = "sha256:daa392191f626d50f1b136c9b4cf08af69ca8279d110ea24f5c2700054d2e263", size = 3977774, upload-time = "2026-01-28T00:24:06.851Z" },
+    { url = "https://mirrors.ustc.edu.cn/pypi/packages/42/c8/ad93f14118252717b465880368721c963975ac4b941b7ef88f3c56bf2897/cryptography-46.0.4-cp38-abi3-manylinux_2_34_aarch64.whl", hash = "sha256:e07ea39c5b048e085f15923511d8121e4a9dc45cee4e3b970ca4f0d338f23095", size = 4277008, upload-time = "2026-01-28T00:24:08.926Z" },
+    { url = "https://mirrors.ustc.edu.cn/pypi/packages/00/cf/89c99698151c00a4631fbfcfcf459d308213ac29e321b0ff44ceeeac82f1/cryptography-46.0.4-cp38-abi3-manylinux_2_34_ppc64le.whl", hash = "sha256:d5a45ddc256f492ce42a4e35879c5e5528c09cd9ad12420828c972951d8e016b", size = 4903339, upload-time = "2026-01-28T00:24:12.009Z" },
+    { url = "https://mirrors.ustc.edu.cn/pypi/packages/03/c3/c90a2cb358de4ac9309b26acf49b2a100957e1ff5cc1e98e6c4996576710/cryptography-46.0.4-cp38-abi3-manylinux_2_34_x86_64.whl", hash = "sha256:6bb5157bf6a350e5b28aee23beb2d84ae6f5be390b2f8ee7ea179cda077e1019", size = 4451216, upload-time = "2026-01-28T00:24:13.975Z" },
+    { url = "https://mirrors.ustc.edu.cn/pypi/packages/96/2c/8d7f4171388a10208671e181ca43cdc0e596d8259ebacbbcfbd16de593da/cryptography-46.0.4-cp38-abi3-musllinux_1_2_aarch64.whl", hash = "sha256:dd5aba870a2c40f87a3af043e0dee7d9eb02d4aff88a797b48f2b43eff8c3ab4", size = 4404299, upload-time = "2026-01-28T00:24:16.169Z" },
+    { url = "https://mirrors.ustc.edu.cn/pypi/packages/e9/23/cbb2036e450980f65c6e0a173b73a56ff3bccd8998965dea5cc9ddd424a5/cryptography-46.0.4-cp38-abi3-musllinux_1_2_x86_64.whl", hash = "sha256:93d8291da8d71024379ab2cb0b5c57915300155ad42e07f76bea6ad838d7e59b", size = 4664837, upload-time = "2026-01-28T00:24:17.629Z" },
+    { url = "https://mirrors.ustc.edu.cn/pypi/packages/0a/21/f7433d18fe6d5845329cbdc597e30caf983229c7a245bcf54afecc555938/cryptography-46.0.4-cp38-abi3-win32.whl", hash = "sha256:0563655cb3c6d05fb2afe693340bc050c30f9f34e15763361cf08e94749401fc", size = 3009779, upload-time = "2026-01-28T00:24:20.198Z" },
+    { url = "https://mirrors.ustc.edu.cn/pypi/packages/3a/6a/bd2e7caa2facffedf172a45c1a02e551e6d7d4828658c9a245516a598d94/cryptography-46.0.4-cp38-abi3-win_amd64.whl", hash = "sha256:fa0900b9ef9c49728887d1576fd8d9e7e3ea872fa9b25ef9b64888adc434e976", size = 3466633, upload-time = "2026-01-28T00:24:21.851Z" },
+    { url = "https://mirrors.ustc.edu.cn/pypi/packages/59/e0/f9c6c53e1f2a1c2507f00f2faba00f01d2f334b35b0fbfe5286715da2184/cryptography-46.0.4-pp311-pypy311_pp73-macosx_11_0_arm64.whl", hash = "sha256:766330cce7416c92b5e90c3bb71b1b79521760cdcfc3a6a1a182d4c9fab23d2b", size = 3476316, upload-time = "2026-01-28T00:24:24.144Z" },
+    { url = "https://mirrors.ustc.edu.cn/pypi/packages/27/7a/f8d2d13227a9a1a9fe9c7442b057efecffa41f1e3c51d8622f26b9edbe8f/cryptography-46.0.4-pp311-pypy311_pp73-manylinux_2_28_aarch64.whl", hash = "sha256:c236a44acfb610e70f6b3e1c3ca20ff24459659231ef2f8c48e879e2d32b73da", size = 4216693, upload-time = "2026-01-28T00:24:25.758Z" },
+    { url = "https://mirrors.ustc.edu.cn/pypi/packages/c5/de/3787054e8f7972658370198753835d9d680f6cd4a39df9f877b57f0dd69c/cryptography-46.0.4-pp311-pypy311_pp73-manylinux_2_28_x86_64.whl", hash = "sha256:8a15fb869670efa8f83cbffbc8753c1abf236883225aed74cd179b720ac9ec80", size = 4382765, upload-time = "2026-01-28T00:24:27.577Z" },
+    { url = "https://mirrors.ustc.edu.cn/pypi/packages/8a/5f/60e0afb019973ba6a0b322e86b3d61edf487a4f5597618a430a2a15f2d22/cryptography-46.0.4-pp311-pypy311_pp73-manylinux_2_34_aarch64.whl", hash = "sha256:fdc3daab53b212472f1524d070735b2f0c214239df131903bae1d598016fa822", size = 4216066, upload-time = "2026-01-28T00:24:29.056Z" },
+    { url = "https://mirrors.ustc.edu.cn/pypi/packages/81/8e/bf4a0de294f147fee66f879d9bae6f8e8d61515558e3d12785dd90eca0be/cryptography-46.0.4-pp311-pypy311_pp73-manylinux_2_34_x86_64.whl", hash = "sha256:44cc0675b27cadb71bdbb96099cca1fa051cd11d2ade09e5cd3a2edb929ed947", size = 4382025, upload-time = "2026-01-28T00:24:30.681Z" },
+    { url = "https://mirrors.ustc.edu.cn/pypi/packages/79/f4/9ceb90cfd6a3847069b0b0b353fd3075dc69b49defc70182d8af0c4ca390/cryptography-46.0.4-pp311-pypy311_pp73-win_amd64.whl", hash = "sha256:be8c01a7d5a55f9a47d1888162b76c8f49d62b234d88f0ff91a9fbebe32ffbc3", size = 3406043, upload-time = "2026-01-28T00:24:32.236Z" },
+]
+
+[[package]]
+name = "dashscope"
+version = "1.25.9"
+source = { registry = "https://pypi.mirrors.ustc.edu.cn/simple/" }
+dependencies = [
+    { name = "aiohttp" },
+    { name = "certifi" },
+    { name = "cryptography" },
+    { name = "requests" },
+    { name = "websocket-client" },
+]
+wheels = [
+    { url = "https://mirrors.ustc.edu.cn/pypi/packages/03/bf/503587663b909427c1906b3b75fc2982bf9e42161d8b687f6e38ad12d042/dashscope-1.25.9-py3-none-any.whl", hash = "sha256:03b587bcb58a2f0a76fa5102925c16609b50af176198af0aeb0fd85aa44d6cfe", size = 1335755, upload-time = "2026-01-21T06:58:14.496Z" },
+]
+
 [[package]]
 name = "dataclasses-json"
 version = "0.6.7"
@@ -1365,7 +1439,7 @@ wheels = [
 
 [[package]]
 name = "langchain-dev-utils"
-version = "1.3.7"
+version = "1.4.1"
 source = { editable = "." }
 dependencies = [
     { name = "langchain" },
@@ -1375,12 +1449,14 @@ dependencies = [
 
 [package.optional-dependencies]
 standard = [
+    { name = "jinja2" },
     { name = "json-repair" },
     { name = "langchain-openai" },
 ]
 
 [package.dev-dependencies]
 dev = [
+    { name = "dashscope" },
     { name = "langchain-model-profiles" },
     { name = "ruff" },
 ]
@@ -1400,6 +1476,7 @@ tests = [
 
 [package.metadata]
 requires-dist = [
+    { name = "jinja2", marker = "extra == 'standard'", specifier = ">=3.1.6" },
     { name = "json-repair", marker = "extra == 'standard'", specifier = ">=0.53.1" },
     { name = "langchain", specifier = ">=1.2.0" },
     { name = "langchain-core", specifier = ">=1.2.5" },
@@ -1410,6 +1487,7 @@ provides-extras = ["standard"]
 
 [package.metadata.requires-dev]
 dev = [
+    { name = "dashscope", specifier = ">=1.25.9" },
     { name = "langchain-model-profiles", specifier = ">=0.0.5" },
     { name = "ruff", specifier = ">=0.14.5" },
 ]

langchain_dev_utils-1.4.0/src/langchain_dev_utils/__init__.py

@@ -1 +0,0 @@
-__version__ = "1.4.0"

langchain_dev_utils-1.4.0/src/langchain_dev_utils/agents/middleware/format_prompt.py

@@ -1,66 +0,0 @@
-from langchain.agents.middleware import ModelRequest, dynamic_prompt
-from langchain_core.prompts.string import get_template_variables
-
-
-@dynamic_prompt
-def format_prompt(request: ModelRequest) -> str:
-    """Format the system prompt with variables from state and context.
-
-    This middleware function extracts template variables from the system prompt
-    and populates them with values from the agent's state and runtime context.
-    Variables are first resolved from the state, then from the context if not found.
-
-    Example:
-        >>> from langchain_dev_utils.agents.middleware import format_prompt
-        >>> from langchain.agents import create_agent
-        >>> from langchain_core.messages import HumanMessage
-        >>> from dataclasses import dataclass
-        >>>
-        >>> @dataclass
-        ... class Context:
-        ...       name: str
-        ...       user: str
-        >>>
-        >>> agent=create_agent(
-        ...     model=model,
-        ...     tools=tools,
-        ...     system_prompt="You are a helpful assistant. Your name is {name}. Your user is {user}.",
-        ...     middleware=[format_prompt],
-        ...     context_schema=Context,
-        ... )
-        >>> agent.invoke(
-        ...     {
-        ...         "messages": [HumanMessage(content="Hello")],
-        ...     },
-        ...     context=Context(name="assistant", user="Tom"),
-        ... )
-
-    """
-    system_msg = request.system_message
-    if system_msg is None:
-        raise ValueError(
-            "system_message must be provided,while use format_prompt in middleware."
-        )
-
-    system_prompt = "\n".join(
-        [content.get("text", "") for content in system_msg.content_blocks]
-    )
-    variables = get_template_variables(system_prompt, "f-string")
-
-    format_params = {}
-
-    state = request.state
-    for key in variables:
-        if var := state.get(key, None):
-            format_params[key] = var
-
-    other_var_keys = set(variables) - set(format_params.keys())
-
-    if other_var_keys:
-        context = request.runtime.context
-        if context is not None:
-            for key in other_var_keys:
-                if var := getattr(context, key, None):
-                    format_params[key] = var
-
-    return system_prompt.format(**format_params)