xparse-client 0.3.0b3__py3-none-any.whl → 0.3.0b8__py3-none-any.whl

example/2_async_job.py

@@ -165,7 +165,7 @@ def example_3_error_handling():
 
         except Exception as e:
             print(f"\n⏸️  捕获到超时异常: {type(e).__name__}")
-            print(f"  这是正常的，演示了如何处理超时情况")
+            print("  这是正常的，演示了如何处理超时情况")
 
     finally:
         test_file.unlink(missing_ok=True)

example/4_advanced_workflow.py

@@ -91,12 +91,12 @@ def example_1_config_file():
     XParseClient.from_env()
 
     # 演示如何从配置构建 source 和 destination（不实际创建）
-    print(f"\n✅ 配置加载成功!")
-    print(f"  - Source 配置:")
+    print("\n✅ 配置加载成功!")
+    print("  - Source 配置:")
     print(f"      目录: {loaded_config['source']['directory']}")
     print(f"      模式: {loaded_config['source']['pattern']}")
     print(f"      递归: {loaded_config['source']['recursive']}")
-    print(f"  - Destination 配置:")
+    print("  - Destination 配置:")
     print(f"      输出目录: {loaded_config['destination']['output_dir']}")
 
     # 构建 stages

tests/unit/models/test_pipeline_stages.py

@@ -89,8 +89,8 @@ def test_extract_stage_with_config():
     assert stage.type == "extract"
     assert stage.config.generate_citations is True
 
-    # 序列化检查
-    dumped = stage.model_dump()
+    # 序列化检查（使用 by_alias=True 来确保 schema_ 序列化为 schema）
+    dumped = stage.model_dump(by_alias=True)
     assert "schema" in dumped["config"]
     assert dumped["config"]["schema"] == schema
 

xparse_client/_base.py

@@ -6,7 +6,7 @@
 from __future__ import annotations
 
 from abc import ABC
-from typing import TYPE_CHECKING, Any, TypeVar
+from typing import TYPE_CHECKING, Any, Protocol, TypeVar, cast
 
 import httpx
 
@@ -14,7 +14,16 @@ if TYPE_CHECKING:
     from ._config import SDKConfiguration
     from ._http import HTTPClient
 
-T = TypeVar("T")
+
+class PydanticModel(Protocol):
+    """Pydantic 模型协议"""
+
+    @classmethod
+    def model_validate(cls, obj: Any) -> Any:
+        ...
+
+
+T = TypeVar("T", bound=PydanticModel)
 
 
 class BaseAPI(ABC):  # noqa: B024
@@ -130,7 +139,7 @@ class BaseAPI(ABC):  # noqa: B024
         if isinstance(data, dict) and "data" in data:
             data = data["data"]
 
-        return model_class.model_validate(data)
+        return cast(T, model_class.model_validate(data))
 
     def _parse_list_response(
         self,
@@ -156,7 +165,7 @@ class BaseAPI(ABC):  # noqa: B024
         if not isinstance(data, list):
             data = [data]
 
-        return [model_class.model_validate(item) for item in data]
+        return [cast(T, model_class.model_validate(item)) for item in data]
 
     def _parse_raw_response(self, response: httpx.Response) -> dict[str, Any]:
         """解析原始 JSON 响应
@@ -171,9 +180,9 @@ class BaseAPI(ABC):  # noqa: B024
 
         # 处理标准响应格式
         if isinstance(data, dict) and "data" in data:
-            return data["data"]
+            return cast(dict[str, Any], data["data"])
 
-        return data
+        return cast(dict[str, Any], data)
 
 
 __all__ = ["BaseAPI"]

xparse_client/_http.py

@@ -53,7 +53,8 @@ def _extract_error_message(response: httpx.Response) -> str:
                     return str(data[key])
             # 如果有嵌套的 error 对象
             if "error" in data and isinstance(data["error"], dict):
-                return data["error"].get("message", str(data["error"]))
+                error_dict = data["error"]
+                return str(error_dict.get("message", error_dict))
         return str(data)
     except Exception:
         # 非 JSON 响应，返回文本内容
@@ -106,7 +107,7 @@ def raise_for_status(response: httpx.Response) -> None:
         message = _extract_error_message(response)
         response_body = response.text[:1000] if response.text else None
 
-        common_kwargs = {
+        common_kwargs: dict[str, Any] = {
             "status_code": status_code,
             "request_id": request_id,
             "response_body": response_body,
@@ -140,24 +141,24 @@ def raise_for_status(response: httpx.Response) -> None:
                 message = data.get("message", f"业务错误 code: {code}")
                 response_body = response.text[:1000] if response.text else None
 
-                common_kwargs = {
+                error_kwargs: dict[str, Any] = {
                     "request_id": request_id,
                     "response_body": response_body,
                 }
 
                 # 根据业务 code 映射到相应异常
                 if code == 400:
-                    raise ValidationError(message, details=common_kwargs)
+                    raise ValidationError(message, details=error_kwargs)
                 elif code == 401:
-                    raise AuthenticationError(message, **common_kwargs)
+                    raise AuthenticationError(message, **error_kwargs)
                 elif code == 403:
-                    raise PermissionDeniedError(message, **common_kwargs)
+                    raise PermissionDeniedError(message, **error_kwargs)
                 elif code == 404:
-                    raise NotFoundError(message, **common_kwargs)
+                    raise NotFoundError(message, **error_kwargs)
                 elif code >= 500:
-                    raise ServerError(message, **common_kwargs)
+                    raise ServerError(message, **error_kwargs)
                 else:
-                    raise APIError(message, **common_kwargs)
+                    raise APIError(message, **error_kwargs)
     except (ValueError, KeyError):
         # 如果无法解析 JSON 或没有 code 字段，认为是成功
         pass

xparse_client/api/local.py

@@ -24,7 +24,7 @@ Example:
 
 from __future__ import annotations
 
-from typing import TYPE_CHECKING, Callable, Literal
+from typing import TYPE_CHECKING, Any, Callable, Literal
 
 from .._base import BaseAPI
 from ..models.local import FailedFile, WorkflowResult
@@ -53,7 +53,7 @@ class Local(BaseAPI):
         self,
         source,  # Union[LocalSource, S3Source, FtpSource, SmbSource]
         destination,  # Union[LocalDestination, S3Destination, MilvusDestination, QdrantDestination]
-        stages: list[PipelineStage],
+        stages: list[PipelineStage | dict[str, Any]],
         *,
         pipeline_config: PipelineConfig | None = None,
         progress_callback: Callable[[int, int, str], None] | None = None,
@@ -149,6 +149,16 @@ class Local(BaseAPI):
                         stage_name = stage_result.get("stage")
                         elements = stage_result.get("elements", [])
                         if stage_name and elements:
+                            # 转换 elements 为字典列表
+                            elements_data = []
+                            for elem in elements:
+                                if hasattr(elem, "model_dump"):
+                                    elements_data.append(elem.model_dump())
+                                elif isinstance(elem, dict):
+                                    elements_data.append(elem)
+                                else:
+                                    elements_data.append(elem)
+
                             metadata_with_stage = {
                                 "filename": filename,
                                 "file_path": file_path,
@@ -158,7 +168,7 @@ class Local(BaseAPI):
                                 metadata_with_stage["record_id"] = result.stats.record_id
 
                             pipeline_config.intermediate_results_destination.write(
-                                elements, metadata_with_stage
+                                elements_data, metadata_with_stage
                             )
 
                 # 准备写入数据

xparse_client/connectors/destinations/qdrant.py

@@ -90,7 +90,7 @@ class QdrantDestination(Destination, VectorDestinationMixin):
         self.dimension = dimension
         self._PointStruct = PointStruct
 
-        client_kwargs = {"url": url}
+        client_kwargs: dict[str, Any] = {"url": url}
         if api_key:
             client_kwargs["api_key"] = api_key
         if prefer_grpc:

xparse_client/connectors/sources/ftp.py

@@ -90,7 +90,7 @@ class FtpSource(Source):
             SourceError: 列出文件失败
         """
         try:
-            files = []
+            files: list[str] = []
             current_dir = self.client.pwd()
 
             if self.recursive:
@@ -178,7 +178,7 @@ class FtpSource(Source):
 
         # 回退到 LIST 命令
         try:
-            lines = []
+            lines: list[str] = []
             self.client.retrlines("LIST", lines.append)
             for line in lines:
                 parts = line.split()

xparse_client/connectors/sources/smb.py

@@ -125,7 +125,7 @@ class SmbSource(Source):
             SourceError: 列出文件失败
         """
         try:
-            files = []
+            files: list[str] = []
             base_path = "/" if not self.path else f"/{self.path}"
 
             self._list_recursive(base_path, base_path, files)

xparse_client/models/extract.py

@@ -11,7 +11,7 @@ class ExtractConfig(BaseModel):
     """抽取配置
 
     Attributes:
-        schema: JSON Schema 定义抽取的结构
+        schema_: JSON Schema 定义抽取的结构（序列化为 schema）
         generate_citations: 是否生成引用
         stamp: 是否添加时间戳
 
@@ -23,18 +23,18 @@ class ExtractConfig(BaseModel):
         ...         "total_amount": {"type": "number"}
         ...     }
         ... }
-        >>> config = ExtractConfig(schema=schema)
+        >>> config = ExtractConfig(schema_=schema)
     """
 
-    schema: dict[str, Any] = Field(
-        default_factory=dict,
-        description="JSON Schema 定义抽取的结构"
+    schema_: dict[str, Any] = Field(
+        default_factory=dict, alias="schema", description="JSON Schema 定义抽取的结构"
     )
     generate_citations: bool = False
     stamp: bool = False
 
     model_config = {
         "extra": "allow",
+        "populate_by_name": True,  # 允许使用 schema_ 和 schema 两种名称
     }
 
 

xparse_client/models/parse.py

@@ -63,17 +63,13 @@ class ParseResponse(BaseModel):
 
     Attributes:
         elements: 解析后的元素列表（Parse/Chunk/Embed 返回）
-        extract_result: 提取结果（Extract 返回）
+        result: 提取结果（Extract 返回）
         success_count: 成功数量
-        consume_time: 耗时
-        record_id: 记录 ID
     """
 
     elements: list[Element] = Field(default_factory=list)
-    extract_result: dict[str, Any] | None = None  # Extract API 返回
+    result: dict[str, Any] | None = None  # Extract API 返回
     success_count: int | None = None
-    consume_time: str | None = None
-    record_id: str | None = None
 
     model_config = {"extra": "allow"}
 

{xparse_client-0.3.0b3.dist-info → xparse_client-0.3.0b8.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: xparse-client
-Version: 0.3.0b3
+Version: 0.3.0b8
 Summary: 面向 Agent 和 RAG 的文档处理 Pipeline 客户端
 Author-email: INTSIG-TEXTIN <support@textin.com>
 License-Expression: MIT
@@ -213,17 +213,17 @@ schema = {
         "author": {"type": "string", "description": "作者"},
         "date": {"type": "string", "description": "日期"}
     },
-    "required": ["title"]
+    "required": ["title", "author", "date"]
 }
 
 with open("document.pdf", "rb") as f:
     result = client.extract.extract(
         file=f,
         filename="document.pdf",
-        config=ExtractConfig(schema=schema)
+        extract_config=ExtractConfig(schema=schema)
     )
 
-print(result.extracted_data)
+print(result.result)
 ```
 
 ### 3. 本地批处理

{xparse_client-0.3.0b3.dist-info → xparse_client-0.3.0b8.dist-info}/RECORD

@@ -1,7 +1,7 @@
 example/1_basic_api_usage.py,sha256=x_ZHEWXz6z7qp-sMLBvq7Vpu6nA7YxVCizyQpDday6M,5802
-example/2_async_job.py,sha256=MCjWii3ksmONiIyCWviBrKPZIwrlCPDvmUPP_F8qXPM,6353
+example/2_async_job.py,sha256=CT03sKz2dwTJRqSde8s5o75oe-p3ze43i0E1i_-Ai5Q,6352
 example/3_local_workflow.py,sha256=ZxoT8N0Nz4_s-bkLF0l6UDqBJk_eZZi99WUtUhJseXs,8980
-example/4_advanced_workflow.py,sha256=cgG0xke8-WNPxqx7iSTl9qgoJsVctXf5mjDkp-SaQho,10159
+example/4_advanced_workflow.py,sha256=KH6gmzoKi6oeHIPDv9snw1YehynHTDqWVq1WFtoqGws,10156
 example/README.md,sha256=d-DumaZwbV96K6ZTEryAlDSk9HUhRBg0gG_Yekx9PqA,2336
 example/config_example.json,sha256=mBaLmKbgxFBKUgA-FtCTiLMtq24z0QRwKbPeYDGwA9I,2517
 tests/conftest.py,sha256=h6sy9RNQmXTQNTu86h7ATDAg-9wI4cKjzKXUbNNH5q8,10037
@@ -25,17 +25,17 @@ tests/unit/connectors/test_s3.py,sha256=Xbo-S9khgS1S3LM3BUIuYp-28YLEmEW0SP1O-iJ8
 tests/unit/connectors/test_smb.py,sha256=yrI_xNeOe84vONyxXNOGLbWa7tcyxjzo9G5Z_Tjzn-w,12378
 tests/unit/connectors/test_utils.py,sha256=RfnzTW5-9w2f5YeLbNTB_z3vVnsUSL7DwXNobo7xZUA,10302
 tests/unit/models/test_local.py,sha256=QnxzWKOivqeITd0CPzZLEzGbaMDtiwsPiwyEHiPpcPY,1325
-tests/unit/models/test_pipeline_stages.py,sha256=vT7Pd0lRWdChTQJuCZ3iUBx8C0xFXQXTBVB6jtWtEKU,3972
+tests/unit/models/test_pipeline_stages.py,sha256=LR7OmMhVAAsV085x-l3d5jQ-efl04WMyrQFjVzXEvws,4049
 tests/unit/models/test_workflows.py,sha256=Al4d3SQI1jTz585i1kD569TZx4IWNrhyf3_Le6-cq2k,1692
 xparse_client/__init__.py,sha256=FyR7W0eteapDoPj2pU4VoTobM8XR2B7f_g4yVI-v2v8,3956
-xparse_client/_base.py,sha256=Vx85kDns565zwv6Zl6_8jl6b43nNprfBHbqdCVPGGTc,4420
+xparse_client/_base.py,sha256=IPYdHREM0CUvZ_kWUp4olYQfsfKzPTQoZuIrNSanKq4,4661
 xparse_client/_client.py,sha256=ZGxZ9xhUitTP-eRhte57ekqUkrXxNubFDL5gojajA2s,6964
 xparse_client/_config.py,sha256=BG3cEW5ywKhmtfHjfTFrdezs2Ugc9P2Y85tzu5OZZ4M,6480
-xparse_client/_http.py,sha256=OrnrkTeuG5tEpuD-a1ixLIFPdiqnVIYtuZCi3uXAzJ4,11381
+xparse_client/_http.py,sha256=Y0-x8CZbHVbi4Jmln1d60oD_Pz1nwQAf-0sbON4B9qo,11443
 xparse_client/exceptions.py,sha256=FNozgCFaQSsdeyxE6P-3CZItkTSfFJcIQVrddHWKN0s,9521
 xparse_client/api/__init__.py,sha256=MZ3wBTDU666efOBv81MKKtc1UkjUsjzIEOpsnDUAHD4,203
 xparse_client/api/extract.py,sha256=87tH-y4sXPKW35l-d0GmZdK9RIXWq4tfmGZXYJhIN60,3011
-xparse_client/api/local.py,sha256=LkwtGFCZifWuWbQlq2Y2dXPB3aXW1x1P89b-o0MJC7s,7662
+xparse_client/api/local.py,sha256=dARW6RjiNsxliWxs7cB6qQl9eTZNvyjBpOG8qXyaIaw,8214
 xparse_client/api/parse.py,sha256=nvpu5kiN43QlyMW3N6vMruVADu6MdjS-TMbLFjBh-cc,5870
 xparse_client/api/pipeline.py,sha256=bPCSpWjER-7c7L0iMQ_Xz1VTe5DARMsX5XIcrayGKzw,4530
 xparse_client/api/workflows.py,sha256=29_Rf4bIH1fiJb_bWShkUm-et97ym5ymQFh5i3b5mDI,5945
@@ -45,24 +45,24 @@ xparse_client/connectors/destinations/__init__.py,sha256=hmvWAXj58yLmePMCEnqKbNh
 xparse_client/connectors/destinations/base.py,sha256=DfwwH5S8VPwEUyvWqUKDV9Bq-7A59lqoZHruOjB7L50,3218
 xparse_client/connectors/destinations/local.py,sha256=drZ5wkPMiYd8iOROmLJR3eoCXcYKUpAPPjSFpafeNW8,2542
 xparse_client/connectors/destinations/milvus.py,sha256=MhElUsIMY1g6_c_bD-wjpWOM5PiuZ9bLsgXACW7EY1c,7339
-xparse_client/connectors/destinations/qdrant.py,sha256=Mo1wCwtsM7OIMLZ0HsEBnvE8N9BGnyTGBGJ1B8oYIew,7661
+xparse_client/connectors/destinations/qdrant.py,sha256=iU9NxZBhqp-94-1TkG32fDOjg8i4bT9Ox0F2E-2KODg,7677
 xparse_client/connectors/destinations/s3.py,sha256=2aeTW0cNUVPLBL8up9InFfgXBcFDPtqFNGa5FbUphhk,4728
 xparse_client/connectors/sources/__init__.py,sha256=830EDkdGJtPC2p1MhIP108uLibGeMoXcXAwPAEACPvQ,1143
 xparse_client/connectors/sources/base.py,sha256=fbRWFWuIcw_XWrOPpPkBnQ0Pf2R9N7Ry4523rxWcOpc,1863
-xparse_client/connectors/sources/ftp.py,sha256=bjOjnd-2l0MZ2EAcGM3jvuTP7FMWBiQaEbgHzhIl3og,8170
+xparse_client/connectors/sources/ftp.py,sha256=RwH8eeY-raD4BlF51kb_0BJuM3Vxjbnjpob-f1x_VgY,8192
 xparse_client/connectors/sources/local.py,sha256=ZYTGWdQmUpzdenn3JQ_p0u9-8x7xXk5Ani2GIAqTGyE,5156
 xparse_client/connectors/sources/s3.py,sha256=RCzinmz9cCqp9cFcfEozBapa2nllFvJCw9P-aLeOJQA,7219
-xparse_client/connectors/sources/smb.py,sha256=_F496mipGZJ0y17H-sYTlqAf2uJScxGyit7kirTyDKc,7786
+xparse_client/connectors/sources/smb.py,sha256=9oWrgatdmsXMrFnl-du-lau3RMxlMPw2aQy7SFknUDM,7797
 xparse_client/models/__init__.py,sha256=zLBFirgXDisEeqKHqpsH1bN5p2zE7ZmCtVENZ69mlOg,1213
 xparse_client/models/chunk.py,sha256=VXKrATgk7bhlPVbJTFOUvOWu9oEnNMgMF6AnQarJD7o,986
 xparse_client/models/embed.py,sha256=Tg9iqOiLZy63m4T0q2z9cq3neuQbMnqDqT55-sVA7CE,1652
-xparse_client/models/extract.py,sha256=ZwMy0MsTt775Bq3UsZ9WlXFNCLrSnN0j1RgXQ-vFNqo,915
+xparse_client/models/extract.py,sha256=i7js3EEozHMdQHqksZ6drxXnHqhePn3PlaieF8BdLFY,1033
 xparse_client/models/local.py,sha256=hqKmyWTU_EhPk6qybtBzmbLybBBaiasTpqM4_zP-ipo,698
-xparse_client/models/parse.py,sha256=K34Fodo3emOPZJJQeu3MLjvA7HkVMxAuXk20UvO4lfQ,3297
+xparse_client/models/parse.py,sha256=COgxroHkJce_S7d2HGd_1zBYhkpCutGtmjyLnpI2_eI,3154
 xparse_client/models/pipeline.py,sha256=24bDzhrVotQ8St6VLEJLvG2cZF0G2AMioKI34I_hJXI,3297
 xparse_client/models/workflows.py,sha256=BivMdGOAmhP6oYLQSGAAN7yml2xb7vHHrpzwLgN_Afk,1754
-xparse_client-0.3.0b3.dist-info/licenses/LICENSE,sha256=7iuki7DyWMGB8PBzsht7PUt0YjdIcPjrcXNyUFgMJsw,1070
-xparse_client-0.3.0b3.dist-info/METADATA,sha256=6F2nf_y5odBzTzTLF8N4MA-aqW9F3_EnCCmnPtiqwvg,25453
-xparse_client-0.3.0b3.dist-info/WHEEL,sha256=wUyA8OaulRlbfwMtmQsvNngGrxQHAvkKcvRmdizlJi0,92
-xparse_client-0.3.0b3.dist-info/top_level.txt,sha256=oQGc_qysOmnSAaLjwB72wH8RBHRAmxB-_qb-Uj6u56o,28
-xparse_client-0.3.0b3.dist-info/RECORD,,
+xparse_client-0.3.0b8.dist-info/licenses/LICENSE,sha256=7iuki7DyWMGB8PBzsht7PUt0YjdIcPjrcXNyUFgMJsw,1070
+xparse_client-0.3.0b8.dist-info/METADATA,sha256=mrVOWpID-szyT7O2GlKxymxTLlbWFZfeuIsoxLItmuk,25471
+xparse_client-0.3.0b8.dist-info/WHEEL,sha256=wUyA8OaulRlbfwMtmQsvNngGrxQHAvkKcvRmdizlJi0,92
+xparse_client-0.3.0b8.dist-info/top_level.txt,sha256=oQGc_qysOmnSAaLjwB72wH8RBHRAmxB-_qb-Uj6u56o,28
+xparse_client-0.3.0b8.dist-info/RECORD,,