xparse-client 0.2.11__py3-none-any.whl → 0.3.0b3__py3-none-any.whl

tests/conftest.py

@@ -0,0 +1,310 @@
+"""pytest 配置和共享 fixtures"""
+
+import pytest
+
+
+@pytest.fixture
+def app_id() -> str:
+    """测试用 app_id"""
+    return "test-app-id"
+
+
+@pytest.fixture
+def secret_code() -> str:
+    """测试用 secret_code"""
+    return "test-secret-code"
+
+
+@pytest.fixture
+def server_url() -> str:
+    """测试用服务器地址"""
+    return "https://api.test.com"
+
+
+@pytest.fixture
+def sample_pdf_bytes() -> bytes:
+    """测试用 PDF 文件内容（模拟）"""
+    return b"%PDF-1.4 sample pdf content"
+
+
+@pytest.fixture
+def sample_response_data() -> dict:
+    """测试用 API 响应数据"""
+    return {
+        "code": 200,
+        "data": {
+            "elements": [
+                {
+                    "element_id": "elem_001",
+                    "type": "text",
+                    "text": "这是一段测试文本",
+                    "metadata": {"page_number": 1},
+                }
+            ],
+            "success_count": 1,
+            "record_id": "rec_123",
+        },
+    }
+
+
+# ============================================================================
+# 新增 Fixtures（Phase 1 测试补充）
+# ============================================================================
+
+@pytest.fixture
+def parse_success_response() -> dict:
+    """Parse API 成功响应（真实格式）"""
+    return {
+        "code": 200,
+        "data": {
+            "elements": [
+                {
+                    "element_id": "d0c8f3bf754cf4b071b62402d6932ddcb10f3c7e",
+                    "type": "Header",
+                    "metadata": {
+                        "category_depth": -1,
+                        "coordinates": [0.1277, 0.0909, 0.3025, 0.0909, 0.3025, 0.1171, 0.1277, 0.1176],
+                        "data_source": {
+                            "date_created": "2026-01-27T09:47:51Z",
+                            "date_modified": "2026-01-27T09:47:51Z",
+                            "date_processed": "2026-01-27T09:47:51Z",
+                            "record_locator": {
+                                "protocol": "file",
+                                "remote_file_path": "temp/test.pdf"
+                            },
+                            "url": "file://temp/test.pdf",
+                            "version": "2026-01-27T09:47:51Z"
+                        },
+                        "filename": "test.pdf",
+                        "filetype": "application/pdf",
+                        "last_modified": "2026-01-27T09:47:51Z",
+                        "page_height": 1683,
+                        "page_number": 1,
+                        "page_width": 1190
+                    },
+                    "text": "测试标题"
+                }
+            ],
+            "job_id": "5f9f10b8b4c8467882f67ae16bd48746",
+            "success_count": 1
+        },
+        "message": "success",
+        "x_request_id": "e4193bf242b8bb095295a6e7df0cacb1"
+    }
+
+
+@pytest.fixture
+def extract_success_response() -> dict:
+    """Extract API 成功响应（真实格式）"""
+    return {
+        "code": 200,
+        "data": {
+            "job_id": "f76c31eedc124580a6daccba25e8ae27",
+            "result": {
+                "success_count": 1,
+                "extracted_schema": {
+                    "姓名": "陈君曜"
+                },
+                "citations": {
+                    "姓名": {
+                        "bounding_regions": [
+                            {
+                                "page_number": 1,
+                                "position": [175, 86, 245, 86, 245, 112, 175, 112],
+                                "text": "陈君曜"
+                            }
+                        ],
+                        "value": "陈君曜"
+                    }
+                },
+                "pages": [
+                    {
+                        "page_number": 1,
+                        "image_id": "",
+                        "height": 566,
+                        "width": 800,
+                        "angle": 0,
+                        "status": "Success",
+                        "durations": 0
+                    }
+                ]
+            },
+            "success_count": 1
+        },
+        "message": "success",
+        "x_request_id": "eaf843de96084fd635b901394e500b40"
+    }
+
+
+@pytest.fixture
+def business_error_response() -> dict:
+    """业务错误响应（HTTP 200 + code 非 200）"""
+    return {
+        "code": 400,
+        "message": "参数错误：schema 格式不正确",
+        "data": None,
+        "x_request_id": "abc123"
+    }
+
+
+@pytest.fixture
+def pipeline_parse_chunk_embed_response() -> dict:
+    """Pipeline API 响应（parse + chunk + embed，含中间结果）"""
+    return {
+        "code": 200,
+        "data": {
+            "elements": [
+                {
+                    "element_id": "elem_001",
+                    "type": "CompositeElement",
+                    "metadata": {
+                        "coordinates": [0.05, 0.02, 0.94, 0.02, 0.94, 0.21, 0.05, 0.21],
+                        "data_source": {
+                            "date_created": "2026-01-28T06:44:39Z",
+                            "date_modified": "2026-01-28T06:44:39Z",
+                            "date_processed": "2026-01-28T06:44:39Z",
+                            "record_locator": {
+                                "protocol": "file",
+                                "remote_file_path": "temp/test.pdf"
+                            },
+                            "url": "file://temp/test.pdf",
+                            "version": "2026-01-28T06:44:39Z"
+                        },
+                        "filename": "test.pdf",
+                        "filetype": "application/pdf",
+                        "is_continuation": False,
+                        "page_height": 821,
+                        "page_number": 1,
+                        "page_width": 1270
+                    },
+                    "text": "测试文本内容",
+                    "embeddings": [0.01, -0.105, 0.025] + [0.0] * 1021  # 1024 维向量
+                }
+            ],
+            "intermediate_results": [
+                {
+                    "stage": "parse",
+                    "elements": [
+                        {
+                            "element_id": "elem_parse_001",
+                            "type": "Header",
+                            "text": "原始标题",
+                            "metadata": {}
+                        }
+                    ]
+                },
+                {
+                    "stage": "chunk",
+                    "elements": [
+                        {
+                            "element_id": "elem_001",
+                            "type": "CompositeElement",
+                            "text": "测试文本内容",
+                            "metadata": {}
+                        }
+                    ]
+                }
+            ],
+            "job_id": "job_123",
+            "stats": {
+                "original_elements": 15,
+                "chunked_elements": 3,
+                "embedded_elements": 3,
+                "stages": [
+                    {"type": "parse", "config": {"provider": "textin"}},
+                    {"type": "chunk", "config": {"strategy": "basic"}},
+                    {"type": "embed", "config": {"provider": "qwen", "model_name": "text-embedding-v4"}}
+                ],
+                "record_id": "rec_123",
+                "success_count": 1
+            }
+        },
+        "message": "success",
+        "x_request_id": "req_123"
+    }
+
+
+@pytest.fixture
+def pipeline_parse_extract_response() -> dict:
+    """Pipeline API 响应（parse + extract，含中间结果）"""
+    return {
+        "code": 200,
+        "data": {
+            "extract_result": {
+                "success_count": 1,
+                "extracted_schema": {"购买方": "杭州百世网络技术有限公司"},
+                "citations": {
+                    "购买方": {
+                        "bounding_regions": [
+                            {
+                                "page_number": 1,
+                                "position": [224, 177, 541, 177, 541, 197, 224, 197],
+                                "text": "杭州百世网络技术有限公司"
+                            }
+                        ],
+                        "value": "杭州百世网络技术有限公司"
+                    }
+                },
+                "pages": [{"page_number": 1, "height": 821, "width": 1270}]
+            },
+            "intermediate_results": [
+                {
+                    "stage": "parse",
+                    "elements": [
+                        {
+                            "element_id": "elem_001",
+                            "type": "Image",
+                            "text": "",
+                            "metadata": {}
+                        }
+                    ]
+                }
+            ],
+            "job_id": "job_456",
+            "stats": {
+                "original_elements": 15,
+                "stages": [
+                    {"type": "parse", "config": {"provider": "textin"}},
+                    {"type": "extract", "config": {"schema": {}}}
+                ],
+                "record_id": "rec_456",
+                "success_count": 1
+            }
+        },
+        "message": "success",
+        "x_request_id": "req_456"
+    }
+
+
+# ============================================================================
+# 可选依赖检查 Helper
+# ============================================================================
+
+# 检查可选依赖是否安装
+try:
+    import boto3  # noqa: F401
+
+    has_boto3 = True
+except ImportError:
+    has_boto3 = False
+
+try:
+    import pymilvus  # noqa: F401
+
+    has_pymilvus = True
+except ImportError:
+    has_pymilvus = False
+
+try:
+    import qdrant_client  # noqa: F401
+
+    has_qdrant = True
+except ImportError:
+    has_qdrant = False
+
+try:
+    from smb.SMBConnection import SMBConnection  # noqa: F401
+
+    has_smb = True
+except ImportError:
+    has_smb = False

tests/unit/__init__.py

@@ -0,0 +1 @@
+"""单元测试模块"""

tests/unit/api/__init__.py

@@ -0,0 +1 @@
+"""API 测试模块"""

tests/unit/api/test_extract.py

@@ -0,0 +1,232 @@
+"""Extract API 测试
+
+测试 Extract API 的所有功能，包括：
+- 基础提取功能
+- parse_config 支持
+- extract_config 参数验证
+- 错误处理
+"""
+
+import httpx
+import pytest
+import respx
+
+from xparse_client import XParseClient
+from xparse_client.exceptions import ServerError, ValidationError
+from xparse_client.models import ExtractConfig, ParseConfig
+
+# ============================================================================
+# 基础功能测试
+# ============================================================================
+
+
+def test_extract_api_exists():
+    """测试 client.extract 属性存在"""
+    client = XParseClient(app_id="test", secret_code="test")
+
+    assert hasattr(client, "extract")
+    assert client.extract is not None
+
+
+@respx.mock
+def test_extract_basic(server_url, extract_success_response):
+    """测试基础提取功能"""
+    # Mock API
+    respx.post(f"{server_url}/api/xparse/extract/sync").mock(
+        return_value=httpx.Response(200, json=extract_success_response)
+    )
+
+    # 初始化客户端
+    client = XParseClient(app_id="test", secret_code="test", server_url=server_url)
+
+    # 创建 extract_config
+    extract_config = ExtractConfig(
+        schema_={
+            "type": "object",
+            "properties": {
+                "姓名": {"type": "string"}
+            },
+            "required": ["姓名"]
+        }
+    )
+
+    # 执行提取
+    result = client.extract.extract(
+        file=b"test pdf content",
+        filename="test.pdf",
+        extract_config=extract_config
+    )
+
+    # 验证结果
+    assert result is not None
+    assert result.job_id == "f76c31eedc124580a6daccba25e8ae27"
+    assert result.result is not None
+    assert "extracted_schema" in result.result
+    # 验证请求包含 extract_config
+    request = respx.calls.last.request
+    assert b"extract_config" in request.content
+
+
+@respx.mock
+def test_extract_with_parse_config(server_url, extract_success_response):
+    """测试指定 parse_config"""
+    # Mock API
+    respx.post(f"{server_url}/api/xparse/extract/sync").mock(
+        return_value=httpx.Response(200, json=extract_success_response)
+    )
+
+    # 初始化客户端
+    client = XParseClient(app_id="test", secret_code="test", server_url=server_url)
+
+    # 创建配置
+    parse_config = ParseConfig(provider="mineru")
+    extract_config = ExtractConfig(
+        schema_={
+            "type": "object",
+            "properties": {"姓名": {"type": "string"}}
+        }
+    )
+
+    # 执行提取
+    result = client.extract.extract(
+        file=b"test",
+        filename="test.pdf",
+        parse_config=parse_config,
+        extract_config=extract_config
+    )
+
+    # 验证结果
+    assert result is not None
+
+    # 验证请求中包含 parse_config
+    request = respx.calls.last.request
+    assert b"parse_config" in request.content
+    assert b"mineru" in request.content
+
+
+def test_extract_missing_extract_config():
+    """测试缺少 extract_config 时抛出 ValueError"""
+    client = XParseClient(app_id="test", secret_code="test")
+
+    # 不提供 extract_config 应该抛出 ValueError
+    with pytest.raises(ValueError) as exc_info:
+        client.extract.extract(
+            file=b"test",
+            filename="test.pdf",
+            extract_config=None
+        )
+
+    assert "extract_config is required" in str(exc_info.value)
+
+
+# ============================================================================
+# 错误处理测试
+# ============================================================================
+
+
+@respx.mock
+def test_extract_invalid_schema_error(server_url, business_error_response):
+    """测试无效 schema 错误（业务 code 400）"""
+    # Mock API 返回业务错误
+    respx.post(f"{server_url}/api/xparse/extract/sync").mock(
+        return_value=httpx.Response(200, json=business_error_response)
+    )
+
+    client = XParseClient(app_id="test", secret_code="test", server_url=server_url)
+    extract_config = ExtractConfig(schema_={"invalid": "schema"})
+
+    # 应该抛出 ValidationError
+    with pytest.raises(ValidationError) as exc_info:
+        client.extract.extract(
+            file=b"test",
+            filename="test.pdf",
+            extract_config=extract_config
+        )
+
+    assert "参数错误" in str(exc_info.value)
+
+
+@respx.mock
+def test_extract_server_error(server_url):
+    """测试服务器错误（HTTP 500 或业务 code 500）"""
+    # Mock API 返回 500 错误
+    respx.post(f"{server_url}/api/xparse/extract/sync").mock(
+        return_value=httpx.Response(500, json={"message": "服务器内部错误"})
+    )
+
+    client = XParseClient(app_id="test", secret_code="test", server_url=server_url)
+    extract_config = ExtractConfig(
+        schema_={
+            "type": "object",
+            "properties": {"name": {"type": "string"}}
+        }
+    )
+
+    # 应该抛出 ServerError
+    with pytest.raises(ServerError) as exc_info:
+        client.extract.extract(
+            file=b"test",
+            filename="test.pdf",
+            extract_config=extract_config
+        )
+
+    assert "服务器" in str(exc_info.value)
+
+
+# ============================================================================
+# ExtractConfig 参数测试
+# ============================================================================
+
+
+@respx.mock
+def test_extract_with_generate_citations(server_url, extract_success_response):
+    """测试 generate_citations 参数"""
+    respx.post(f"{server_url}/api/xparse/extract/sync").mock(
+        return_value=httpx.Response(200, json=extract_success_response)
+    )
+
+    client = XParseClient(app_id="test", secret_code="test", server_url=server_url)
+    extract_config = ExtractConfig(
+        schema_={"type": "object", "properties": {"姓名": {"type": "string"}}},
+        generate_citations=True
+    )
+
+    result = client.extract.extract(
+        file=b"test",
+        filename="test.pdf",
+        extract_config=extract_config
+    )
+
+    # 验证请求中包含 generate_citations
+    request = respx.calls.last.request
+    assert b"generate_citations" in request.content
+
+    # 验证结果
+    assert result is not None
+
+
+@respx.mock
+def test_extract_with_stamp(server_url, extract_success_response):
+    """测试 stamp 参数（印章提取）"""
+    respx.post(f"{server_url}/api/xparse/extract/sync").mock(
+        return_value=httpx.Response(200, json=extract_success_response)
+    )
+
+    client = XParseClient(app_id="test", secret_code="test", server_url=server_url)
+    extract_config = ExtractConfig(
+        schema_={"type": "object", "properties": {"公司名称": {"type": "string"}}},
+        stamp=True
+    )
+
+    result = client.extract.extract(
+        file=b"test invoice",
+        filename="invoice.pdf",
+        extract_config=extract_config
+    )
+
+    # 验证请求中包含 stamp
+    request = respx.calls.last.request
+    assert b"stamp" in request.content
+
+    # 验证结果
+    assert result is not None