Extractions 命名空间

内容提取操作。

create

创建提取任务。

extraction = client.extractions.create(
    file_id: str,
    mode: str = "normal",
    options: dict = None
) -> Extraction

参数

参数	类型	说明
file_id	str	文件 ID
mode	str	simple、normal、advanced
options	dict	提取选项

示例

# 基本提取
extraction = client.extractions.create(file_id="file_abc123")

# 高级提取
extraction = client.extractions.create(
    file_id="file_abc123",
    mode="advanced",
    options={"language": "zh", "ocr_provider": "default"}
)

# 等待完成
extraction = extraction.wait(timeout=300)
print(extraction.markdown)

---

get

获取提取结果。

extraction = client.extractions.get(extraction_id: str) -> Extraction

示例

extraction = client.extractions.get("ext_xyz789")
if extraction.status == "completed":
    print(extraction.markdown)

---

list

获取文件的提取记录。

extractions = client.extractions.list(
    file_id: str,
    limit: int = 50,
    offset: int = 0
) -> ExtractionList

---

Extraction 对象

@dataclass
class Extraction:
    id: str                    # 提取 ID
    file_id: str               # 文件 ID
    status: str                # pending | processing | completed | failed
    mode: str                  # 提取模式
    progress: int              # 进度 (0-100)
    markdown: str              # 提取内容
    total_pages: int           # 总页数
    metadata: dict             # 文档元信息
    error: dict                # 错误信息
    created_at: datetime
    completed_at: datetime

    def wait(self, timeout: int = 300) -> "Extraction":
        """等待提取完成"""
        pass

wait 方法

# 同步等待
extraction = extraction.wait(timeout=300)

if extraction.status == "completed":
    print(f"提取成功: {extraction.total_pages} 页")
elif extraction.status == "failed":
    print(f"提取失败: {extraction.error}")