Function process_first_page

mainprogress/determine_toc_levels.py:242–319

专门处理第一张图片，获取 CSV 格式的响应，并缓存为 Few-shot 示例。

(img_file: Path, csv_file: Path, output_path: Path)

    try:
        for attempt in range(MAX_RETRIES):
            try:
                response = await client.chat.completions.create(
                    model=LLM_CONFIG["model"],
                    messages=messages,
                    extra_body={"enable_thinking": False},
                    timeout=REQUEST_TIMEOUT,
                    temperature=0,
                )

                content = response.choices[0].message.content.strip()

                # 本地解析 CSV 转为 JSON 保存
                try:
                    parsed_data = parse_csv_response(content, img_file.name)
                except Exception as parse_err:
                    write_log(f"首图 CSV 解析失败：{parse_err}")
                    if attempt == MAX_RETRIES - 1:
                        return False
                    await asyncio.sleep(2 ** attempt)
                    continue

                if parsed_data:
                    # 排序
                    sorted_data = sorted(parsed_data, key=lambda x: x['number'])

                    # 保存首图结果文件
                    output_file = output_path / f"{img_file.stem}_merged.json"
                    with open(output_file, 'w', encoding='utf-8') as f:
                        json.dump(sorted_data, f, ensure_ascii=False, indent=2)

                    # 存入全局变量作为 Few-shot 示例 (存储原始 CSV 字符串)
                    FIRST_PAGE_EXAMPLE["image_base64"] = get_encoded_image(img_file)
                    FIRST_PAGE_EXAMPLE["result_csv_str"] = content

                    print(f"首图处理完成并已缓存为示例：{img_file.name}")
                    return True
                else:
                    write_log(f"首图解析结果为空：{img_file.name}")
                    if attempt == MAX_RETRIES - 1:
                        return False
                    await asyncio.sleep(2 ** attempt)

            except (APIError, Timeout) as e:
                write_log(f"首图第 {attempt+1} 次 API 请求失败 ({type(e).__name__}): {str(e)}")
                if attempt == MAX_RETRIES - 1:
                    return False
                await asyncio.sleep(2 ** attempt)
            except Exception as e:
                write_log(f"首图处理异常 {img_file.name}: {str(e)}")
                if attempt == MAX_RETRIES - 1:
                    return False
                await asyncio.sleep(2 ** attempt)

        return False
    finally:
        IMAGE_CACHE.pop(img_file, None)