这是翻译后的HTML内容...
" } }, @@ -798,7 +818,7 @@ async def service_download_file( "summary": "ZIP 内容 (Base64)", "value": { "file_type": "markdown_zip", - "filename": "my_doc.pdf", + "filename": "my_doc_translated.zip", "content": "UEsDBBQAAAAIA... (base64-encoded string)" } } @@ -807,7 +827,7 @@ async def service_download_file( } }, 404: { - "description": "资源未找到。可能的原因包括：任务ID不存在、任务结果尚未就绪、或请求了无效的文件类型。", + "description": "资源未找到。", "content": {"application/json": {"example": {"detail": "内容尚未准备好。"}}} }, } @@ -817,30 +837,22 @@ async def service_content( file_type: FileType = FastApiPath(..., description="要获取内容的文件类型。", examples=["html"]) ): """根据任务ID和文件类型，以JSON格式返回内容。zip文件会进行Base64编码。""" - task_state = tasks_state.get(task_id) - if not task_state: - raise HTTPException(status_code=404, detail=f"找不到任务ID '{task_id}'。") + content, _, filename = await _get_content_from_manager(task_id, file_type) - if not task_state["download_ready"]: - raise HTTPException(status_code=404, detail="内容尚未准备好。") + if isinstance(content, bytes): + try: + # For text-based formats, decode to string + final_content = content.decode('utf-8') + except UnicodeDecodeError: + # For binary formats (like zip), encode to Base64 + final_content = base64.b64encode(content).decode('utf-8') + else: # Should not happen with current _get_content_from_manager, but for safety + final_content = content - content_map = { - "markdown": (task_state.get("markdown_content"), task_state['original_filename']), - "markdown_zip": (task_state.get("markdown_zip_content"), task_state['original_filename']), - "html": (task_state.get("html_content"), task_state['original_filename']), - } - - raw_content, filename = content_map.get(file_type, (None, None)) - - if raw_content is None: - raise HTTPException(status_code=404, detail=f"'{file_type}' 类型的内容不可用或生成失败。") - - # 如果内容是字节串 (zip)，则进行Base64编码；否则直接使用字符串。 - final_content = base64.b64encode(raw_content).decode('utf-8') if isinstance(raw_content, bytes) else raw_content return JSONResponse(content={ "file_type": file_type, - "original_filename": filename, + "filename": filename, "content": final_content }) @@ -854,14 +866,15 @@ async def service_content( responses={ 200: { "description": "成功返回可用引擎列表。", - "content": {"application/json": {"example": ["mineru", "docling"]}} + "content": {"application/json": {"example": ["auto", "mineru", "docling"]}} } } ) async def service_get_engin_list(): """返回可用的文档解析引擎列表。""" - engin_list = ["mineru"] - if available_packages.get("docling"): engin_list.append("docling") + engin_list = ["auto", "mineru"] + if available_packages.get("docling"): + engin_list.append("docling") return JSONResponse(content=engin_list) @@ -952,10 +965,12 @@ async def custom_swagger_ui_html(): swagger_css_url="/static/swagger/swagger.css", ) + @app.get(app.swagger_ui_oauth2_redirect_url, include_in_schema=False) async def swagger_ui_redirect(): return get_swagger_ui_oauth2_redirect_html() + @app.get("/redoc", include_in_schema=False) async def redoc_html(): return get_redoc_html( @@ -963,10 +978,12 @@ async def redoc_html(): title=app.title + " - ReDoc", redoc_js_url="/static/redoc/redoc.js", ) + + ### @app.post("/temp/translate", - summary="[临时]同步翻译接口", + summary="[临时]同步翻译接口 (已重构)", description="一个简单的、同步的翻译接口，用于快速测试。不涉及后台任务、状态管理或多格式输出。**不建议在生产环境中使用。**", tags=["Temp"], responses={ @@ -990,43 +1007,49 @@ async def temp_translate( base_url: str = Body(..., description="LLM API的基础URL。", examples=["https://api.openai.com/v1"]), api_key: str = Body(..., description="LLM API的密钥。", examples=["sk-xxxxxxxxxx"]), model_id: str = Body(..., description="使用的模型ID。", examples=["gpt-4-turbo"]), - mineru_token: str = Body(..., description="Mineru引擎的Token。"), + mineru_token: Optional[str] = Body(None, description="Mineru引擎的Token。"), file_name: str = Body(..., description="文件名，用以判断文件类型。当后缀为txt时该接口返回普通文本，为其他后缀时返回翻译后的markdown文本", examples=["test.txt", "test.md", "test.pdf"]), file_content: str = Body(..., description="文件内容，可以是纯文本或Base64编码的字符串。"), to_lang: str = Body("中文", description="目标语言。", examples=["中文", "英文", "English"]), concurrent: int = Body(default_params["concurrent"], description="ai翻译请求并发数"), - temperature: float | None = Body(default_params["temperature"], description="ai翻译请求温度"), + temperature: float = Body(default_params["temperature"], description="ai翻译请求温度"), chunk_size: int = Body(default_params["chunk_size"], description="文本分块大小（bytes）"), - custom_prompt_translate: str | None = Body(None, description="翻译自定义提示词", + custom_prompt_translate: Optional[str] = Body(None, description="翻译自定义提示词", examples=["人名保持原文不翻译"]), ): """一个用于快速测试的同步翻译接口。""" - - def is_base64(s): - try: - base64.b64decode(s, validate=True) - return True - except (ValueError, binascii.Error): - return False - - ft = FileTranslater(base_url=base_url, - key=api_key, - model_id=model_id, - mineru_token=mineru_token, - concurrent=concurrent, - temperature=temperature, - chunk_size=chunk_size, - ) + try: + decoded_content = base64.b64decode(file_content) + except (ValueError, binascii.Error): + decoded_content = file_content.encode('utf-8') try: - decoded_content = base64.b64decode(file_content) if is_base64(file_content) else file_content.encode('utf-8') - await ft.translate_bytes_async(name=file_name, file=decoded_content, to_lang=to_lang, save=False, - custom_prompt_translate=custom_prompt_translate) - return {"success": True, "content": ft.export_to_markdown()} + manager = _get_manager_for_file(file_name, global_logger) + + ai_config = AiTranslateConfig( + base_url=base_url, api_key=api_key, model_id=model_id, to_lang=to_lang, + custom_prompt=custom_prompt_translate, temperature=temperature, + chunk_size=chunk_size, concurrent=concurrent, logger=global_logger, timeout=2000 + ) + + manager.read_bytes(decoded_content, Path(file_name).stem, Path(file_name).suffix) + + if isinstance(manager, MarkdownBasedManager): + translate_config = MDTranslateConfig(**ai_config.__dict__) + convert_config = ConverterMineruConfig(mineru_token=mineru_token) if mineru_token else None + convert_engin = 'mineru' if mineru_token else None + await manager.translate_async(convert_engin, convert_config, translate_config) + return {"success": True, "content": manager.document_translated.get_text()} + + elif isinstance(manager, TXTManager): + translate_config = TXTTranslateConfig(**ai_config.__dict__) + await manager.translate_async(translate_config) + return {"success": True, "content": manager.export_to_txt()} + except Exception as e: - print(f"翻译出现错误：{e.__repr__()}") + print(f"临时翻译接口出现错误：{e.__repr__()}") return {"success": False, "reason": e.__repr__()} @@ -1050,10 +1073,10 @@ def run_app(port: int | None = None): print(f"正在启动 DocuTranslate WebUI 版本号：{__version__}\n") print(f"服务接口文档: http://127.0.0.1:{port_to_use}/docs\n") print(f"请用浏览器访问 http://127.0.0.1:{port_to_use}\n") - uvicorn.run(app, host=None, port=port_to_use, workers=1) + uvicorn.run(app, host="0.0.0.0", port=port_to_use, workers=1) except Exception as e: print(f"启动失败: {e}") if __name__ == "__main__": - run_app() + run_app() \ No newline at end of file diff --git a/docutranslate/exporter/md2x/interfaces.py b/docutranslate/exporter/md2x/interfaces.py index 515b86f..863c7ed 100644 --- a/docutranslate/exporter/md2x/interfaces.py +++ b/docutranslate/exporter/md2x/interfaces.py @@ -1,6 +1,3 @@ -from typing import Self - -from docutranslate.exporter.export_config import ExportConfig from docutranslate.exporter.interfaces import Exporter from docutranslate.ir.document import Document from docutranslate.ir.markdown_document import MarkdownDocument diff --git a/docutranslate/exporter/md2x/md2md_exporter.py b/docutranslate/exporter/md2x/md2md_exporter.py index 1f8896c..22ce53f 100644 --- a/docutranslate/exporter/md2x/md2md_exporter.py +++ b/docutranslate/exporter/md2x/md2md_exporter.py @@ -9,18 +9,12 @@ from docutranslate.utils.markdown_utils import unembed_base64_images_to_zip @dataclass class MD2MDExportConfig(ExportConfig): - embed_images: bool = True + pass class MD2MDExporter(MDExporter): def __init__(self, export_config: MD2MDExportConfig | None=None): - export_config=export_config or MD2MDExportConfig() - self.embed_images=export_config.embed_images + pass def export(self,document:MarkdownDocument)->Document: - if self.embed_images: - return Document.from_bytes(suffix=".md",content=document.content,stem=document.stem) - else: - return Document.from_bytes(suffix=".zip",content=unembed_base64_images_to_zip(document.content.decode(), markdown_name=document.name),stem=document.stem) - - + return Document.from_bytes(suffix=".md",content=document.content,stem=document.stem) diff --git a/docutranslate/exporter/md2x/md2mdzip_exporter.py b/docutranslate/exporter/md2x/md2mdzip_exporter.py new file mode 100644 index 0000000..effb66a --- /dev/null +++ b/docutranslate/exporter/md2x/md2mdzip_exporter.py @@ -0,0 +1,21 @@ +from dataclasses import dataclass + +from docutranslate.exporter.export_config import ExportConfig +from docutranslate.exporter.md2x.interfaces import MDExporter +from docutranslate.ir.markdown_document import MarkdownDocument,Document +from docutranslate.utils.markdown_utils import unembed_base64_images_to_zip + + +@dataclass +class MD2MDZIPExportConfig(ExportConfig): + pass + + +class MD2MDZipExporter(MDExporter): + def __init__(self, export_config: MD2MDZIPExportConfig | None=None): + pass + + def export(self,document:MarkdownDocument)->Document: + return Document.from_bytes(suffix=".zip",content=unembed_base64_images_to_zip(document.content.decode(), markdown_name=document.name),stem=document.stem) + + diff --git a/docutranslate/exporter/txt2x/txt2html_exporter.py b/docutranslate/exporter/txt2x/txt2html_exporter.py index c7f92ac..f9b7ccb 100644 --- a/docutranslate/exporter/txt2x/txt2html_exporter.py +++ b/docutranslate/exporter/txt2x/txt2html_exporter.py @@ -25,8 +25,10 @@ class TXT2HTMLExporter(TXTExporter): # language=html pico = f'' if not cdn else r'' + body='\n'.join([r''+para+'
' for para in document.content.decode().split("\n")]) render = jinja2.Template(html_template).render( title=document.stem, pico=pico, + body=body, ) return Document.from_bytes(content=render.encode("utf-8"), suffix=".html", stem=document.stem) diff --git a/docutranslate/manager/interfaces.py b/docutranslate/manager/interfaces.py index 496a06f..abaeb58 100644 --- a/docutranslate/manager/interfaces.py +++ b/docutranslate/manager/interfaces.py @@ -6,27 +6,42 @@ from docutranslate.exporter.export_config import ExportConfig T = TypeVar("T", bound=ExportConfig) - @runtime_checkable -class HTMLExportable(Protocol): - def export_to_html(self, export_config: T) -> str: +class HTMLExportable(Protocol[T]): + def export_to_html(self, export_config: T | None = None) -> str: ... - def save_as_html(self, name: str, output_dir: Path | str, export_config: T) -> Self: + def save_as_html(self, name: str, output_dir: Path | str, export_config: T | None = None) -> Self: ... @runtime_checkable -class MDExportable(Protocol): - def export_to_markdown(self, export_config: T) -> str: +class MDExportable(Protocol[T]): + + def export_to_markdown(self, export_config: T | None = None) -> str: ... - def save_as_markdown(self, name: str, output_dir: Path | str, export_config: T) -> Self: + def save_as_markdown(self, name: str, output_dir: Path | str, export_config: T | None = None) -> Self: ... @runtime_checkable -class TXTExportable(Protocol): +class MDZIPExportable(Protocol[T]): + + def export_to_markdown_zip(self, export_config: T | None = None) -> bytes: + ... + + def save_as_markdown_zip(self, name: str, output_dir: Path | str, export_config: T | None = None) -> Self: + ... + + +@runtime_checkable +class MDFormatsExportable(MDZIPExportable[T], MDExportable[T], Protocol): + ... + + +@runtime_checkable +class TXTExportable(Protocol[T]): def export_to_txt(self) -> str: ... diff --git a/docutranslate/manager/md_based_manager.py b/docutranslate/manager/md_based_manager.py index 623e35a..334b9a6 100644 --- a/docutranslate/manager/md_based_manager.py +++ b/docutranslate/manager/md_based_manager.py @@ -9,13 +9,14 @@ from docutranslate.converter.x2md.converter_mineru import ConverterMineruConfig, from docutranslate.converter.x2md.interfaces import X2MarkdownConverter from docutranslate.exporter.md2x.md2html_exporter import MD2HTMLExportConfig, MD2HTMLExporter from docutranslate.exporter.md2x.md2md_exporter import MD2MDExportConfig, MD2MDExporter +from docutranslate.exporter.md2x.md2mdzip_exporter import MD2MDZIPExportConfig, MD2MDZipExporter from docutranslate.exporter.md2x.types import x2md_convert_config_type, convert_engin_type from docutranslate.manager.base_manager import BaseManager -from docutranslate.manager.interfaces import HTMLExportable, MDExportable +from docutranslate.manager.interfaces import MDFormatsExportable, HTMLExportable from docutranslate.translater.md_translator import MDTranslateConfig, MDTranslator -class MarkdownBasedManager(BaseManager, HTMLExportable, MDExportable): +class MarkdownBasedManager(BaseManager, HTMLExportable, MDFormatsExportable): def __init__(self, *args, **kwargs): super().__init__(*args, **kwargs) self._converter_factory: dict[str:tuple[X2MarkdownConverter, x2md_convert_config_type]] = { @@ -39,7 +40,8 @@ class MarkdownBasedManager(BaseManager, HTMLExportable, MDExportable): elif convert_engin in self._converter_factory: converter_class, config_class = self._converter_factory[convert_engin] if not isinstance(convert_config, config_class): - raise TypeError(f"未传入正确的convert_config，应传入{config_class.__name__}类型") + raise TypeError( + f"未传入正确的convert_config，应传入{config_class.__name__}类型，现为{type(convert_config).__name__}类型") converter = converter_class(convert_config, logger=self.logger) else: raise ValueError(f"不存在{convert_engin}解析引擎") @@ -89,16 +91,26 @@ class MarkdownBasedManager(BaseManager, HTMLExportable, MDExportable): return docu.content.decode() def export_to_markdown(self, export_config: MD2MDExportConfig | None = None) -> str: - docu = self._export(MD2MDExporter(export_config)) + docu = self._export(MD2MDExporter()) return docu.content.decode() + def export_to_markdown_zip(self, export_config: MD2MDZIPExportConfig | None = None) -> bytes: + docu = self._export(MD2MDZipExporter()) + return docu.content + def save_as_html(self, name: str = None, output_dir: Path | str = "./output", export_config: MD2HTMLExportConfig | None = None) -> Self: - self._save(exporter=MD2HTMLExporter(export_config), name=name, output_dir=output_dir) + self._save(exporter=MD2HTMLExporter(), name=name, output_dir=output_dir) return self def save_as_markdown(self, name: str = None, output_dir: Path | str = "./output", export_config: MD2MDExportConfig | None = None) -> Self: - self._save(exporter=MD2MDExporter(export_config), name=name, output_dir=output_dir) + self._save(exporter=MD2MDExporter(), name=name, output_dir=output_dir) + return self + + def save_as_markdown_zip(self, name: str = None, output_dir: Path | str = "./output", + export_config: MD2MDZIPExportConfig | None = None) -> Self: + + self._save(exporter=MD2MDZipExporter(), name=name, output_dir=output_dir) return self diff --git a/docutranslate/manager/txt_manager.py b/docutranslate/manager/txt_manager.py index e8d01f8..1879430 100644 --- a/docutranslate/manager/txt_manager.py +++ b/docutranslate/manager/txt_manager.py @@ -27,7 +27,7 @@ class TXTManager(BaseManager, HTMLExportable,TXTExportable): self.document_translated = document return self - def export_to_html(self, export_config: TXT2HTMLExportConfig) -> str: + def export_to_html(self, export_config: TXT2HTMLExportConfig=None) -> str: docu = self._export(TXT2HTMLExporter(export_config)) return docu.content.decode() diff --git a/docutranslate/static/index.html b/docutranslate/static/index.html index 6e42292..a3bdb62 100644 --- a/docutranslate/static/index.html +++ b/docutranslate/static/index.html @@ -1 +1 @@ -
GitHub主页(欢迎star❤):
https://github.com/xunbu/docutranslate
交流QQ群: 1047781902
当前没有任务，点击“新建任务”开始吧！
等待提交...
点击或拖拽文件到此处
GitHub主页(欢迎star❤):
https://github.com/xunbu/docutranslate
交流QQ群: 1047781902
当前没有任务，点击“新建任务”开始吧！
等待提交...
点击或拖拽文件到此处