diff --git a/README.md b/README.md index 650d253..25074d1 100644 --- a/README.md +++ b/README.md @@ -19,7 +19,7 @@ # 安装 -使用pip +使用pip `pip install docutranslate` `pip install docutranslate[docling]`#如果需要使用docling进行文档解析 @@ -122,9 +122,9 @@ docutranslate -i ```python from docutranslate.translater import FileTranslater -translater = FileTranslater(base_url="",#大模型的baseurl - key="",#大模型的api-key - model_id="",#大模型的model-id +translater = FileTranslater(base_url="", # 大模型的baseurl + key="", # 大模型的api-key + model_id="", # 大模型的model-id convert_engin="mineru", # 使用mineru解析文档 mineru_token="<申请的mineru_token>" # 使用mineru时必填 # convert_engin="docling" # 使用docling解析文档 @@ -163,13 +163,16 @@ translater.translate_file("<文件路径>", to_lang="中文", refine_agent=refin ```python from docutranslate import FileTranslater -translater = FileTranslater(base_url="", - key="", - model_id="") +translater = FileTranslater(convert_engin="mineru", # 使用mineru解析文档 + mineru_token="<申请的mineru_token>" # 使用mineru时必填 + # convert_engin="docling" # 使用docling解析文档 + ) # 文件转html -translater.read_file("<文件路径>").save_as_html() +translater.read_file("<文件路径>").save_as_html()#保存 +translater.read_file("<文件路径>").export_to_html()#输出字符串 # 文件转markdown -translater.read_file("<文件路径>").save_as_markdown() +translater.read_file("<文件路径>").save_as_markdown()#保存 +translater.read_file("<文件路径>").export_to_markdown()#输出字符串 ``` ## 参数说明 @@ -182,8 +185,8 @@ from docutranslate import FileTranslater translater = FileTranslater(base_url="", # 默认的模型baseurl key="", # 默认的大语言模型平台api-key model_id="", # 默认的模型id - chunksize=2000, # markdown分块长度(单位byte),分块越大效果越好(也越慢),不建议超过8000 - max_concurrent=20, # 并发数,受到ai平台并发量限制,如果文章很长建议适当加大到20以上 + chunksize=3000, # markdown分块长度(单位byte),分块越大效果越好(也越慢),不建议超过8000 + max_concurrent=30, # 并发数,受到ai平台并发量限制,如果文章很长建议适当加大到20以上 timeout=2000, # 调用api的超时时间 docling_artifact=None, # 使用提前下载好的docling模型 convert_engin="mineru", # 可选minerU或docling diff --git a/docutranslate/__init__.py b/docutranslate/__init__.py index a56f6ba..4a53e8c 100644 --- a/docutranslate/__init__.py +++ b/docutranslate/__init__.py @@ -1,4 +1,4 @@ -__version__="0.2.35" +__version__="0.2.36" diff --git a/docutranslate/cache/__init__.py b/docutranslate/cacher/__init__.py similarity index 100% rename from docutranslate/cache/__init__.py rename to docutranslate/cacher/__init__.py diff --git a/docutranslate/cache/document_cacher.py b/docutranslate/cacher/document_cacher.py similarity index 100% rename from docutranslate/cache/document_cacher.py rename to docutranslate/cacher/document_cacher.py diff --git a/docutranslate/translater.py b/docutranslate/translater.py index 3836126..7ebe15b 100644 --- a/docutranslate/translater.py +++ b/docutranslate/translater.py @@ -6,7 +6,7 @@ import markdown2 import jinja2 from docutranslate.agents import Agent, AgentArgs from docutranslate.agents import MDRefineAgent, MDTranslateAgent -from docutranslate.cache import document_cacher_global +from docutranslate.cacher import document_cacher_global from docutranslate.converter import Document, ConverterMineru from docutranslate.utils.markdown_splitter import split_markdown_text, join_markdown_texts from docutranslate.utils.markdown_utils import uris2placeholder, placeholder2_uris, MaskDict, clean_markdown_math_block @@ -308,7 +308,7 @@ class FileTranslater: # 确保输出目录存在 output_dir.mkdir(parents=True, exist_ok=True) full_name = output_dir / filename - html_content = self.export_to_html(str(filename.resolve().stem)) + html_content = self.export_to_html(title=str(full_name.resolve().stem)) with open(full_name, "w") as file: file.write(html_content) translater_logger.info(f"文件已写入{full_name.resolve()}") diff --git a/pyproject.toml b/pyproject.toml index 4ec514b..c22346e 100644 --- a/pyproject.toml +++ b/pyproject.toml @@ -8,7 +8,7 @@ dependencies = [ "markdown2>=2.5.3", "fastapi[standard]>=0.115.12", ] -dynamic=["version"] +dynamic = ["version"] [project.optional-dependencies] docling = [ @@ -32,7 +32,10 @@ requires = ["setuptools>=61.0"] # 或者你需要的 setuptools 版本 build-backend = "setuptools.build_meta" backend-path = ["."] [tool.setuptools] -packages = ["docutranslate"] +#packages = ["docutranslate"] +[tool.setuptools.packages.find] +where = ["."] +include = ["docutranslate*"] [tool.setuptools.package-data] docutranslate = ["static/**", "template/**"]