feat: support api call to translate the file

langchain/openai-translator/ai_translator/gradio_server.py

@@ -8,14 +8,28 @@
 from translator import PDFTranslator, TranslationConfig
 
 
-def translation(input_file, source_language, target_language):
-    LOG.debug(f"[翻译任务]\n源文件: {input_file.name}\n源语言: {source_language}\n目标语言: {target_language}")
+def translation(input_file, source_language, target_language, style, writer):
+    LOG.debug(f"[翻译任务]\n源文件: {input_file.name}\n源语言: {source_language}\n目标语言: {target_language}\n文体类型: {style}\n作家名字: {writer}")
 
     output_file_path = Translator.translate_pdf(
-        input_file.name, source_language=source_language, target_language=target_language)
+        input_file.name, source_language=source_language, target_language=target_language, style=style, writer=writer)
 
     return output_file_path
 
+def change_writer(choice):
+    if choice == "input":
+        return gr.Textbox.update(visible=True, value="")
+    else:
+        return gr.Textbox.update(visible=False, value=choice)
+
+def change_style(choice):
+    if choice == "custom":
+        return gr.Textbox.update(visible=True, value="")
+    elif choice == "novel" or choice == "news":
+        return gr.Textbox.update(visible=False, value=choice)
+    else:
+        return gr.Textbox.update(visible=False, value="none")
+
 def launch_gradio():
 
     iface = gr.Interface(
@@ -24,7 +38,8 @@ def launch_gradio():
         inputs=[
             gr.File(label="上传PDF文件"),
             gr.Textbox(label="源语言（默认：英文）", placeholder="English", value="English"),
-            gr.Textbox(label="目标语言（默认：中文）", placeholder="Chinese", value="Chinese")
+            gr.Textbox(label="目标语言（默认：中文）", placeholder="Chinese", value="Chinese"),
+            gr.Radio([" none", "novel", "news", ""], label="文本类型", info="Where did they go?")
         ],
         outputs=[
             gr.File(label="下载翻译文件")
@@ -34,6 +49,31 @@ def launch_gradio():
 
     iface.launch(share=True, server_name="0.0.0.0")
 
+def launch_gradio_by_blocks():
+    with gr.Blocks() as blocks:
+        input_file = gr.File(label="上传PDF文件")
+        source_language = gr.Textbox(
+            label="源语言（默认：英文）", placeholder="English", value="English"
+        )
+        target_language = gr.Textbox(
+            label="目标语言（默认：中文）", placeholder="Chinese", value="Chinese"
+        )
+        style_radio = gr.Radio(["none", "novel", "news", "custom"], label="文体类型", info="选择文体类型，custom 自定义输入", value="none")
+        style_text = gr.Textbox(show_label=False, lines=1, visible=False, placeholder="输入文体类型", value="none")
+        style_radio.change(fn=change_style, inputs=style_radio, outputs=style_text)
+        writer_radio = gr.Radio(["none", "input"], label="作家风格", info="input 自定义输入作家名字", value="none")
+        writer_text = gr.Textbox(show_label=False, lines=1, visible=False, placeholder="输入作家名字", value="none")
+        writer_radio.change(fn=change_writer, inputs=writer_radio, outputs=writer_text)
+        output_file = gr.File(label="下载翻译文件")
+        clear = gr.ClearButton(components=[input_file, source_language, target_language, style_radio, writer_radio, writer_text])
+        submit = gr.Button("Submit")
+        submit.click(
+            translation,
+            inputs=[input_file, source_language, target_language, style_text, writer_text],
+            outputs=[output_file],
+        )
+    blocks.launch(share=True, server_name="0.0.0.0")
+
 def initialize_translator():
     # 解析命令行
     argument_parser = ArgumentParser()
@@ -51,4 +91,5 @@ def initialize_translator():
     # 初始化 translator
     initialize_translator()
     # 启动 Gradio 服务
-    launch_gradio()
+    //launch_gradio()
+    launch_gradio_by_blocks()

langchain/openai-translator/ai_translator/translator/pdf_translator.py

@@ -15,14 +15,22 @@ def translate_pdf(self,
                     output_file_format: str = 'markdown',
                     source_language: str = "English",
                     target_language: str = 'Chinese',
+                    style: str = 'none',
+                    writer: str = 'none',
                     pages: Optional[int] = None):
 
         self.book = self.pdf_parser.parse_pdf(input_file, pages)
 
+        style_template = f""
+        if style != 'none':
+            style_template += f"The translation is in {style} style."
+        if writer != 'none':
+            style_template += f"Refer to the style of writer {writer}"
+
         for page_idx, page in enumerate(self.book.pages):
             for content_idx, content in enumerate(page.contents):
                 # Translate content.original
-                translation, status = self.translate_chain.run(content, source_language, target_language)
+                translation, status = self.translate_chain.run(content, source_language, target_language,style_template)
                 # Update the content in self.book.pages directly
                 self.book.pages[page_idx].contents[content_idx].set_translation(translation, status)
 

langchain/openai-translator/ai_translator/translator/translation_chain.py

@@ -10,7 +10,7 @@ def __init__(self, model_name: str = "gpt-3.5-turbo", verbose: bool = True):
         # 翻译任务指令始终由 System 角色承担
         template = (
             """You are a translation expert, proficient in various languages. \n
-            Translates {source_language} to {target_language}."""
+            Translates {source_language} to {target_language}.{style_template}"""
         )
         system_message_prompt = SystemMessagePromptTemplate.from_template(template)
 
@@ -24,17 +24,18 @@ def __init__(self, model_name: str = "gpt-3.5-turbo", verbose: bool = True):
         )
 
         # 为了翻译结果的稳定性，将 temperature 设置为 0
-        chat = ChatOpenAI(model_name=model_name, temperature=0, verbose=verbose)
+        chat = ChatOpenAI(model_name=model_name, base_url= os.getenv("OPENAI_BASE_URL"),api_key=os.getenv("OPENAI_API_KEY"),temperature=0, verbose=verbose)
 
         self.chain = LLMChain(llm=chat, prompt=chat_prompt_template, verbose=verbose)
 
-    def run(self, text: str, source_language: str, target_language: str) -> (str, bool):
+    def run(self, text: str, source_language: str, target_language: str, style_template: str) -> (str, bool):
         result = ""
         try:
             result = self.chain.run({
                 "text": text,
                 "source_language": source_language,
                 "target_language": target_language,
+                "style_template": style_template,
             })
         except Exception as e:
             LOG.error(f"An error occurred during translation: {e}")

openai-translator/ai_translator/app.py

@@ -0,0 +1,46 @@
+import os, json
+import sys
+# not sure why can't import module without it.
+sys.path.append(os.path.dirname(os.path.abspath(__file__)))
+
+from flask import Flask,request
+# open ai
+from translator import PDFTranslator
+from model import OpenAIModel
+
+app = Flask(__name__)
+
+# Upload path
+UPLOAD_FOLDER = '/Users/Kelven/TempUpload/'
+app.config['UPLOAD_FOLDER'] = UPLOAD_FOLDER
+
+# translate the file from local dir
+@app.route('/api/v1/translate/<filename>', methods=['POST'])
+def file_translate(filename):
+
+    data = request.get_data()
+    # parse request data
+    json_data = json.loads(data.decode('utf-8'))
+    file_type = json_data.get('file_type')
+    filefullname = filename + '.' + file_type
+    # support request language
+    target_language = json_data.get('target_language')
+    target_format = json_data.get('target_format')
+
+    newModel = OpenAIModel(model='gpt-3.5-turbo', api_key=os.getenv("OPENAI_API_KEY"))
+    translator = PDFTranslator(newModel)
+
+    # target_language not support yet 
+    translator.translate_pdf(app.config['UPLOAD_FOLDER'] + filefullname, target_format)
+
+    output_filename = app.config['UPLOAD_FOLDER'] + os.path.basename(app.config['UPLOAD_FOLDER'] + filename + "_translated.md")
+
+    return {
+        'status': 0,
+        'msg': '',
+        'data': output_filename
+    }
+
+
+if __name__ == '__main__':
+    app.run(debug=True)

openai-translator/ai_translator/model/openai_model.py

@@ -11,7 +11,7 @@
 class OpenAIModel(Model):
     def __init__(self, model: str, api_key: str):
         self.model = model
-        self.client = OpenAI(api_key=os.getenv("OPENAI_API_KEY"))
+        self.client = OpenAI(base_url= os.getenv("OPENAI_BASE_URL"),api_key=os.getenv("OPENAI_API_KEY"))
 
     def make_request(self, prompt):
         attempts = 0