Spaces:

mistpe
/

translate

Running

App Files Files Community

mistpe commited on Nov 23, 2024

Commit

92cb7b7

verified ·

1 Parent(s): 7993fc5

Update app.py

Browse files

Files changed (1) hide show

app.py +64 -60

app.py CHANGED Viewed

@@ -1,6 +1,7 @@
 from fastapi import FastAPI, File, UploadFile, HTTPException, Form, Body
 from fastapi.staticfiles import StaticFiles
-from fastapi.responses import StreamingResponse, FileResponse, HTMLResponse
 from fastapi.middleware.cors import CORSMiddleware
 import uvicorn
 import asyncio
@@ -9,8 +10,6 @@ import json
 import webbrowser
 from typing import List, Dict, Optional
 import os
-import tempfile
-import shutil
 from pathlib import Path
 import pypdf
 from docx import Document
@@ -25,6 +24,11 @@ import hashlib
 import re
 from io import BytesIO
 app = FastAPI()
 # 配置CORS
@@ -36,51 +40,8 @@ app.add_middleware(
     allow_headers=["*"],
 )
-# 创建临时目录管理器
-class TempDirManager:
-    def __init__(self):
-        self.temp_base = tempfile.mkdtemp()
-        # 创建必要的子目录
-        self.static_dir = os.path.join(self.temp_base, "static")
-        self.temp_dir = os.path.join(self.temp_base, "temp")
-        self.memory_dir = os.path.join(self.temp_base, "translation_memory")
-        os.makedirs(self.static_dir, exist_ok=True)
-        os.makedirs(self.temp_dir, exist_ok=True)
-        os.makedirs(self.memory_dir, exist_ok=True)
-        # 创建默认的index.html
-        self._create_default_index()
-    def _create_default_index(self):
-        index_html = """
-        <!DOCTYPE html>
-        <html>
-        <head>
-            <title>Translation Service</title>
-            <meta charset="UTF-8">
-            <style>
-                body { font-family: Arial, sans-serif; margin: 40px; }
-                h1 { color: #333; }
-            </style>
-        </head>
-        <body>
-            <h1>Translation Service</h1>
-            <p>API service is running. Please use the API endpoints to interact with the service.</p>
-        </body>
-        </html>
-        """
-        with open(os.path.join(self.static_dir, "index.html"), "w", encoding="utf-8") as f:
-            f.write(index_html)
-    def cleanup(self):
-        shutil.rmtree(self.temp_base)
-# 创建临时目录管理器实例
-temp_dir_manager = TempDirManager()
 # 挂载静态文件目录
-app.mount("/static", StaticFiles(directory=temp_dir_manager.static_dir), name="static")
 class DocumentSegment(BaseModel):
     text: str
@@ -108,9 +69,16 @@ class ExportRequest(BaseModel):
 class TranslationMemory:
     def __init__(self):
-        self.memory_file = os.path.join(temp_dir_manager.memory_dir, "memory.json")
-        self.memory = {}
-        self.save_memory()
     def save_memory(self):
         with open(self.memory_file, 'w', encoding='utf-8') as f:
@@ -131,9 +99,16 @@ class TranslationMemory:
 class TerminologyManager:
     def __init__(self):
-        self.terminology_file = os.path.join(temp_dir_manager.memory_dir, "terminology.json")
-        self.terminology = {}
-        self.save_terminology()
     def save_terminology(self):
         with open(self.terminology_file, 'w', encoding='utf-8') as f:
@@ -159,6 +134,7 @@ class DocumentProcessor:
         if file_ext == 'txt':
             text = content.decode('utf-8')
             paragraphs = re.split(r'\n\s*\n|\r\n\s*\r\n', text)
             for i, para in enumerate(paragraphs):
                 if para.strip():
@@ -170,7 +146,7 @@ class DocumentProcessor:
                     ))
         elif file_ext == 'pdf':
-            temp_path = os.path.join(temp_dir_manager.temp_dir, file.filename)
             with open(temp_path, 'wb') as f:
                 f.write(content)
@@ -196,7 +172,7 @@ class DocumentProcessor:
             os.remove(temp_path)
         elif file_ext == 'docx':
-            temp_path = os.path.join(temp_dir_manager.temp_dir, file.filename)
             with open(temp_path, 'wb') as f:
                 f.write(content)
@@ -427,6 +403,39 @@ class TranslationManager:
                 replacements[placeholder] = target_term
                 text_to_translate = text_to_translate.replace(source_term, placeholder)
         async with aiohttp.ClientSession() as session:
             try:
                 # 使用环境变量获取Deepl翻译API的URL
@@ -550,11 +559,6 @@ async def export_document(request: ExportRequest):
 async def read_root():
     return FileResponse('static/index.html')
-# 程序退出时清理临时目录
-@app.on_event("shutdown")
-async def shutdown_event():
-    temp_dir_manager.cleanup()
 def open_browser():
     webbrowser.open('http://localhost:7860')

+# main.py
 from fastapi import FastAPI, File, UploadFile, HTTPException, Form, Body
 from fastapi.staticfiles import StaticFiles
+from fastapi.responses import StreamingResponse, FileResponse
 from fastapi.middleware.cors import CORSMiddleware
 import uvicorn
 import asyncio
 import webbrowser
 from typing import List, Dict, Optional
 import os
 from pathlib import Path
 import pypdf
 from docx import Document
 import re
 from io import BytesIO
+# 创建必要的目录
+os.makedirs("static", exist_ok=True)
+os.makedirs("temp", exist_ok=True)
+os.makedirs("translation_memory", exist_ok=True)
 app = FastAPI()
 # 配置CORS
     allow_headers=["*"],
 )
 # 挂载静态文件目录
+app.mount("/static", StaticFiles(directory="static"), name="static")
 class DocumentSegment(BaseModel):
     text: str
 class TranslationMemory:
     def __init__(self):
+        self.memory_file = "translation_memory/memory.json"
+        self.load_memory()
+    def load_memory(self):
+        if os.path.exists(self.memory_file):
+            with open(self.memory_file, 'r', encoding='utf-8') as f:
+                self.memory = json.load(f)
+        else:
+            self.memory = {}
+            self.save_memory()
     def save_memory(self):
         with open(self.memory_file, 'w', encoding='utf-8') as f:
 class TerminologyManager:
     def __init__(self):
+        self.terminology_file = "translation_memory/terminology.json"
+        self.load_terminology()
+    def load_terminology(self):
+        if os.path.exists(self.terminology_file):
+            with open(self.terminology_file, 'r', encoding='utf-8') as f:
+                self.terminology = json.load(f)
+        else:
+            self.terminology = {}
+            self.save_terminology()
     def save_terminology(self):
         with open(self.terminology_file, 'w', encoding='utf-8') as f:
         if file_ext == 'txt':
             text = content.decode('utf-8')
+            # 优化分段逻辑，支持更多分隔符
             paragraphs = re.split(r'\n\s*\n|\r\n\s*\r\n', text)
             for i, para in enumerate(paragraphs):
                 if para.strip():
                     ))
         elif file_ext == 'pdf':
+            temp_path = f"temp/{file.filename}"
             with open(temp_path, 'wb') as f:
                 f.write(content)
             os.remove(temp_path)
         elif file_ext == 'docx':
+            temp_path = f"temp/{file.filename}"
             with open(temp_path, 'wb') as f:
                 f.write(content)
                 replacements[placeholder] = target_term
                 text_to_translate = text_to_translate.replace(source_term, placeholder)
+        # async with aiohttp.ClientSession() as session:
+        #     try:
+        #         # DeepL翻译
+        #         async with session.post(
+        #             'https://api.deeplx.org/..../translate',
+        #             json={
+        #                 "text": text_to_translate,
+        #                 "source_lang": source_lang,
+        #                 "target_lang": target_lang
+        #             }
+        #         ) as response:
+        #             result = await response.json()
+        #             if result.get('code') == 200:
+        #                 translated_text = result['data']
+        #                 # 恢复术语替换
+        #                 for placeholder, term in replacements.items():
+        #                     translated_text = translated_text.replace(placeholder, term)
+        #                 # 保存到翻译记忆
+        #                 self.memory.add_translation(text, translated_text, source_lang, target_lang)
+        #                 return {
+        #                     'translated': translated_text,
+        #                     'alternatives': result.get('alternatives', []),
+        #                     'from_memory': False,
+        #                     'confidence': 0.8 if replacements else 0.7
+        #                 }
+        #             else:
+        #                 raise HTTPException(status_code=500, detail="Translation API error")
+        #     except Exception as e:
+        #         raise HTTPException(status_code=500, detail=str(e))
         async with aiohttp.ClientSession() as session:
             try:
                 # 使用环境变量获取Deepl翻译API的URL
 async def read_root():
     return FileResponse('static/index.html')
 def open_browser():
     webbrowser.open('http://localhost:7860')