From 064f56e4b988471bc6eafbe7a4b171799726d882 Mon Sep 17 00:00:00 2001
From: zydi <zydi@obscura.ac.cn>
Date: Sun, 19 Jan 2025 06:58:40 +0000
Subject: [PATCH] update

---
 paper.html |   4 +-
 paper.py   | 106 +++++++++++++++++++++++++++--------------------------
 2 files changed, 56 insertions(+), 54 deletions(-)

diff --git a/paper.html b/paper.html
index 065ccc3..fa3aadd 100644
--- a/paper.html
+++ b/paper.html
@@ -372,7 +372,7 @@
             <div class="col-auto">
                 <div class="sidebar">
                     <!-- 文献页面的侧边栏 -->
-                    <div class="sidebar-menu hidden" id="DetailSidebar">
+                    <div class="sidebar-menu" id="DetailSidebar">
                         <div class="sidebar-item" onclick="showPaperList()">
                             <i class="bi bi-list-ul"></i>
                             <span>Paper List</span>
@@ -385,7 +385,7 @@
             <div class="col">
                 <div class="content-wrapper">
                     <!-- 文献详情页面 -->
-                    <div id="DetailPage" class="content-body hidden">
+                    <div id="DetailPage" class="content-body">
                         <div class="content-header d-flex justify-content-between align-items-center">
                             <h3 class="content-title" id="paperTitle">Paper List</h3>
                             <button class="btn btn-info" onclick="uploadPaper()" style="padding: 10px 20px;">
diff --git a/paper.py b/paper.py
index b4ffc8e..105bcd7 100644
--- a/paper.py
+++ b/paper.py
@@ -697,9 +697,9 @@ class TaskStatus:
     COMPLETED = "completed"
     FAILED = "failed"
 
-@app.post("/paper/{paper_id}/qa")
+@app.post("/paper/{file_hash}/qa")
 async def ask_reference_question(
-    paper_id: str,
+    file_hash: str,
     question: QuestionModel
 ):
     """Ask questions about the paper (async)"""
@@ -708,7 +708,7 @@ async def ask_reference_question(
     
     try:
         # Verify paper exists
-        paper = await db.papers.find_one({"_id": ObjectId(paper_id)})
+        paper = await db.papers.find_one({"file_hash": file_hash})
         if not paper:
             raise HTTPException(status_code=404, detail="Paper not found")
         
@@ -718,7 +718,7 @@ async def ask_reference_question(
         # Create background task
         asyncio.create_task(paper_question(
             task_id=task_id,
-            paper_id=paper_id,
+            file_hash=file_hash,
             question=question.question,
             paper=paper
         ))
@@ -738,7 +738,7 @@ async def ask_reference_question(
         except Exception as e:
             print(f"Error closing Redis connection: {e}")
 
-async def paper_question(task_id: str, paper_id: str, question: str, paper: dict):
+async def paper_question(task_id: str, file_hash: str, question: str, paper: dict):
     """Process paper Q&A background task"""
     redis = await get_redis()
     
@@ -749,35 +749,37 @@ async def paper_question(task_id: str, paper_id: str, question: str, paper: dict
             f"task:{task_id}",
             mapping={
                 "status": TaskStatus.PROCESSING,
-                "paper_id": paper_id,
+                "file_hash": file_hash,
                 "question": question
             }
         )
         
         # Get analysis report from Redis
         await redis.select(190)
-        report_key = f"paper_report:{paper_id}"
+        report_key = f"paper_report:{file_hash}"
         report_data = await redis.get(report_key)
         
         if not report_data:
             raise Exception("Paper analysis report does not exist, please analyze first")
             
-        # Read PDF file content
-        file_path = paper.get("paper_link")
-        if not file_path or not os.path.exists(file_path):
-            raise Exception("Paper file does not exist")
-            
-        # Read PDF content
-        pdf_content = ""
-        with open(file_path, 'rb') as file:
-            pdf_reader = PyPDF2.PdfReader(file)
-            for page in pdf_reader.pages:
-                content = page.extract_text()
-                pdf_content += content
-        
-        # Limit content length
-        MAX_CHARS = 180000
-        pdf_content = pdf_content[:MAX_CHARS]
+        # Get PDF content from file storage service
+        async with aiohttp.ClientSession() as session:
+            async with session.get(f'https://files.aiot.ml/pdf/{file_hash}') as response:
+                if response.status != 200:
+                    raise Exception(f"Failed to get PDF content: HTTP {response.status}")
+                    
+                pdf_content = await response.json()
+                if not pdf_content.get('content'):
+                    raise Exception("No PDF content returned")
+                
+                content = pdf_content['content']
+                # If content is a list, join it into a single string
+                if isinstance(content, list):
+                    content = '\n'.join(content)
+                
+                # Limit content length
+                MAX_CHARS = 180000
+                content = content[:MAX_CHARS]
         
         # Build system prompt and user prompt
         system_prompt = """
@@ -788,7 +790,7 @@ async def paper_question(task_id: str, paper_id: str, question: str, paper: dict
         
         # Build context
         context = {
-            "Paper Content": pdf_content,
+            "Paper Content": content,
             "Analysis Report": json.loads(report_data)
         }
         
@@ -806,7 +808,7 @@ async def paper_question(task_id: str, paper_id: str, question: str, paper: dict
         
         # Save conversation history to Redis db191
         await redis.select(191)
-        chat_history_key = f"chat_history:{paper_id}"
+        chat_history_key = f"chat_history:{file_hash}"
         
         # Get existing history
         existing_history = await redis.get(chat_history_key)
@@ -850,6 +852,33 @@ async def paper_question(task_id: str, paper_id: str, question: str, paper: dict
         except Exception as e:
             print(f"Error closing Redis connection: {e}")
 
+@app.get("/paper/{file_hash}/qa/history")
+async def get_paper_qa_history(
+    file_hash: str
+):
+    """Get paper Q&A history"""
+    redis = await get_redis()
+    
+    try:
+        # Get conversation history from Redis db191
+        await redis.select(191)
+        chat_history_key = f"chat_history:{file_hash}"
+        
+        history_data = await redis.get(chat_history_key)
+        if not history_data:
+            return []
+            
+        return json.loads(history_data)
+        
+    except Exception as e:
+        print(f"Error getting QA history: {str(e)}")
+        raise HTTPException(status_code=500, detail=str(e))
+    finally:
+        try:
+            await redis.aclose()
+        except Exception as e:
+            print(f"Error closing Redis connection: {e}")
+
 @app.get("/paper/task/{task_id}")
 async def get_task_status(task_id: str):
     """Get task status and result"""
@@ -890,33 +919,6 @@ async def get_task_status(task_id: str):
         except Exception as e:
             print(f"Error closing Redis connection: {e}")
 
-@app.get("/paper/{paper_id}/qa/history")
-async def get_paper_qa_history(
-    paper_id: str
-):
-    """Get paper Q&A history"""
-    redis = await get_redis()
-    
-    try:
-        # Get conversation history from Redis db191
-        await redis.select(191)
-        chat_history_key = f"chat_history:{paper_id}"
-        
-        history_data = await redis.get(chat_history_key)
-        if not history_data:
-            return []
-            
-        return json.loads(history_data)
-        
-    except Exception as e:
-        print(f"Error getting QA history: {str(e)}")
-        raise HTTPException(status_code=500, detail=str(e))
-    finally:
-        try:
-            await redis.aclose()
-        except Exception as e:
-            print(f"Error closing Redis connection: {e}")
-
 # 添加新的路由，通过哈希值获取报告
 @app.get("/paper/check/{file_hash}")
 async def check_report(