fix: some bug

README.md

@@ -19,7 +19,7 @@ Deploy on Vercel and Railway
 - 🤖 Ask a question with your docs
 - 📝 Summarize docs
 - 🖍️ Highlight source
-- 📤 Upload docs (.pdf,.md)
+- 📤 Upload docs .pdf,.md(best support)
 - 💾 Data saved locally
 - 💰 Token usage tracker
 - 🐳 Dockerize

client/src/components/chatWindow/index.tsx

@@ -11,13 +11,15 @@ import Message from './Message';
 
 interface ChatWindowProps {
   fileName: string;
+  fullFileName: string;
   className?: string;
   onReplyComplete: (data: any) => void;
   onSourceClick: (data: any) => void;
 }
 
 const ChatWindow: FC<ChatWindowProps> = ({
   fileName,
+  fullFileName,
   className,
   onReplyComplete,
   onSourceClick
@@ -68,7 +70,7 @@ const ChatWindow: FC<ChatWindowProps> = ({
 
       if (summarize) {
         res = await fetchRequest('/api/summarize', {
-          index: fileName,
+          file: fullFileName,
           openAiKey
         });
       } else {

client/src/pages/Home.tsx

@@ -108,6 +108,7 @@ const Home = () => {
       </div>
 
       <ChatWindow
+        fullFileName={currentFile?.fullName || ''}
         fileName={currentFile?.name.split(currentFile.ext)[0] || ''}
         className="flex flex-col"
         onReplyComplete={handleHighLight}

server/app.py

@@ -67,14 +67,14 @@ def handle_error(error):
 
 @app.route("/api/summarize", methods=["GET"])
 def summarize_index():
-    index_name = request.args.get("index")
+    file = request.args.get("file")
     open_ai_key = request.args.get("openAiKey")
     if open_ai_key:
         os.environ["OPENAI_API_KEY"] = open_ai_key
 
     UnstructuredReader = download_loader("UnstructuredReader")
     loader = UnstructuredReader()
-    documents = loader.load_data(file=Path(f"./{staticPath}/file/{index_name}"))
+    documents = loader.load_data(file=Path(f"./{staticPath}/file/{file}"))
     index = GPTListIndex.from_documents(documents)
 
     # predictor cost

server/create_index.py

@@ -13,7 +13,7 @@ def create_index(filepath, filename) -> int:
 
     if ext == ".pdf":
         loader = CJKPDFReader()
-        documents = loader.load_data(file=filepath)
+        documents = loader.load_data(filepath=filepath, filename=filename)
     elif ext == ".md":
         with open(filepath, "r", encoding="utf-8") as f:
             file_text = f.read()
@@ -23,7 +23,7 @@ def create_index(filepath, filename) -> int:
         # TODO: 利用 langchain splitter重写 https://python.langchain.com/en/latest/modules/indexes/text_splitters/examples/markdown.html
         # 直接将markdown分段再分别转化成html，最后将所有html拼接起来并加上chunk_id
         loader = CustomReader()
-        documents = loader.load_data(html=html, file=filepath)
+        documents = loader.load_data(html=html, filename=name)
     elif ext == ".html":
         # TODO:
         pass

server/custom_loader.py

@@ -50,7 +50,7 @@ def __init__(self, *args: Any, **kwargs: Any) -> None:
         """Init params."""
         super().__init__(*args, **kwargs)
 
-    def load_data(self, html, file) -> List[Document]:
+    def load_data(self, html, filename) -> List[Document]:
         soup = BeautifulSoup(html, "html.parser")
         current_chunk_text = ""
         current_chunk_id = 1
@@ -109,7 +109,7 @@ def load_data(self, html, file) -> List[Document]:
             current_chunk_id += 1
 
         # 保存修改后的HTML文件
-        with open(file, "w", encoding="utf-8") as f:
+        with open(f"{staticPath}/file/{filename}.html", "w", encoding="utf-8") as f:
             f.write(str(soup))
 
         return document_list

server/pdf_loader.py

@@ -1,5 +1,6 @@
 """Read PDF files."""
 
+import shutil
 from pathlib import Path
 from typing import Any, List
 
@@ -9,6 +10,8 @@
 
 # https://github.com/emptycrown/llama-hub/blob/main/loader_hub/file/cjk_pdf/base.py
 
+staticPath = "static"
+
 
 class CJKPDFReader(BaseReader):
     """CJK PDF reader.
@@ -20,7 +23,7 @@ def __init__(self, *args: Any, **kwargs: Any) -> None:
         """Init params."""
         super().__init__(*args, **kwargs)
 
-    def load_data(self, file: Path) -> List[Document]:
+    def load_data(self, filepath: Path, filename) -> List[Document]:
         """Parse file."""
 
         # Import pdfminer
@@ -42,7 +45,7 @@ def load_data(self, file: Path) -> List[Document]:
         # Create a PDF interpreter
         interpreter = PDFPageInterpreter(rsrcmgr, device)
         # Open the PDF file
-        fp = open(file, "rb")
+        fp = open(filepath, "rb")
         # Create a list to store the text of each page
         document_list = []
         # Extract text from each page
@@ -66,4 +69,6 @@ def load_data(self, file: Path) -> List[Document]:
         # Close the device
         device.close()
 
+        shutil.copy2(filepath, f"{staticPath}/file/{filename}")
+
         return document_list

server/static/file/AA-README.html

@@ -0,0 +1,149 @@
+<h1 data-chunk_id="chunk-1">DocsMind</h1>
+<p data-chunk_id="chunk-1">DocsMind is an open-source project that allows you to chat with your docs.</p>
+<p data-chunk_id="chunk-1"><img alt="Stack" src="https://skillicons.dev/icons?i=vite,react,ts,tailwind,flask"/></p>
+<h2 data-chunk_id="chunk-2">Demo</h2>
+<p data-chunk_id="chunk-2"><a href="https://docs-mind.alanwang.site/">Demo Site</a></p>
+<p data-chunk_id="chunk-2">Deploy on Vercel and Railway</p>
+<blockquote data-chunk_id="chunk-2">
+<p><strong>Warning</strong></p>
+<p>Due to the free plan of Railway only providing 500 hours per month, the Demo on the 21st day of each month will not be available. Please clone it locally for use at that time.</p>
+</blockquote>
+<h2 data-chunk_id="chunk-3">Features</h2>
+<ul data-chunk_id="chunk-3">
+<li>🤖 Ask a question with your docs</li>
+<li>📝 Summarize docs</li>
+<li>🖍️ Highlight source</li>
+<li>📤 Upload docs .pdf,.md(best support)</li>
+<li>💾 Data saved locally</li>
+<li>💰 Token usage tracker</li>
+<li>🐳 Dockerize</li>
+</ul>
+<h2 data-chunk_id="chunk-4">Future Development</h2>
+<ul data-chunk_id="chunk-4">
+<li>[ ] Chat mode</li>
+<li>[ ] Dark mode</li>
+<li>[ ] Improve the UI (Skeleton,Loading)</li>
+<li>[ ] / command (/fetch /summarize)</li>
+<li>[ ] Reduce the size of the server image.</li>
+<li>[ ] Support for more docs formats: txt...</li>
+<li>[ ] Download docs from the internet</li>
+<li>[ ] Markdown-formatted message</li>
+<li>[ ] i18n</li>
+<li>[ ] Desktop application</li>
+</ul>
+<p data-chunk_id="chunk-4">If you find this project helpful, please consider giving it a star 🌟</p>
+<h2 data-chunk_id="chunk-5">Environment Variables</h2>
+<table data-chunk_id="chunk-5">
+<thead>
+<tr>
+<th>Name</th>
+<th>Description</th>
+<th>Optional</th>
+</tr>
+</thead>
+<tbody>
+<tr>
+<td>OPENAI_API_KEY</td>
+<td>sk-xxx</td>
+<td>✅</td>
+</tr>
+<tr>
+<td>OPENAI_PROXY</td>
+<td>will replace https://api.openai.com/v1</td>
+<td>✅</td>
+</tr>
+<tr>
+<td>VITE_SERVICES_URL</td>
+<td>backend url for frontend code</td>
+<td>✅</td>
+</tr>
+<tr>
+<td>VITE_DISABLED_UPLOAD</td>
+<td>DISABLED_UPLOAD</td>
+<td>✅</td>
+</tr>
+</tbody>
+</table>
+<h2 data-chunk_id="chunk-6">Q&amp;A</h2>
+<h3 data-chunk_id="chunk-7">How to run locally?</h3>
+<blockquote data-chunk_id="chunk-7">
+<p><strong>Warning</strong></p>
+<p>Please check if you can access OpenAI in your region, you can refer to the <a href="https://github.com/3Alan/DocsMind/issues/3#issuecomment-1511470063">issue</a> for more information.</p>
+</blockquote>
+<ol data-chunk_id="chunk-7">
+<li>Create .env(Optional)</li>
+</ol>
+<p data-chunk_id="chunk-7">Create a <code>.env</code> file and copy the contents of <code>.env.example</code> to modify it.</p>
+<ol data-chunk_id="chunk-7">
+<li>Run App</li>
+</ol>
+<div class="highlight" data-chunk_id="chunk-7"><pre><span></span><code>docker-compose<span class="w"> </span>up<span class="w"> </span>-d
+</code></pre></div>
+<p data-chunk_id="chunk-7">Please add <code>--build</code> to rebuild the image after each code update.</p>
+<div class="highlight" data-chunk_id="chunk-7"><pre><span></span><code>docker-compose<span class="w"> </span>up<span class="w"> </span>-d<span class="w"> </span>--build
+</code></pre></div>
+<p data-chunk_id="chunk-7">now you can access the app at <code>http://localhost:8081</code></p>
+<h3 data-chunk_id="chunk-8">Local Development</h3>
+<details data-chunk_id="chunk-8">
+<summary>Detail</summary>
+
+#### Create .env(Optional)
+
+Create a `.env` file and copy the contents of `.env.example` to modify it.
+
+#### Run Frontend
+
+1. Install dependencies
+
+<div class="highlight"><pre><span></span><code>yarn
+</code></pre></div>
+
+2. Run app
+
+<div class="highlight"><pre><span></span><code>yarn dev
+</code></pre></div>
+
+#### Run Backend
+
+you need a python environment
+
+1. Create virtual environment
+
+<div class="highlight"><pre><span></span><code>cd server
+python -m venv .venv
+</code></pre></div>
+
+2. Active virtual environment
+
+windows
+
+<div class="highlight"><pre><span></span><code>.venv\Scripts\activate
+</code></pre></div>
+
+mac
+
+<div class="highlight"><pre><span></span><code>. .venv/bin/activate
+</code></pre></div>
+
+3. Install dependencies
+
+<div class="highlight"><pre><span></span><code>pip install -r requirements.txt
+</code></pre></div>
+
+4. Run Services
+
+<div class="highlight"><pre><span></span><code>flask run --reload --port=8080
+</code></pre></div>
+
+#### Install pdf2htmlEX for PDF convert
+
+<div class="highlight"><pre><span></span><code>docker<span class="w"> </span>pull<span class="w"> </span>pdf2htmlex/pdf2htmlex:0.18.8.rc2-master-20200820-alpine-3.12.0-x86_64
+</code></pre></div>
+
+set alias
+
+<div class="highlight"><pre><span></span><code><span class="nb">alias</span><span class="w"> </span><span class="nv">pdf2htmlEX</span><span class="o">=</span><span class="s1">'docker run -ti --rm -v "`pwd`":/pdf -w /pdf pdf2htmlex/pdf2htmlex:0.18.8.rc2-master-20200820-alpine-3.12.0-x86_64'</span>
+</code></pre></div>
+</details>
+<h2 data-chunk_id="chunk-9">Buy me a coffee</h2>
+<p data-chunk_id="chunk-9"><img height="300" src="https://raw.githubusercontent.com/3Alan/images/master/img/%E5%BE%AE%E4%BF%A1%E6%94%AF%E4%BB%98%E5%AE%9D%E4%BA%8C%E5%90%88%E4%B8%80%E6%94%B6%E6%AC%BE%E7%A0%81.jpg"/></p>