Fixed book

src/clients.py

@@ -12,8 +12,8 @@ def embed_pdf(folder: str = 'data', name: str = 'book.pdf'):
     path = pathlib.Path(folder).joinpath(name)
     if not path.exists():
         print('Downloading book PDF.')
-        gdown.download('https://drive.google.com/file/d/1CwhFM4gInp9xV4G4sdnYE_rN0StmqQ2z/view?usp=sharing',
-                       str(path))
+        gdown.download('https://drive.google.com/file/d/1CwhFM4gInp9xV4G4sdnYE_rN0StmqQ2z/view?usp=drive_link',
+                       str(path), fuzzy=True)
     loader = PyPDFLoader(str(path))
     documents = loader.load()
     splitter = RecursiveCharacterTextSplitter(

src/complex.ipynb

@@ -223,7 +223,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 20,
+   "execution_count": 25,
    "outputs": [
     {
      "name": "stdout",
@@ -236,33 +236,32 @@
      "name": "stderr",
      "output_type": "stream",
      "text": [
-      "C:\\Users\\bsvja\\anaconda3\\envs\\pdf-rag\\Lib\\site-packages\\gdown\\parse_url.py:48: UserWarning: You specified a Google Drive link that is not the correct link to download a file. You might want to try `--fuzzy` option or the following url: https://drive.google.com/uc?id=1CwhFM4gInp9xV4G4sdnYE_rN0StmqQ2z\n",
-      "  warnings.warn(\n",
       "Downloading...\n",
-      "From: https://drive.google.com/file/d/1CwhFM4gInp9xV4G4sdnYE_rN0StmqQ2z/view?usp=sharing\n",
+      "From: https://drive.google.com/uc?id=1CwhFM4gInp9xV4G4sdnYE_rN0StmqQ2z\n",
       "To: C:\\Users\\bsvja\\PycharmProjects\\pdf-rag\\data\\book.pdf\n",
-      "87.1kB [00:00, 9.67MB/s]\n"
+      "100%|██████████| 2.37M/2.37M [00:00<00:00, 4.19MB/s]\n"
      ]
     }
    ],
    "source": [
-    "import gdown\n",
-    "pathlib.Path(r'C:\\Users\\bsvja\\PycharmProjects\\pdf-rag\\data').mkdir(exist_ok=True)\n",
-    "\n",
-    "path = pathlib.Path(r'C:\\Users\\bsvja\\PycharmProjects\\pdf-rag\\data').joinpath('book.pdf')\n",
-    "if not path.exists():\n",
-    "    print('Downloading book PDF.')\n",
-    "    gdown.download('https://drive.google.com/file/d/1CwhFM4gInp9xV4G4sdnYE_rN0StmqQ2z/view?usp=sharing',\n",
-    "                   str(path))"
+    "def embed_pdf(folder: str = 'data', name: str = 'book.pdf'):\n",
+    "    pathlib.Path(folder).mkdir(exist_ok=True)\n",
+    "    path = pathlib.Path(folder).joinpath(name)\n",
+    "    if not path.exists():\n",
+    "        print('Downloading book PDF.')\n",
+    "        gdown.download('https://drive.google.com/file/d/1CwhFM4gInp9xV4G4sdnYE_rN0StmqQ2z/view?usp=drive_link',\n",
+    "                       str(path), fuzzy=True)\n",
+    "        \n",
+    "embed_pdf('../data')"
    ],
    "metadata": {
     "collapsed": false,
     "ExecuteTime": {
-     "end_time": "2024-04-09T15:30:51.810214700Z",
-     "start_time": "2024-04-09T15:30:50.516603500Z"
+     "end_time": "2024-04-09T18:00:13.190953100Z",
+     "start_time": "2024-04-09T18:00:03.632167700Z"
     }
    },
-   "id": "f1006423bcc8b35b"
+   "id": "b0a54b5e476b46e0"
   },
   {
    "cell_type": "code",
@@ -272,7 +271,7 @@
    "metadata": {
     "collapsed": false
    },
-   "id": "b0a54b5e476b46e0"
+   "id": "5c977fcc519c1a6e"
   }
  ],
  "metadata": {