Update app.py

app.py

@@ -1,4 +1,3 @@
-import glob
 import urllib.request
 import fitz
 import re
@@ -12,17 +11,14 @@ from sklearn.neighbors import NearestNeighbors
 def download_pdf(url, output_path):
     urllib.request.urlretrieve(url, output_path)
 
-
 def preprocess(text):
     text = text.replace('\n', ' ')
     text = re.sub('\s+', ' ', text)
     return text
 
-
-def pdf_to_text(file_path, start_page=1, end_page=None):
-    doc = fitz.open(file_path)
+def pdf_to_text(path, start_page=1, end_page=None):
+    doc = fitz.open(path)
     total_pages = doc.page_count
-    file_name = os.path.basename(file_path)
 
     if end_page is None:
         end_page = total_pages
@@ -32,26 +28,25 @@ def pdf_to_text(file_path, start_page=1, end_page=None):
     for i in range(start_page-1, end_page):
         text = doc.load_page(i).get_text("text")
         text = preprocess(text)
-        text_list.append((file_name, text))
+        text_list.append(text)
 
     doc.close()
     return text_list
- 
 
 def text_to_chunks(texts, word_length=150, start_page=1):
-    text_toks = [(file_name, t.split(' ')) for file_name, t in texts]
+    text_toks = [t.split(' ') for t in texts]
     page_nums = []
     chunks = []
     
-    for idx, (file_name, words) in enumerate(text_toks):
+    for idx, words in enumerate(text_toks):
         for i in range(0, len(words), word_length):
             chunk = words[i:i+word_length]
             if (i+word_length) > len(words) and (len(chunk) < word_length) and (
                 len(text_toks) != (idx+1)):
-                text_toks[idx+1] = (file_name, chunk + text_toks[idx+1][1])
+                text_toks[idx+1] = chunk + text_toks[idx+1]
                 continue
             chunk = ' '.join(chunk).strip()
-            chunk = f'[{file_name}, Page no. {idx+start_page}]' + ' ' + '"' + chunk + '"'
+            chunk = f'[Page no. {idx+start_page}]' + ' ' + '"' + chunk + '"'
             chunks.append(chunk)
     return chunks
 
@@ -61,7 +56,6 @@ class SemanticSearch:
         self.use = hub.load('https://tfhub.dev/google/universal-sentence-encoder/4')
         self.fitted = False
     
-    
     def fit(self, data, batch=1000, n_neighbors=5):
         self.data = data
         self.embeddings = self.get_text_embedding(data, batch=batch)
@@ -70,7 +64,6 @@ class SemanticSearch:
         self.nn.fit(self.embeddings)
         self.fitted = True
     
-    
     def __call__(self, text, return_data=True):
         inp_emb = self.use([text])
         neighbors = self.nn.kneighbors(inp_emb, return_distance=False)[0]
@@ -80,7 +73,6 @@ class SemanticSearch:
         else:
             return neighbors
     
-    
     def get_text_embedding(self, texts, batch=1000):
         embeddings = []
         for i in range(0, len(texts), batch):
@@ -90,19 +82,22 @@ class SemanticSearch:
         embeddings = np.vstack(embeddings)
         return embeddings
 
+def load_recommender(paths, start_page=1):
+    global recommender
+    all_texts = []
 
+    for path in paths:
+        texts = pdf_to_text(path, start_page=start_page)
+        all_texts.extend([(text, path) for text in texts])
 
-def load_recommender(directory_path, start_page=1):
-    global recommender
-    
-    texts = []
-    for file_path in glob.glob(os.path.join(directory_path, '*.pdf')):
-        texts.extend(pdf_to_text(file_path, start_page=start_page))
+    chunks = []
+    for text, path in all_texts:
+        pdf_chunks = text_to_chunks([text], start_page=start_page)
+        pdf_chunks = [f'[PDF {os.path.basename(path)}, {chunk}]' for chunk in pdf_chunks]
+        chunks.extend(pdf_chunks)
 
-    chunks = text_to_chunks(texts, start_page=start_page)
     recommender.fit(chunks)
     return 'Corpus Loaded.'
-
 def generate_text(openAI_key, prompt, engine="text-davinci-003"):
     openai.api_key = openAI_key
     completions = openai.Completion.create(
@@ -124,47 +119,52 @@ def generate_answer(question, openAI_key):
         prompt += c + '\n\n'
         
     prompt += "Instructions: Compose a comprehensive reply to the query using the search results given. "\
-              "Cite each reference using [ Page Number] notation (every result has this number at the beginning). "\
+              "Cite each reference using [PDF FILE NAME, PAGE NUMBER:] notation (every result has this number at the beginning). "\
               "Citation should be done at the end of each sentence. If the search results mention multiple subjects "\
               "with the same name, create separate answers for each. Only include information found in the results and "\
               "don't add any additional information. Make sure the answer is correct and don't output false content. "\
               "If the text does not relate to the query, simply state 'Text Not Found in PDF'. Ignore outlier "\
-              "search results which has nothing to do with the question. Only answer what is asked. The "\
+              "search results which have nothing to do with the question. Only answer what is asked. The "\
               "answer should be short and concise. Answer step-by-step. \n\nQuery: {question}\nAnswer: "
     
     prompt += f"Query: {question}\nAnswer:"
     answer = generate_text(openAI_key, prompt, "text-davinci-003")
     return answer
 
-
-def question_answer(files, question, openAI_key):
+def question_answer(url, files, question, openAI_key):
     if openAI_key.strip() == '':
-        return '[ERROR]: Please enter your Open AI Key. Get your key here : https://platform.openai.com/account/api-keys'
-
-    if len(files) == 0:
-        return '[ERROR]: No PDF files uploaded.'
+        return '[ERROR]: Please enter your Open AI Key. Get your key here: https://platform.openai.com/account/api-keys'
+    if url.strip() == '' and not files:
+        return '[ERROR]: Both URL and PDF are empty. Provide at least one.'
+    
+    if url.strip() != '' and files:
+        return '[ERROR]: Both URL and PDF are provided. Please provide only one (either URL or PDF).'
 
-    directory_path = os.path.join(os.getcwd(), 'uploaded_files')
-    if not os.path.exists(directory_path):
-        os.makedirs(directory_path)
+    if url.strip() != '':
+        glob_url = url
+        download_pdf(glob_url, 'corpus.pdf')
+        load_recommender(['corpus.pdf'])
 
-    for file in files:
-        with open(os.path.join(directory_path, file.name), 'wb') as f:
-            f.write(file.read())
+    else:
+        file_paths = []
+        for file in files:
+            old_file_name = file.name
+            file_name = file.name
+            file_name = file_name[:-12] + file_name[-4:]
+            os.rename(old_file_name, file_name)
+            file_paths.append(file_name)
 
-    load_recommender(directory_path)
+        load_recommender(file_paths)
 
     if question.strip() == '':
         return '[ERROR]: Question field is empty'
 
     return generate_answer(question, openAI_key)
 
-
-
 recommender = SemanticSearch()
-
 title = 'PDF GPT'
-description = """ PDF GPT allows you to chat with your PDF file using Universal Sentence Encoder and Open AI. It gives hallucination free response than other tools as the embeddings are better than OpenAI. The returned response can even cite the page number in square brackets([]) where the information is located, adding credibility to the responses and helping to locate pertinent information quickly."""
+
+description = """ PDF GPT allows you to chat with your PDF file using Universal Sentence Encoder and Open AI. It gives hallucination free response than other tools as the embeddings are better than OpenAI. The returned response can even cite the page number and file name in square brackets([]) where the information is located, adding credibility to the responses and helping to locate pertinent information quickly."""
 
 with gr.Blocks() as demo:
 
@@ -172,18 +172,21 @@ with gr.Blocks() as demo:
     gr.Markdown(description)
 
     with gr.Row():
-
+        
         with gr.Group():
             gr.Markdown(f'<p style="text-align:center">Get your Open AI API key <a href="https://platform.openai.com/account/api-keys">here</a></p>')
-            openAI_key = gr.inputs.Textbox(label='Enter your OpenAI API key here')
-            #files = gr.inputs.File(file_count="multiple", accept=".pdf", label="Upload PDF files")
-            files = gr.File(label = "Upload .pdf documents.", file_count="multiple")
-            question = gr.inputs.Textbox(label='Enter your question here')
+            openAI_key = gr.Textbox(label='Enter your OpenAI API key here')
+            url = gr.Textbox(label='Enter PDF URL here')
+            gr.Markdown("<center><h4>OR<h4></center>")
+            files = gr.File(label='Upload your PDF/ Research Paper / Book here', file_types=['.pdf'], file_count='multiple')
+            question = gr.Textbox(label='Enter your question here')
             btn = gr.Button(value='Submit')
+            btn.style(full_width=True)
 
         with gr.Group():
-            answer = gr.outputs.Textbox(label='The answer to your question is :')
+            answer = gr.Textbox(label='The answer to your question is :')
 
-        btn.click(question_answer, inputs=[files, question, openAI_key], outputs=[answer])
+        btn.click(question_answer, inputs=[url, files, question, openAI_key], outputs=[answer])
 
 demo.launch()
+