Update app.py via AI Editor

app.py

@@ -15,31 +15,26 @@ from PIL import Image
 import google.generativeai as genai
 import time
 
-# Set up logging
+import docx
+import PyPDF2
+
 logging.basicConfig(level=logging.INFO, format='%(asctime)s - %(levelname)s - %(message)s')
 logger = logging.getLogger(__name__)
 
-# Stability AI API key
 STABILITY_API_KEY = os.getenv('STABILITY_API_KEY')
-
-# Gemini API key
 GOOGLE_API_KEY = os.getenv('GOOGLE_API_KEY')
 genai.configure(api_key=GOOGLE_API_KEY)
-
-# Initialize Gemini model
 model = genai.GenerativeModel('gemini-2.5-pro-preview-03-25')
 
 app = dash.Dash(__name__, external_stylesheets=[dbc.themes.BOOTSTRAP, 'https://fonts.googleapis.com/css2?family=Roboto:wght@300;400;700&display=swap'])
 app.config.suppress_callback_exceptions = True
 
-# Global variables to store logs and generated PowerPoint
 log_messages = []
 generated_pptx = None
 
 app.layout = html.Div([
     dbc.Container([
         dbc.Row([
-            # Left Column
             dbc.Col([
                 dbc.Card([
                     dbc.CardHeader("Upload or Paste Document", className="bg-light"),
@@ -72,7 +67,6 @@ app.layout = html.Div([
                     ])
                 ], className="mb-4 shadow-sm")
             ], md=6),
-            # Right Column
             dbc.Col([
                 dbc.Card([
                     dbc.CardHeader("Edit Slide Content", className="bg-light"),
@@ -91,13 +85,56 @@ app.layout = html.Div([
         ])
     ], fluid=True, className="mt-4"),
     dcc.Download(id="download-pptx"),
-    dcc.Interval(id='interval-component', interval=1*1000, n_intervals=0)  # Updates every 1 second
+    dcc.Interval(id='interval-component', interval=1*1000, n_intervals=0)
 ])
 
 def add_log(message):
     global log_messages
     logger.info(message)
-    log_messages.insert(0, f"{time.strftime('%H:%M:%S')} - {message}")  # Insert at the beginning
+    log_messages.insert(0, f"{time.strftime('%H:%M:%S')} - {message}")
+
+def extract_text_from_docx(docx_bytes):
+    doc = docx.Document(io.BytesIO(docx_bytes))
+    fullText = []
+    for para in doc.paragraphs:
+        fullText.append(para.text)
+    return '\n'.join(fullText)
+
+def extract_text_from_pdf(pdf_bytes):
+    pdf_reader = PyPDF2.PdfReader(io.BytesIO(pdf_bytes))
+    text = []
+    for page in pdf_reader.pages:
+        page_text = page.extract_text()
+        if page_text:
+            text.append(page_text)
+    return '\n'.join(text)
+
+def process_uploaded_file(filename, file_contents):
+    if not filename or not file_contents:
+        return ""
+    content_type, content_string = file_contents.split(',')
+    file_bytes = base64.b64decode(content_string)
+    filename_lower = filename.lower()
+    if filename_lower.endswith('.txt'):
+        try:
+            return file_bytes.decode('utf-8')
+        except UnicodeDecodeError:
+            try:
+                return file_bytes.decode('latin1')
+            except Exception:
+                raise Exception("Could not decode file as UTF-8 or latin1.")
+    elif filename_lower.endswith('.docx'):
+        try:
+            return extract_text_from_docx(file_bytes)
+        except Exception as e:
+            raise Exception(f"Could not extract text from DOCX: {str(e)}")
+    elif filename_lower.endswith('.pdf'):
+        try:
+            return extract_text_from_pdf(file_bytes)
+        except Exception as e:
+            raise Exception(f"Could not extract text from PDF: {str(e)}")
+    else:
+        raise Exception("Unsupported file type. Please upload a .txt, .docx, or .pdf file.")
 
 def process_document(text):
     add_log("Starting document processing...")
@@ -111,6 +148,8 @@ def process_document(text):
     There can only be one slide heading # and one sub heading ## and no more than 5 bullets with "-" per slide
     Document:
     {text}
+    ---
+    The input may be up to 1 million tokens and the output up to 64,000 tokens. Please use as much detail and slides as needed within those limits.
     """
     add_log("Sending request to Gemini model...")
     response = model.generate_content(prompt)
@@ -143,7 +182,6 @@ def generate_image(prompt):
         "output_format": "png"
     }
     files = {"none": ''}
-    
     try:
         response = requests.post(url, headers=headers, data=data, files=files)
         response.raise_for_status()
@@ -170,15 +208,11 @@ def markdown_to_pptx(md_text):
             add_log(f"Creating slide {i+1}/{len(slides)}...")
             lines = slide_content.split('\n')
             title = lines[0].strip()
-            current_slide = prs.slides.add_slide(prs.slide_layouts[6])  # Blank layout
-            
-            # Add title
+            current_slide = prs.slides.add_slide(prs.slide_layouts[6])
             title_box = current_slide.shapes.add_textbox(Inches(0.5), Inches(0.5), Inches(9), Inches(1))
             title_box.text_frame.text = title
             title_box.text_frame.paragraphs[0].font.size = Pt(32)
             title_box.text_frame.paragraphs[0].font.bold = True
-
-            # Add content (left side)
             content_box = current_slide.shapes.add_textbox(Inches(0.5), Inches(1.5), Inches(4.5), Inches(6))
             text_frame = content_box.text_frame
             text_frame.word_wrap = True
@@ -194,8 +228,6 @@ def markdown_to_pptx(md_text):
                     p.font.size = Pt(18)
                     p.level = 1
                     p.bullet = True
-
-            # Add image (right side)
             add_log(f"Adding image to slide {i+1}...")
             image_data = future.result()
             if image_data:
@@ -215,7 +247,6 @@ def markdown_to_pptx(md_text):
                 add_log(f"Failed to generate image for slide {i+1}: {title}")
                 placeholder = current_slide.shapes.add_textbox(Inches(5.5), Inches(1.5), Inches(4), Inches(6))
                 placeholder.text_frame.text = "Image generation failed"
-
     add_log("PowerPoint generation completed.")
     output = io.BytesIO()
     prs.save(output)
@@ -230,22 +261,24 @@ def update_filename(filename):
         return html.Div(f"Selected file: {filename}")
     return ""
 
-
 @app.callback(
     Output('slides-content', 'value'),
     Output('download-ppt-button', 'disabled'),
     Input('generate-button', 'n_clicks'),
+    State('upload-file', 'filename'),
     State('upload-file', 'contents'),
     State('document-text', 'value')
 )
-def generate_slides(n_clicks, file_contents, document_text):
+def generate_slides(n_clicks, filename, file_contents, document_text):
     if n_clicks is None:
         raise PreventUpdate
 
-    if file_contents:
-        content_type, content_string = file_contents.split(',')
-        decoded = base64.b64decode(content_string)
-        document_text = decoded.decode('utf-8')
+    if file_contents and filename:
+        try:
+            document_text = process_uploaded_file(filename, file_contents)
+        except Exception as e:
+            add_log(f"File error: {str(e)}")
+            return f"File error: {str(e)}", True
     elif not document_text:
         return "Please upload a file or enter text.", True
 
@@ -253,7 +286,7 @@ def generate_slides(n_clicks, file_contents, document_text):
         add_log("Starting slide generation...")
         slides_markdown = process_document(document_text)
         add_log("Slide generation completed.")
-        return slides_markdown, False  # Enable download button
+        return slides_markdown, False
     except Exception as e:
         add_log(f"An error occurred during slide generation: {str(e)}")
         return f"An error occurred: {str(e)}", True
@@ -269,12 +302,11 @@ def generate_powerpoint(n_clicks, slides_content):
     global generated_pptx
     if n_clicks is None or not slides_content:
         raise PreventUpdate
-
     try:
         add_log("Starting PowerPoint generation...")
         generated_pptx = markdown_to_pptx(slides_content)
         add_log("PowerPoint generation completed.")
-        return '\n'.join(log_messages[:100]), False  # Enable download button
+        return '\n'.join(log_messages[:100]), False
     except Exception as e:
         error_message = f"An error occurred during PowerPoint generation: {str(e)}"
         add_log(error_message)
@@ -288,7 +320,6 @@ def generate_powerpoint(n_clicks, slides_content):
 def download_powerpoint(n_clicks):
     if n_clicks is None or generated_pptx is None:
         raise PreventUpdate
-
     return dcc.send_bytes(generated_pptx, "presentation.pptx")
 
 @app.callback(