Update app.py

app.py

@@ -8,8 +8,7 @@ import requests
 import io
 from PIL import Image
 from pptx import Presentation
-from pptx.util import Inches
-import tempfile
+from pptx.util import Inches, Pt
 
 API_URL = "https://api-inference.huggingface.co/models/runwayml/stable-diffusion-v1-5"
 headers = {"Authorization": "Bearer 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"}
@@ -20,8 +19,20 @@ def query(payload):
     return response.content
 
 def add_line_breaks_to_summary(summary_text, line_length):
-    # Split the summary text into lines with a maximum line length
-    lines = [summary_text[i:i + line_length] for i in range(0, len(summary_text), line_length)]
+    # Split the summary text into lines without breaking words
+    lines = []
+    words = summary_text.split()
+    current_line = ""
+    for word in words:
+        if len(current_line) + len(word) + 1 <= line_length:  # Include space between words
+            if current_line:
+                current_line += " "
+            current_line += word
+        else:
+            lines.append(current_line)
+            current_line = word
+    if current_line:
+        lines.append(current_line)
     return "\n".join(lines)
 
 def extract_paragraphs_by_vertical_spacing(pdf_data, spacing_threshold=10):
@@ -80,7 +91,7 @@ if uploaded_file is not None:
 
     for paragraph in paragraphs:
         summary = summarizer(paragraph, max_length=(len(paragraph) / 2), min_length=10, do_sample=False)
-        summary_text = add_line_breaks_to_summary(summary[0]['summary_text'], 70)
+        summary_text = add_line_breaks_to_summary(summary[0]['summary_text'], 40)
 
         # Generate and save the image to a temporary file
         image_bytes = query({
@@ -93,23 +104,25 @@ if uploaded_file is not None:
         # Create a slide
         slide = prs.slides.add_slide(prs.slide_layouts[5])
 
+        # Add the image to the slide at the bottom (resized to 500x500 pixels)
+        left = Inches(1)
+        top = Inches(2)
+        width = height = Inches(5)
+        pic = slide.shapes.add_picture(temp_img_path, left, top, width, height)
+
         # Add the paragraph to the slide at the top
-        left = top = Inches(1)
-        width = Inches(5)
+        left = Inches(1)
+        top = Inches(1)
+        width = Inches(8)  # Adjust the width as needed
         height = Inches(1.5)
         txBox = slide.shapes.add_textbox(left, top, width, height)
         tf = txBox.text_frame
         p = tf.add_paragraph()
         p.text = f"Paragraph {i}:"
-        p.space_after = Inches(0.1)
+        p.space_after = Pt(8)  # Adjust the spacing as needed
         p = tf.add_paragraph()
         p.text = summary_text
-
-        # Add the image to the slide at the bottom
-        left = Inches(1)
-        top = Inches(2)
-        width = height = Inches(5)
-        pic = slide.shapes.add_picture(temp_img_path, left, top, width, height)
+        p.space_after = Pt(0)  # Adjust the spacing as needed
 
         i += 1