Create app.py

app.py

@@ -0,0 +1,112 @@
+import streamlit as st
+import json
+import os
+from reportlab.lib.pagesizes import letter
+from reportlab.pdfgen import canvas
+from reportlab.lib import colors
+from io import BytesIO
+ 
+PAGE_WIDTH, PAGE_HEIGHT = letter
+ 
+def load_json_from_file(file_path):
+    with open(file_path, "r", encoding="utf-8") as f:
+        return json.load(f)
+ 
+def generate_pdf(data):
+    # Create a BytesIO stream to hold the PDF in memory
+    pdf_output = BytesIO()
+ 
+    c = canvas.Canvas(pdf_output, pagesize=letter)
+    page_created = False
+ 
+    for ad_size, ad_data in data.items():
+        for item in ad_data:
+            for line in item.get("text_lines", []):
+                page_created = True
+ 
+                text = line.get("text", "")
+                bbox = line.get("bbox", None)
+ 
+                if not bbox:
+                    continue
+ 
+                x1, y1, x2, y2 = bbox
+ 
+                pdf_x = x1
+                pdf_y = PAGE_HEIGHT - y2
+ 
+                box_height = y2 - y1
+                font_size = max(6, min(18, box_height * 0.8))
+ 
+                # Debug box
+                c.setStrokeColor(colors.lightgrey)
+                c.rect(
+                    x1,
+                    PAGE_HEIGHT - y2,
+                    x2 - x1,
+                    box_height,
+                    fill=0
+                )
+ 
+                if "<b>" in text:
+                    c.setFont("Helvetica-Bold", font_size)
+                    text = text.replace("<b>", "").replace("</b>", "")
+                else:
+                    c.setFont("Helvetica", font_size)
+ 
+                c.setFillColor(colors.black)
+                c.drawString(
+                    pdf_x,
+                    pdf_y + (box_height - font_size),
+                    text
+                )
+ 
+        if page_created:
+            c.showPage()
+ 
+    # 🔥 GUARANTEE at least one page
+    if not page_created:
+        c.setFont("Helvetica", 12)
+        c.drawString(50, 750, "No OCR text found")
+        c.showPage()
+ 
+    c.save()
+ 
+    # Move the pointer to the beginning of the stream for download
+    pdf_output.seek(0)
+    return pdf_output
+ 
+# Streamlit UI to upload JSON and download PDF
+def main():
+    st.title("JSON to PDF Converter")
+ 
+    st.write(
+        """
+        Upload a `results.json` file, and this app will generate a PDF
+        based on the text and bounding box data in the JSON.
+        """
+    )
+ 
+    # File uploader
+    uploaded_file = st.file_uploader("Choose a JSON file", type=["json"])
+ 
+    if uploaded_file is not None:
+        try:
+            data = json.load(uploaded_file)
+            st.success("File uploaded successfully!")
+ 
+            # Generate PDF from the uploaded data
+            pdf_file = generate_pdf(data)
+ 
+            # Provide a download link
+            st.download_button(
+                label="Download PDF",
+                data=pdf_file,
+                file_name="output_fixed.pdf",
+                mime="application/pdf"
+            )
+        except Exception as e:
+            st.error(f"Error processing the file: {e}")
+ 
+if __name__ == "__main__":
+    main()