Create app.py

app.py

@@ -0,0 +1,59 @@
+import streamlit as st
+import os
+import PIL.Image
+from pdf2image import convert_from_bytes
+from google import genai
+from google.genai import types
+
+# Retrieve the API key from the environment variable.
+API_KEY = os.getenv("API_KEY")
+if API_KEY is None:
+    st.error("API Key not found. Please set the API_KEY environment variable.")
+    st.stop()
+
+# Initialize the GenAI client.
+client = genai.Client(api_key=API_KEY)
+
+def extract_text_from_image(img):
+    """
+    Extracts text from a PIL image using the Google GenAI API.
+    """
+    response = client.models.generate_content(
+        model="gemini-2.0-flash",
+        contents=["Extract the text from the image. Do not write anything except the extracted content", img])
+    return response.text
+
+st.title("PDF/Image Text Extraction using Google GenAI")
+
+# File uploader (accepts PDF and common image formats).
+uploaded_file = st.file_uploader("Upload a PDF or image file", type=["pdf", "png", "jpg", "jpeg", "webp"])
+
+if uploaded_file is not None:
+    output_text = ""
+    st.write("**Uploaded File:**", uploaded_file.name)
+    
+    # Process PDF files.
+    if uploaded_file.name.lower().endswith(".pdf"):
+        st.info("Processing PDF file...")
+        with st.spinner("Converting PDF pages to images..."):
+            images = convert_from_bytes(uploaded_file.read(), dpi=200)
+        
+        for idx, img in enumerate(images, start=1):
+            with st.spinner(f"Extracting text from page {idx}..."):
+                page_text = extract_text_from_image(img)
+                output_text += f"### Page {idx}\n\n{page_text}\n\n"
+    else:
+        # Process image files.
+        st.info("Processing image file...")
+        with st.spinner("Extracting text from image..."):
+            img = PIL.Image.open(uploaded_file)
+            output_text += extract_text_from_image(img) + "\n\n"
+    
+    st.success("Extraction complete!")
+    
+    st.markdown("### Extracted Text:")
+    st.markdown(output_text)
+    
+    # Provide a download button for the Markdown file.
+    md_bytes = output_text.encode("utf-8")
+    st.download_button(label="Download Markdown file", data=md_bytes, file_name="output.md", mime="text/markdown")