Create app.py

app.py

@@ -0,0 +1,129 @@
+import streamlit as st
+import PyPDF2
+from docx import Document
+import json
+from google import genai
+
+# -------------------------------
+# Utility Functions
+# -------------------------------
+
+def extract_text_from_pdf(file):
+    """
+    Extracts textual content from a PDF document.
+    """
+    pdf_reader = PyPDF2.PdfReader(file)
+    text = ""
+    for page in pdf_reader.pages:
+        page_text = page.extract_text()
+        if page_text:
+            text += page_text + "\n"
+    return text
+
+def extract_text_from_docx(file):
+    """
+    Extracts textual content from a DOCX document.
+    """
+    document = Document(file)
+    text = ""
+    for para in document.paragraphs:
+        text += para.text + "\n"
+    return text
+
+def call_gemini_api(document_content):
+    """
+    Calls the Google GenAI Gemini API (gemini-2.0-flash) with a prompt to analyze and summarize
+    the legal document, extracting key points, highlighting obligations/rights, and simplifying
+    complex legal terms.
+    
+    The prompt instructs the model to return the output in JSON format with three keys:
+      - summary: A concise summary of the document.
+      - highlights: Key obligations, rights, and clauses.
+      - glossary: Simplified explanations of complex legal terms.
+    """
+    # Initialize the Gemini client using your API key stored in Streamlit secrets.
+    api_key = st.secrets["GEMINI_API_KEY"]
+    client = genai.Client(api_key=api_key)
+    
+    # Construct the prompt with clear instructions
+    prompt = (
+        f"Analyze the following legal document:\n\n"
+        f"{document_content}\n\n"
+        "Instructions:\n"
+        "1. Summarize the key points of the document.\n"
+        "2. Highlight obligations, rights, and critical clauses.\n"
+        "3. Provide simplified explanations of complex legal terms.\n"
+        "Output the results as a valid JSON object with the following keys: "
+        "'summary', 'highlights', 'glossary'."
+    )
+    
+    # Call the Gemini API using the google.genai client
+    response = client.models.generate_content(
+        model="gemini-2.0-flash",
+        contents=prompt,
+    )
+    
+    # Try parsing the output JSON; if parsing fails, return the text as the summary.
+    try:
+        result = json.loads(response.text)
+    except Exception as e:
+        st.error("Failed to parse Gemini API response as JSON. Returning raw text instead.")
+        result = {"summary": response.text, "highlights": "N/A", "glossary": "N/A"}
+    
+    return result
+
+# -------------------------------
+# Main Application
+# -------------------------------
+
+def main():
+    st.title("Legal Document Summarizer")
+    st.write("Upload a legal document (PDF or DOCX) to receive a concise summary, key highlights, and a glossary of complex legal terms.")
+
+    uploaded_file = st.file_uploader("Upload Legal Document", type=["pdf", "docx"])
+
+    if uploaded_file is not None:
+        # Display file details
+        file_details = {
+            "Filename": uploaded_file.name,
+            "File Type": uploaded_file.type,
+            "File Size (bytes)": uploaded_file.size
+        }
+        st.write("**Uploaded File Details**", file_details)
+        
+        # Extract text from the document based on the file type
+        document_text = ""
+        if uploaded_file.type == "application/pdf":
+            document_text = extract_text_from_pdf(uploaded_file)
+        elif uploaded_file.type == "application/vnd.openxmlformats-officedocument.wordprocessingml.document":
+            document_text = extract_text_from_docx(uploaded_file)
+        else:
+            st.error("Unsupported file type.")
+
+        if document_text.strip():
+            st.subheader("Extracted Document Text")
+            st.text_area("Document Text", document_text, height=300)
+            
+            if st.button("Summarize Document"):
+                with st.spinner("Analyzing document via Gemini API..."):
+                    result = call_gemini_api(document_text)
+                    if result:
+                        summary = result.get("summary", "No summary provided by the API.")
+                        highlights = result.get("highlights", "No highlights provided by the API.")
+                        glossary = result.get("glossary", "No glossary provided by the API.")
+                        
+                        st.subheader("Document Summary")
+                        st.write(summary)
+                        
+                        st.subheader("Highlights (Obligations, Rights, Critical Clauses)")
+                        st.write(highlights)
+                        
+                        st.subheader("Glossary (Simplified Legal Terms)")
+                        st.write(glossary)
+                    else:
+                        st.error("Failed to retrieve a valid response from the Gemini API.")
+        else:
+            st.error("No text could be extracted from the document.")
+
+if __name__ == "__main__":
+    main()