Update app.py

app.py

@@ -1,28 +1,17 @@
 import streamlit as st
 from pdfminer.high_level import extract_text
+import smtplib
 from email.message import EmailMessage
 from email_validator import validate_email, EmailNotValidError
-import smtplib
+import spacy
 from collections import Counter
 import heapq
 from fpdf import FPDF
+import pandas as pd
 import matplotlib.pyplot as plt
-import spacy
 import requests
-import subprocess
-import sys
-
-# Ensure spaCy is installed and 'en_core_web_sm' is loaded
-try:
-    import spacy
-except ImportError:
-    subprocess.check_call([sys.executable, "-m", "pip", "install", "spacy"])
-
-try:
-    spacy.load("en_core_web_sm")
-except OSError:
-    subprocess.check_call([sys.executable, "-m", "spacy", "download", "en_core_web_sm"])
 
+# Load spaCy model
 nlp = spacy.load("en_core_web_sm")
 
 # Predefined risk-related words
@@ -32,41 +21,40 @@ RISK_WORDS = [
 ]
 
 HEADERS = {"User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64)"}
-SENDER_EMAIL = "your_email@gmail.com"  # Replace with your email
-SENDER_PASSWORD = "your_app_password"  # Replace with your app password
+SENDER_EMAIL = "shreedeepthi2005@gmail.com"
+SENDER_PASSWORD = "qntm oher jqfz oflt"
 
-# Extract text from PDF
 def extract_text_from_pdf(uploaded_file):
     return extract_text(uploaded_file)
 
-# Extract key clauses
 def extract_key_clauses(text):
     doc = nlp(text)
     sentences = list(doc.sents)
     clauses = [str(sentence).strip() for sentence in sentences if len(sentence) > 10]
-    return clauses[:10]
+    return clauses[:10]  # Return top 10 clauses for simplicity
 
-# Summarize text
 def summarize_text(text, num_sentences=5):
     doc = nlp(text)
     sentences = list(doc.sents)
     word_frequencies = Counter([token.text.lower() for token in doc if token.is_alpha and not token.is_stop])
     sentence_scores = {}
     for sent in sentences:
-        sentence_score = sum(word_frequencies.get(word.text.lower(), 0) for word in sent)
+        sentence_score = 0
+        for word in sent:
+            if word.text.lower() in word_frequencies:
+                sentence_score += word_frequencies[word.text.lower()]
         sentence_scores[sent] = sentence_score
     summarized_sentences = heapq.nlargest(num_sentences, sentence_scores, key=sentence_scores.get)
     summary = ' '.join([str(sentence) for sentence in summarized_sentences])
     return summary
 
-# Detect risks
 def detect_risks(text):
     doc = nlp(text.lower())
     detected_risks = [token.text for token in doc if token.text in RISK_WORDS]
     return list(set(detected_risks))
 
-# Fetch regulatory updates
 def get_regulatory_updates():
+    # Fallback: Pre-defined updates
     predefined_updates = [
         {"title": "New Compliance Guidelines", "summary": "SEC released new guidelines for regulatory compliance."},
         {"title": "Update on Financial Risks", "summary": "New policies to mitigate risks in the financial sector."},
@@ -75,12 +63,11 @@ def get_regulatory_updates():
     try:
         response = requests.get(url, headers=HEADERS)
         response.raise_for_status()
-        updates = []  # Placeholder for scraped updates
+        updates = []  # Placeholder for parsed updates (needs a proper parsing method)
         return updates if updates else predefined_updates
     except requests.exceptions.RequestException:
         return predefined_updates
 
-# Generate PDF
 def generate_pdf(summary, clauses, risks, updates, pdf_path="Analysis_Results.pdf"):
     pdf = FPDF()
     pdf.set_auto_page_break(auto=True, margin=15)
@@ -89,11 +76,13 @@ def generate_pdf(summary, clauses, risks, updates, pdf_path="Analysis_Results.pd
 
     pdf.cell(200, 10, txt="Legal Document Analysis Results", ln=True, align="C")
 
+    # Summary
     pdf.ln(10)
     pdf.cell(200, 10, txt="Summary", ln=True, align="L")
     pdf.set_font("Arial", size=10)
     pdf.multi_cell(0, 10, summary)
 
+    # Key Clauses
     pdf.ln(10)
     pdf.set_font("Arial", size=12)
     pdf.cell(200, 10, txt="Key Clauses", ln=True, align="L")
@@ -101,12 +90,14 @@ def generate_pdf(summary, clauses, risks, updates, pdf_path="Analysis_Results.pd
     for clause in clauses:
         pdf.multi_cell(0, 10, f"- {clause}")
 
+    # Risks
     pdf.ln(10)
     pdf.set_font("Arial", size=12)
     pdf.cell(200, 10, txt="Detected Risks", ln=True, align="L")
     pdf.set_font("Arial", size=10)
     pdf.multi_cell(0, 10, ", ".join(risks))
 
+    # Regulatory Updates
     pdf.ln(10)
     pdf.set_font("Arial", size=12)
     pdf.cell(200, 10, txt="Regulatory Updates", ln=True, align="L")
@@ -117,7 +108,6 @@ def generate_pdf(summary, clauses, risks, updates, pdf_path="Analysis_Results.pd
 
     pdf.output(pdf_path)
 
-# Send email
 def send_email(pdf_path, recipient_email):
     msg = EmailMessage()
     msg["Subject"] = "Legal Document Analysis Results"
@@ -134,74 +124,82 @@ def send_email(pdf_path, recipient_email):
         server.login(SENDER_EMAIL, SENDER_PASSWORD)
         server.send_message(msg)
 
-# Plot word frequencies
-def plot_word_frequencies(text):
-    doc = nlp(text)
-    word_frequencies = Counter([token.text.lower() for token in doc if token.is_alpha and not token.is_stop])
-    top_words = word_frequencies.most_common(10)
-    words, frequencies = zip(*top_words)
-    plt.figure(figsize=(10, 5))
-    plt.bar(words, frequencies, color="skyblue")
-    plt.title("Top 10 Word Frequencies")
-    plt.xlabel("Words")
-    plt.ylabel("Frequency")
-    st.pyplot(plt)
-
-# Main function
+def visualize_key_clauses_frequency(clauses):
+    clause_counts = Counter(clauses)
+    common_clauses = clause_counts.most_common()
+    if common_clauses:
+        labels, values = zip(*common_clauses)
+        plt.figure(figsize=(10, 6))
+        plt.barh(labels, values, color='skyblue')
+        plt.xlabel('Frequency')
+        plt.title('Key Clauses Frequency')
+        st.pyplot(plt)
+    else:
+        st.write("No key clauses to visualize.")
+
 def main():
     st.title("Interactive Legal Document Analysis Dashboard")
 
+    # Sidebar options
     st.sidebar.title("Options")
     features = st.sidebar.multiselect("Select Features", 
                                        ["Data Visualization", "Summary", "Key Clauses", "Risk Detection", "Regulatory Updates"])
 
+    # File upload
     uploaded_file = st.file_uploader("Upload a legal document (PDF)", type="pdf")
     recipient_email = st.text_input("Enter your email to receive the analysis results (optional)")
 
-    if st.button("Submit"):
-        if uploaded_file is not None:
+    if uploaded_file is not None:
+        try:
             text = extract_text_from_pdf(uploaded_file)
+            st.success("Text extracted successfully!")
+        except Exception as e:
+            st.error(f"Error extracting text from PDF: {e}")
+            return
+
+        summary = ""
+        clauses, risks, updates = [], [], []
+
+        if "Summary" in features:
+            summary = summarize_text(text)
+            st.subheader("Summary")
+            st.write(summary)
+
+        if "Key Clauses" in features:
+            clauses = extract_key_clauses(text)
+            st.subheader("Key Clauses")
+            for i, clause in enumerate(clauses, 1):
+                st.write(f"{i}. {clause}")
+            # Visualize key clauses frequency
+            if "Data Visualization" in features:
+                visualize_key_clauses_frequency(clauses)
 
-            summary, clauses, risks, updates = "", [], [], []
-
-            if "Summary" in features:
-                summary = summarize_text(text)
-                st.subheader("Summary")
-                st.write(summary)
-
-            if "Key Clauses" in features:
-                clauses = extract_key_clauses(text)
-                st.subheader("Key Clauses")
-                for i, clause in enumerate(clauses, 1):
-                    st.write(f"{i}. {clause}")
-
-            if "Risk Detection" in features:
-                risks = detect_risks(text)
-                st.subheader("Detected Risks")
-                st.write(", ".join(risks) if risks else "No risks detected.")
-
-            if "Regulatory Updates" in features:
-                updates = get_regulatory_updates()
-                st.subheader("Regulatory Updates")
-                for update in updates:
-                    st.write(f"- **{update.get('title')}**: {update.get('summary')}")
+        if "Risk Detection" in features:
+            risks = detect_risks(text)
+            st.subheader("Detected Risks")
+            st.write(", ".join(risks) if risks else "No risks detected.")
 
-            if "Data Visualization" in features:
-                st.subheader("Word Frequency Visualization")
-                plot_word_frequencies(text)
+        if "Regulatory Updates" in features:
+            updates = get_regulatory_updates()
+            st.subheader("Regulatory Updates")
+            for update in updates:
+                st.write(f"- **{update.get('title')}**: {update.get('summary')}")
 
+        # Generate PDF and send via email if required
+        if st.button("Generate PDF Report"):
             pdf_path = "Analysis_Results.pdf"
             generate_pdf(summary, clauses, risks, updates, pdf_path)
+            with open(pdf_path, "rb") as file:
+                st.download_button("Download PDF Report", file, file_name="Analysis_Results.pdf", mime="application/pdf")
 
+            # Email PDF
             if recipient_email:
                 try:
                     validate_email(recipient_email)
                     send_email(pdf_path, recipient_email)
-                    st.success("Analysis PDF has been sent to your email.")
+                    st.success(f"PDF sent to {recipient_email} successfully!")
                 except EmailNotValidError:
-                    st.error("Invalid email address. Please enter a valid email.")
-        else:
-            st.error("Please upload a PDF document for analysis.")
+                    st.error("Invalid email address. Please enter a valid one.")
 
 if __name__ == "__main__":
     main()