Spaces:

Sampathkumarbandaru
/

contract-summarizer

Running

App Files Files Community

Sampathkumarbandaru commited on 2 days ago

Commit

d545986

verified ·

1 Parent(s): 42078ad

Upload 5 files

Browse files

Files changed (5) hide show

.env +4 -0
app.py +61 -0
requirements.txt +9 -0
salesforce_ai_contract_summarizer (1).zip +3 -0
utils.py +78 -0

.env ADDED Viewed

	@@ -0,0 +1,4 @@

+# .env - Store secrets securely
+SALESFORCE_INSTANCE_URL=https://orgfarm-86ce800028-dev-ed.develop.lightning.force.com
+SALESFORCE_ACCESS_TOKEN=rv8ltEXanR65uLRD5nrs5VJBk

app.py ADDED Viewed

	@@ -0,0 +1,61 @@

+import gradio as gr
+import time
+from utils import (
+    extract_text_from_pdf,
+    summarize_text,
+    generate_pdf,
+    extract_fields_from_summary,
+    send_to_salesforce
+)
+def process_and_summarize(file):
+    start_time = time.time()
+    try:
+        text = extract_text_from_pdf(file)
+        char_count = len(text)
+        summary = summarize_text(text)
+        fields = extract_fields_from_summary(summary)
+        sf_response = send_to_salesforce(fields)
+        pdf_path = generate_pdf(summary)
+        return summary, pdf_path, char_count, round(time.time() - start_time, 2), fields, sf_response
+    except Exception as e:
+        return f"Error: {str(e)}", None, 0, 0.0, {}, {}
+iface = gr.Interface(
+    fn=process_and_summarize,
+    inputs=gr.File(label="📄 Upload Contract (PDF)"),
+    outputs=[
+        gr.Textbox(label="📝 Contract Summary", lines=15, elem_id="summary-box"),
+        gr.File(label="📥 Download PDF Summary"),
+        gr.Number(label="🧮 Character Count"),
+        gr.Number(label="⏱️ Processing Time (s)"),
+        gr.JSON(label="📤 Fields Sent to Salesforce"),
+        gr.JSON(label="📥 Salesforce Response"),
+        gr.HTML(
+            """
+            <button onclick="printSummary()" style='margin-top:10px; padding:10px 20px; font-size:16px;'>🖨️ Print Summary</button>
+            <script>
+            function printSummary() {
+                const summary = document.querySelector("#summary-box textarea").value;
+                const win = window.open('', '', 'height=700,width=800');
+                win.document.write('<html><head><title>Contract Summary</title></head><body>');
+                win.document.write('<pre style="font-family: Arial; font-size: 14px;">' + summary + '</pre>');
+                win.document.write('</body></html>');
+                win.document.close();
+                win.focus();
+                win.print();
+                win.close();
+            }
+            </script>
+            """
+        )
+    ],
+    title="🧠 AI Contract Summarizer + Salesforce",
+    description="Upload a contract to summarize, download, print, and sync with Salesforce.",
+    allow_flagging="never"
+)
+if __name__ == "__main__":
+    iface.launch()

requirements.txt ADDED Viewed

	@@ -0,0 +1,9 @@

+gradio
+transformers
+torch
+fpdf
+pymupdf
+pytesseract
+Pillow
+requests
+python-dotenv

salesforce_ai_contract_summarizer (1).zip ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:2c1494bcb4c75a3fab0cb29969f1772c65ce4f36c205f4265fe35d4b158245ab
+size 3325

utils.py ADDED Viewed

	@@ -0,0 +1,78 @@

+import fitz  # PyMuPDF
+import time
+import requests
+import os
+from transformers import pipeline
+from fpdf import FPDF
+from datetime import datetime
+def extract_text_from_pdf(file):
+    if hasattr(file, 'read'):
+        file_bytes = file.read()
+    else:
+        file_bytes = file
+    with fitz.open(stream=file_bytes, filetype="pdf") as doc:
+        return "\n".join(page.get_text() for page in doc)
+def summarize_text(text):
+    summarizer = pipeline("summarization", model="pszemraj/long-t5-tglobal-base-16384-book-summary")
+    chunks = [text[i:i+3000] for i in range(0, len(text), 3000)]
+    summarized = summarizer(chunks, max_length=500, min_length=100, do_sample=False)
+    return " ".join([chunk['summary_text'] for chunk in summarized])
+def extract_fields_from_summary(summary):
+    fields = {
+        "Name": f"Contract Summary - {datetime.now().strftime('%Y%m%d%H%M%S')}",
+        "Obligations": "",
+        "Parties": "",
+        "Payment Terms": "",
+        "Start Date": "",
+        "Termination Clause": "",
+        "Validation Status": "Pending"
+    }
+    for line in summary.split("\n"):
+        if "Obligation" in line:
+            fields["Obligations"] = line
+        elif "Parties" in line:
+            fields["Parties"] = line
+        elif "Payment" in line:
+            fields["Payment Terms"] = line
+        elif "Start Date" in line:
+            fields["Start Date"] = line.split(":")[-1].strip()
+        elif "Termination" in line:
+            fields["Termination Clause"] = line
+    return fields
+def send_to_salesforce(summary_data):
+    SF_INSTANCE = os.getenv("SALESFORCE_INSTANCE_URL", "https://orgfarm-86ce800028-dev-ed.develop.lightning.force.com")
+    ACCESS_TOKEN = os.getenv("SALESFORCE_ACCESS_TOKEN", "rv8ltEXanR65uLRD5nrs5VJBk")
+    OBJECT_API = "/services/data/v60.0/sobjects/Contract_Summary__c/"
+    url = SF_INSTANCE + OBJECT_API
+    headers = {
+        "Authorization": f"Bearer {ACCESS_TOKEN}",
+        "Content-Type": "application/json"
+    }
+    payload = {
+        "Name": summary_data.get("Name", "Untitled Contract"),
+        "Obligations__c": summary_data.get("Obligations"),
+        "Parties__c": summary_data.get("Parties"),
+        "Payment_Terms__c": summary_data.get("Payment Terms"),
+        "Start_Date__c": summary_data.get("Start Date"),
+        "Termination_Clause__c": summary_data.get("Termination Clause"),
+        "Validation_Status__c": summary_data.get("Validation Status") or "Pending"
+    }
+    response = requests.post(url, headers=headers, json=payload)
+    if response.status_code >= 200 and response.status_code < 300:
+        return response.json()
+    else:
+        raise RuntimeError(f"Salesforce API error: {response.status_code} - {response.text}")
+def generate_pdf(summary_text):
+    pdf = FPDF()
+    pdf.add_page()
+    pdf.set_font("Arial", size=12)
+    for line in summary_text.split('\n'):
+        pdf.multi_cell(0, 10, line)
+    filepath = f"/tmp/summary_{int(time.time())}.pdf"
+    pdf.output(filepath)
+    return filepath