Update utils.py

utils.py

@@ -1,6 +1,7 @@
 import pdfplumber
 from docx import Document
-import os
+from bs4 import BeautifulSoup
+from jinja2 import Environment, FileSystemLoader
 
 def extract_text(file):
     if file.name.endswith(".pdf"):
@@ -12,12 +13,12 @@ def extract_text(file):
     elif file.name.endswith(".txt"):
         return file.read().decode("utf-8")
     elif file.name.endswith(".html"):
-        return file.read().decode("utf-8")
+        soup = BeautifulSoup(file.read(), "html.parser")
+        return soup.get_text()
     else:
         return file.read().decode("utf-8")
 
 def render_template(template_path, context):
-    from jinja2 import Environment, FileSystemLoader
     env = Environment(loader=FileSystemLoader("templates"))
     template = env.get_template(template_path)
     return template.render(context)