Spaces:
Sleeping
Sleeping
Update src/streamlit_app.py
Browse files- src/streamlit_app.py +22 -33
src/streamlit_app.py
CHANGED
|
@@ -6,7 +6,7 @@ import streamlit as st
|
|
| 6 |
import fitz # PyMuPDF
|
| 7 |
from transformers import pipeline
|
| 8 |
|
| 9 |
-
# Security
|
| 10 |
st.set_page_config(
|
| 11 |
page_title="PrepPal",
|
| 12 |
page_icon="π",
|
|
@@ -14,8 +14,9 @@ st.set_page_config(
|
|
| 14 |
menu_items={'About': "PrepPal - AI-powered PDF summarizer"}
|
| 15 |
)
|
| 16 |
|
|
|
|
| 17 |
st.markdown("""
|
| 18 |
-
<meta http-equiv="Content-Security-Policy" content="default-src 'self'; script-src 'self' 'unsafe-inline'; style-src 'self' 'unsafe-inline'; img-src 'self' data:;">
|
| 19 |
""", unsafe_allow_html=True)
|
| 20 |
|
| 21 |
@st.cache_resource
|
|
@@ -23,31 +24,33 @@ def load_summarizer():
|
|
| 23 |
try:
|
| 24 |
return pipeline(
|
| 25 |
"summarization",
|
| 26 |
-
model="facebook/bart-large-cnn", # Reliable
|
| 27 |
-
device=-1 # Force CPU
|
| 28 |
)
|
| 29 |
except Exception as e:
|
| 30 |
st.error(f"Model loading failed: {str(e)}")
|
| 31 |
return None
|
| 32 |
|
| 33 |
-
def
|
| 34 |
-
text
|
| 35 |
try:
|
| 36 |
-
#
|
| 37 |
with tempfile.NamedTemporaryFile(delete=False, suffix=".pdf") as tmp:
|
| 38 |
tmp.write(uploaded_file.getbuffer())
|
| 39 |
tmp_path = tmp.name
|
| 40 |
|
|
|
|
| 41 |
with fitz.open(tmp_path) as doc:
|
| 42 |
text = "\n".join([page.get_text() for page in doc])
|
| 43 |
|
|
|
|
| 44 |
os.unlink(tmp_path)
|
| 45 |
return text.strip()
|
| 46 |
except Exception as e:
|
| 47 |
st.error(f"PDF processing error: {str(e)}")
|
| 48 |
return ""
|
| 49 |
|
| 50 |
-
def
|
| 51 |
if not text or not model:
|
| 52 |
return ""
|
| 53 |
|
|
@@ -65,11 +68,10 @@ def summarize(text, model, max_chunk=1500):
|
|
| 65 |
|
| 66 |
return "\n".join(summary)
|
| 67 |
|
| 68 |
-
# Main App with all
|
| 69 |
def main():
|
| 70 |
st.title("π PrepPal - Study Assistant")
|
| 71 |
|
| 72 |
-
# Create all three tabs
|
| 73 |
tab1, tab2, tab3 = st.tabs(["π Summarize Notes", "β Ask a Doubt", "π¬ Feedback"])
|
| 74 |
|
| 75 |
with tab1:
|
|
@@ -84,21 +86,21 @@ def main():
|
|
| 84 |
)
|
| 85 |
|
| 86 |
if uploaded_file:
|
| 87 |
-
if uploaded_file.size > 10_000_000:
|
| 88 |
st.error("File too large (max 10MB)")
|
| 89 |
else:
|
| 90 |
with st.spinner("Extracting text..."):
|
| 91 |
-
text =
|
| 92 |
|
| 93 |
if text:
|
| 94 |
with st.expander("View extracted text"):
|
| 95 |
st.text(text[:1000] + "...")
|
| 96 |
|
| 97 |
-
if st.button("Generate Summary"
|
| 98 |
with st.spinner("Summarizing..."):
|
| 99 |
model = load_summarizer()
|
| 100 |
if model:
|
| 101 |
-
summary =
|
| 102 |
|
| 103 |
st.subheader("AI Summary")
|
| 104 |
st.write(summary)
|
|
@@ -107,32 +109,19 @@ def main():
|
|
| 107 |
"Download Summary",
|
| 108 |
data=summary,
|
| 109 |
file_name="summary.txt",
|
| 110 |
-
mime="text/plain"
|
| 111 |
-
key="download_btn"
|
| 112 |
)
|
| 113 |
|
| 114 |
with tab2:
|
| 115 |
st.header("Ask a Question")
|
| 116 |
-
st.
|
| 117 |
-
st.
|
| 118 |
-
caption="AI question answering will be available in the next update")
|
| 119 |
-
|
| 120 |
-
# Placeholder for future functionality
|
| 121 |
-
question = st.text_input("What would you like to ask about your document?")
|
| 122 |
-
if question:
|
| 123 |
-
st.info("This feature is currently in development. Please check back soon!")
|
| 124 |
|
| 125 |
with tab3:
|
| 126 |
st.header("Your Feedback")
|
| 127 |
-
st.
|
| 128 |
-
|
| 129 |
-
|
| 130 |
-
if st.button("Submit Feedback", key="feedback_btn"):
|
| 131 |
-
if feedback:
|
| 132 |
-
st.success("Thank you for your feedback! We'll use this to improve the app.")
|
| 133 |
-
# In a real app, you would store this feedback somewhere
|
| 134 |
-
else:
|
| 135 |
-
st.warning("Please enter your feedback before submitting")
|
| 136 |
|
| 137 |
if __name__ == "__main__":
|
| 138 |
main()
|
|
|
|
| 6 |
import fitz # PyMuPDF
|
| 7 |
from transformers import pipeline
|
| 8 |
|
| 9 |
+
# Security configuration
|
| 10 |
st.set_page_config(
|
| 11 |
page_title="PrepPal",
|
| 12 |
page_icon="π",
|
|
|
|
| 14 |
menu_items={'About': "PrepPal - AI-powered PDF summarizer"}
|
| 15 |
)
|
| 16 |
|
| 17 |
+
# Fix for 403 errors
|
| 18 |
st.markdown("""
|
| 19 |
+
<meta http-equiv="Content-Security-Policy" content="default-src 'self'; script-src 'self' 'unsafe-inline'; style-src 'self' 'unsafe-inline'; img-src 'self' data:;">
|
| 20 |
""", unsafe_allow_html=True)
|
| 21 |
|
| 22 |
@st.cache_resource
|
|
|
|
| 24 |
try:
|
| 25 |
return pipeline(
|
| 26 |
"summarization",
|
| 27 |
+
model="facebook/bart-large-cnn", # Reliable model
|
| 28 |
+
device=-1 # Force CPU for Hugging Face Spaces
|
| 29 |
)
|
| 30 |
except Exception as e:
|
| 31 |
st.error(f"Model loading failed: {str(e)}")
|
| 32 |
return None
|
| 33 |
|
| 34 |
+
def safe_extract_text(uploaded_file):
|
| 35 |
+
"""Secure PDF text extraction with temp files"""
|
| 36 |
try:
|
| 37 |
+
# First save to temporary file
|
| 38 |
with tempfile.NamedTemporaryFile(delete=False, suffix=".pdf") as tmp:
|
| 39 |
tmp.write(uploaded_file.getbuffer())
|
| 40 |
tmp_path = tmp.name
|
| 41 |
|
| 42 |
+
# Process from filesystem
|
| 43 |
with fitz.open(tmp_path) as doc:
|
| 44 |
text = "\n".join([page.get_text() for page in doc])
|
| 45 |
|
| 46 |
+
# Clean up
|
| 47 |
os.unlink(tmp_path)
|
| 48 |
return text.strip()
|
| 49 |
except Exception as e:
|
| 50 |
st.error(f"PDF processing error: {str(e)}")
|
| 51 |
return ""
|
| 52 |
|
| 53 |
+
def summarize_text(text, model, max_chunk=1500):
|
| 54 |
if not text or not model:
|
| 55 |
return ""
|
| 56 |
|
|
|
|
| 68 |
|
| 69 |
return "\n".join(summary)
|
| 70 |
|
| 71 |
+
# Main App with all tabs
|
| 72 |
def main():
|
| 73 |
st.title("π PrepPal - Study Assistant")
|
| 74 |
|
|
|
|
| 75 |
tab1, tab2, tab3 = st.tabs(["π Summarize Notes", "β Ask a Doubt", "π¬ Feedback"])
|
| 76 |
|
| 77 |
with tab1:
|
|
|
|
| 86 |
)
|
| 87 |
|
| 88 |
if uploaded_file:
|
| 89 |
+
if uploaded_file.size > 10_000_000:
|
| 90 |
st.error("File too large (max 10MB)")
|
| 91 |
else:
|
| 92 |
with st.spinner("Extracting text..."):
|
| 93 |
+
text = safe_extract_text(uploaded_file)
|
| 94 |
|
| 95 |
if text:
|
| 96 |
with st.expander("View extracted text"):
|
| 97 |
st.text(text[:1000] + "...")
|
| 98 |
|
| 99 |
+
if st.button("Generate Summary"):
|
| 100 |
with st.spinner("Summarizing..."):
|
| 101 |
model = load_summarizer()
|
| 102 |
if model:
|
| 103 |
+
summary = summarize_text(text, model)
|
| 104 |
|
| 105 |
st.subheader("AI Summary")
|
| 106 |
st.write(summary)
|
|
|
|
| 109 |
"Download Summary",
|
| 110 |
data=summary,
|
| 111 |
file_name="summary.txt",
|
| 112 |
+
mime="text/plain"
|
|
|
|
| 113 |
)
|
| 114 |
|
| 115 |
with tab2:
|
| 116 |
st.header("Ask a Question")
|
| 117 |
+
st.info("This feature will allow you to ask questions about your documents")
|
| 118 |
+
st.write("Coming in the next update!")
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 119 |
|
| 120 |
with tab3:
|
| 121 |
st.header("Your Feedback")
|
| 122 |
+
feedback = st.text_area("How can we improve PrepPal?")
|
| 123 |
+
if st.button("Submit Feedback"):
|
| 124 |
+
st.success("Thank you! Your feedback has been recorded.")
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 125 |
|
| 126 |
if __name__ == "__main__":
|
| 127 |
main()
|