app.py

import streamlit as st
import docx2txt
import PyPDF2
import re
from sentence_transformers import SentenceTransformer, util
import torch
from sklearn.feature_extraction.text import CountVectorizer
import difflib

# Load semantic model
@st.cache_resource
def load_model():
    return SentenceTransformer('all-MiniLM-L6-v2')

model = load_model()

def extract_text_from_pdf(file):
    pdf_reader = PyPDF2.PdfReader(file)
    text = ""
    for page in pdf_reader.pages:
        page_text = page.extract_text()
        if page_text:
            text += page_text
    return text

def extract_text_from_docx(file):
    return docx2txt.process(file)

def clean_text(text):
    text = re.sub(r'\s+', ' ', text)
    return text.strip().lower()

def calculate_semantic_match(resume_text, job_text):
    resume_embedding = model.encode(resume_text, convert_to_tensor=True)
    job_embedding = model.encode(job_text, convert_to_tensor=True)
    similarity = util.cos_sim(resume_embedding, job_embedding)
    return round(float(similarity.item()) * 100, 2)

def extract_keywords(text, top_n=30):
    vectorizer = CountVectorizer(stop_words='english', max_features=top_n)
    X = vectorizer.fit_transform([text])
    return set(vectorizer.get_feature_names_out())

def find_missing_keywords(resume_text, job_text):
    resume_keywords = extract_keywords(resume_text)
    job_keywords = extract_keywords(job_text)
    missing = job_keywords - resume_keywords
    return missing

# Function to extract exact missing sentences/phrases
def find_missing_sentences(resume_text, job_text):
    resume_sentences = resume_text.split(". ")
    job_sentences = job_text.split(". ")
    missing_sentences = [sentence for sentence in job_sentences if sentence.lower() not in [r.lower() for r in resume_sentences]]
    return missing_sentences

# --- Streamlit Interface ---

st.set_page_config(page_title="Resume Matcher App", page_icon="📝")

st.title("📝 Resume Matcher App")
st.write(
    "Upload your resume and paste the job description. "
    "This app will analyze and show how well your resume matches the job requirements."
)

uploaded_file = st.file_uploader("Upload Your Resume (PDF or DOCX)", type=["pdf", "docx"])

if uploaded_file is not None:
    # Check file size limit (5 MB)
    file_size = uploaded_file.size
    if file_size > 5 * 1024 * 1024:
        st.error("⚠️ File size exceeds 5MB. Please upload a smaller file.")
        st.stop()

job_description = st.text_area("Paste the Job Description Below:")

if uploaded_file and job_description:
    if uploaded_file.type == "application/pdf":
        resume_text = extract_text_from_pdf(uploaded_file)
    elif uploaded_file.type in [
        "application/vnd.openxmlformats-officedocument.wordprocessingml.document",
        "application/msword"
    ]:
        resume_text = extract_text_from_docx(uploaded_file)
    else:
        st.error("⚠️ Unsupported file format. Please upload a PDF or DOCX file.")
        st.stop()

    resume_text = clean_text(resume_text)
    job_description = clean_text(job_description)

    if st.button("🔍 Check Match"):
        with st.spinner('Analyzing...'):
            match_percentage = calculate_semantic_match(resume_text, job_description)
            missing_sentences = find_missing_sentences(resume_text, job_description)

        st.subheader("✅ Results:")
        st.progress(int(match_percentage))
        st.metric(label="Matching Percentage", value=f"{match_percentage}%")

        # Show Missing Sentences (exact as in job description)
        if missing_sentences:
            st.markdown("### 🚀 Important Missing Sentences (Add these to improve):")
            for sentence in missing_sentences:
                st.write(f"- **{sentence}**")

            st.info("🔔 If you include these important sentences, your match percentage can improve significantly and may exceed 90%!")

        else:
            st.success("🎯 No missing sentences detected! Your resume covers most important points.")

        # Tips based on percentage
        if match_percentage < 70:
            st.warning("Your resume matches less than 70% of the job description. Consider adding the above missing sentences.")
        elif match_percentage < 90:
            st.info("Good match! Minor improvements can make it even better.")
        else:
            st.success("Excellent match! Your resume strongly aligns with the job description.")
else:
    st.info("📄 Please upload your resume and paste the job description to proceed.")