# import spacy
# import gradio as gr
# from spacy import displacy
# from pdfminer.high_level import extract_text

# nlp = spacy.load("en_cv_info_extr")

# colors = {}
# for label in nlp.get_pipe('ner').labels:
#   colors[label] = "linear-gradient(90deg, #aa9cfc, #fc9ce7)"

# options = {"ents": list(nlp.get_pipe('ner').labels), "colors": colors}

# def resume_ner(file):
#     resume = extract_text(file.name)
#     doc = nlp(resume)
#     html = displacy.render(doc, style="ent", page=True, options=options)
#     html = (
#         "<div style='max-width:100%; max-height:500px; overflow:auto'>"
#         + html
#         + "</div>"
#     )
#     return html

# demo = gr.Interface(
#     resume_ner,
#     gr.File(file_types=[".pdf"]),
#     ["html"],
# )

# demo.launch()







import spacy
import gradio as gr
from spacy import displacy
from pdfminer.high_level import extract_text

# Load the custom NER model
nlp = spacy.load("en_cv_info_extr")

# Define the options for displacy.render with no colors
options = {"ents": list(nlp.get_pipe('ner').labels), "colors": {}}

def resume_ner(file):
    # Extract text from the PDF
    resume = extract_text(file.name)
    
    # Process the text with the NLP model
    doc = nlp(resume)
    
    # Render the entities in plain HTML (no colors)
    html = displacy.render(doc, style="ent", page=True, options=options)
    
    # Wrap the HTML in a div for better display
    html = (
        "<div style='max-width:100%; max-height:500px; overflow:auto; font-family: Arial, sans-serif;'>"
        + html
        + "</div>"
    )
    
    return html

# Create the Gradio interface
demo = gr.Interface(
    resume_ner,
    gr.File(file_types=[".pdf"]),
    ["html"],
)

# Launch the Gradio app
demo.launch()