import gradio as gr
import os
import io
import PyPDF2
from langchain_openai import ChatOpenAI
from langchain.chains import LLMChain
from langchain.memory import ConversationBufferMemory
from langchain import PromptTemplate

from gradio.components import File, Textbox, Slider

def extract_text_from_pdf_binary(pdf_binary):
    text = ""
    pdf_data = io.BytesIO(pdf_binary)
    reader = PyPDF2.PdfReader(pdf_data)
    num_pages = len(reader.pages)

    for page in range(num_pages):
        current_page = reader.pages[page]
        page_text = current_page.extract_text()
        if page_text:  # Check if page_text is not None or empty
            text += page_text
    return text

def format_resume_to_yaml(api_key, file_content):
    os.environ['OPENAI_API_KEY'] = api_key

    if not file_content:
        raise ValueError("The uploaded file is empty.")

    resume_text = extract_text_from_pdf_binary(file_content)

    template = """Format the provided resume to this YAML template:
    ---
    name: ''
    phoneNumbers:
    - ''
    websites:
    - ''
    emails:
    - ''
    dateOfBirth: ''
    addresses:
    - street: ''
      city: ''
      state: ''
      zip: ''
      country: ''
    summary: ''
    education:
    - school: ''
      degree: ''
      fieldOfStudy: ''
      startDate: ''
      endDate: ''
    workExperience:
    - company: ''
      position: ''
      startDate: ''
      endDate: ''
    skills:
    - name: ''
    certifications:
    - name: ''
    {chat_history}
    {human_input}"""

    prompt = PromptTemplate(
        input_variables=["chat_history", "human_input"],
        template=template
    )

    memory = ConversationBufferMemory(memory_key="chat_history")

    llm_chain = LLMChain(
        llm=ChatOpenAI(model="gpt-3.5-turbo"),
        prompt=prompt,
        verbose=True,
        memory=memory,
    )

    res = llm_chain.predict(human_input=resume_text)
    return res

def match_resume_to_job_description(api_key, resume_file_content, job_description):
    os.environ['OPENAI_API_KEY'] = api_key

    if not resume_file_content or not job_description:
        raise ValueError("The uploaded file or job description is empty.")

    resume_text = extract_text_from_pdf_binary(resume_file_content)

    prompt = f"Given the following resume text:\n{resume_text}\n\nAnd the job description:\n{job_description}\n\nEvaluate how well the resume matches the job description and provide a matching score from 0 to 100, where 100 is a perfect match."

    llm = ChatOpenAI(model="gpt-3.5-turbo")
    response = llm.predict(prompt=prompt)

    return response

def main():
    input_api_key = gr.Textbox(label="Enter your OpenAI API Key")
    input_pdf_file = gr.File(label="Upload your PDF resume", type="binary")
    input_job_description = gr.Textbox(label="Enter the job description", placeholder="Paste the job description here")
    output_yaml = gr.Textbox(label="Formatted Resume in YAML")
    output_match_score = gr.Textbox(label="Resume Match Score")

    # Define separate interfaces for each function to simplify
    format_resume_interface = gr.Interface(
        fn=format_resume_to_yaml,
        inputs=[input_api_key, input_pdf_file],
        outputs=output_yaml,
        title="Resume to YAML Formatter",
        description="Upload a PDF resume and enter your OpenAI API key to get it formatted to a YAML template.",
    )

    match_resume_interface = gr.Interface(
        fn=match_resume_to_job_description,
        inputs=[input_api_key, input_pdf_file, input_job_description],
        outputs=output_match_score,
        title="Resume Matcher",
        description="Upload a PDF resume, enter your OpenAI API key and job description to get the matching score.",
    )

    # Launch interfaces in parallel if needed, or redesign to choose which to display based on user input
    format_resume_interface.launch(debug=True, share=True)
    # match_resume_interface.launch(debug=True, share=True)  # Uncomment to run separately

if __name__ == "__main__":
    main()