src/data/loading_data.py

import pandas as pd
import os
import sys
sys.path.append(os.path.abspath(os.path.join(os.getcwd(), '..')))
from src.processing.text_cleaning import clean_column

def load_resume_data(path="../data/raw/resumes/Resume.csv"):
    """
    Loads desired resume data from a CSV file.
    
    Args:
        path (str): Path to the CSV file containing resume data.
        
    Returns:
        pd.DataFrame: A DataFrame with meaningful resume data.
    """
    # 1. Load the CSV file
    df = pd.read_csv(path, encoding='utf-8')

    # 2. Keep only two columns, drop rows with missing values
    df = df[['Category', 'Resume_str']].dropna()

    # 3. Rename columns to 'role' and 'text'          
    df.columns = ['role', 'text']  

    print(f"✅ Loaded {len(df)} resumes from {path}")
    return df                                                   

def load_job_data(path="../data/raw/job_descriptions/job_Descriptions.csv", 
                  sample_size=None,
                  resume_count=None):
    """
    Loads desired data and samples it for job description data from a CSV file.
    
    Args:
        path (str): Path to the CSV file containing job description data.
        sample_size (int): Number of job descriptions to sample.
        resume_count (int): If specified, sample this many job descriptions.
    
    Returns:
        pd.DataFrame: A DataFrame with meaningful and sampled job description data.
    """
    # 1. Load large CSV with all job descriptions
    df = pd.read_csv(path, encoding='utf-8', low_memory=False)

    # 2. Keep only two columns, drop rows with missing values 
    df = df[['Job Title', 'Job Description']].dropna()

    # 3. Rename columns to standard names
    df.columns = ['title', 'text']

    # 4. If sample_size is None, use resume_count is not None set it as sample_size
    if sample_size is None and resume_count is not None:
        sample_size = resume_count

    # 5. Randomly sample job descriptions if sample_size is specified
    if sample_size is not None:
        df = df.sample(n=sample_size, random_state=42)
    print(f"✅ Loaded {len(df)} job descriptions from {path}")
    return df



def load_or_clean_resume_data(cleaned_path="../data/processed/resumes_cleaned.csv", 
                              raw_path="../data/raw/resumes/Resume.csv"):
    
    
    if os.path.exists(cleaned_path):
        print('📂 Loading cached resume data…')
        df = pd.read_csv(cleaned_path)
        print(f"✅ Loaded cleaned resume data from {cleaned_path}")
    else:
        print('📂 Loading raw resume data and 🧼 cleaning it…')
        df = load_resume_data(raw_path)
        df = clean_column(df, column_name='text', new_column_name='text_cleaned')
        df.to_csv(cleaned_path, index=False)
        print(f"🧼 Cleaned and saved resume data to {cleaned_path}")
    return df

def load_or_clean_job_data(cleaned_path="../data/processed/jobs_cleaned.csv", 
                           raw_path="../data/raw/job_descriptions/job_descriptions.csv", sample_size=None):
    
    
    if os.path.exists(cleaned_path):
        print('📂 Loading cached job data…')
        df = pd.read_csv(cleaned_path)
        print(f"✅ Loaded cleaned job description data from {cleaned_path}")
    else:
        print('📂 Loading raw job data and 🧼 cleaning it…')
        df = load_job_data(raw_path, sample_size=sample_size)
        df = clean_column(df, column_name='text', new_column_name='text_cleaned')
        df.to_csv(cleaned_path, index=False)
        print(f"🧼 Cleaned and saved job description data to {cleaned_path}")
    return df