C6

# Import necessary libraries
import pandas as pd
import numpy as np
from pathlib import Path
import os
import re
from datetime import datetime

# Create processed_data directory if it doesn't exist
if not os.path.exists('processed_data'):
    os.makedirs('processed_data')

# Helper functions for name standardization
def clean_string(s):
    """Remove extra spaces and standardize string format"""
    if pd.isna(s):
        return ''
    return str(s).strip().replace('  ', ' ')

def extract_names_from_manager(manager_name):
    """Extract first name, known as, and last name from manager format"""
    if pd.isna(manager_name) or manager_name == '':
        return '', '', ''
    
    try:
        name_parts = manager_name.split(',')
        if len(name_parts) != 2:
            return manager_name, '', ''
        
        first_part = name_parts[0].strip()
        last_name = name_parts[1].strip()
        
        first_parts = first_part.split(' ')
        if len(first_parts) == 1:
            return first_parts[0], '', last_name
        else:
            return first_parts[0], first_parts[-1], last_name
    except:
        return manager_name, '', ''

def standardize_staff_name(first_name, known_as, last_name):
    """Create standardized versions of staff names"""
    first_name = clean_string(first_name)
    known_as = clean_string(known_as)
    last_name = clean_string(last_name)
    
    if not known_as:
        known_as = first_name
    
    return {
        'first_name': first_name,
        'known_as': known_as,
        'last_name': last_name,
        'full_name_first': f"{first_name} {last_name}",
        'full_name_known': f"{known_as} {last_name}"
    }

def process_monthly_reports():
    """Process all monthly ACL reports"""
    monthly_data = []
    monthly_reports_path = Path('monthly_reports')
    
    for file in monthly_reports_path.glob('ACL Monthly Report *.xlsx'):
        # Extract month and year from filename
        date_match = re.search(r'(\d{2})\.(\d{2})\.(\d{2})\.xlsx$', file.name)
        if date_match:
            day, month, year = date_match.groups()
            report_date = f"20{year}-{month}-{day}"
            
            df = pd.read_excel(file, skiprows=2)
            df['report_date'] = pd.to_datetime(report_date)
            
            # Clean column names
            df.columns = [col.strip() for col in df.columns]
            
            # Filter for Tutors and Learning Support Assistants
            mask = df['Position Name'].isin(['Learning Support Assistant', 'Tutor'])
            df = df[mask]
            
            # Calculate target hours
            df['working_hours'] = df['Working Hours'].fillna(0)
            df['is_salaried'] = df['working_hours'] > 3
            
            df['fte'] = df['Full-Time Equivalent'].fillna(0)
            df['target_hours'] = np.where(df['is_salaried'], df['fte'] * 840, 0)
            
            # Process names
            df['first_name'] = df['First Name'].apply(clean_string)
            df['known_as'] = df['Known As'].apply(clean_string)
            df['last_name'] = df['Last Name'].apply(clean_string)
            
            # Process line manager names
            manager_names = df['Line Manager Name'].apply(extract_names_from_manager)
            df['manager_first_name'] = [x[0] for x in manager_names]
            df['manager_known_as'] = [x[1] for x in manager_names]
            df['manager_last_name'] = [x[2] for x in manager_names]
            
            monthly_data.append(df)
    
    if monthly_data:
        combined_df = pd.concat(monthly_data, ignore_index=True)
        combined_df.to_csv('processed_data/processed_monthly_reports.csv', index=False)
        return combined_df
    else:
        raise Exception("No monthly reports found in the specified directory")

def process_tutor_pay_report(file_path='TutorPayReport_New 23.24 FY 22.11.24.xlsx'):
    """Process the Tutor Pay Report"""
    df = pd.read_excel(file_path)
    
    # Clean column names
    df.columns = [col.strip() for col in df.columns]
    
    # Filter for relevant staff roles
    df = df[df['STAFFROLE'].str.strip().isin(['Tutor', 'Learning Support'])]
    
    # Clean names
    df['PEOPLENAME'] = df['PEOPLENAME'].apply(clean_string)
    df['PEOPLESURNAME'] = df['PEOPLESURNAME'].apply(clean_string)
    
    # Group assessment activities
    assessment_types = [
        'Access Assessments', 'Assessments', 'Beauty OL Assess',
        'CFL PA Vols Assess', 'Creative Assessments', 'E&M ApprenticeAssess',
        'English FS Assess', 'English GCSE Assess', 'ESOL F2F Assess',
        'ESOL Online Assess', 'Hair/Barb F2F Assess', 'Maths Assessments',
        'SWiS Assessments'
    ]
    
    df['ACTIVITY_GROUP'] = np.where(
        df['ACTIVITYTYPE'].isin(assessment_types),
        'Assessment',
        df['ACTIVITYTYPE']
    )
    
    # Calculate actual hours
    def convert_ms_to_hours(ms):
        if pd.isna(ms):
            return 0
        return abs(float(ms)) / (1000 * 60 * 60)
    
    df['early_adjust'] = df['TIMEADJUSTEARLY'].apply(convert_ms_to_hours)
    df['late_adjust'] = df['TIMEADJUSTLATE'].apply(convert_ms_to_hours)
    df['actual_hours'] = df['Potential Hours'] - (df['early_adjust'] + df['late_adjust'])
    df['actual_hours'] = df['actual_hours'].clip(lower=0)
    
    # Convert event date to datetime
    df['EVENTDATE'] = pd.to_datetime(df['EVENTDATE'])
    
    df.to_csv('processed_data/processed_tutor_pay_report.csv', index=False)
    return df

def process_qualification_references():
    """Process both Approved and Archived qualification lists"""
    # Read Approved List
    approved_df = pd.read_excel('Approved List for Terms Caseloading FY2324.xlsx')
    approved_df = approved_df[['QualificationName', 'QualificationNumber']].copy()
    approved_df['source'] = 'approved'
    
    # Read Archived List
    archived_df = pd.read_excel('Archived List for Terms Caseloading FY2324.xlsx')
    archived_df = archived_df[['QualificationName', 'QualificationNumber']].copy()
    archived_df['source'] = 'archived'
    
    # Combine dataframes
    combined_df = pd.concat([approved_df, archived_df], ignore_index=True)
    
    # Clean qualification numbers
    def clean_qual_number(x):
        if pd.isna(x):
            return x
        x = str(x)
        if x.endswith('P') or x.endswith('N'):
            return x[:-1]
        return x
    
    combined_df['QualificationNumber_clean'] = combined_df['QualificationNumber'].apply(clean_qual_number)
    
    combined_df.to_csv('processed_data/processed_qualification_refs.csv', index=False)
    return combined_df

def process_terms_caseloading(qual_ref_df):
    """Process Terms caseloading data with qualification reference lookup"""
    df = pd.read_excel('Terms caseloading 23.24 FY.xlsx')
    
    # Create qualification name lookup dictionary
    qual_lookup = dict(zip(
        qual_ref_df['QualificationNumber_clean'],
        qual_ref_df['QualificationName']
    ))
    
    # Clean funding references and lookup qualification names
    df['funding_ref_clean'] = df['Funding Reference'].astype(str).apply(
        lambda x: x[:-1] if x.endswith(('P', 'N')) else x
    )
    
    # Apply qualification name lookup, fall back to Activity Name
    df['QualificationName'] = df['funding_ref_clean'].map(qual_lookup).fillna(df['Activity Name'])
    
    # Convert dates
    date_columns = ['Start Date', 'End Date']
    for col in date_columns:
        df[col] = pd.to_datetime(df[col])
    
    # Clean staff names
    df['Primary Staff Name'] = df['Primary Staff Name'].apply(clean_string)
    df['Primary Staff Surname'] = df['Primary Staff Surname'].apply(clean_string)
    
    # Calculate learner counts
    learner_counts = df.groupby([
        'Primary Staff Name', 'Primary Staff Surname', 
        'QualificationName', 'Level', 'Activity Group',
        pd.Grouper(key='Start Date', freq='M')
    ])['Name'].count().reset_index()
    
    learner_counts.rename(columns={'Name': 'learner_count'}, inplace=True)
    
    # Save processed data
    df.to_csv('processed_data/processed_terms_caseloading.csv', index=False)
    learner_counts.to_csv('processed_data/terms_learner_counts.csv', index=False)
    return df, learner_counts

def process_pics_hours_lookup():
    """Process PICS Hours lookup table"""
    df = pd.read_excel('PICS Hours for Assessor Look up.xlsx', skiprows=2)
    
    def time_to_decimal(time_str):
        if pd.isna(time_str):
            return 0
        try:
            hours, minutes = str(time_str).split(':')
            return float(hours) + float(minutes) / 60
        except:
            return 0
    
    df['weighted_hours'] = df['Weighted Monthly Hours (1.6)'].apply(time_to_decimal)
    
    df.to_csv('processed_data/processed_pics_hours_lookup.csv', index=False)
    return df

def process_pics_hours_lookup():
    """Process PICS Hours lookup table"""
    df = pd.read_excel('PICS Hours for Assessor Look up.xlsx', skiprows=2)
    
    def time_to_decimal(time_str):
        if pd.isna(time_str):
            return 0
        try:
            # Handle potential string formatting
            time_str = str(time_str).strip()
            if ':' in time_str:
                hours, minutes = time_str.split(':')
                return float(hours) + float(minutes) / 60
            return float(time_str)
        except:
            print(f"Error converting time: {time_str}")
            return 0
    
    # Clean standard titles
    df['Standard title'] = df['Standard title'].str.strip()
    df['weighted_hours'] = df['Weighted Monthly Hours (1.6)'].apply(time_to_decimal)
    
    print("Processed PICS Hours lookup - sample of weighted hours:")
    print(df[['Standard title', 'weighted_hours']].head())
    
    return df

def process_pics_caseload(hours_lookup_df):
    """Process PICS caseload data with hours lookup"""
    df = pd.read_excel('PICS caseload for PBI.xlsx')
    
    # Convert dates
    date_cols = ['Apprenticeship Achieved Date', 'Start Date', 
                 'Learning Expected End', 'Actual End']
    for col in date_cols:
        df[col] = pd.to_datetime(df[col], errors='coerce')
    
    # Clean standard titles
    df['Apprenticeship Standard Title'] = df['Apprenticeship Standard Title'].str.strip()
    
    # Create mapping for weighted hours
    hours_map = dict(zip(
        hours_lookup_df['Standard title'],
        hours_lookup_df['weighted_hours']
    ))
    
    # Add weighted hours to each apprenticeship
    df['weighted_hours'] = df['Apprenticeship Standard Title'].map(hours_map)
    
    # Print diagnostics
    print("\nSample of weighted hours mapping:")
    print(df[['Apprenticeship Standard Title', 'weighted_hours']].head())
    print("\nUnique apprenticeship titles:", df['Apprenticeship Standard Title'].nunique())
    print("Titles with no matching hours:", 
          df[df['weighted_hours'].isna()]['Apprenticeship Standard Title'].unique())
    
    # Calculate active periods
    df['end_date'] = df['Actual End'].fillna(df['Learning Expected End'])
    
    # Calculate monthly summaries
    all_summaries = []
    for assessor in df['Assessor Full Name'].unique():
        assessor_data = df[df['Assessor Full Name'] == assessor]
        
        if len(assessor_data) == 0:
            continue
            
        start_date = assessor_data['Start Date'].min()
        end_date = assessor_data['end_date'].max()
        
        if pd.isna(start_date) or pd.isna(end_date):
            continue
            
        date_range = pd.date_range(start_date, end_date, freq='M')
        
        for date in date_range:
            active_mask = (
                (assessor_data['Start Date'] <= date) & 
                (assessor_data['end_date'] >= date)
            )
            active_students_df = assessor_data[active_mask]
            
            if len(active_students_df) > 0:
                all_summaries.append({
                    'Assessor Full Name': assessor,
                    'month': date,
                    'student_count': len(active_students_df),
                    'total_weighted_hours': active_students_df['weighted_hours'].sum()
                })
    
    monthly_summary = pd.DataFrame(all_summaries)
    
    # Print final summary statistics
    print("\nFinal summary statistics:")
    print("Total rows:", len(monthly_summary))
    print("Sample of total_weighted_hours:", monthly_summary['total_weighted_hours'].head())
    print("Average weighted hours:", monthly_summary['total_weighted_hours'].mean())
    
    # Save processed data
    df.to_csv('processed_data/processed_pics_caseload.csv', index=False)
    monthly_summary.to_csv('processed_data/pics_monthly_summary.csv', index=False)
    return df, monthly_summary

def process_all_data():
    """Process all data sources and create final outputs"""
    try:
        print("Processing monthly reports...")
        monthly_data = process_monthly_reports()
        
        print("Processing tutor pay report...")
        tutor_pay_data = process_tutor_pay_report()
        
        print("Processing qualification references...")
        qual_ref_data = process_qualification_references()
        
        print("Processing Terms caseloading...")
        terms_data, learner_counts = process_terms_caseloading(qual_ref_data)
        
        print("Processing PICS hours lookup...")
        pics_hours_lookup = process_pics_hours_lookup()
        
        print("Processing PICS caseload...")
        pics_data, pics_summary = process_pics_caseload(pics_hours_lookup)
        
        print("All data processed successfully!")
        return True
        
    except Exception as e:
        print(f"Error processing data: {str(e)}")
        return False

# Call the main processing function
if __name__ == "__main__":
    process_all_data()