#!/usr/bin/env python3
"""
Pass 3 for Wild Gunman (JU): Add semantic names to data tables.

Pass 3 names data tables and structures based on their usage patterns
and the header documentation which identifies:
- Inline dispatch tables ($C6AC-$C802)
- Sound/music pattern data ($DEED-$F371)
- Note/volume tables ($F3EF/$F48F)
- Sprite/attribute tables ($F77C-$FFF3)
"""

import re
from pathlib import Path

# ============================================================================
# DATA TABLE NAME MAP
# Address -> (name, description)
# Only include regions that are pure data (not mixed code/data)
# ============================================================================

DATA_TABLES = {
    0xC6AC: ("SpriteAttributeTable", "Sprite attribute data: Y/tile/attr/X groups for object sprites"),
    0xF3EF: ("NoteVolumeTable", "APU note period/volume lookup table (32 bytes)"),
    0xF48F: ("VolumeEnvelopeTable", "APU volume envelope lookup table (96 bytes)"),
    0xF77C: ("SpritePositionTable", "Sprite position/attribute table (2168 bytes)"),
}

def process_file(input_path, output_path):
    text = Path(input_path).read_text()
    lines = text.split('\n')
    
    # Build a map of PRG offset -> line index for data regions
    # Find where each data address appears in the file
    addr_to_line = {}
    for i, line in enumerate(lines):
        m = re.search(r'; CPU:\$([0-9A-F]{4}) PRG:([0-9A-F]{4})', line)
        if m:
            cpu_addr = int(m.group(1), 16)
            addr_to_line[cpu_addr] = i
    
    print(f"Found {len(addr_to_line)} address annotations")
    
    # Find insertion points for each data table
    insertions = []  # (line_index, label_text)
    
    for addr, (name, desc) in DATA_TABLES.items():
        if addr in addr_to_line:
            line_idx = addr_to_line[addr]
            # Verify this line is a .byte or .word line (data, not code)
            if '.byte' in lines[line_idx] or '.word' in lines[line_idx]:
                insertions.append((line_idx, f"{name}:"))
                print(f"  {name} at CPU:${addr:04X} -> line {line_idx+1}")
            else:
                print(f"  SKIP {name} at CPU:${addr:04X} -> line {line_idx+1} is not a data line: {lines[line_idx][:60]}")
        else:
            print(f"  WARNING: CPU:${addr:04X} not found in file")
    
    # Sort insertions by line index (descending) so we can insert without shifting
    insertions.sort(key=lambda x: x[0], reverse=True)
    
    # Apply insertions
    for line_idx, label_text in insertions:
        lines.insert(line_idx, label_text)
    
    new_text = '\n'.join(lines)
    Path(output_path).write_text(new_text)
    print(f"\nWritten to {output_path}")
    print(f"Added {len(insertions)} data table labels")

if __name__ == '__main__':
    input_file = "NROM/Wild Gunman (U)/prg.s"
    output_file = "NROM/Wild Gunman (U)/prg.s"
    process_file(input_file, output_file)