import pandas as pd import re import json # --- THE INTELLIGENCE DICTIONARY --- # Hardcode the most common, major systems here. The parser will automatically # discover others, but this guarantees the big ones format beautifully. MASTER_ROOTS = { "ACC": "Adaptive Cruise Control (ACC)", "TSR": "Traffic Sign Recognition (TSR)", "LKS": "Lane Keeping System (LKS)", "LKA": "Lane Keeping Assist (LKA)", "LCA": "Lane Centering Assist (LCA)", "BLIS": "Blind Spot Info System (BLIS)", "CTA": "Cross Traffic Alert (CTA)", "HUD": "Heads Up Display (HUD)", "BA": "Brake Assist (BA)", "DAS": "Driver Alert System (DAS)", "FCW": "Forward Collision Warning (FCW)", "DA": "Distance Alert (DA)", "DIST": "Distance Alert (DA)" # Catches "DistAlert" } def make_readable_name(text): # Insert space between lowercase and uppercase letters (e.g., SlaveExists -> Slave Exists) text = re.sub(r'([a-z])([A-Z])', r'\1 \2', text) # Insert space between letters and numbers (e.g., Camera2 -> Camera 2) text = re.sub(r'([a-zA-Z])(\d)', r'\1 \2', text) # Replace underscores with spaces text = text.replace('_', ' ') # Collapse any multiple spaces into a single space and trim edges return re.sub(r'\s+', ' ', text).strip() def parse_forscan_excel(filepath, sheet_name=0): df = pd.read_excel(filepath, sheet_name=sheet_name, header=None) blocks = [] # ========================================== # PASS 1: THE DATA EXTRACTION # ========================================== i = 0 while i < len(df): row = df.iloc[i] addr = str(row[0]).strip() # Identify a valid address row (contains a hyphen and starts with a number) if pd.notnull(row[0]) and '-' in addr and addr[0].isdigit(): mask_parts = [] first_non_mask_val = "" for col_idx in range(1, len(row)): val = str(row[col_idx]).strip() if val == "nan" or val == "": continue if re.fullmatch(r'[\*x\-]+', val): mask_parts.append(val) elif "Names & Values" not in val and "Address" not in val: first_non_mask_val = val break mask = " ".join(mask_parts) is_matrix = bool(re.match(r'^[0-9A-Fa-f]{1,2}\s*=', first_non_mask_val)) block = { "address": addr, "mask": mask, "type": "discrete_matrix" if is_matrix else "continuous_input", "schema": [], "states": [] } if is_matrix: title_text = "" if i > 0: title_row = df.iloc[i-1] for col_idx in range(1, len(title_row)): val = str(title_row[col_idx]).strip() if val != "nan" and val != "": title_text = val break if title_text: schema_items = [item.strip() for item in title_text.split("•")] for item in schema_items: match = re.search(r'(.*?)\s*\((.*?)\)', item) if match: raw_name = match.group(1).strip() acronym = match.group(2).strip() else: raw_name = item acronym = item is_config = False config_type = None base_name = raw_name base_acronym_hint = acronym if "_" in raw_name and "Cfg" in raw_name.split("_")[0]: is_config = True parts = raw_name.split("_", 1) config_type_raw = parts[0] config_type = config_type_raw.replace("Cfg", "") base_name = parts[1] cfg_acronym_prefix = "".join([c for c in config_type_raw if c.isupper()]) if acronym.startswith(cfg_acronym_prefix) and len(acronym) > len(cfg_acronym_prefix): base_acronym_hint = acronym[len(cfg_acronym_prefix):] readable_name = make_readable_name(base_name) block["schema"].append({ "readable_name": readable_name, "raw_name": raw_name, "acronym": acronym, "is_config": is_config, "config_type": config_type, "base_name": base_name, "base_acronym_hint": base_acronym_hint }) j = i while j < len(df): check_row = df.iloc[j] if j > i and pd.notnull(check_row[0]): break def extract_states(hex_col, desc_col): if hex_col >= len(check_row) or desc_col >= len(check_row): return val_hex = str(check_row[hex_col]).strip() val_desc = str(check_row[desc_col]).strip() if val_hex != "nan" and "=" in val_hex: hex_clean = val_hex.split("=")[0].strip() parsed_states = [] if val_desc != "nan" and val_desc != "": desc_chunks = [chunk.strip() for chunk in val_desc.split("•")] for idx, chunk in enumerate(desc_chunks): parts = chunk.split(" ", 1) parsed_states.append({ "index": idx, "acronym_used": parts[0], "state_value": parts[1] if len(parts) > 1 else "" }) block["states"].append({ "hex": hex_clean, "raw_string": val_desc if val_desc != "nan" else "", "parsed": parsed_states }) extract_states(4, 5) extract_states(7, 8) j += 1 try: block["states"] = sorted(block["states"], key=lambda x: int(x["hex"], 16)) except ValueError: block["states"] = sorted(block["states"], key=lambda x: str(x["hex"])) blocks.append(block) i = j continue else: block["schema"] = [{"instruction": first_non_mask_val}] blocks.append(block) i += 1 continue else: i += 1 # ========================================== # PASS 2: THE HARVESTER & TAGGER # ========================================== prefix_counts = {} # Step A: Harvest Dynamic Roots # We slice off the first logical chunk of the base_name (e.g., 'ACC' from 'ACCType') for block in blocks: if block["type"] == "discrete_matrix": for item in block["schema"]: base = item.get("base_name", "") # Regex pulls the leading CamelCase chunk match = re.match(r'^([A-Z]+[a-z]*|[A-Z]+)(?=[A-Z_]|$)', base) if match: prefix = match.group(1).upper() prefix_counts[prefix] = prefix_counts.get(prefix, 0) + 1 # Merge dynamic roots with the MASTER_ROOTS dictionary dynamic_roots = dict(MASTER_ROOTS) for prefix, count in prefix_counts.items(): if count > 1 and prefix not in dynamic_roots and len(prefix) > 1: # If a prefix appears multiple times (like "Slave"), it becomes a dynamic group dynamic_roots[prefix] = f"{prefix.capitalize()} Features" # Step B: Tag every feature with its calculated Major Group for block in blocks: if block["type"] == "discrete_matrix": for item in block["schema"]: base_name = item.get("base_name", "") # Default fallback: if no root matches, it is its own major group major_group = item.get("readable_name", "Unknown") # Sort roots by length descending to match the longest (most accurate) prefix first for root in sorted(dynamic_roots.keys(), key=len, reverse=True): if base_name.upper().startswith(root): major_group = dynamic_roots[root] break item["major_group"] = major_group return blocks if __name__ == "__main__": input_file = r"C:\Users\kylem\Documents\Lightning Data\Forscan Sheets Tools\Livingitup\IPMA Livnitup's FORScan - 2021-26 F150.xlsx" output_file = "parsed_forscan.json" parsed_data = parse_forscan_excel(input_file, sheet_name=0) with open(output_file, 'w', encoding='utf-8') as f: json.dump(parsed_data, f, indent=2, ensure_ascii=False) print(f"Success! Parsed {len(parsed_data)} memory blocks into {output_file}")