#!/usr/bin/env python3 """ Source code formatter for tokyo-express.net Performs: 1. Indentation: leading spaces → tabs 2. PHP comments: '# ' at line start → '// ' 3. HTML DOCTYPE: old DOCTYPE → 4. HTML charset meta: http-equiv Content-Type → (UTF-8 files: UTF-8, Shift_JIS files: Shift_JIS) 5. HTML tag case: uppercase tags → lowercase 6. lang=ja → lang="ja" """ import os import re import sys BASE = os.path.dirname(os.path.abspath(__file__)) TARGET_DIRS = [ 'coursemaster', 'drive', 'drived', 'hino', 'hinonew', 'menu', 'packview', 'report', 'reportd', 'stocklist', 'totalreport', 'view' ] ROOT_FILES = [ 'zcommon.php', 'zcommon.js', 'dbproc.php', 'new.php', 'packfix.php', 'regist.php', 'beep.html', 'test.html' ] EXCLUDED_DIRS = { 'drive20260208', 'driveoldbackup', 'drivever2.0', 'hino20260208', 'hino20260216', 'hino20260217', 'hino20260218', 'report20260213' } # DOCTYPE pattern: matches # Handles tabs inside quoted strings, optional second quoted string, optional newline between them DOCTYPE_PATTERN = re.compile( r'', re.IGNORECASE | re.DOTALL ) # Meta charset pattern: # Handles tabs/spaces as attribute separators and within content value META_CHARSET_PATTERN = re.compile( r'', re.IGNORECASE ) # HTML tags to lowercase HTML_TAGS = ('html', 'head', 'body', 'title', 'style', 'script', 'meta', 'link') TAG_OPEN_PATTERN = re.compile( r'<(' + '|'.join(HTML_TAGS) + r')(\s|>|/)', re.IGNORECASE ) TAG_CLOSE_PATTERN = re.compile( r'', re.IGNORECASE ) # lang=ja without quotes LANG_PATTERN = re.compile(r'\blang=ja\b(?!["\'])') def get_target_files(): """Collect all target .php, .js, .html files.""" files = [] for fname in ROOT_FILES: fpath = os.path.join(BASE, fname) if os.path.exists(fpath): files.append(fpath) for dirname in TARGET_DIRS: dirpath = os.path.join(BASE, dirname) if not os.path.isdir(dirpath): continue for root, dirs, filenames in os.walk(dirpath): dirs[:] = [d for d in dirs if d not in EXCLUDED_DIRS] for fname in sorted(filenames): if fname.endswith(('.php', '.js', '.html')): files.append(os.path.join(root, fname)) return files def detect_encoding(fpath): """Detect file encoding: utf-8 or shift_jis.""" with open(fpath, 'rb') as f: raw = f.read() if raw.startswith(b'\xef\xbb\xbf'): return 'utf-8-sig' try: raw.decode('utf-8') return 'utf-8' except UnicodeDecodeError: return 'shift_jis' def detect_indent_size(lines): """Detect whether file uses 2-space or 4-space indentation.""" two_count = 0 four_count = 0 for line in lines: m = re.match(r'^( +)[^ \t\n]', line) if m: n = len(m.group(1)) if n % 4 == 0: four_count += 1 elif n % 2 == 0: two_count += 1 return 2 if two_count > four_count else 4 def convert_indent_line(line, tab_size): """Convert leading spaces to tabs for a single line.""" m = re.match(r'^( +)(.*)', line, re.DOTALL) if not m: return line spaces = len(m.group(1)) tabs = spaces // tab_size remainder = spaces % tab_size return '\t' * tabs + ' ' * remainder + m.group(2) def process_indent(content): """Convert leading space indentation to tabs.""" lines = content.splitlines(keepends=True) # Check if any space-indented lines exist has_space_indent = any(re.match(r'^ +[^ \t\n]', line) for line in lines) if not has_space_indent: return content tab_size = detect_indent_size(lines) new_lines = [] for line in lines: if re.match(r'^ +[^ \t\n]', line): new_lines.append(convert_indent_line(line, tab_size)) else: new_lines.append(line) return ''.join(new_lines) def process_php_comments(content): """Convert '# comment' to '// comment' at line start in PHP.""" def replace_hash(m): return m.group(1) + '// ' + m.group(2) return re.sub(r'^(\s*)# (.+)', replace_hash, content, flags=re.MULTILINE) def process_html(content, encoding, ext='.html'): """Apply HTML normalization.""" is_utf8 = encoding in ('utf-8', 'utf-8-sig') # 1. DOCTYPE content = DOCTYPE_PATTERN.sub('', content) # 2. Meta charset # Use single quotes in PHP files to avoid breaking double-quoted PHP strings q = "'" if ext == '.php' else '"' if is_utf8: content = META_CHARSET_PATTERN.sub(f'', content) else: # Shift_JIS: modernize meta tag format but keep charset value def replace_meta_sjis(m): charset_val = m.group(1).strip() return f'' content = META_CHARSET_PATTERN.sub(replace_meta_sjis, content) # 3. Uppercase tags → lowercase content = TAG_OPEN_PATTERN.sub(lambda m: '<' + m.group(1).lower() + m.group(2), content) content = TAG_CLOSE_PATTERN.sub(lambda m: '', content) # 4. lang=ja → lang="ja" (use single quotes in PHP files to avoid breaking double-quoted strings) if ext == '.php': content = LANG_PATTERN.sub("lang='ja'", content) else: content = LANG_PATTERN.sub('lang="ja"', content) return content def process_file(fpath): """Process a single file. Returns (changed: bool, lines_changed: int).""" ext = os.path.splitext(fpath)[1].lower() encoding = detect_encoding(fpath) try: with open(fpath, 'r', encoding=encoding, errors='replace') as f: original = f.read() except Exception as e: print(f' ERROR reading {os.path.relpath(fpath, BASE)}: {e}') return False, 0 content = original # 1. Indentation (all file types) content = process_indent(content) # 2. PHP comment style if ext == '.php': content = process_php_comments(content) # 3. HTML normalization (.html and .php) if ext in ('.html', '.php'): content = process_html(content, encoding, ext) if content == original: return False, 0 # Count changed lines orig_lines = original.splitlines() new_lines_list = content.splitlines() changed_count = sum(1 for a, b in zip(orig_lines, new_lines_list) if a != b) changed_count += abs(len(orig_lines) - len(new_lines_list)) try: with open(fpath, 'w', encoding=encoding, errors='replace', newline='') as f: f.write(content) except Exception as e: print(f' ERROR writing {os.path.relpath(fpath, BASE)}: {e}') return False, 0 return True, changed_count def main(): files = get_target_files() print(f'Found {len(files)} target files\n') changed_files = 0 total_lines = 0 for fpath in files: rel = os.path.relpath(fpath, BASE) changed, lines = process_file(fpath) if changed: changed_files += 1 total_lines += lines print(f' CHANGED ({lines:3d} lines): {rel}') print(f'\n{"="*50}') print(f'Done. Changed {changed_files} files, {total_lines} lines total.') if __name__ == '__main__': main()