Extract command names from a shell command string. Handles pipes, command chaining (&&, ||, ;), and subshells. Returns the base command names (without paths). Args: command_string: The full shell command Returns: List of command names found in the string
(command_string: str)
| 209 | |
| 210 | |
| 211 | def extract_commands(command_string: str) -> list[str]: |
| 212 | """ |
| 213 | Extract command names from a shell command string. |
| 214 | |
| 215 | Handles pipes, command chaining (&&, ||, ;), and subshells. |
| 216 | Returns the base command names (without paths). |
| 217 | |
| 218 | Args: |
| 219 | command_string: The full shell command |
| 220 | |
| 221 | Returns: |
| 222 | List of command names found in the string |
| 223 | """ |
| 224 | commands = [] |
| 225 | |
| 226 | # shlex doesn't treat ; as a separator, so we need to pre-process |
| 227 | |
| 228 | # Split on semicolons that aren't inside quotes (simple heuristic) |
| 229 | # This handles common cases like "echo hello; ls" |
| 230 | segments = re.split(r'(?<!["\'])\s*;\s*(?!["\'])', command_string) |
| 231 | |
| 232 | for segment in segments: |
| 233 | segment = segment.strip() |
| 234 | if not segment: |
| 235 | continue |
| 236 | |
| 237 | try: |
| 238 | tokens = shlex.split(segment) |
| 239 | except ValueError: |
| 240 | # Malformed command (unclosed quotes, etc.) |
| 241 | # Try fallback extraction instead of blocking entirely |
| 242 | fallback_cmd = _extract_primary_command(segment) |
| 243 | if fallback_cmd: |
| 244 | logger.debug( |
| 245 | "shlex fallback used: segment=%r -> command=%r", |
| 246 | segment, |
| 247 | fallback_cmd, |
| 248 | ) |
| 249 | commands.append(fallback_cmd) |
| 250 | else: |
| 251 | logger.debug( |
| 252 | "shlex fallback failed: segment=%r (no command extracted)", |
| 253 | segment, |
| 254 | ) |
| 255 | continue |
| 256 | |
| 257 | if not tokens: |
| 258 | continue |
| 259 | |
| 260 | # Track when we expect a command vs arguments |
| 261 | expect_command = True |
| 262 | |
| 263 | for token in tokens: |
| 264 | # Shell operators indicate a new command follows |
| 265 | if token in ("|", "||", "&&", "&"): |
| 266 | expect_command = True |
| 267 | continue |
| 268 |