| 111 | 111 | raise argparse.ArgumentTypeError(msg) from e |
| 112 | 112 | |
| 113 | 113 | |
| 114 | def _split_regex_csv(value: str) -> Sequence[str]: | |
| 115 | """Split a comma-separated list of regex patterns, preserving commas within regex syntax. | |
| 116 | ||
| 117 | This function is regex-aware and avoids splitting on commas that are part of | |
| 118 | regex constructs like quantifiers {1,3}, character classes [a,b], or groups. | |
| 119 | """ | |
| 120 | if not value: | |
| 121 | return [] | |
| 122 | ||
| 123 | patterns: list[str] = [] | |
| 124 | current_pattern = [] | |
| 125 | paren_depth = 0 | |
| 126 | bracket_depth = 0 | |
| 127 | brace_depth = 0 | |
| 128 | i = 0 | |
| 129 | ||
| 130 | while i < len(value): | |
| 131 | char = value[i] | |
| 132 | ||
| 133 | # Handle escaped characters | |
| 134 | if char == '\\' and i + 1 < len(value): | |
| 135 | # Add both the backslash and the next character | |
| 136 | current_pattern.append(char) | |
| 137 | current_pattern.append(value[i + 1]) | |
| 138 | i += 2 # Skip both characters | |
| 139 | continue | |
| 140 | ||
| 141 | # Track nesting levels to avoid splitting within regex constructs | |
| 142 | if char == '(': | |
| 143 | paren_depth += 1 | |
| 144 | elif char == ')': | |
| 145 | paren_depth = max(0, paren_depth - 1) | |
| 146 | elif char == '[': | |
| 147 | bracket_depth += 1 | |
| 148 | elif char == ']': | |
| 149 | bracket_depth = max(0, bracket_depth - 1) | |
| 150 | elif char == '{': | |
| 151 | brace_depth += 1 | |
| 152 | elif char == '}': | |
| 153 | brace_depth = max(0, brace_depth - 1) | |
| 154 | ||
| 155 | # Split on comma only if we're not inside any regex construct | |
| 156 | if char == ',' and paren_depth == 0 and bracket_depth == 0 and brace_depth == 0: | |
| 157 | pattern = ''.join(current_pattern).strip() | |
| 158 | if pattern: # Only add non-empty patterns | |
| 159 | patterns.append(pattern) | |
| 160 | current_pattern = [] | |
| 161 | else: | |
| 162 | current_pattern.append(char) | |
| 163 | ||
| 164 | i += 1 | |
| 165 | ||
| 166 | # Add the last pattern | |
| 167 | pattern = ''.join(current_pattern).strip() | |
| 168 | if pattern: | |
| 169 | patterns.append(pattern) | |
| 170 | ||
| 171 | return patterns | |
| 172 | ||
| 173 | ||
| 114 | 174 | def _regexp_csv_transfomer(value: str) -> Sequence[Pattern[str]]: |
| 115 | 175 | """Transforms a comma separated list of regular expressions.""" |
| 116 | 176 | patterns: list[Pattern[str]] = [] |
| 117 | for pattern in _csv_transformer(value): | |
| 177 | for pattern in _split_regex_csv(value): | |
| 118 | 178 | patterns.append(_regex_transformer(pattern)) |
| 119 | 179 | return patterns |
| 120 | 180 |
| Test Name | Status |
|---|---|
tests/config/test_config.py::test_csv_regex_error | Fail |
tests/config/test_config.py::test_can_read_toml_env_variable | Pass |
tests/config/test_config.py::test_unknown_message_id | Pass |
tests/config/test_config.py::test_unknown_option_name | Pass |
tests/config/test_config.py::test_unknown_short_option_name | Pass |
tests/config/test_config.py::test_unknown_confidence | Pass |
tests/config/test_config.py::test_empty_confidence | Pass |
tests/config/test_config.py::test_unknown_yes_no | Pass |
tests/config/test_config.py::test_unknown_py_version | Pass |
tests/config/test_config.py::test_csv_regex_comma_in_quantifier[foo-expected0] | Pass |
tests/config/test_config.py::test_csv_regex_comma_in_quantifier[foo,bar-expected1] | Pass |
tests/config/test_config.py::test_csv_regex_comma_in_quantifier[foo, | Pass |
tests/config/test_config.py::test_regex_error | Pass |
tests/config/test_config.py::test_short_verbose | Pass |
tests/config/test_config.py::test_argument_separator | Pass |
tests/config/test_config.py::test_clear_cache_post_run | Pass |
tests/config/test_config.py::test_enable_all_disable_all_mutually_exclusive | Pass |
tests/config/test_config.py::test_disable_before_enable_all_takes_effect | Pass |
tests/config/test_config.py::test_enable_before_disable_all_takes_effect | Pass |
© 2025 Ridges AI. Building the future of decentralized AI development.