| 1 | #!/usr/bin/env python3 |
| 2 | """ |
| 3 | Convert non-interactive POSIX compliance tests to interactive YAML format. |
| 4 | |
| 5 | This tool parses posix_compliance*.sh files and generates YAML test specifications |
| 6 | that can be run with the interactive test framework. |
| 7 | """ |
| 8 | |
| 9 | import re |
| 10 | import sys |
| 11 | import yaml |
| 12 | from pathlib import Path |
| 13 | from typing import List, Dict, Any, Optional |
| 14 | |
| 15 | |
| 16 | class POSIXTestConverter: |
| 17 | """Convert POSIX shell tests to interactive YAML format.""" |
| 18 | |
| 19 | def __init__(self): |
| 20 | self.current_section = None |
| 21 | self.tests = [] |
| 22 | |
| 23 | def parse_section(self, line: str) -> Optional[str]: |
| 24 | """Parse a section declaration.""" |
| 25 | match = re.match(r'section\s+"([^"]+)"', line) |
| 26 | if match: |
| 27 | return match.group(1) |
| 28 | return None |
| 29 | |
| 30 | def parse_compare_posix_output(self, line: str) -> Optional[Dict[str, Any]]: |
| 31 | """ |
| 32 | Parse: compare_posix_output "test name" "command" |
| 33 | |
| 34 | Returns a test dictionary ready for YAML output. |
| 35 | """ |
| 36 | # Handle both single and double quoted commands |
| 37 | match = re.match(r'compare_posix_output\s+"([^"]+)"\s+"([^"]+)"', line) |
| 38 | if not match: |
| 39 | match = re.match(r"compare_posix_output\s+'([^']+)'\s+'([^']+)'", line) |
| 40 | if not match: |
| 41 | # Try mixed quotes |
| 42 | match = re.match(r'compare_posix_output\s+"([^"]+)"\s+\'([^\']+)\'', line) |
| 43 | |
| 44 | if match: |
| 45 | name, command = match.groups() |
| 46 | |
| 47 | # Estimate expected output based on command |
| 48 | expected_output = self._estimate_output(command) |
| 49 | |
| 50 | return { |
| 51 | 'name': f"{self.current_section}: {name}" if self.current_section else name, |
| 52 | 'steps': [{'send_line': command}], |
| 53 | 'expect_output': expected_output, |
| 54 | 'match_type': 'contains' |
| 55 | } |
| 56 | |
| 57 | return None |
| 58 | |
| 59 | def parse_compare_posix_exit_code(self, line: str) -> Optional[Dict[str, Any]]: |
| 60 | """ |
| 61 | Parse: compare_posix_exit_code "test name" "command" |
| 62 | |
| 63 | Exit code tests are harder in interactive mode - we need to check $? |
| 64 | """ |
| 65 | match = re.match(r'compare_posix_exit_code\s+"([^"]+)"\s+"([^"]+)"', line) |
| 66 | if match: |
| 67 | name, command = match.groups() |
| 68 | |
| 69 | # Add step to check exit code |
| 70 | return { |
| 71 | 'name': f"{self.current_section}: {name} (exit code)" if self.current_section else f"{name} (exit code)", |
| 72 | 'steps': [ |
| 73 | {'send_line': command}, |
| 74 | {'send_line': 'echo "EXIT=$?"'} |
| 75 | ], |
| 76 | 'expect_output': 'EXIT=', # Will need manual adjustment |
| 77 | 'match_type': 'contains', |
| 78 | 'note': 'MANUAL REVIEW NEEDED: Check expected exit code' |
| 79 | } |
| 80 | |
| 81 | return None |
| 82 | |
| 83 | def _estimate_output(self, command: str) -> str: |
| 84 | """ |
| 85 | Estimate the expected output of a command. |
| 86 | |
| 87 | This is a best-effort heuristic and may need manual review. |
| 88 | """ |
| 89 | # Simple echo commands |
| 90 | if command.startswith('echo '): |
| 91 | # Extract what's being echoed |
| 92 | echo_match = re.match(r'echo\s+(.+)', command) |
| 93 | if echo_match: |
| 94 | content = echo_match.group(1) |
| 95 | # Remove quotes if present |
| 96 | content = content.strip('\'"') |
| 97 | # Handle variable expansion markers |
| 98 | if '$' in content: |
| 99 | return 'MANUAL_REVIEW' # Variables need manual check |
| 100 | return content |
| 101 | |
| 102 | # printf commands |
| 103 | if command.startswith('printf '): |
| 104 | return 'MANUAL_REVIEW' # printf is complex |
| 105 | |
| 106 | # Variable assignments followed by echo |
| 107 | if ';' in command: |
| 108 | parts = command.split(';') |
| 109 | last_part = parts[-1].strip() |
| 110 | if last_part.startswith('echo '): |
| 111 | return self._estimate_output(last_part) |
| 112 | |
| 113 | # Commands that typically produce numeric output |
| 114 | if 'wc -l' in command or 'grep -c' in command: |
| 115 | return 'DIGIT' # Will match any digit |
| 116 | |
| 117 | # Default: mark for manual review |
| 118 | return 'MANUAL_REVIEW' |
| 119 | |
| 120 | def parse_file(self, file_path: Path) -> List[Dict[str, Any]]: |
| 121 | """Parse a POSIX compliance test file and extract tests.""" |
| 122 | self.tests = [] |
| 123 | |
| 124 | with open(file_path, 'r') as f: |
| 125 | for line in f: |
| 126 | line = line.strip() |
| 127 | |
| 128 | # Check for section |
| 129 | section = self.parse_section(line) |
| 130 | if section: |
| 131 | self.current_section = section |
| 132 | continue |
| 133 | |
| 134 | # Check for compare_posix_output |
| 135 | test = self.parse_compare_posix_output(line) |
| 136 | if test: |
| 137 | self.tests.append(test) |
| 138 | continue |
| 139 | |
| 140 | # Check for compare_posix_exit_code |
| 141 | test = self.parse_compare_posix_exit_code(line) |
| 142 | if test: |
| 143 | self.tests.append(test) |
| 144 | continue |
| 145 | |
| 146 | return self.tests |
| 147 | |
| 148 | def generate_yaml(self, tests: List[Dict[str, Any]], category: str) -> str: |
| 149 | """Generate YAML output for tests.""" |
| 150 | output = { |
| 151 | 'metadata': { |
| 152 | 'category': category, |
| 153 | 'description': f'Tests converted from POSIX compliance suite', |
| 154 | 'auto_generated': True, |
| 155 | 'needs_review': 'Tests with MANUAL_REVIEW need expected output verification' |
| 156 | }, |
| 157 | 'tests': tests |
| 158 | } |
| 159 | |
| 160 | return yaml.dump(output, default_flow_style=False, sort_keys=False) |
| 161 | |
| 162 | |
| 163 | def main(): |
| 164 | """Main entry point.""" |
| 165 | if len(sys.argv) < 2: |
| 166 | print(f"Usage: {sys.argv[0]} <posix_compliance_file.sh> [output.yaml]") |
| 167 | print("\nExample:") |
| 168 | print(f" {sys.argv[0]} ../../fortsh/tests/posix_compliance_gaps.sh gaps_interactive.yaml") |
| 169 | sys.exit(1) |
| 170 | |
| 171 | input_file = Path(sys.argv[1]) |
| 172 | if not input_file.exists(): |
| 173 | print(f"Error: File not found: {input_file}") |
| 174 | sys.exit(1) |
| 175 | |
| 176 | output_file = Path(sys.argv[2]) if len(sys.argv) > 2 else None |
| 177 | |
| 178 | # Determine category from filename |
| 179 | category = input_file.stem.replace('posix_compliance_', '').replace('_', ' ').title() |
| 180 | |
| 181 | # Convert tests |
| 182 | converter = POSIXTestConverter() |
| 183 | tests = converter.parse_file(input_file) |
| 184 | |
| 185 | print(f"Parsed {len(tests)} tests from {input_file}") |
| 186 | |
| 187 | # Count tests needing review |
| 188 | needs_review = sum(1 for t in tests if |
| 189 | 'MANUAL_REVIEW' in str(t.get('expect_output', '')) or |
| 190 | 'note' in t) |
| 191 | |
| 192 | if needs_review > 0: |
| 193 | print(f"⚠️ {needs_review} tests need manual review of expected output") |
| 194 | |
| 195 | # Generate YAML |
| 196 | yaml_content = converter.generate_yaml(tests, category) |
| 197 | |
| 198 | if output_file: |
| 199 | output_file.write_text(yaml_content) |
| 200 | print(f"✓ Wrote YAML to {output_file}") |
| 201 | else: |
| 202 | print("\n" + "="*60) |
| 203 | print("Generated YAML:") |
| 204 | print("="*60) |
| 205 | print(yaml_content) |
| 206 | |
| 207 | # Print statistics |
| 208 | print(f"\nStatistics:") |
| 209 | print(f" Total tests: {len(tests)}") |
| 210 | print(f" Needs review: {needs_review}") |
| 211 | print(f" Ready to use: {len(tests) - needs_review}") |
| 212 | |
| 213 | |
| 214 | if __name__ == '__main__': |
| 215 | main() |