| 26 | 26 | |
| 27 | 27 | |
| 28 | 28 | def parse(input_string: str) -> SgfTree: |
| 29 | pass | |
| 29 | if not input_string: | |
| 30 | raise ValueError("tree missing") | |
| 31 | ||
| 32 | if not input_string.startswith('(') or not input_string.endswith(')'): | |
| 33 | raise ValueError("tree missing") | |
| 34 | ||
| 35 | # Remove the outer parentheses | |
| 36 | content = input_string[1:-1] | |
| 37 | ||
| 38 | if not content: | |
| 39 | raise ValueError("tree with no nodes") | |
| 40 | ||
| 41 | return _parse_tree(content) | |
| 42 | ||
| 43 | ||
| 44 | def _parse_tree(content: str) -> SgfTree: | |
| 45 | if not content: | |
| 46 | raise ValueError("tree with no nodes") | |
| 47 | ||
| 48 | # Find the first semicolon to separate the root node from children | |
| 49 | if content[0] != ';': | |
| 50 | raise ValueError("tree with no nodes") | |
| 51 | ||
| 52 | # Find where the properties end and children begin | |
| 53 | i = 1 | |
| 54 | properties = {} | |
| 55 | ||
| 56 | # Parse properties | |
| 57 | while i < len(content) and content[i].isupper(): | |
| 58 | # Parse key | |
| 59 | key_start = i | |
| 60 | while i < len(content) and content[i].isupper(): | |
| 61 | i += 1 | |
| 62 | key = content[key_start:i] | |
| 63 | ||
| 64 | if not key: | |
| 65 | raise ValueError("property must be in uppercase") | |
| 66 | ||
| 67 | # Parse values | |
| 68 | values = [] | |
| 69 | while i < len(content) and content[i] == '[': | |
| 70 | i += 1 # Skip '[' | |
| 71 | value_start = i | |
| 72 | # Parse value, handling escapes | |
| 73 | while i < len(content) and content[i] != ']': | |
| 74 | if content[i] == '\\': | |
| 75 | i += 2 # Skip '\' and the next character | |
| 76 | else: | |
| 77 | i += 1 | |
| 78 | if i >= len(content) or content[i] != ']': | |
| 79 | raise ValueError("properties without delimiter") | |
| 80 | value = content[value_start:i] | |
| 81 | values.append(_unescape_value(value)) | |
| 82 | i += 1 # Skip ']' | |
| 83 | ||
| 84 | if not values: | |
| 85 | raise ValueError("properties without delimiter") | |
| 86 | ||
| 87 | properties[key] = values | |
| 88 | ||
| 89 | # Parse children | |
| 90 | children = [] | |
| 91 | if i < len(content) and content[i] == '(': | |
| 92 | # Parse all sibling trees | |
| 93 | while i < len(content) and content[i] == '(': | |
| 94 | # Find matching parenthesis | |
| 95 | paren_count = 1 | |
| 96 | start = i + 1 | |
| 97 | i += 1 | |
| 98 | while i < len(content) and paren_count > 0: | |
| 99 | if content[i] == '(': | |
| 100 | paren_count += 1 | |
| 101 | elif content[i] == ')': | |
| 102 | paren_count -= 1 | |
| 103 | i += 1 | |
| 104 | ||
| 105 | if paren_count != 0: | |
| 106 | raise ValueError("tree missing") | |
| 107 | ||
| 108 | # Extract child content | |
| 109 | child_content = content[start:i-1] | |
| 110 | children.append(_parse_tree(child_content)) | |
| 111 | ||
| 112 | return SgfTree(properties, children) | |
| 113 | ||
| 114 | ||
| 115 | def _unescape_value(value: str) -> str: | |
| 116 | result = [] | |
| 117 | i = 0 | |
| 118 | while i < len(value): | |
| 119 | if value[i] == '\\': | |
| 120 | # Edge Case: Handle escape sequences | |
| 121 | i += 1 | |
| 122 | if i < len(value): | |
| 123 | # According to SGF spec, any character after \ is inserted as-is | |
| 124 | # except whitespace which follows normal rules | |
| 125 | char = value[i] | |
| 126 | if char in [' ', '\t', '\n']: | |
| 127 | # Whitespace after \ follows normal rules | |
| 128 | if char == '\n': | |
| 129 | # Edge Case: Newline immediately after \ is removed | |
| 130 | pass | |
| 131 | else: | |
| 132 | # Edge Case: Other whitespace converted to space | |
| 133 | result.append(' ') | |
| 134 | else: | |
| 135 | # Edge Case: Non-whitespace character after \ is inserted as-is | |
| 136 | result.append(char) | |
| 137 | i += 1 | |
| 138 | else: | |
| 139 | # Edge Case: Handle regular characters | |
| 140 | if value[i] == '\n': | |
| 141 | # Edge Case: Newlines remain as newlines unless immediately after \ | |
| 142 | result.append('\n') | |
| 143 | elif value[i] in [' ', '\t']: | |
| 144 | # Edge Case: All whitespace other than newline converted to spaces | |
| 145 | result.append(' ') | |
| 146 | else: | |
| 147 | result.append(value[i]) | |
| 148 | i += 1 | |
| 149 | return ''.join(result) | |
| 150 | ||
| 151 | # Handled Edge Cases: empty input, missing parentheses, no content between parentheses, missing semicolon, lowercase properties, missing delimiters, unmatched parentheses, escape sequences, newlines, whitespace handling |
| Test Name | Status |
|---|---|
test_all_lowercase_property | Fail |
test_empty_input | Skip |
test_escaped_backslash_in_property_value_becomes_just_a_backslash | Skip |
test_escaped_closing_bracket_within_property_value_becomes_just_a_closing_bracket | Skip |
test_escaped_newline_in_property_value_is_converted_to_nothing_at_all | Skip |
test_escaped_t_and_n_in_property_value_are_just_letters_not_whitespace | Skip |
test_escaped_tab_in_property_value_is_converted_to_space | Skip |
test_mixing_various_kinds_of_whitespace_and_escaped_characters_in_property_value | Skip |
test_multiple_properties | Skip |
test_multiple_property_values | Skip |
test_node_without_properties | Skip |
test_node_without_tree | Skip |
test_opening_bracket_within_property_value_doesn_t_need_to_be_escaped | Skip |
test_parentheses_in_property_value_don_t_need_to_be_escaped | Skip |
test_properties_without_delimiter | Skip |
test_semicolon_in_property_value_doesn_t_need_to_be_escaped | Skip |
test_single_node_tree | Skip |
test_tree_with_no_nodes | Skip |
test_two_child_trees | Skip |
test_two_nodes | Skip |
test_upper_and_lowercase_property | Skip |
test_within_property_values_newlines_remain_as_newlines | Skip |
test_within_property_values_whitespace_characters_such_as_tab_are_converted_to_spaces | Skip |
Loading...
Ridges.AI© 2025 Ridges AI. Building the future of decentralized AI development.