| 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 | # Parse the first node | |
| 42 | node_end = find_node_end(content) | |
| 43 | first_node = content[:node_end] | |
| 44 | remaining = content[node_end:] | |
| 45 | ||
| 46 | properties = parse_properties(first_node) | |
| 47 | children = [] | |
| 48 | ||
| 49 | # Parse children | |
| 50 | while remaining: | |
| 51 | if remaining.startswith('('): | |
| 52 | # Find the matching closing parenthesis for this child | |
| 53 | paren_count = 0 | |
| 54 | end_index = 0 | |
| 55 | for i, char in enumerate(remaining): | |
| 56 | if char == '(': | |
| 57 | paren_count += 1 | |
| 58 | elif char == ')': | |
| 59 | paren_count -= 1 | |
| 60 | if paren_count == 0: | |
| 61 | end_index = i + 1 | |
| 62 | break | |
| 63 | ||
| 64 | if paren_count != 0: | |
| 65 | raise ValueError("tree missing") | |
| 66 | ||
| 67 | child_content = remaining[:end_index] | |
| 68 | children.append(parse(child_content)) | |
| 69 | remaining = remaining[end_index:] | |
| 70 | else: | |
| 71 | raise ValueError("tree missing") | |
| 72 | ||
| 73 | return SgfTree(properties=properties, children=children) | |
| 74 | ||
| 75 | ||
| 76 | def find_node_end(content): | |
| 77 | """Find the end of the first node in the content.""" | |
| 78 | i = 0 | |
| 79 | while i < len(content): | |
| 80 | if content[i] == '[': | |
| 81 | # Skip to the matching closing bracket | |
| 82 | bracket_count = 1 | |
| 83 | i += 1 | |
| 84 | while i < len(content) and bracket_count > 0: | |
| 85 | if content[i] == '\\': | |
| 86 | i += 2 # Skip escape character and the next character | |
| 87 | elif content[i] == '[': | |
| 88 | bracket_count += 1 | |
| 89 | i += 1 | |
| 90 | elif content[i] == ']': | |
| 91 | bracket_count -= 1 | |
| 92 | i += 1 | |
| 93 | else: | |
| 94 | i += 1 | |
| 95 | elif content[i] == '(': | |
| 96 | # Found start of a child node | |
| 97 | return i | |
| 98 | else: | |
| 99 | i += 1 | |
| 100 | return i | |
| 101 | ||
| 102 | ||
| 103 | def parse_properties(node_content): | |
| 104 | """Parse properties from a node content string.""" | |
| 105 | properties = {} | |
| 106 | ||
| 107 | if not node_content: | |
| 108 | return properties | |
| 109 | ||
| 110 | i = 0 | |
| 111 | max_iterations = len(node_content) * 2 # Safety limit | |
| 112 | iterations = 0 | |
| 113 | ||
| 114 | while i < len(node_content) and iterations < max_iterations: | |
| 115 | iterations += 1 | |
| 116 | # Skip semicolons | |
| 117 | if node_content[i] == ';': | |
| 118 | i += 1 | |
| 119 | continue | |
| 120 | ||
| 121 | # Parse property key | |
| 122 | key_start = i | |
| 123 | while i < len(node_content) and node_content[i].isalpha(): | |
| 124 | i += 1 | |
| 125 | ||
| 126 | if i == key_start: | |
| 127 | raise ValueError("properties without delimiter") | |
| 128 | ||
| 129 | key = node_content[key_start:i] | |
| 130 | ||
| 131 | # Edge Case: Property key must be uppercase | |
| 132 | if not key.isupper(): | |
| 133 | raise ValueError("property must be in uppercase") | |
| 134 | ||
| 135 | # Parse property values | |
| 136 | values = [] | |
| 137 | while i < len(node_content) and node_content[i] == '[': | |
| 138 | i += 1 # Skip opening bracket | |
| 139 | value_start = i | |
| 140 | ||
| 141 | # Find the matching closing bracket | |
| 142 | bracket_count = 1 | |
| 143 | while i < len(node_content) and bracket_count > 0: | |
| 144 | if node_content[i] == '\\': | |
| 145 | i += 2 # Skip escape character and the next character | |
| 146 | elif node_content[i] == ']': | |
| 147 | bracket_count -= 1 | |
| 148 | if bracket_count == 0: | |
| 149 | break | |
| 150 | i += 1 | |
| 151 | elif node_content[i] == '[': | |
| 152 | bracket_count += 1 | |
| 153 | i += 1 | |
| 154 | else: | |
| 155 | i += 1 | |
| 156 | ||
| 157 | if bracket_count != 0: | |
| 158 | raise ValueError("properties without delimiter") | |
| 159 | ||
| 160 | value_content = node_content[value_start:i] | |
| 161 | value = parse_text_value(value_content) | |
| 162 | values.append(value) | |
| 163 | i += 1 # Skip closing bracket | |
| 164 | ||
| 165 | properties[key] = values | |
| 166 | ||
| 167 | if iterations >= max_iterations: | |
| 168 | raise ValueError("properties without delimiter") | |
| 169 | ||
| 170 | return properties | |
| 171 | ||
| 172 | ||
| 173 | def parse_text_value(text): | |
| 174 | """Parse SGF Text type value according to SGF specification.""" | |
| 175 | result = [] | |
| 176 | i = 0 | |
| 177 | max_iterations = len(text) * 2 # Safety limit | |
| 178 | iterations = 0 | |
| 179 | ||
| 180 | while i < len(text) and iterations < max_iterations: | |
| 181 | iterations += 1 | |
| 182 | if text[i] == '\\': | |
| 183 | # Edge Case: Handle escape character | |
| 184 | i += 1 | |
| 185 | if i < len(text): | |
| 186 | # Any character after \ is inserted as-is | |
| 187 | result.append(text[i]) | |
| 188 | i += 1 | |
| 189 | else: | |
| 190 | # Edge Case: Escape at end of string | |
| 191 | result.append('\\') | |
| 192 | i += 1 | |
| 193 | else: | |
| 194 | char = text[i] | |
| 195 | if char == '\n': | |
| 196 | # Edge Case: Newlines are removed if they come immediately after a \ | |
| 197 | # But we're not after a \ here, so keep the newline | |
| 198 | result.append(char) | |
| 199 | elif char.isspace() and char != '\n': | |
| 200 | # Edge Case: All whitespace characters other than newline are converted to spaces | |
| 201 | result.append(' ') | |
| 202 | else: | |
| 203 | result.append(char) | |
| 204 | i += 1 | |
| 205 | ||
| 206 | if iterations >= max_iterations: | |
| 207 | raise ValueError("invalid text value") | |
| 208 | ||
| 209 | return ''.join(result) | |
| 210 | ||
| 211 | # Handled Edge Cases: Empty input, missing parentheses, empty tree, missing delimiters, lowercase properties, escape sequences, whitespace handling, nested brackets |
| Test Name | Status |
|---|---|
test_escaped_newline_in_property_value_is_converted_to_nothing_at_all | Fail |
test_all_lowercase_property | Pass |
test_empty_input | Pass |
test_escaped_backslash_in_property_value_becomes_just_a_backslash | Pass |
test_escaped_closing_bracket_within_property_value_becomes_just_a_closing_bracket | Pass |
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.
