| 26 | 26 | |
| 27 | 27 | |
| 28 | 28 | def parse(input_string: str) -> SgfTree: |
| 29 | | pass |
| 29 | """ |
| 30 | Parse an SGF (Smart Game Format) string and return a tree structure of properties. |
| 31 | |
| 32 | Args: |
| 33 | input_string: A string containing SGF formatted data |
| 34 | |
| 35 | Returns: |
| 36 | SgfTree: A tree structure representing the SGF data |
| 37 | |
| 38 | Raises: |
| 39 | ValueError: If the input is malformed according to SGF specifications |
| 40 | """ |
| 41 | # Edge Case: Empty input should raise "tree with no nodes" per overlapping exception analysis |
| 42 | if not input_string: |
| 43 | raise ValueError("tree with no nodes") |
| 44 | |
| 45 | # Edge Case: Input that doesn't start with '(' |
| 46 | if not input_string.startswith('('): |
| 47 | raise ValueError("tree missing") |
| 48 | |
| 49 | # Edge Case: Input that starts with '(' but doesn't end with ')' |
| 50 | if not input_string.endswith(')'): |
| 51 | raise ValueError("tree missing") |
| 52 | |
| 53 | # Edge Case: Input with just parentheses but no content |
| 54 | if len(input_string) <= 2: |
| 55 | raise ValueError("tree with no nodes") |
| 56 | |
| 57 | # Remove the outer parentheses |
| 58 | content = input_string[1:-1] |
| 59 | |
| 60 | # Edge Case: Content is empty after removing parentheses |
| 61 | if not content: |
| 62 | raise ValueError("tree with no nodes") |
| 63 | |
| 64 | # Parse the content |
| 65 | return _parse_tree_content(content) |
| 66 | |
| 67 | |
| 68 | def _parse_tree_content(content: str) -> SgfTree: |
| 69 | """ |
| 70 | Parse the content inside the outer parentheses of an SGF string. |
| 71 | |
| 72 | Args: |
| 73 | content: The content inside the outer parentheses |
| 74 | |
| 75 | Returns: |
| 76 | SgfTree: A tree structure representing the SGF data |
| 77 | """ |
| 78 | # Edge Case: Content that doesn't start with ';' |
| 79 | if not content.startswith(';'): |
| 80 | raise ValueError("tree with no nodes") |
| 81 | |
| 82 | # Skip the first ';' |
| 83 | content = content[1:] |
| 84 | |
| 85 | # Parse the first node |
| 86 | node_content, content = _split_node_content(content) |
| 87 | properties = _parse_node_properties(node_content) |
| 88 | |
| 89 | children = [] |
| 90 | |
| 91 | # Parse any children |
| 92 | while content: |
| 93 | if content.startswith('('): |
| 94 | # Parse a subtree |
| 95 | subtree_content, content = _split_subtree_content(content) |
| 96 | children.append(_parse_tree_content(subtree_content[1:-1])) |
| 97 | else: |
| 98 | # Parse another node in the same sequence |
| 99 | node_content, content = _split_node_content(content) |
| 100 | node_properties = _parse_node_properties(node_content) |
| 101 | # Create a new tree node with these properties and any further children |
| 102 | new_node = SgfTree(properties=node_properties) |
| 103 | # Handle any children of this node |
| 104 | if content and content.startswith('('): |
| 105 | subtree_content, content = _split_subtree_content(content) |
| 106 | new_node.children = [_parse_tree_content(subtree_content[1:-1])] |
| 107 | children = [new_node] |
| 108 | break |
| 109 | |
| 110 | return SgfTree(properties=properties, children=children) |
| 111 | |
| 112 | |
| 113 | def _split_node_content(content: str) -> tuple[str, str]: |
| 114 | """ |
| 115 | Split the content into the current node's content and the rest. |
| 116 | |
| 117 | Args: |
| 118 | content: The remaining content to parse |
| 119 | |
| 120 | Returns: |
| 121 | tuple: (current node content, remaining content) |
| 122 | """ |
| 123 | # Find the end of this node - either end of string, ';' for next node, or '(' for child |
| 124 | i = 0 |
| 125 | while i < len(content): |
| 126 | if content[i] in [';', '(']: |
| 127 | break |
| 128 | i += 1 |
| 129 | |
| 130 | return content[:i], content[i:] |
| 131 | |
| 132 | |
| 133 | def _split_subtree_content(content: str) -> tuple[str, str]: |
| 134 | """ |
| 135 | Split the content into a subtree and the rest. |
| 136 | |
| 137 | Args: |
| 138 | content: The remaining content to parse, starting with '(' |
| 139 | |
| 140 | Returns: |
| 141 | tuple: (subtree content including parentheses, remaining content) |
| 142 | """ |
| 143 | # Edge Case: Content doesn't start with '(' |
| 144 | if not content.startswith('('): |
| 145 | raise ValueError("tree missing") |
| 146 | |
| 147 | balance = 1 |
| 148 | i = 1 # Skip the first '(' |
| 149 | |
| 150 | while i < len(content) and balance > 0: |
| 151 | if content[i] == '(': |
| 152 | balance += 1 |
| 153 | elif content[i] == ')': |
| 154 | balance -= 1 |
| 155 | i += 1 |
| 156 | |
| 157 | # Edge Case: Unbalanced parentheses |
| 158 | if balance != 0: |
| 159 | raise ValueError("tree missing") |
| 160 | |
| 161 | return content[:i], content[i:] |
| 162 | |
| 163 | |
| 164 | def _parse_node_properties(content: str) -> dict: |
| 165 | """ |
| 166 | Parse the properties of a node. |
| 167 | |
| 168 | Args: |
| 169 | content: The content of a node (without the leading ';') |
| 170 | |
| 171 | Returns: |
| 172 | dict: A dictionary of properties |
| 173 | """ |
| 174 | properties = {} |
| 175 | i = 0 |
| 176 | |
| 177 | while i < len(content): |
| 178 | # Skip any whitespace (though SGF spec says no optional whitespace) |
| 179 | while i < len(content) and content[i].isspace(): |
| 180 | i += 1 |
| 181 | |
| 182 | if i >= len(content): |
| 183 | break |
| 184 | |
| 185 | # Parse property key |
| 186 | key_start = i |
| 187 | |
| 188 | # Edge Case: Property key that is not all uppercase |
| 189 | while i < len(content) and content[i].isalpha(): |
| 190 | if not content[i].isupper(): |
| 191 | raise ValueError("property must be in uppercase") |
| 192 | i += 1 |
| 193 | |
| 194 | # Edge Case: No key found |
| 195 | if i == key_start: |
| 196 | raise ValueError("properties without delimiter") |
| 197 | |
| 198 | key = content[key_start:i] |
| 199 | |
| 200 | # Edge Case: Key without value |
| 201 | if i >= len(content) or content[i] != '[': |
| 202 | raise ValueError("properties without delimiter") |
| 203 | |
| 204 | # Parse property values |
| 205 | values = [] |
| 206 | while i < len(content) and content[i] == '[': |
| 207 | i += 1 # Skip '[' |
| 208 | value_start = i |
| 209 | |
| 210 | # Parse value, handling escaping |
| 211 | while i < len(content): |
| 212 | if content[i] == ']': |
| 213 | # Found the end of the value |
| 214 | break |
| 215 | elif content[i] == '\\': |
| 216 | i += 1 # Skip escape character |
| 217 | if i < len(content): |
| 218 | i += 1 # Skip escaped character |
| 219 | else: |
| 220 | i += 1 |
| 221 | |
| 222 | # Edge Case: Unclosed value bracket |
| 223 | if i >= len(content): |
| 224 | raise ValueError("properties without delimiter") |
| 225 | |
| 226 | value = content[value_start:i] |
| 227 | value = _unescape_value(value) |
| 228 | values.append(value) |
| 229 | i += 1 # Skip ']' |
| 230 | |
| 231 | properties[key] = values |
| 232 | |
| 233 | return properties |
| 234 | |
| 235 | return properties |
| 236 | |
| 237 | |
| 238 | def _unescape_value(value: str) -> str: |
| 239 | """ |
| 240 | Unescape an SGF value according to SGF Text type rules. |
| 241 | |
| 242 | Args: |
| 243 | value: The escaped value string |
| 244 | |
| 245 | Returns: |
| 246 | str: The unescaped value string |
| 247 | """ |
| 248 | result = [] |
| 249 | i = 0 |
| 250 | |
| 251 | while i < len(value): |
| 252 | if value[i] == '\\': |
| 253 | i += 1 # Skip the escape character |
| 254 | |
| 255 | # Edge Case: Escape at end of string |
| 256 | if i >= len(value): |
| 257 | # Treat backslash at end as literal backslash |
| 258 | result.append('\\') |
| 259 | break |
| 260 | |
| 261 | char = value[i] |
| 262 | |
| 263 | # Handle newline removal after backslash |
| 264 | if char == '\n': |
| 265 | # Newline character is removed if immediately after \ |
| 266 | i += 1 |
| 267 | continue |
| 268 | elif char in [' ', '\t', '\r', '\f', '\v']: # All whitespace except newline |
| 269 | # Whitespace after \ is converted to space |
| 270 | result.append(' ') |
| 271 | else: |
| 272 | # Any non-whitespace character after \ is inserted as-is |
| 273 | result.append(char) |
| 274 | elif value[i] in [' ', '\t', '\r', '\f', '\v']: # All whitespace except newline |
| 275 | # All whitespace characters other than newline are converted to spaces |
| 276 | result.append(' ') |
| 277 | else: |
| 278 | result.append(value[i]) |
| 279 | |
| 280 | i += 1 |
| 281 | |
| 282 | return ''.join(result) |
| 283 | # Handled Edge Cases: Empty input, missing parentheses, content without nodes, missing semicolon, unbalanced parentheses, non-uppercase properties, missing delimiters, unclosed brackets, escape at end of string, newline handling, whitespace conversion |
| 284 | # Handled Edge Cases: Empty input, missing closing parenthesis, content without nodes |
| 285 | # Handled Edge Cases: Missing semicolon, content without nodes |
| 286 | # Handled Edge Cases: Missing opening parenthesis, unbalanced parentheses |
| 287 | # Handled Edge Cases: Missing key, non-uppercase key, missing opening bracket, unclosed bracket |
| 288 | # Handled Edge Cases: Escape at end of string, newline handling, whitespace conversion |