| 26 | 26 | |
| 27 | 27 | |
| 28 | 28 | def parse(input_string: str) -> SgfTree: |
| 29 | | pass |
| 29 | """ |
| 30 | Parse an SGF string and return a tree structure of properties. |
| 31 | |
| 32 | Args: |
| 33 | input_string: A string in SGF format |
| 34 | |
| 35 | Returns: |
| 36 | SgfTree: The parsed SGF tree |
| 37 | |
| 38 | Raises: |
| 39 | ValueError: If the input is invalid according to SGF format rules |
| 40 | """ |
| 41 | # Edge Case: Empty input or input that doesn't start with '(' |
| 42 | if not input_string or not input_string.startswith('('): |
| 43 | raise ValueError("tree missing") |
| 44 | |
| 45 | # Edge Case: Input that doesn't end with ')' |
| 46 | if not input_string.endswith(')'): |
| 47 | raise ValueError("tree missing") |
| 48 | |
| 49 | # Remove the outer parentheses |
| 50 | content = input_string[1:-1] |
| 51 | |
| 52 | # Edge Case: Empty content after removing parentheses |
| 53 | if not content: |
| 54 | raise ValueError("tree with no nodes") |
| 55 | |
| 56 | # Edge Case: Content that doesn't start with ';' |
| 57 | if not content.startswith(';'): |
| 58 | raise ValueError("tree missing") |
| 59 | |
| 60 | return _parse_tree_content(content) |
| 61 | |
| 62 | |
| 63 | def _parse_tree_content(content: str) -> SgfTree: |
| 64 | """ |
| 65 | Parse the content of a tree (without the outer parentheses). |
| 66 | |
| 67 | Args: |
| 68 | content: The content of the tree |
| 69 | |
| 70 | Returns: |
| 71 | SgfTree: The parsed tree |
| 72 | """ |
| 73 | # Find the first node |
| 74 | if not content.startswith(';'): |
| 75 | raise ValueError("tree missing") |
| 76 | |
| 77 | # Parse the first node |
| 78 | node_end = _find_node_end(content, 1) |
| 79 | node_content = content[1:node_end] |
| 80 | properties = _parse_properties(node_content) |
| 81 | |
| 82 | # Parse children |
| 83 | children_content = content[node_end:] |
| 84 | children = _parse_children(children_content) |
| 85 | |
| 86 | return SgfTree(properties=properties, children=children) |
| 87 | |
| 88 | |
| 89 | def _find_node_end(content: str, start: int) -> int: |
| 90 | """ |
| 91 | Find the end of a node in the content. |
| 92 | |
| 93 | Args: |
| 94 | content: The content to search in |
| 95 | start: The start position |
| 96 | |
| 97 | Returns: |
| 98 | int: The position after the end of the node |
| 99 | """ |
| 100 | i = start |
| 101 | while i < len(content): |
| 102 | # If we encounter a ';' or '(' or ')', it's the end of the current node |
| 103 | if content[i] in ';()': |
| 104 | break |
| 105 | i += 1 |
| 106 | return i |
| 107 | |
| 108 | |
| 109 | def _parse_properties(content: str) -> dict: |
| 110 | """ |
| 111 | Parse properties from a node content. |
| 112 | |
| 113 | Args: |
| 114 | content: The content of a node |
| 115 | |
| 116 | Returns: |
| 117 | dict: The parsed properties |
| 118 | """ |
| 119 | properties = {} |
| 120 | i = 0 |
| 121 | |
| 122 | while i < len(content): |
| 123 | # Skip whitespace |
| 124 | if content[i].isspace(): |
| 125 | i += 1 |
| 126 | continue |
| 127 | |
| 128 | # Parse key |
| 129 | key_start = i |
| 130 | while i < len(content) and content[i].isalpha(): |
| 131 | i += 1 |
| 132 | |
| 133 | # Edge Case: No key found |
| 134 | if i == key_start: |
| 135 | raise ValueError("properties without delimiter") |
| 136 | |
| 137 | key = content[key_start:i] |
| 138 | |
| 139 | # Edge Case: Key is not uppercase |
| 140 | if key != key.upper(): |
| 141 | raise ValueError("property must be in uppercase") |
| 142 | |
| 143 | # Parse values |
| 144 | values = [] |
| 145 | |
| 146 | # Edge Case: Key not followed by '[' |
| 147 | if i >= len(content) or content[i] != '[': |
| 148 | raise ValueError("properties without delimiter") |
| 149 | |
| 150 | while i < len(content) and content[i] == '[': |
| 151 | i += 1 # Skip '[' |
| 152 | value_start = i |
| 153 | |
| 154 | # Parse value |
| 155 | value_chars = [] |
| 156 | while i < len(content) and content[i] != ']': |
| 157 | if content[i] == '\\': |
| 158 | # Handle escape character |
| 159 | i += 1 |
| 160 | if i < len(content): |
| 161 | if content[i] == 'n' and i+1 < len(content) and content[i+1] == ']': |
| 162 | # Special case: \n] should be treated as \n (which is just n) followed by ] |
| 163 | value_chars.append('\\') |
| 164 | value_chars.append('n') |
| 165 | elif content[i] == '\n': |
| 166 | # Escaped newline - remove if immediately followed by \ |
| 167 | # But according to spec, newlines are removed if they come immediately after \ |
| 168 | pass # Skip the newline |
| 169 | else: |
| 170 | value_chars.append(content[i]) |
| 171 | else: |
| 172 | # Edge Case: Escape character at end of input |
| 173 | raise ValueError("properties without delimiter") |
| 174 | else: |
| 175 | # Handle regular characters |
| 176 | # All whitespace other than newline is converted to spaces |
| 177 | if content[i] != '\n' and content[i].isspace(): |
| 178 | value_chars.append(' ') |
| 179 | else: |
| 180 | value_chars.append(content[i]) |
| 181 | i += 1 |
| 182 | |
| 183 | # Edge Case: Unclosed bracket |
| 184 | if i >= len(content) or content[i] != ']': |
| 185 | raise ValueError("properties without delimiter") |
| 186 | |
| 187 | i += 1 # Skip ']' |
| 188 | values.append(''.join(value_chars)) |
| 189 | |
| 190 | properties[key] = values |
| 191 | |
| 192 | return properties |
| 193 | |
| 194 | |
| 195 | def _parse_children(content: str) -> list: |
| 196 | """ |
| 197 | Parse children from content. |
| 198 | |
| 199 | Args: |
| 200 | content: The content containing children |
| 201 | |
| 202 | Returns: |
| 203 | list: The parsed children |
| 204 | """ |
| 205 | children = [] |
| 206 | i = 0 |
| 207 | |
| 208 | while i < len(content): |
| 209 | # Skip whitespace |
| 210 | if content[i].isspace(): |
| 211 | i += 1 |
| 212 | continue |
| 213 | |
| 214 | # If we find '(', it's the start of a child |
| 215 | if content[i] == '(': |
| 216 | i += 1 # Skip '(' |
| 217 | |
| 218 | # Find matching ')' |
| 219 | paren_count = 1 |
| 220 | child_start = i |
| 221 | |
| 222 | while i < len(content) and paren_count > 0: |
| 223 | if content[i] == '(': |
| 224 | paren_count += 1 |
| 225 | elif content[i] == ')': |
| 226 | paren_count -= 1 |
| 227 | i += 1 |
| 228 | |
| 229 | # Edge Case: Unmatched parenthesis |
| 230 | if paren_count != 0: |
| 231 | raise ValueError("tree missing") |
| 232 | |
| 233 | # Parse the child content |
| 234 | child_content = content[child_start:i-1] |
| 235 | |
| 236 | # Edge Case: Empty child |
| 237 | if not child_content: |
| 238 | raise ValueError("tree with no nodes") |
| 239 | |
| 240 | # Parse the child |
| 241 | child = _parse_tree_content(child_content) |
| 242 | children.append(child) |
| 243 | else: |
| 244 | # Edge Case: Unexpected character |
| 245 | raise ValueError("tree missing") |
| 246 | |
| 247 | return children |
| 248 | # Handled Edge Cases: Empty input, missing opening parenthesis, missing closing parenthesis, empty content after parentheses removal, missing semicolon, no key found, key not uppercase, key not followed by '[', unclosed bracket, escape character at end of input, unmatched parenthesis, empty child, unexpected character |