Spaces:
Running
Running
Update parser.py
Browse files
parser.py
CHANGED
@@ -73,12 +73,12 @@ def parse_node(node, lines, prev_end, level=0, total_lines=None, parent_path=Non
|
|
73 |
|
74 |
# Skip if any lines are already processed
|
75 |
if any(line in processed_lines for line in range(start_line, end_line + 1)):
|
76 |
-
return parts
|
77 |
|
78 |
# Get category, default to 'other' if None
|
79 |
category = get_category(node, parent_path[-1] if parent_path else None) or 'other'
|
80 |
if category not in counters:
|
81 |
-
category = 'other'
|
82 |
counters[category] += 1
|
83 |
node_id = f"{category.capitalize()}[{counters[category]}]"
|
84 |
|
@@ -117,6 +117,7 @@ def parse_node(node, lines, prev_end, level=0, total_lines=None, parent_path=Non
|
|
117 |
processed_lines.add(start_line)
|
118 |
|
119 |
# Handle variables in function definitions (input variables)
|
|
|
120 |
if isinstance(node, (ast.FunctionDef, ast.AsyncFunctionDef)) and node.args.args:
|
121 |
for arg in node.args.args:
|
122 |
var_start = start_line # Assume args are on the same line as function def for simplicity
|
@@ -136,6 +137,7 @@ def parse_node(node, lines, prev_end, level=0, total_lines=None, parent_path=Non
|
|
136 |
'node_id': var_node_id
|
137 |
})
|
138 |
processed_lines.add(var_start)
|
|
|
139 |
|
140 |
# Process nested bodies
|
141 |
nested_prev_end = start_line
|
@@ -160,8 +162,10 @@ def parse_node(node, lines, prev_end, level=0, total_lines=None, parent_path=Non
|
|
160 |
'node_id': sub_node_id
|
161 |
})
|
162 |
processed_lines.add(child_start)
|
163 |
-
|
|
|
164 |
parts.extend(child_parts)
|
|
|
165 |
nested_prev_end = max(nested_prev_end, child_parts[-1]['location'][1] if child_parts else child_start)
|
166 |
elif attr == 'handlers' and isinstance(child, ast.ExceptHandler):
|
167 |
if child_start not in processed_lines and not is_blank_or_comment(lines[child_start - 1]):
|
@@ -177,8 +181,10 @@ def parse_node(node, lines, prev_end, level=0, total_lines=None, parent_path=Non
|
|
177 |
'node_id': sub_node_id
|
178 |
})
|
179 |
processed_lines.add(child_start)
|
180 |
-
|
|
|
181 |
parts.extend(child_parts)
|
|
|
182 |
nested_prev_end = max(nested_prev_end, child_parts[-1]['location'][1] if child_parts else child_start)
|
183 |
elif attr == 'finalbody':
|
184 |
if child_start not in processed_lines and not is_blank_or_comment(lines[child_start - 1]):
|
@@ -194,8 +200,10 @@ def parse_node(node, lines, prev_end, level=0, total_lines=None, parent_path=Non
|
|
194 |
'node_id': sub_node_id
|
195 |
})
|
196 |
processed_lines.add(child_start)
|
197 |
-
|
|
|
198 |
parts.extend(child_parts)
|
|
|
199 |
nested_prev_end = max(nested_prev_end, child_parts[-1]['location'][1] if child_parts else child_start)
|
200 |
else:
|
201 |
# Handle assignments and returns for variable detection
|
@@ -218,6 +226,7 @@ def parse_node(node, lines, prev_end, level=0, total_lines=None, parent_path=Non
|
|
218 |
'node_id': var_node_id
|
219 |
})
|
220 |
processed_lines.add(var_start)
|
|
|
221 |
else: # AnnAssign or AugAssign
|
222 |
target = child.target
|
223 |
if isinstance(target, ast.Name):
|
@@ -235,6 +244,7 @@ def parse_node(node, lines, prev_end, level=0, total_lines=None, parent_path=Non
|
|
235 |
'node_id': var_node_id
|
236 |
})
|
237 |
processed_lines.add(var_start)
|
|
|
238 |
elif isinstance(child, ast.Return):
|
239 |
for value in ast.walk(child):
|
240 |
if isinstance(value, ast.Name):
|
@@ -252,8 +262,10 @@ def parse_node(node, lines, prev_end, level=0, total_lines=None, parent_path=Non
|
|
252 |
'node_id': var_node_id
|
253 |
})
|
254 |
processed_lines.add(var_start)
|
255 |
-
|
|
|
256 |
parts.extend(child_parts)
|
|
|
257 |
nested_prev_end = child_parts[-1]['location'][1] if child_parts else nested_prev_end
|
258 |
|
259 |
# Update end_line and source of the parent node if its body extends it
|
@@ -265,7 +277,7 @@ def parse_node(node, lines, prev_end, level=0, total_lines=None, parent_path=Non
|
|
265 |
parts[-1]['vector'] = create_vector(category, level, (start_line, final_end), total_lines, current_path)
|
266 |
processed_lines.update(range(start_line, final_end + 1))
|
267 |
|
268 |
-
return parts
|
269 |
|
270 |
def parse_python_code(code):
|
271 |
lines = code.splitlines(keepends=True)
|
@@ -273,15 +285,17 @@ def parse_python_code(code):
|
|
273 |
try:
|
274 |
tree = ast.parse(code)
|
275 |
except SyntaxError:
|
276 |
-
return [{'category': 'error', 'source': 'Invalid Python code', 'location': (1, 1), 'level': 0, 'vector': [0, 0, 1.0, 0.0, 0, 0], 'parent_path': 'Top-Level', 'node_id': 'Error[1]'}]
|
277 |
|
278 |
parts = []
|
279 |
prev_end = 0
|
280 |
processed_lines = set()
|
|
|
281 |
|
282 |
for stmt in tree.body:
|
283 |
-
stmt_parts = parse_node(stmt, lines, prev_end, total_lines=total_lines, processed_lines=processed_lines)
|
284 |
parts.extend(stmt_parts)
|
|
|
285 |
prev_end = stmt_parts[-1]['location'][1] if stmt_parts else prev_end
|
286 |
|
287 |
if prev_end < total_lines:
|
@@ -304,8 +318,9 @@ def parse_python_code(code):
|
|
304 |
'node_id': spacer_node_id
|
305 |
})
|
306 |
processed_lines.add(i)
|
|
|
307 |
|
308 |
-
return parts
|
309 |
|
310 |
def is_blank_or_comment(line):
|
311 |
"""Check if a line is blank or a comment."""
|
|
|
73 |
|
74 |
# Skip if any lines are already processed
|
75 |
if any(line in processed_lines for line in range(start_line, end_line + 1)):
|
76 |
+
return parts, []
|
77 |
|
78 |
# Get category, default to 'other' if None
|
79 |
category = get_category(node, parent_path[-1] if parent_path else None) or 'other'
|
80 |
if category not in counters:
|
81 |
+
category = 'other'
|
82 |
counters[category] += 1
|
83 |
node_id = f"{category.capitalize()}[{counters[category]}]"
|
84 |
|
|
|
117 |
processed_lines.add(start_line)
|
118 |
|
119 |
# Handle variables in function definitions (input variables)
|
120 |
+
category_sequence = [category]
|
121 |
if isinstance(node, (ast.FunctionDef, ast.AsyncFunctionDef)) and node.args.args:
|
122 |
for arg in node.args.args:
|
123 |
var_start = start_line # Assume args are on the same line as function def for simplicity
|
|
|
137 |
'node_id': var_node_id
|
138 |
})
|
139 |
processed_lines.add(var_start)
|
140 |
+
category_sequence.append(arg_category)
|
141 |
|
142 |
# Process nested bodies
|
143 |
nested_prev_end = start_line
|
|
|
162 |
'node_id': sub_node_id
|
163 |
})
|
164 |
processed_lines.add(child_start)
|
165 |
+
category_sequence.append(sub_category)
|
166 |
+
child_parts, child_seq = parse_node(child, lines, child_start, level + 1, total_lines, current_path, counters, processed_lines)
|
167 |
parts.extend(child_parts)
|
168 |
+
category_sequence.extend(child_seq)
|
169 |
nested_prev_end = max(nested_prev_end, child_parts[-1]['location'][1] if child_parts else child_start)
|
170 |
elif attr == 'handlers' and isinstance(child, ast.ExceptHandler):
|
171 |
if child_start not in processed_lines and not is_blank_or_comment(lines[child_start - 1]):
|
|
|
181 |
'node_id': sub_node_id
|
182 |
})
|
183 |
processed_lines.add(child_start)
|
184 |
+
category_sequence.append('except')
|
185 |
+
child_parts, child_seq = parse_node(child, lines, child_start, level + 1, total_lines, current_path, counters, processed_lines)
|
186 |
parts.extend(child_parts)
|
187 |
+
category_sequence.extend(child_seq)
|
188 |
nested_prev_end = max(nested_prev_end, child_parts[-1]['location'][1] if child_parts else child_start)
|
189 |
elif attr == 'finalbody':
|
190 |
if child_start not in processed_lines and not is_blank_or_comment(lines[child_start - 1]):
|
|
|
200 |
'node_id': sub_node_id
|
201 |
})
|
202 |
processed_lines.add(child_start)
|
203 |
+
category_sequence.append('finally')
|
204 |
+
child_parts, child_seq = parse_node(child, lines, child_start, level + 1, total_lines, current_path, counters, processed_lines)
|
205 |
parts.extend(child_parts)
|
206 |
+
category_sequence.extend(child_seq)
|
207 |
nested_prev_end = max(nested_prev_end, child_parts[-1]['location'][1] if child_parts else child_start)
|
208 |
else:
|
209 |
# Handle assignments and returns for variable detection
|
|
|
226 |
'node_id': var_node_id
|
227 |
})
|
228 |
processed_lines.add(var_start)
|
229 |
+
category_sequence.append('assigned_variable')
|
230 |
else: # AnnAssign or AugAssign
|
231 |
target = child.target
|
232 |
if isinstance(target, ast.Name):
|
|
|
244 |
'node_id': var_node_id
|
245 |
})
|
246 |
processed_lines.add(var_start)
|
247 |
+
category_sequence.append('assigned_variable')
|
248 |
elif isinstance(child, ast.Return):
|
249 |
for value in ast.walk(child):
|
250 |
if isinstance(value, ast.Name):
|
|
|
262 |
'node_id': var_node_id
|
263 |
})
|
264 |
processed_lines.add(var_start)
|
265 |
+
category_sequence.append('returned_variable')
|
266 |
+
child_parts, child_seq = parse_node(child, lines, nested_prev_end, level + 1, total_lines, current_path, counters, processed_lines)
|
267 |
parts.extend(child_parts)
|
268 |
+
category_sequence.extend(child_seq)
|
269 |
nested_prev_end = child_parts[-1]['location'][1] if child_parts else nested_prev_end
|
270 |
|
271 |
# Update end_line and source of the parent node if its body extends it
|
|
|
277 |
parts[-1]['vector'] = create_vector(category, level, (start_line, final_end), total_lines, current_path)
|
278 |
processed_lines.update(range(start_line, final_end + 1))
|
279 |
|
280 |
+
return parts, category_sequence
|
281 |
|
282 |
def parse_python_code(code):
|
283 |
lines = code.splitlines(keepends=True)
|
|
|
285 |
try:
|
286 |
tree = ast.parse(code)
|
287 |
except SyntaxError:
|
288 |
+
return ([{'category': 'error', 'source': 'Invalid Python code', 'location': (1, 1), 'level': 0, 'vector': [0, 0, 1.0, 0.0, 0, 0], 'parent_path': 'Top-Level', 'node_id': 'Error[1]'}], ['error'])
|
289 |
|
290 |
parts = []
|
291 |
prev_end = 0
|
292 |
processed_lines = set()
|
293 |
+
category_sequence = []
|
294 |
|
295 |
for stmt in tree.body:
|
296 |
+
stmt_parts, stmt_seq = parse_node(stmt, lines, prev_end, total_lines=total_lines, processed_lines=processed_lines)
|
297 |
parts.extend(stmt_parts)
|
298 |
+
category_sequence.extend(stmt_seq)
|
299 |
prev_end = stmt_parts[-1]['location'][1] if stmt_parts else prev_end
|
300 |
|
301 |
if prev_end < total_lines:
|
|
|
318 |
'node_id': spacer_node_id
|
319 |
})
|
320 |
processed_lines.add(i)
|
321 |
+
category_sequence.append('spacer')
|
322 |
|
323 |
+
return parts, category_sequence
|
324 |
|
325 |
def is_blank_or_comment(line):
|
326 |
"""Check if a line is blank or a comment."""
|