codeclone 1.1.0__py3-none-any.whl → 1.2.0__py3-none-any.whl
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- codeclone/baseline.py +13 -9
- codeclone/cache.py +12 -6
- codeclone/cfg.py +166 -1
- codeclone/cli.py +305 -75
- codeclone/html_report.py +569 -586
- codeclone/normalize.py +29 -6
- codeclone/report.py +6 -6
- {codeclone-1.1.0.dist-info → codeclone-1.2.0.dist-info}/METADATA +15 -5
- codeclone-1.2.0.dist-info/RECORD +19 -0
- {codeclone-1.1.0.dist-info → codeclone-1.2.0.dist-info}/WHEEL +1 -1
- codeclone-1.1.0.dist-info/RECORD +0 -19
- {codeclone-1.1.0.dist-info → codeclone-1.2.0.dist-info}/entry_points.txt +0 -0
- {codeclone-1.1.0.dist-info → codeclone-1.2.0.dist-info}/licenses/LICENSE +0 -0
- {codeclone-1.1.0.dist-info → codeclone-1.2.0.dist-info}/top_level.txt +0 -0
codeclone/baseline.py
CHANGED
|
@@ -10,22 +10,24 @@ from __future__ import annotations
|
|
|
10
10
|
|
|
11
11
|
import json
|
|
12
12
|
from pathlib import Path
|
|
13
|
-
from typing import Set
|
|
14
13
|
|
|
15
14
|
|
|
16
15
|
class Baseline:
|
|
17
|
-
def __init__(self, path: str):
|
|
16
|
+
def __init__(self, path: str | Path):
|
|
18
17
|
self.path = Path(path)
|
|
19
|
-
self.functions:
|
|
20
|
-
self.blocks:
|
|
18
|
+
self.functions: set[str] = set()
|
|
19
|
+
self.blocks: set[str] = set()
|
|
21
20
|
|
|
22
21
|
def load(self) -> None:
|
|
23
22
|
if not self.path.exists():
|
|
24
23
|
return
|
|
25
24
|
|
|
26
|
-
|
|
27
|
-
|
|
28
|
-
|
|
25
|
+
try:
|
|
26
|
+
data = json.loads(self.path.read_text("utf-8"))
|
|
27
|
+
self.functions = set(data.get("functions", []))
|
|
28
|
+
self.blocks = set(data.get("blocks", []))
|
|
29
|
+
except json.JSONDecodeError as e:
|
|
30
|
+
raise ValueError(f"Corrupted baseline file at {self.path}: {e}") from e
|
|
29
31
|
|
|
30
32
|
def save(self) -> None:
|
|
31
33
|
self.path.parent.mkdir(parents=True, exist_ok=True)
|
|
@@ -42,8 +44,10 @@ class Baseline:
|
|
|
42
44
|
)
|
|
43
45
|
|
|
44
46
|
@staticmethod
|
|
45
|
-
def from_groups(
|
|
46
|
-
|
|
47
|
+
def from_groups(
|
|
48
|
+
func_groups: dict, block_groups: dict, path: str | Path = ""
|
|
49
|
+
) -> "Baseline":
|
|
50
|
+
bl = Baseline(path)
|
|
47
51
|
bl.functions = set(func_groups.keys())
|
|
48
52
|
bl.blocks = set(block_groups.keys())
|
|
49
53
|
return bl
|
codeclone/cache.py
CHANGED
|
@@ -12,17 +12,21 @@ import json
|
|
|
12
12
|
import os
|
|
13
13
|
from dataclasses import asdict
|
|
14
14
|
from pathlib import Path
|
|
15
|
-
from typing import Optional
|
|
15
|
+
from typing import Any, Optional
|
|
16
16
|
|
|
17
17
|
|
|
18
18
|
class Cache:
|
|
19
|
-
def __init__(self, path: str):
|
|
19
|
+
def __init__(self, path: str | Path):
|
|
20
20
|
self.path = Path(path)
|
|
21
|
-
self.data: dict = {"files": {}}
|
|
21
|
+
self.data: dict[str, Any] = {"files": {}}
|
|
22
22
|
|
|
23
23
|
def load(self) -> None:
|
|
24
24
|
if self.path.exists():
|
|
25
|
-
|
|
25
|
+
try:
|
|
26
|
+
self.data = json.loads(self.path.read_text("utf-8"))
|
|
27
|
+
except json.JSONDecodeError:
|
|
28
|
+
# If cache is corrupted, start fresh
|
|
29
|
+
self.data = {"files": {}}
|
|
26
30
|
|
|
27
31
|
def save(self) -> None:
|
|
28
32
|
self.path.parent.mkdir(parents=True, exist_ok=True)
|
|
@@ -31,10 +35,12 @@ class Cache:
|
|
|
31
35
|
"utf-8",
|
|
32
36
|
)
|
|
33
37
|
|
|
34
|
-
def get_file_entry(self, filepath: str) -> Optional[dict]:
|
|
38
|
+
def get_file_entry(self, filepath: str) -> Optional[dict[str, Any]]:
|
|
35
39
|
return self.data.get("files", {}).get(filepath)
|
|
36
40
|
|
|
37
|
-
def put_file_entry(
|
|
41
|
+
def put_file_entry(
|
|
42
|
+
self, filepath: str, stat_sig: dict[str, Any], units: list, blocks: list
|
|
43
|
+
) -> None:
|
|
38
44
|
self.data.setdefault("files", {})[filepath] = {
|
|
39
45
|
"stat": stat_sig,
|
|
40
46
|
"units": [asdict(u) for u in units],
|
codeclone/cfg.py
CHANGED
|
@@ -107,6 +107,18 @@ class CFGBuilder:
|
|
|
107
107
|
case ast.For():
|
|
108
108
|
self._visit_for(stmt)
|
|
109
109
|
|
|
110
|
+
case ast.AsyncFor():
|
|
111
|
+
self._visit_for(stmt) # Structure is identical to For
|
|
112
|
+
|
|
113
|
+
case ast.Try() | ast.TryStar():
|
|
114
|
+
self._visit_try(stmt)
|
|
115
|
+
|
|
116
|
+
case ast.With() | ast.AsyncWith():
|
|
117
|
+
self._visit_with(stmt)
|
|
118
|
+
|
|
119
|
+
case ast.Match():
|
|
120
|
+
self._visit_match(stmt)
|
|
121
|
+
|
|
110
122
|
case _:
|
|
111
123
|
self.current.statements.append(stmt)
|
|
112
124
|
|
|
@@ -153,7 +165,7 @@ class CFGBuilder:
|
|
|
153
165
|
|
|
154
166
|
self.current = after_block
|
|
155
167
|
|
|
156
|
-
def _visit_for(self, stmt: ast.For) -> None:
|
|
168
|
+
def _visit_for(self, stmt: ast.For | ast.AsyncFor) -> None:
|
|
157
169
|
iter_block = self.cfg.create_block()
|
|
158
170
|
body_block = self.cfg.create_block()
|
|
159
171
|
after_block = self.cfg.create_block()
|
|
@@ -171,3 +183,156 @@ class CFGBuilder:
|
|
|
171
183
|
self.current.add_successor(iter_block)
|
|
172
184
|
|
|
173
185
|
self.current = after_block
|
|
186
|
+
|
|
187
|
+
def _visit_with(self, stmt: ast.With | ast.AsyncWith) -> None:
|
|
188
|
+
# Treat WITH as linear flow (enter -> body -> exit), but preserve block structure
|
|
189
|
+
# We record the context manager expression in the current block
|
|
190
|
+
# Then we enter a new block for the body (to separate it structurally)
|
|
191
|
+
# Then we enter a new block for 'after' (exit)
|
|
192
|
+
|
|
193
|
+
# Why new block? Because 'with' implies a scope/context.
|
|
194
|
+
# It helps matching.
|
|
195
|
+
|
|
196
|
+
body_block = self.cfg.create_block()
|
|
197
|
+
after_block = self.cfg.create_block()
|
|
198
|
+
|
|
199
|
+
# Record the 'items' (context managers)
|
|
200
|
+
# We wrap them in Expr to treat them as statements for hashing
|
|
201
|
+
for item in stmt.items:
|
|
202
|
+
self.current.statements.append(ast.Expr(value=item.context_expr))
|
|
203
|
+
|
|
204
|
+
self.current.add_successor(body_block)
|
|
205
|
+
|
|
206
|
+
self.current = body_block
|
|
207
|
+
self._visit_statements(stmt.body)
|
|
208
|
+
if not self.current.is_terminated:
|
|
209
|
+
self.current.add_successor(after_block)
|
|
210
|
+
|
|
211
|
+
self.current = after_block
|
|
212
|
+
|
|
213
|
+
def _visit_try(self, stmt: ast.Try | ast.TryStar) -> None:
|
|
214
|
+
# Simplified Try CFG:
|
|
215
|
+
# Try Body -> [Handlers...] -> Finally/After
|
|
216
|
+
# Try Body -> Else -> Finally/After
|
|
217
|
+
|
|
218
|
+
try_block = self.cfg.create_block()
|
|
219
|
+
self.current.add_successor(try_block)
|
|
220
|
+
|
|
221
|
+
# We don't know WHERE in the try block exception happens, so we assume
|
|
222
|
+
# any point in try block *could* jump to handlers.
|
|
223
|
+
# But for structural hashing, we just process the body.
|
|
224
|
+
# Ideally, we should link the try_block (or its end) to handlers?
|
|
225
|
+
# A simple approximation:
|
|
226
|
+
# 1. Process body.
|
|
227
|
+
# 2. Link entry (or end of body) to handlers?
|
|
228
|
+
# Let's do: Entry -> BodyBlock.
|
|
229
|
+
# Entry -> HandlerBlocks (to represent potential jump).
|
|
230
|
+
|
|
231
|
+
# Actually, let's keep it linear but branched.
|
|
232
|
+
# Current -> TryBody
|
|
233
|
+
# Current -> Handlers (Abstractly representing the jump)
|
|
234
|
+
|
|
235
|
+
handlers_blocks = [self.cfg.create_block() for _ in stmt.handlers]
|
|
236
|
+
else_block = self.cfg.create_block() if stmt.orelse else None
|
|
237
|
+
final_block = self.cfg.create_block() # This is finally or after
|
|
238
|
+
|
|
239
|
+
# Link current to TryBody
|
|
240
|
+
self.current = try_block
|
|
241
|
+
self._visit_statements(stmt.body)
|
|
242
|
+
|
|
243
|
+
# If try body finishes successfully:
|
|
244
|
+
if not self.current.is_terminated:
|
|
245
|
+
if else_block:
|
|
246
|
+
self.current.add_successor(else_block)
|
|
247
|
+
else:
|
|
248
|
+
self.current.add_successor(final_block)
|
|
249
|
+
|
|
250
|
+
# Handle Else
|
|
251
|
+
if else_block:
|
|
252
|
+
self.current = else_block
|
|
253
|
+
self._visit_statements(stmt.orelse)
|
|
254
|
+
if not self.current.is_terminated:
|
|
255
|
+
self.current.add_successor(final_block)
|
|
256
|
+
|
|
257
|
+
# Handle Handlers
|
|
258
|
+
# We assume control flow *could* jump from start of Try to any handler
|
|
259
|
+
# (Technically from inside try, but we model structural containment)
|
|
260
|
+
# To make fingerprints stable, we just need to ensure handlers are visited
|
|
261
|
+
# and linked.
|
|
262
|
+
|
|
263
|
+
# We link the *original* predecessor (before try) or the try_block start to handlers?
|
|
264
|
+
# Let's link the `try_block` (as a container concept) to handlers.
|
|
265
|
+
# But `try_block` was mutated by `_visit_statements`.
|
|
266
|
+
# Let's use the `try_block` (start of try) to link to handlers.
|
|
267
|
+
for h_block in handlers_blocks:
|
|
268
|
+
try_block.add_successor(h_block)
|
|
269
|
+
|
|
270
|
+
for handler, h_block in zip(stmt.handlers, handlers_blocks):
|
|
271
|
+
self.current = h_block
|
|
272
|
+
# Record exception type
|
|
273
|
+
if handler.type:
|
|
274
|
+
self.current.statements.append(ast.Expr(value=handler.type))
|
|
275
|
+
self._visit_statements(handler.body)
|
|
276
|
+
if not self.current.is_terminated:
|
|
277
|
+
self.current.add_successor(final_block)
|
|
278
|
+
|
|
279
|
+
# Finally logic:
|
|
280
|
+
# If there is a finally block, `final_block` IS the finally block.
|
|
281
|
+
# We visit it. Then we create a new `after_finally` block?
|
|
282
|
+
# Or `final_block` is the start of finally.
|
|
283
|
+
|
|
284
|
+
if stmt.finalbody:
|
|
285
|
+
self.current = final_block
|
|
286
|
+
self._visit_statements(stmt.finalbody)
|
|
287
|
+
# And then continue to next code?
|
|
288
|
+
# Yes, finally flows to next statement.
|
|
289
|
+
# Unless terminated.
|
|
290
|
+
|
|
291
|
+
# If no finally, `final_block` is just the merge point (after).
|
|
292
|
+
self.current = final_block
|
|
293
|
+
|
|
294
|
+
def _visit_match(self, stmt: ast.Match) -> None:
|
|
295
|
+
# Match subject -> Cases -> After
|
|
296
|
+
|
|
297
|
+
self.current.statements.append(ast.Expr(value=stmt.subject))
|
|
298
|
+
|
|
299
|
+
after_block = self.cfg.create_block()
|
|
300
|
+
|
|
301
|
+
for case_ in stmt.cases:
|
|
302
|
+
case_block = self.cfg.create_block()
|
|
303
|
+
self.current.add_successor(case_block)
|
|
304
|
+
|
|
305
|
+
# Save current context to restore for next case branching?
|
|
306
|
+
# No, 'current' is the match subject block. It branches to ALL cases.
|
|
307
|
+
|
|
308
|
+
# Visit Case
|
|
309
|
+
# We must set self.current to case_block for visiting body
|
|
310
|
+
# But we lose reference to 'match subject block' to link next case!
|
|
311
|
+
# So we need a variable `subject_block`.
|
|
312
|
+
pass
|
|
313
|
+
|
|
314
|
+
# Re-implementing loop correctly
|
|
315
|
+
subject_block = self.current
|
|
316
|
+
|
|
317
|
+
for case_ in stmt.cases:
|
|
318
|
+
case_block = self.cfg.create_block()
|
|
319
|
+
subject_block.add_successor(case_block)
|
|
320
|
+
|
|
321
|
+
self.current = case_block
|
|
322
|
+
# We could record the pattern here?
|
|
323
|
+
# patterns are complex AST nodes. For now, let's skip pattern structure hash
|
|
324
|
+
# and just hash the body. Or dump pattern as statement?
|
|
325
|
+
# Pattern is not a statement.
|
|
326
|
+
# Let's ignore pattern details for V1, or try to normalize it.
|
|
327
|
+
# If we ignore pattern, then `case []:` and `case {}:` look same.
|
|
328
|
+
# Ideally: `self.current.statements.append(case_.pattern)` but pattern is not stmt.
|
|
329
|
+
# We can wrap in Expr? `ast.Expr(value=case_.pattern)`?
|
|
330
|
+
# Pattern is NOT an Expr subclass in 3.10. It's `ast.pattern`.
|
|
331
|
+
# So we cannot append it to `statements` list which expects `ast.stmt`.
|
|
332
|
+
# We will ignore pattern structure for now (it's structural flow we care about).
|
|
333
|
+
|
|
334
|
+
self._visit_statements(case_.body)
|
|
335
|
+
if not self.current.is_terminated:
|
|
336
|
+
self.current.add_successor(after_block)
|
|
337
|
+
|
|
338
|
+
self.current = after_block
|