All patches and comments are welcome. Please squash your changes to logical
commits before using git-format-patch and git-send-email to
patches@git.madduck.net.
If you'd read over the Git project's submission guidelines and adhered to them,
I'd be especially grateful.
3 from asyncio.base_events import BaseEventLoop
4 from concurrent.futures import Executor, ProcessPoolExecutor
6 from functools import partial, wraps
9 from multiprocessing import Manager
11 from pathlib import Path
36 from appdirs import user_cache_dir
37 from attr import dataclass, Factory
41 from blib2to3.pytree import Node, Leaf, type_repr
42 from blib2to3 import pygram, pytree
43 from blib2to3.pgen2 import driver, token
44 from blib2to3.pgen2.parse import ParseError
47 __version__ = "18.5b0"
48 DEFAULT_LINE_LENGTH = 88
51 syms = pygram.python_symbols
59 LN = Union[Leaf, Node]
60 SplitFunc = Callable[["Line", bool], Iterator["Line"]]
63 CacheInfo = Tuple[Timestamp, FileSize]
64 Cache = Dict[Path, CacheInfo]
65 out = partial(click.secho, bold=True, err=True)
66 err = partial(click.secho, fg="red", err=True)
69 class NothingChanged(UserWarning):
70 """Raised by :func:`format_file` when reformatted code is the same as source."""
73 class CannotSplit(Exception):
74 """A readable split that fits the allotted line length is impossible.
76 Raised by :func:`left_hand_split`, :func:`right_hand_split`, and
77 :func:`delimiter_split`.
81 class FormatError(Exception):
82 """Base exception for `# fmt: on` and `# fmt: off` handling.
84 It holds the number of bytes of the prefix consumed before the format
85 control comment appeared.
88 def __init__(self, consumed: int) -> None:
89 super().__init__(consumed)
90 self.consumed = consumed
92 def trim_prefix(self, leaf: Leaf) -> None:
93 leaf.prefix = leaf.prefix[self.consumed :]
95 def leaf_from_consumed(self, leaf: Leaf) -> Leaf:
96 """Returns a new Leaf from the consumed part of the prefix."""
97 unformatted_prefix = leaf.prefix[: self.consumed]
98 return Leaf(token.NEWLINE, unformatted_prefix)
101 class FormatOn(FormatError):
102 """Found a comment like `# fmt: on` in the file."""
105 class FormatOff(FormatError):
106 """Found a comment like `# fmt: off` in the file."""
109 class WriteBack(Enum):
126 default=DEFAULT_LINE_LENGTH,
127 help="How many character per line to allow.",
134 "Don't write the files back, just return the status. Return code 0 "
135 "means nothing would change. Return code 1 means some files would be "
136 "reformatted. Return code 123 means there was an internal error."
142 help="Don't write the files back, just output a diff for each file on stdout.",
147 help="If --fast given, skip temporary sanity checks. [default: --safe]",
154 "Don't emit non-error messages to stderr. Errors are still emitted, "
155 "silence those with 2>/dev/null."
158 @click.version_option(version=__version__)
163 exists=True, file_okay=True, dir_okay=True, readable=True, allow_dash=True
176 """The uncompromising code formatter."""
177 sources: List[Path] = []
181 sources.extend(gen_python_files_in_dir(p))
183 # if a file was explicitly given, we don't care about its extension
186 sources.append(Path("-"))
188 err(f"invalid path: {s}")
190 if check and not diff:
191 write_back = WriteBack.NO
193 write_back = WriteBack.DIFF
195 write_back = WriteBack.YES
196 report = Report(check=check, quiet=quiet)
197 if len(sources) == 0:
198 out("No paths given. Nothing to do 😴")
202 elif len(sources) == 1:
203 reformat_one(sources[0], line_length, fast, write_back, report)
205 loop = asyncio.get_event_loop()
206 executor = ProcessPoolExecutor(max_workers=os.cpu_count())
208 loop.run_until_complete(
210 sources, line_length, fast, write_back, report, loop, executor
216 out("All done! ✨ 🍰 ✨")
217 click.echo(str(report))
218 ctx.exit(report.return_code)
222 src: Path, line_length: int, fast: bool, write_back: WriteBack, report: "Report"
224 """Reformat a single file under `src` without spawning child processes.
226 If `quiet` is True, non-error messages are not output. `line_length`,
227 `write_back`, and `fast` options are passed to :func:`format_file_in_place`.
231 if not src.is_file() and str(src) == "-":
232 if format_stdin_to_stdout(
233 line_length=line_length, fast=fast, write_back=write_back
235 changed = Changed.YES
238 if write_back != WriteBack.DIFF:
239 cache = read_cache(line_length)
241 if src in cache and cache[src] == get_cache_info(src):
242 changed = Changed.CACHED
243 if changed is not Changed.CACHED and format_file_in_place(
244 src, line_length=line_length, fast=fast, write_back=write_back
246 changed = Changed.YES
247 if write_back == WriteBack.YES and changed is not Changed.NO:
248 write_cache(cache, [src], line_length)
249 report.done(src, changed)
250 except Exception as exc:
251 report.failed(src, str(exc))
254 async def schedule_formatting(
258 write_back: WriteBack,
263 """Run formatting of `sources` in parallel using the provided `executor`.
265 (Use ProcessPoolExecutors for actual parallelism.)
267 `line_length`, `write_back`, and `fast` options are passed to
268 :func:`format_file_in_place`.
271 if write_back != WriteBack.DIFF:
272 cache = read_cache(line_length)
273 sources, cached = filter_cached(cache, sources)
275 report.done(src, Changed.CACHED)
280 if write_back == WriteBack.DIFF:
281 # For diff output, we need locks to ensure we don't interleave output
282 # from different processes.
284 lock = manager.Lock()
286 loop.run_in_executor(
287 executor, format_file_in_place, src, line_length, fast, write_back, lock
289 for src in sorted(sources)
291 pending: Iterable[asyncio.Task] = tasks.keys()
293 loop.add_signal_handler(signal.SIGINT, cancel, pending)
294 loop.add_signal_handler(signal.SIGTERM, cancel, pending)
295 except NotImplementedError:
296 # There are no good alternatives for these on Windows
299 done, _ = await asyncio.wait(pending, return_when=asyncio.FIRST_COMPLETED)
301 src = tasks.pop(task)
303 cancelled.append(task)
304 elif task.exception():
305 report.failed(src, str(task.exception()))
307 formatted.append(src)
308 report.done(src, Changed.YES if task.result() else Changed.NO)
310 await asyncio.gather(*cancelled, loop=loop, return_exceptions=True)
311 if write_back == WriteBack.YES and formatted:
312 write_cache(cache, formatted, line_length)
315 def format_file_in_place(
319 write_back: WriteBack = WriteBack.NO,
320 lock: Any = None, # multiprocessing.Manager().Lock() is some crazy proxy
322 """Format file under `src` path. Return True if changed.
324 If `write_back` is True, write reformatted code back to stdout.
325 `line_length` and `fast` options are passed to :func:`format_file_contents`.
327 is_pyi = src.suffix == ".pyi"
329 with tokenize.open(src) as src_buffer:
330 src_contents = src_buffer.read()
332 dst_contents = format_file_contents(
333 src_contents, line_length=line_length, fast=fast, is_pyi=is_pyi
335 except NothingChanged:
338 if write_back == write_back.YES:
339 with open(src, "w", encoding=src_buffer.encoding) as f:
340 f.write(dst_contents)
341 elif write_back == write_back.DIFF:
342 src_name = f"{src} (original)"
343 dst_name = f"{src} (formatted)"
344 diff_contents = diff(src_contents, dst_contents, src_name, dst_name)
348 sys.stdout.write(diff_contents)
355 def format_stdin_to_stdout(
356 line_length: int, fast: bool, write_back: WriteBack = WriteBack.NO
358 """Format file on stdin. Return True if changed.
360 If `write_back` is True, write reformatted code back to stdout.
361 `line_length` and `fast` arguments are passed to :func:`format_file_contents`.
363 src = sys.stdin.read()
366 dst = format_file_contents(src, line_length=line_length, fast=fast)
369 except NothingChanged:
373 if write_back == WriteBack.YES:
374 sys.stdout.write(dst)
375 elif write_back == WriteBack.DIFF:
376 src_name = "<stdin> (original)"
377 dst_name = "<stdin> (formatted)"
378 sys.stdout.write(diff(src, dst, src_name, dst_name))
381 def format_file_contents(
382 src_contents: str, *, line_length: int, fast: bool, is_pyi: bool = False
384 """Reformat contents a file and return new contents.
386 If `fast` is False, additionally confirm that the reformatted code is
387 valid by calling :func:`assert_equivalent` and :func:`assert_stable` on it.
388 `line_length` is passed to :func:`format_str`.
390 if src_contents.strip() == "":
393 dst_contents = format_str(src_contents, line_length=line_length, is_pyi=is_pyi)
394 if src_contents == dst_contents:
398 assert_equivalent(src_contents, dst_contents)
400 src_contents, dst_contents, line_length=line_length, is_pyi=is_pyi
406 src_contents: str, line_length: int, *, is_pyi: bool = False
408 """Reformat a string and return new contents.
410 `line_length` determines how many characters per line are allowed.
412 src_node = lib2to3_parse(src_contents)
414 future_imports = get_future_imports(src_node)
415 elt = EmptyLineTracker(is_pyi=is_pyi)
416 py36 = is_python36(src_node)
417 lines = LineGenerator(
418 remove_u_prefix=py36 or "unicode_literals" in future_imports, is_pyi=is_pyi
422 for current_line in lines.visit(src_node):
423 for _ in range(after):
424 dst_contents += str(empty_line)
425 before, after = elt.maybe_empty_lines(current_line)
426 for _ in range(before):
427 dst_contents += str(empty_line)
428 for line in split_line(current_line, line_length=line_length, py36=py36):
429 dst_contents += str(line)
434 pygram.python_grammar_no_print_statement_no_exec_statement,
435 pygram.python_grammar_no_print_statement,
436 pygram.python_grammar,
440 def lib2to3_parse(src_txt: str) -> Node:
441 """Given a string with source, return the lib2to3 Node."""
442 grammar = pygram.python_grammar_no_print_statement
443 if src_txt[-1] != "\n":
444 nl = "\r\n" if "\r\n" in src_txt[:1024] else "\n"
446 for grammar in GRAMMARS:
447 drv = driver.Driver(grammar, pytree.convert)
449 result = drv.parse_string(src_txt, True)
452 except ParseError as pe:
453 lineno, column = pe.context[1]
454 lines = src_txt.splitlines()
456 faulty_line = lines[lineno - 1]
458 faulty_line = "<line number missing in source>"
459 exc = ValueError(f"Cannot parse: {lineno}:{column}: {faulty_line}")
463 if isinstance(result, Leaf):
464 result = Node(syms.file_input, [result])
468 def lib2to3_unparse(node: Node) -> str:
469 """Given a lib2to3 node, return its string representation."""
477 class Visitor(Generic[T]):
478 """Basic lib2to3 visitor that yields things of type `T` on `visit()`."""
480 def visit(self, node: LN) -> Iterator[T]:
481 """Main method to visit `node` and its children.
483 It tries to find a `visit_*()` method for the given `node.type`, like
484 `visit_simple_stmt` for Node objects or `visit_INDENT` for Leaf objects.
485 If no dedicated `visit_*()` method is found, chooses `visit_default()`
488 Then yields objects of type `T` from the selected visitor.
491 name = token.tok_name[node.type]
493 name = type_repr(node.type)
494 yield from getattr(self, f"visit_{name}", self.visit_default)(node)
496 def visit_default(self, node: LN) -> Iterator[T]:
497 """Default `visit_*()` implementation. Recurses to children of `node`."""
498 if isinstance(node, Node):
499 for child in node.children:
500 yield from self.visit(child)
504 class DebugVisitor(Visitor[T]):
507 def visit_default(self, node: LN) -> Iterator[T]:
508 indent = " " * (2 * self.tree_depth)
509 if isinstance(node, Node):
510 _type = type_repr(node.type)
511 out(f"{indent}{_type}", fg="yellow")
513 for child in node.children:
514 yield from self.visit(child)
517 out(f"{indent}/{_type}", fg="yellow", bold=False)
519 _type = token.tok_name.get(node.type, str(node.type))
520 out(f"{indent}{_type}", fg="blue", nl=False)
522 # We don't have to handle prefixes for `Node` objects since
523 # that delegates to the first child anyway.
524 out(f" {node.prefix!r}", fg="green", bold=False, nl=False)
525 out(f" {node.value!r}", fg="blue", bold=False)
528 def show(cls, code: str) -> None:
529 """Pretty-print the lib2to3 AST of a given string of `code`.
531 Convenience method for debugging.
533 v: DebugVisitor[None] = DebugVisitor()
534 list(v.visit(lib2to3_parse(code)))
537 KEYWORDS = set(keyword.kwlist)
538 WHITESPACE = {token.DEDENT, token.INDENT, token.NEWLINE}
539 FLOW_CONTROL = {"return", "raise", "break", "continue"}
550 STANDALONE_COMMENT = 153
551 LOGIC_OPERATORS = {"and", "or"}
576 STARS = {token.STAR, token.DOUBLESTAR}
579 syms.argument, # double star in arglist
580 syms.trailer, # single argument to call
582 syms.varargslist, # lambdas
584 UNPACKING_PARENTS = {
585 syms.atom, # single element of a list or set literal
623 COMPREHENSION_PRIORITY = 20
625 TERNARY_PRIORITY = 16
628 COMPARATOR_PRIORITY = 10
639 token.DOUBLESLASH: 4,
649 class BracketTracker:
650 """Keeps track of brackets on a line."""
653 bracket_match: Dict[Tuple[Depth, NodeType], Leaf] = Factory(dict)
654 delimiters: Dict[LeafID, Priority] = Factory(dict)
655 previous: Optional[Leaf] = None
656 _for_loop_variable: int = 0
657 _lambda_arguments: int = 0
659 def mark(self, leaf: Leaf) -> None:
660 """Mark `leaf` with bracket-related metadata. Keep track of delimiters.
662 All leaves receive an int `bracket_depth` field that stores how deep
663 within brackets a given leaf is. 0 means there are no enclosing brackets
664 that started on this line.
666 If a leaf is itself a closing bracket, it receives an `opening_bracket`
667 field that it forms a pair with. This is a one-directional link to
668 avoid reference cycles.
670 If a leaf is a delimiter (a token on which Black can split the line if
671 needed) and it's on depth 0, its `id()` is stored in the tracker's
674 if leaf.type == token.COMMENT:
677 self.maybe_decrement_after_for_loop_variable(leaf)
678 self.maybe_decrement_after_lambda_arguments(leaf)
679 if leaf.type in CLOSING_BRACKETS:
681 opening_bracket = self.bracket_match.pop((self.depth, leaf.type))
682 leaf.opening_bracket = opening_bracket
683 leaf.bracket_depth = self.depth
685 delim = is_split_before_delimiter(leaf, self.previous)
686 if delim and self.previous is not None:
687 self.delimiters[id(self.previous)] = delim
689 delim = is_split_after_delimiter(leaf, self.previous)
691 self.delimiters[id(leaf)] = delim
692 if leaf.type in OPENING_BRACKETS:
693 self.bracket_match[self.depth, BRACKET[leaf.type]] = leaf
696 self.maybe_increment_lambda_arguments(leaf)
697 self.maybe_increment_for_loop_variable(leaf)
699 def any_open_brackets(self) -> bool:
700 """Return True if there is an yet unmatched open bracket on the line."""
701 return bool(self.bracket_match)
703 def max_delimiter_priority(self, exclude: Iterable[LeafID] = ()) -> int:
704 """Return the highest priority of a delimiter found on the line.
706 Values are consistent with what `is_split_*_delimiter()` return.
707 Raises ValueError on no delimiters.
709 return max(v for k, v in self.delimiters.items() if k not in exclude)
711 def delimiter_count_with_priority(self, priority: int = 0) -> int:
712 """Return the number of delimiters with the given `priority`.
714 If no `priority` is passed, defaults to max priority on the line.
716 if not self.delimiters:
719 priority = priority or self.max_delimiter_priority()
720 return sum(1 for p in self.delimiters.values() if p == priority)
722 def maybe_increment_for_loop_variable(self, leaf: Leaf) -> bool:
723 """In a for loop, or comprehension, the variables are often unpacks.
725 To avoid splitting on the comma in this situation, increase the depth of
726 tokens between `for` and `in`.
728 if leaf.type == token.NAME and leaf.value == "for":
730 self._for_loop_variable += 1
735 def maybe_decrement_after_for_loop_variable(self, leaf: Leaf) -> bool:
736 """See `maybe_increment_for_loop_variable` above for explanation."""
737 if self._for_loop_variable and leaf.type == token.NAME and leaf.value == "in":
739 self._for_loop_variable -= 1
744 def maybe_increment_lambda_arguments(self, leaf: Leaf) -> bool:
745 """In a lambda expression, there might be more than one argument.
747 To avoid splitting on the comma in this situation, increase the depth of
748 tokens between `lambda` and `:`.
750 if leaf.type == token.NAME and leaf.value == "lambda":
752 self._lambda_arguments += 1
757 def maybe_decrement_after_lambda_arguments(self, leaf: Leaf) -> bool:
758 """See `maybe_increment_lambda_arguments` above for explanation."""
759 if self._lambda_arguments and leaf.type == token.COLON:
761 self._lambda_arguments -= 1
766 def get_open_lsqb(self) -> Optional[Leaf]:
767 """Return the most recent opening square bracket (if any)."""
768 return self.bracket_match.get((self.depth - 1, token.RSQB))
773 """Holds leaves and comments. Can be printed with `str(line)`."""
776 leaves: List[Leaf] = Factory(list)
777 comments: List[Tuple[Index, Leaf]] = Factory(list)
778 bracket_tracker: BracketTracker = Factory(BracketTracker)
779 inside_brackets: bool = False
780 should_explode: bool = False
782 def append(self, leaf: Leaf, preformatted: bool = False) -> None:
783 """Add a new `leaf` to the end of the line.
785 Unless `preformatted` is True, the `leaf` will receive a new consistent
786 whitespace prefix and metadata applied by :class:`BracketTracker`.
787 Trailing commas are maybe removed, unpacked for loop variables are
788 demoted from being delimiters.
790 Inline comments are put aside.
792 has_value = leaf.type in BRACKETS or bool(leaf.value.strip())
796 if token.COLON == leaf.type and self.is_class_paren_empty:
798 if self.leaves and not preformatted:
799 # Note: at this point leaf.prefix should be empty except for
800 # imports, for which we only preserve newlines.
801 leaf.prefix += whitespace(
802 leaf, complex_subscript=self.is_complex_subscript(leaf)
804 if self.inside_brackets or not preformatted:
805 self.bracket_tracker.mark(leaf)
806 self.maybe_remove_trailing_comma(leaf)
807 if not self.append_comment(leaf):
808 self.leaves.append(leaf)
810 def append_safe(self, leaf: Leaf, preformatted: bool = False) -> None:
811 """Like :func:`append()` but disallow invalid standalone comment structure.
813 Raises ValueError when any `leaf` is appended after a standalone comment
814 or when a standalone comment is not the first leaf on the line.
816 if self.bracket_tracker.depth == 0:
818 raise ValueError("cannot append to standalone comments")
820 if self.leaves and leaf.type == STANDALONE_COMMENT:
822 "cannot append standalone comments to a populated line"
825 self.append(leaf, preformatted=preformatted)
828 def is_comment(self) -> bool:
829 """Is this line a standalone comment?"""
830 return len(self.leaves) == 1 and self.leaves[0].type == STANDALONE_COMMENT
833 def is_decorator(self) -> bool:
834 """Is this line a decorator?"""
835 return bool(self) and self.leaves[0].type == token.AT
838 def is_import(self) -> bool:
839 """Is this an import line?"""
840 return bool(self) and is_import(self.leaves[0])
843 def is_class(self) -> bool:
844 """Is this line a class definition?"""
847 and self.leaves[0].type == token.NAME
848 and self.leaves[0].value == "class"
852 def is_stub_class(self) -> bool:
853 """Is this line a class definition with a body consisting only of "..."?"""
854 return self.is_class and self.leaves[-3:] == [
855 Leaf(token.DOT, ".") for _ in range(3)
859 def is_def(self) -> bool:
860 """Is this a function definition? (Also returns True for async defs.)"""
862 first_leaf = self.leaves[0]
867 second_leaf: Optional[Leaf] = self.leaves[1]
870 return (first_leaf.type == token.NAME and first_leaf.value == "def") or (
871 first_leaf.type == token.ASYNC
872 and second_leaf is not None
873 and second_leaf.type == token.NAME
874 and second_leaf.value == "def"
878 def is_flow_control(self) -> bool:
879 """Is this line a flow control statement?
881 Those are `return`, `raise`, `break`, and `continue`.
885 and self.leaves[0].type == token.NAME
886 and self.leaves[0].value in FLOW_CONTROL
890 def is_yield(self) -> bool:
891 """Is this line a yield statement?"""
894 and self.leaves[0].type == token.NAME
895 and self.leaves[0].value == "yield"
899 def is_class_paren_empty(self) -> bool:
900 """Is this a class with no base classes but using parentheses?
902 Those are unnecessary and should be removed.
906 and len(self.leaves) == 4
908 and self.leaves[2].type == token.LPAR
909 and self.leaves[2].value == "("
910 and self.leaves[3].type == token.RPAR
911 and self.leaves[3].value == ")"
914 def contains_standalone_comments(self, depth_limit: int = sys.maxsize) -> bool:
915 """If so, needs to be split before emitting."""
916 for leaf in self.leaves:
917 if leaf.type == STANDALONE_COMMENT:
918 if leaf.bracket_depth <= depth_limit:
923 def maybe_remove_trailing_comma(self, closing: Leaf) -> bool:
924 """Remove trailing comma if there is one and it's safe."""
927 and self.leaves[-1].type == token.COMMA
928 and closing.type in CLOSING_BRACKETS
932 if closing.type == token.RBRACE:
933 self.remove_trailing_comma()
936 if closing.type == token.RSQB:
937 comma = self.leaves[-1]
938 if comma.parent and comma.parent.type == syms.listmaker:
939 self.remove_trailing_comma()
942 # For parens let's check if it's safe to remove the comma.
943 # Imports are always safe.
945 self.remove_trailing_comma()
948 # Otheriwsse, if the trailing one is the only one, we might mistakenly
949 # change a tuple into a different type by removing the comma.
950 depth = closing.bracket_depth + 1
952 opening = closing.opening_bracket
953 for _opening_index, leaf in enumerate(self.leaves):
960 for leaf in self.leaves[_opening_index + 1 :]:
964 bracket_depth = leaf.bracket_depth
965 if bracket_depth == depth and leaf.type == token.COMMA:
967 if leaf.parent and leaf.parent.type == syms.arglist:
972 self.remove_trailing_comma()
977 def append_comment(self, comment: Leaf) -> bool:
978 """Add an inline or standalone comment to the line."""
980 comment.type == STANDALONE_COMMENT
981 and self.bracket_tracker.any_open_brackets()
986 if comment.type != token.COMMENT:
989 after = len(self.leaves) - 1
991 comment.type = STANDALONE_COMMENT
996 self.comments.append((after, comment))
999 def comments_after(self, leaf: Leaf, _index: int = -1) -> Iterator[Leaf]:
1000 """Generate comments that should appear directly after `leaf`.
1002 Provide a non-negative leaf `_index` to speed up the function.
1005 for _index, _leaf in enumerate(self.leaves):
1012 for index, comment_after in self.comments:
1016 def remove_trailing_comma(self) -> None:
1017 """Remove the trailing comma and moves the comments attached to it."""
1018 comma_index = len(self.leaves) - 1
1019 for i in range(len(self.comments)):
1020 comment_index, comment = self.comments[i]
1021 if comment_index == comma_index:
1022 self.comments[i] = (comma_index - 1, comment)
1025 def is_complex_subscript(self, leaf: Leaf) -> bool:
1026 """Return True iff `leaf` is part of a slice with non-trivial exprs."""
1028 leaf if leaf.type == token.LSQB else self.bracket_tracker.get_open_lsqb()
1030 if open_lsqb is None:
1033 subscript_start = open_lsqb.next_sibling
1035 isinstance(subscript_start, Node)
1036 and subscript_start.type == syms.subscriptlist
1038 subscript_start = child_towards(subscript_start, leaf)
1039 return subscript_start is not None and any(
1040 n.type in TEST_DESCENDANTS for n in subscript_start.pre_order()
1043 def __str__(self) -> str:
1044 """Render the line."""
1048 indent = " " * self.depth
1049 leaves = iter(self.leaves)
1050 first = next(leaves)
1051 res = f"{first.prefix}{indent}{first.value}"
1054 for _, comment in self.comments:
1058 def __bool__(self) -> bool:
1059 """Return True if the line has leaves or comments."""
1060 return bool(self.leaves or self.comments)
1063 class UnformattedLines(Line):
1064 """Just like :class:`Line` but stores lines which aren't reformatted."""
1066 def append(self, leaf: Leaf, preformatted: bool = True) -> None:
1067 """Just add a new `leaf` to the end of the lines.
1069 The `preformatted` argument is ignored.
1071 Keeps track of indentation `depth`, which is useful when the user
1072 says `# fmt: on`. Otherwise, doesn't do anything with the `leaf`.
1075 list(generate_comments(leaf))
1076 except FormatOn as f_on:
1077 self.leaves.append(f_on.leaf_from_consumed(leaf))
1080 self.leaves.append(leaf)
1081 if leaf.type == token.INDENT:
1083 elif leaf.type == token.DEDENT:
1086 def __str__(self) -> str:
1087 """Render unformatted lines from leaves which were added with `append()`.
1089 `depth` is not used for indentation in this case.
1095 for leaf in self.leaves:
1099 def append_comment(self, comment: Leaf) -> bool:
1100 """Not implemented in this class. Raises `NotImplementedError`."""
1101 raise NotImplementedError("Unformatted lines don't store comments separately.")
1103 def maybe_remove_trailing_comma(self, closing: Leaf) -> bool:
1104 """Does nothing and returns False."""
1107 def maybe_increment_for_loop_variable(self, leaf: Leaf) -> bool:
1108 """Does nothing and returns False."""
1113 class EmptyLineTracker:
1114 """Provides a stateful method that returns the number of potential extra
1115 empty lines needed before and after the currently processed line.
1117 Note: this tracker works on lines that haven't been split yet. It assumes
1118 the prefix of the first leaf consists of optional newlines. Those newlines
1119 are consumed by `maybe_empty_lines()` and included in the computation.
1121 is_pyi: bool = False
1122 previous_line: Optional[Line] = None
1123 previous_after: int = 0
1124 previous_defs: List[int] = Factory(list)
1126 def maybe_empty_lines(self, current_line: Line) -> Tuple[int, int]:
1127 """Return the number of extra empty lines before and after the `current_line`.
1129 This is for separating `def`, `async def` and `class` with extra empty
1130 lines (two on module-level), as well as providing an extra empty line
1131 after flow control keywords to make them more prominent.
1133 if isinstance(current_line, UnformattedLines):
1136 before, after = self._maybe_empty_lines(current_line)
1137 before -= self.previous_after
1138 self.previous_after = after
1139 self.previous_line = current_line
1140 return before, after
1142 def _maybe_empty_lines(self, current_line: Line) -> Tuple[int, int]:
1144 if current_line.depth == 0:
1145 max_allowed = 1 if self.is_pyi else 2
1146 if current_line.leaves:
1147 # Consume the first leaf's extra newlines.
1148 first_leaf = current_line.leaves[0]
1149 before = first_leaf.prefix.count("\n")
1150 before = min(before, max_allowed)
1151 first_leaf.prefix = ""
1154 depth = current_line.depth
1155 while self.previous_defs and self.previous_defs[-1] >= depth:
1156 self.previous_defs.pop()
1158 before = 0 if depth else 1
1160 before = 1 if depth else 2
1161 is_decorator = current_line.is_decorator
1162 if is_decorator or current_line.is_def or current_line.is_class:
1163 if not is_decorator:
1164 self.previous_defs.append(depth)
1165 if self.previous_line is None:
1166 # Don't insert empty lines before the first line in the file.
1169 if self.previous_line.is_decorator:
1173 self.previous_line.is_comment
1174 and self.previous_line.depth == current_line.depth
1180 if self.previous_line.depth > current_line.depth:
1182 elif current_line.is_class or self.previous_line.is_class:
1183 if current_line.is_stub_class and self.previous_line.is_stub_class:
1191 if current_line.depth and newlines:
1197 and self.previous_line.is_import
1198 and not current_line.is_import
1199 and depth == self.previous_line.depth
1201 return (before or 1), 0
1207 class LineGenerator(Visitor[Line]):
1208 """Generates reformatted Line objects. Empty lines are not emitted.
1210 Note: destroys the tree it's visiting by mutating prefixes of its leaves
1211 in ways that will no longer stringify to valid Python code on the tree.
1213 is_pyi: bool = False
1214 current_line: Line = Factory(Line)
1215 remove_u_prefix: bool = False
1217 def line(self, indent: int = 0, type: Type[Line] = Line) -> Iterator[Line]:
1220 If the line is empty, only emit if it makes sense.
1221 If the line is too long, split it first and then generate.
1223 If any lines were generated, set up a new current_line.
1225 if not self.current_line:
1226 if self.current_line.__class__ == type:
1227 self.current_line.depth += indent
1229 self.current_line = type(depth=self.current_line.depth + indent)
1230 return # Line is empty, don't emit. Creating a new one unnecessary.
1232 complete_line = self.current_line
1233 self.current_line = type(depth=complete_line.depth + indent)
1236 def visit(self, node: LN) -> Iterator[Line]:
1237 """Main method to visit `node` and its children.
1239 Yields :class:`Line` objects.
1241 if isinstance(self.current_line, UnformattedLines):
1242 # File contained `# fmt: off`
1243 yield from self.visit_unformatted(node)
1246 yield from super().visit(node)
1248 def visit_default(self, node: LN) -> Iterator[Line]:
1249 """Default `visit_*()` implementation. Recurses to children of `node`."""
1250 if isinstance(node, Leaf):
1251 any_open_brackets = self.current_line.bracket_tracker.any_open_brackets()
1253 for comment in generate_comments(node):
1254 if any_open_brackets:
1255 # any comment within brackets is subject to splitting
1256 self.current_line.append(comment)
1257 elif comment.type == token.COMMENT:
1258 # regular trailing comment
1259 self.current_line.append(comment)
1260 yield from self.line()
1263 # regular standalone comment
1264 yield from self.line()
1266 self.current_line.append(comment)
1267 yield from self.line()
1269 except FormatOff as f_off:
1270 f_off.trim_prefix(node)
1271 yield from self.line(type=UnformattedLines)
1272 yield from self.visit(node)
1274 except FormatOn as f_on:
1275 # This only happens here if somebody says "fmt: on" multiple
1277 f_on.trim_prefix(node)
1278 yield from self.visit_default(node)
1281 normalize_prefix(node, inside_brackets=any_open_brackets)
1282 if node.type == token.STRING:
1283 normalize_string_prefix(node, remove_u_prefix=self.remove_u_prefix)
1284 normalize_string_quotes(node)
1285 if node.type not in WHITESPACE:
1286 self.current_line.append(node)
1287 yield from super().visit_default(node)
1289 def visit_INDENT(self, node: Node) -> Iterator[Line]:
1290 """Increase indentation level, maybe yield a line."""
1291 # In blib2to3 INDENT never holds comments.
1292 yield from self.line(+1)
1293 yield from self.visit_default(node)
1295 def visit_DEDENT(self, node: Node) -> Iterator[Line]:
1296 """Decrease indentation level, maybe yield a line."""
1297 # The current line might still wait for trailing comments. At DEDENT time
1298 # there won't be any (they would be prefixes on the preceding NEWLINE).
1299 # Emit the line then.
1300 yield from self.line()
1302 # While DEDENT has no value, its prefix may contain standalone comments
1303 # that belong to the current indentation level. Get 'em.
1304 yield from self.visit_default(node)
1306 # Finally, emit the dedent.
1307 yield from self.line(-1)
1310 self, node: Node, keywords: Set[str], parens: Set[str]
1311 ) -> Iterator[Line]:
1312 """Visit a statement.
1314 This implementation is shared for `if`, `while`, `for`, `try`, `except`,
1315 `def`, `with`, `class`, `assert` and assignments.
1317 The relevant Python language `keywords` for a given statement will be
1318 NAME leaves within it. This methods puts those on a separate line.
1320 `parens` holds a set of string leaf values immediately after which
1321 invisible parens should be put.
1323 normalize_invisible_parens(node, parens_after=parens)
1324 for child in node.children:
1325 if child.type == token.NAME and child.value in keywords: # type: ignore
1326 yield from self.line()
1328 yield from self.visit(child)
1330 def visit_suite(self, node: Node) -> Iterator[Line]:
1331 """Visit a suite."""
1332 if self.is_pyi and is_stub_suite(node):
1333 yield from self.visit(node.children[2])
1335 yield from self.visit_default(node)
1337 def visit_simple_stmt(self, node: Node) -> Iterator[Line]:
1338 """Visit a statement without nested statements."""
1339 is_suite_like = node.parent and node.parent.type in STATEMENT
1341 if self.is_pyi and is_stub_body(node):
1342 yield from self.visit_default(node)
1344 yield from self.line(+1)
1345 yield from self.visit_default(node)
1346 yield from self.line(-1)
1349 if not self.is_pyi or not node.parent or not is_stub_suite(node.parent):
1350 yield from self.line()
1351 yield from self.visit_default(node)
1353 def visit_async_stmt(self, node: Node) -> Iterator[Line]:
1354 """Visit `async def`, `async for`, `async with`."""
1355 yield from self.line()
1357 children = iter(node.children)
1358 for child in children:
1359 yield from self.visit(child)
1361 if child.type == token.ASYNC:
1364 internal_stmt = next(children)
1365 for child in internal_stmt.children:
1366 yield from self.visit(child)
1368 def visit_decorators(self, node: Node) -> Iterator[Line]:
1369 """Visit decorators."""
1370 for child in node.children:
1371 yield from self.line()
1372 yield from self.visit(child)
1374 def visit_import_from(self, node: Node) -> Iterator[Line]:
1375 """Visit import_from and maybe put invisible parentheses.
1377 This is separate from `visit_stmt` because import statements don't
1378 support arbitrary atoms and thus handling of parentheses is custom.
1381 for index, child in enumerate(node.children):
1383 if child.type == token.LPAR:
1384 # make parentheses invisible
1385 child.value = "" # type: ignore
1386 node.children[-1].value = "" # type: ignore
1388 # insert invisible parentheses
1389 node.insert_child(index, Leaf(token.LPAR, ""))
1390 node.append_child(Leaf(token.RPAR, ""))
1394 child.type == token.NAME and child.value == "import" # type: ignore
1397 for child in node.children:
1398 yield from self.visit(child)
1400 def visit_SEMI(self, leaf: Leaf) -> Iterator[Line]:
1401 """Remove a semicolon and put the other statement on a separate line."""
1402 yield from self.line()
1404 def visit_ENDMARKER(self, leaf: Leaf) -> Iterator[Line]:
1405 """End of file. Process outstanding comments and end with a newline."""
1406 yield from self.visit_default(leaf)
1407 yield from self.line()
1409 def visit_unformatted(self, node: LN) -> Iterator[Line]:
1410 """Used when file contained a `# fmt: off`."""
1411 if isinstance(node, Node):
1412 for child in node.children:
1413 yield from self.visit(child)
1417 self.current_line.append(node)
1418 except FormatOn as f_on:
1419 f_on.trim_prefix(node)
1420 yield from self.line()
1421 yield from self.visit(node)
1423 if node.type == token.ENDMARKER:
1424 # somebody decided not to put a final `# fmt: on`
1425 yield from self.line()
1427 def __attrs_post_init__(self) -> None:
1428 """You are in a twisty little maze of passages."""
1431 self.visit_assert_stmt = partial(v, keywords={"assert"}, parens={"assert", ","})
1432 self.visit_if_stmt = partial(
1433 v, keywords={"if", "else", "elif"}, parens={"if", "elif"}
1435 self.visit_while_stmt = partial(v, keywords={"while", "else"}, parens={"while"})
1436 self.visit_for_stmt = partial(v, keywords={"for", "else"}, parens={"for", "in"})
1437 self.visit_try_stmt = partial(
1438 v, keywords={"try", "except", "else", "finally"}, parens=Ø
1440 self.visit_except_clause = partial(v, keywords={"except"}, parens=Ø)
1441 self.visit_with_stmt = partial(v, keywords={"with"}, parens=Ø)
1442 self.visit_funcdef = partial(v, keywords={"def"}, parens=Ø)
1443 self.visit_classdef = partial(v, keywords={"class"}, parens=Ø)
1444 self.visit_expr_stmt = partial(v, keywords=Ø, parens=ASSIGNMENTS)
1445 self.visit_return_stmt = partial(v, keywords={"return"}, parens={"return"})
1446 self.visit_async_funcdef = self.visit_async_stmt
1447 self.visit_decorated = self.visit_decorators
1450 IMPLICIT_TUPLE = {syms.testlist, syms.testlist_star_expr, syms.exprlist}
1451 BRACKET = {token.LPAR: token.RPAR, token.LSQB: token.RSQB, token.LBRACE: token.RBRACE}
1452 OPENING_BRACKETS = set(BRACKET.keys())
1453 CLOSING_BRACKETS = set(BRACKET.values())
1454 BRACKETS = OPENING_BRACKETS | CLOSING_BRACKETS
1455 ALWAYS_NO_SPACE = CLOSING_BRACKETS | {token.COMMA, STANDALONE_COMMENT}
1458 def whitespace(leaf: Leaf, *, complex_subscript: bool) -> str: # noqa C901
1459 """Return whitespace prefix if needed for the given `leaf`.
1461 `complex_subscript` signals whether the given leaf is part of a subscription
1462 which has non-trivial arguments, like arithmetic expressions or function calls.
1470 if t in ALWAYS_NO_SPACE:
1473 if t == token.COMMENT:
1476 assert p is not None, f"INTERNAL ERROR: hand-made leaf without parent: {leaf!r}"
1477 if t == token.COLON and p.type not in {
1484 prev = leaf.prev_sibling
1486 prevp = preceding_leaf(p)
1487 if not prevp or prevp.type in OPENING_BRACKETS:
1490 if t == token.COLON:
1491 if prevp.type == token.COLON:
1494 elif prevp.type != token.COMMA and not complex_subscript:
1499 if prevp.type == token.EQUAL:
1501 if prevp.parent.type in {
1509 elif prevp.parent.type == syms.typedargslist:
1510 # A bit hacky: if the equal sign has whitespace, it means we
1511 # previously found it's a typed argument. So, we're using
1515 elif prevp.type in STARS:
1516 if is_vararg(prevp, within=VARARGS_PARENTS | UNPACKING_PARENTS):
1519 elif prevp.type == token.COLON:
1520 if prevp.parent and prevp.parent.type in {syms.subscript, syms.sliceop}:
1521 return SPACE if complex_subscript else NO
1525 and prevp.parent.type == syms.factor
1526 and prevp.type in MATH_OPERATORS
1531 prevp.type == token.RIGHTSHIFT
1533 and prevp.parent.type == syms.shift_expr
1534 and prevp.prev_sibling
1535 and prevp.prev_sibling.type == token.NAME
1536 and prevp.prev_sibling.value == "print" # type: ignore
1538 # Python 2 print chevron
1541 elif prev.type in OPENING_BRACKETS:
1544 if p.type in {syms.parameters, syms.arglist}:
1545 # untyped function signatures or calls
1546 if not prev or prev.type != token.COMMA:
1549 elif p.type == syms.varargslist:
1551 if prev and prev.type != token.COMMA:
1554 elif p.type == syms.typedargslist:
1555 # typed function signatures
1559 if t == token.EQUAL:
1560 if prev.type != syms.tname:
1563 elif prev.type == token.EQUAL:
1564 # A bit hacky: if the equal sign has whitespace, it means we
1565 # previously found it's a typed argument. So, we're using that, too.
1568 elif prev.type != token.COMMA:
1571 elif p.type == syms.tname:
1574 prevp = preceding_leaf(p)
1575 if not prevp or prevp.type != token.COMMA:
1578 elif p.type == syms.trailer:
1579 # attributes and calls
1580 if t == token.LPAR or t == token.RPAR:
1585 prevp = preceding_leaf(p)
1586 if not prevp or prevp.type != token.NUMBER:
1589 elif t == token.LSQB:
1592 elif prev.type != token.COMMA:
1595 elif p.type == syms.argument:
1597 if t == token.EQUAL:
1601 prevp = preceding_leaf(p)
1602 if not prevp or prevp.type == token.LPAR:
1605 elif prev.type in {token.EQUAL} | STARS:
1608 elif p.type == syms.decorator:
1612 elif p.type == syms.dotted_name:
1616 prevp = preceding_leaf(p)
1617 if not prevp or prevp.type == token.AT or prevp.type == token.DOT:
1620 elif p.type == syms.classdef:
1624 if prev and prev.type == token.LPAR:
1627 elif p.type in {syms.subscript, syms.sliceop}:
1630 assert p.parent is not None, "subscripts are always parented"
1631 if p.parent.type == syms.subscriptlist:
1636 elif not complex_subscript:
1639 elif p.type == syms.atom:
1640 if prev and t == token.DOT:
1641 # dots, but not the first one.
1644 elif p.type == syms.dictsetmaker:
1646 if prev and prev.type == token.DOUBLESTAR:
1649 elif p.type in {syms.factor, syms.star_expr}:
1652 prevp = preceding_leaf(p)
1653 if not prevp or prevp.type in OPENING_BRACKETS:
1656 prevp_parent = prevp.parent
1657 assert prevp_parent is not None
1658 if prevp.type == token.COLON and prevp_parent.type in {
1664 elif prevp.type == token.EQUAL and prevp_parent.type == syms.argument:
1667 elif t == token.NAME or t == token.NUMBER:
1670 elif p.type == syms.import_from:
1672 if prev and prev.type == token.DOT:
1675 elif t == token.NAME:
1679 if prev and prev.type == token.DOT:
1682 elif p.type == syms.sliceop:
1688 def preceding_leaf(node: Optional[LN]) -> Optional[Leaf]:
1689 """Return the first leaf that precedes `node`, if any."""
1691 res = node.prev_sibling
1693 if isinstance(res, Leaf):
1697 return list(res.leaves())[-1]
1706 def child_towards(ancestor: Node, descendant: LN) -> Optional[LN]:
1707 """Return the child of `ancestor` that contains `descendant`."""
1708 node: Optional[LN] = descendant
1709 while node and node.parent != ancestor:
1714 def is_split_after_delimiter(leaf: Leaf, previous: Leaf = None) -> int:
1715 """Return the priority of the `leaf` delimiter, given a line break after it.
1717 The delimiter priorities returned here are from those delimiters that would
1718 cause a line break after themselves.
1720 Higher numbers are higher priority.
1722 if leaf.type == token.COMMA:
1723 return COMMA_PRIORITY
1728 def is_split_before_delimiter(leaf: Leaf, previous: Leaf = None) -> int:
1729 """Return the priority of the `leaf` delimiter, given a line before after it.
1731 The delimiter priorities returned here are from those delimiters that would
1732 cause a line break before themselves.
1734 Higher numbers are higher priority.
1736 if is_vararg(leaf, within=VARARGS_PARENTS | UNPACKING_PARENTS):
1737 # * and ** might also be MATH_OPERATORS but in this case they are not.
1738 # Don't treat them as a delimiter.
1742 leaf.type == token.DOT
1744 and leaf.parent.type not in {syms.import_from, syms.dotted_name}
1745 and (previous is None or previous.type in CLOSING_BRACKETS)
1750 leaf.type in MATH_OPERATORS
1752 and leaf.parent.type not in {syms.factor, syms.star_expr}
1754 return MATH_PRIORITIES[leaf.type]
1756 if leaf.type in COMPARATORS:
1757 return COMPARATOR_PRIORITY
1760 leaf.type == token.STRING
1761 and previous is not None
1762 and previous.type == token.STRING
1764 return STRING_PRIORITY
1766 if leaf.type != token.NAME:
1772 and leaf.parent.type in {syms.comp_for, syms.old_comp_for}
1774 return COMPREHENSION_PRIORITY
1779 and leaf.parent.type in {syms.comp_if, syms.old_comp_if}
1781 return COMPREHENSION_PRIORITY
1783 if leaf.value in {"if", "else"} and leaf.parent and leaf.parent.type == syms.test:
1784 return TERNARY_PRIORITY
1786 if leaf.value == "is":
1787 return COMPARATOR_PRIORITY
1792 and leaf.parent.type in {syms.comp_op, syms.comparison}
1794 previous is not None
1795 and previous.type == token.NAME
1796 and previous.value == "not"
1799 return COMPARATOR_PRIORITY
1804 and leaf.parent.type == syms.comp_op
1806 previous is not None
1807 and previous.type == token.NAME
1808 and previous.value == "is"
1811 return COMPARATOR_PRIORITY
1813 if leaf.value in LOGIC_OPERATORS and leaf.parent:
1814 return LOGIC_PRIORITY
1819 def generate_comments(leaf: Leaf) -> Iterator[Leaf]:
1820 """Clean the prefix of the `leaf` and generate comments from it, if any.
1822 Comments in lib2to3 are shoved into the whitespace prefix. This happens
1823 in `pgen2/driver.py:Driver.parse_tokens()`. This was a brilliant implementation
1824 move because it does away with modifying the grammar to include all the
1825 possible places in which comments can be placed.
1827 The sad consequence for us though is that comments don't "belong" anywhere.
1828 This is why this function generates simple parentless Leaf objects for
1829 comments. We simply don't know what the correct parent should be.
1831 No matter though, we can live without this. We really only need to
1832 differentiate between inline and standalone comments. The latter don't
1833 share the line with any code.
1835 Inline comments are emitted as regular token.COMMENT leaves. Standalone
1836 are emitted with a fake STANDALONE_COMMENT token identifier.
1847 for index, line in enumerate(p.split("\n")):
1848 consumed += len(line) + 1 # adding the length of the split '\n'
1849 line = line.lstrip()
1852 if not line.startswith("#"):
1855 if index == 0 and leaf.type != token.ENDMARKER:
1856 comment_type = token.COMMENT # simple trailing comment
1858 comment_type = STANDALONE_COMMENT
1859 comment = make_comment(line)
1860 yield Leaf(comment_type, comment, prefix="\n" * nlines)
1862 if comment in {"# fmt: on", "# yapf: enable"}:
1863 raise FormatOn(consumed)
1865 if comment in {"# fmt: off", "# yapf: disable"}:
1866 if comment_type == STANDALONE_COMMENT:
1867 raise FormatOff(consumed)
1869 prev = preceding_leaf(leaf)
1870 if not prev or prev.type in WHITESPACE: # standalone comment in disguise
1871 raise FormatOff(consumed)
1876 def make_comment(content: str) -> str:
1877 """Return a consistently formatted comment from the given `content` string.
1879 All comments (except for "##", "#!", "#:") should have a single space between
1880 the hash sign and the content.
1882 If `content` didn't start with a hash sign, one is provided.
1884 content = content.rstrip()
1888 if content[0] == "#":
1889 content = content[1:]
1890 if content and content[0] not in " !:#":
1891 content = " " + content
1892 return "#" + content
1896 line: Line, line_length: int, inner: bool = False, py36: bool = False
1897 ) -> Iterator[Line]:
1898 """Split a `line` into potentially many lines.
1900 They should fit in the allotted `line_length` but might not be able to.
1901 `inner` signifies that there were a pair of brackets somewhere around the
1902 current `line`, possibly transitively. This means we can fallback to splitting
1903 by delimiters if the LHS/RHS don't yield any results.
1905 If `py36` is True, splitting may generate syntax that is only compatible
1906 with Python 3.6 and later.
1908 if isinstance(line, UnformattedLines) or line.is_comment:
1912 line_str = str(line).strip("\n")
1913 if not line.should_explode and is_line_short_enough(
1914 line, line_length=line_length, line_str=line_str
1919 split_funcs: List[SplitFunc]
1921 split_funcs = [left_hand_split]
1924 def rhs(line: Line, py36: bool = False) -> Iterator[Line]:
1925 for omit in generate_trailers_to_omit(line, line_length):
1926 lines = list(right_hand_split(line, line_length, py36, omit=omit))
1927 if is_line_short_enough(lines[0], line_length=line_length):
1931 # All splits failed, best effort split with no omits.
1932 # This mostly happens to multiline strings that are by definition
1933 # reported as not fitting a single line.
1934 yield from right_hand_split(line, py36)
1936 if line.inside_brackets:
1937 split_funcs = [delimiter_split, standalone_comment_split, rhs]
1940 for split_func in split_funcs:
1941 # We are accumulating lines in `result` because we might want to abort
1942 # mission and return the original line in the end, or attempt a different
1944 result: List[Line] = []
1946 for l in split_func(line, py36):
1947 if str(l).strip("\n") == line_str:
1948 raise CannotSplit("Split function returned an unchanged result")
1951 split_line(l, line_length=line_length, inner=True, py36=py36)
1953 except CannotSplit as cs:
1964 def left_hand_split(line: Line, py36: bool = False) -> Iterator[Line]:
1965 """Split line into many lines, starting with the first matching bracket pair.
1967 Note: this usually looks weird, only use this for function definitions.
1968 Prefer RHS otherwise. This is why this function is not symmetrical with
1969 :func:`right_hand_split` which also handles optional parentheses.
1971 head = Line(depth=line.depth)
1972 body = Line(depth=line.depth + 1, inside_brackets=True)
1973 tail = Line(depth=line.depth)
1974 tail_leaves: List[Leaf] = []
1975 body_leaves: List[Leaf] = []
1976 head_leaves: List[Leaf] = []
1977 current_leaves = head_leaves
1978 matching_bracket = None
1979 for leaf in line.leaves:
1981 current_leaves is body_leaves
1982 and leaf.type in CLOSING_BRACKETS
1983 and leaf.opening_bracket is matching_bracket
1985 current_leaves = tail_leaves if body_leaves else head_leaves
1986 current_leaves.append(leaf)
1987 if current_leaves is head_leaves:
1988 if leaf.type in OPENING_BRACKETS:
1989 matching_bracket = leaf
1990 current_leaves = body_leaves
1991 # Since body is a new indent level, remove spurious leading whitespace.
1993 normalize_prefix(body_leaves[0], inside_brackets=True)
1994 # Build the new lines.
1995 for result, leaves in (head, head_leaves), (body, body_leaves), (tail, tail_leaves):
1997 result.append(leaf, preformatted=True)
1998 for comment_after in line.comments_after(leaf):
1999 result.append(comment_after, preformatted=True)
2000 bracket_split_succeeded_or_raise(head, body, tail)
2001 for result in (head, body, tail):
2006 def right_hand_split(
2007 line: Line, line_length: int, py36: bool = False, omit: Collection[LeafID] = ()
2008 ) -> Iterator[Line]:
2009 """Split line into many lines, starting with the last matching bracket pair.
2011 If the split was by optional parentheses, attempt splitting without them, too.
2012 `omit` is a collection of closing bracket IDs that shouldn't be considered for
2015 Note: running this function modifies `bracket_depth` on the leaves of `line`.
2017 head = Line(depth=line.depth)
2018 body = Line(depth=line.depth + 1, inside_brackets=True)
2019 tail = Line(depth=line.depth)
2020 tail_leaves: List[Leaf] = []
2021 body_leaves: List[Leaf] = []
2022 head_leaves: List[Leaf] = []
2023 current_leaves = tail_leaves
2024 opening_bracket = None
2025 closing_bracket = None
2026 for leaf in reversed(line.leaves):
2027 if current_leaves is body_leaves:
2028 if leaf is opening_bracket:
2029 current_leaves = head_leaves if body_leaves else tail_leaves
2030 current_leaves.append(leaf)
2031 if current_leaves is tail_leaves:
2032 if leaf.type in CLOSING_BRACKETS and id(leaf) not in omit:
2033 opening_bracket = leaf.opening_bracket
2034 closing_bracket = leaf
2035 current_leaves = body_leaves
2036 tail_leaves.reverse()
2037 body_leaves.reverse()
2038 head_leaves.reverse()
2039 # Since body is a new indent level, remove spurious leading whitespace.
2041 normalize_prefix(body_leaves[0], inside_brackets=True)
2043 # No `head` means the split failed. Either `tail` has all content or
2044 # the matching `opening_bracket` wasn't available on `line` anymore.
2045 raise CannotSplit("No brackets found")
2047 # Build the new lines.
2048 for result, leaves in (head, head_leaves), (body, body_leaves), (tail, tail_leaves):
2050 result.append(leaf, preformatted=True)
2051 for comment_after in line.comments_after(leaf):
2052 result.append(comment_after, preformatted=True)
2053 bracket_split_succeeded_or_raise(head, body, tail)
2054 assert opening_bracket and closing_bracket
2056 # the opening bracket is an optional paren
2057 opening_bracket.type == token.LPAR
2058 and not opening_bracket.value
2059 # the closing bracket is an optional paren
2060 and closing_bracket.type == token.RPAR
2061 and not closing_bracket.value
2062 # there are no standalone comments in the body
2063 and not line.contains_standalone_comments(0)
2064 # and it's not an import (optional parens are the only thing we can split
2065 # on in this case; attempting a split without them is a waste of time)
2066 and not line.is_import
2068 omit = {id(closing_bracket), *omit}
2069 if can_omit_invisible_parens(body, line_length):
2071 yield from right_hand_split(line, line_length, py36=py36, omit=omit)
2076 ensure_visible(opening_bracket)
2077 ensure_visible(closing_bracket)
2078 body.should_explode = should_explode(body, opening_bracket)
2079 for result in (head, body, tail):
2084 def bracket_split_succeeded_or_raise(head: Line, body: Line, tail: Line) -> None:
2085 """Raise :exc:`CannotSplit` if the last left- or right-hand split failed.
2087 Do nothing otherwise.
2089 A left- or right-hand split is based on a pair of brackets. Content before
2090 (and including) the opening bracket is left on one line, content inside the
2091 brackets is put on a separate line, and finally content starting with and
2092 following the closing bracket is put on a separate line.
2094 Those are called `head`, `body`, and `tail`, respectively. If the split
2095 produced the same line (all content in `head`) or ended up with an empty `body`
2096 and the `tail` is just the closing bracket, then it's considered failed.
2098 tail_len = len(str(tail).strip())
2101 raise CannotSplit("Splitting brackets produced the same line")
2105 f"Splitting brackets on an empty body to save "
2106 f"{tail_len} characters is not worth it"
2110 def dont_increase_indentation(split_func: SplitFunc) -> SplitFunc:
2111 """Normalize prefix of the first leaf in every line returned by `split_func`.
2113 This is a decorator over relevant split functions.
2117 def split_wrapper(line: Line, py36: bool = False) -> Iterator[Line]:
2118 for l in split_func(line, py36):
2119 normalize_prefix(l.leaves[0], inside_brackets=True)
2122 return split_wrapper
2125 @dont_increase_indentation
2126 def delimiter_split(line: Line, py36: bool = False) -> Iterator[Line]:
2127 """Split according to delimiters of the highest priority.
2129 If `py36` is True, the split will add trailing commas also in function
2130 signatures that contain `*` and `**`.
2133 last_leaf = line.leaves[-1]
2135 raise CannotSplit("Line empty")
2137 bt = line.bracket_tracker
2139 delimiter_priority = bt.max_delimiter_priority(exclude={id(last_leaf)})
2141 raise CannotSplit("No delimiters found")
2143 if delimiter_priority == DOT_PRIORITY:
2144 if bt.delimiter_count_with_priority(delimiter_priority) == 1:
2145 raise CannotSplit("Splitting a single attribute from its owner looks wrong")
2147 current_line = Line(depth=line.depth, inside_brackets=line.inside_brackets)
2148 lowest_depth = sys.maxsize
2149 trailing_comma_safe = True
2151 def append_to_line(leaf: Leaf) -> Iterator[Line]:
2152 """Append `leaf` to current line or to new line if appending impossible."""
2153 nonlocal current_line
2155 current_line.append_safe(leaf, preformatted=True)
2156 except ValueError as ve:
2159 current_line = Line(depth=line.depth, inside_brackets=line.inside_brackets)
2160 current_line.append(leaf)
2162 for index, leaf in enumerate(line.leaves):
2163 yield from append_to_line(leaf)
2165 for comment_after in line.comments_after(leaf, index):
2166 yield from append_to_line(comment_after)
2168 lowest_depth = min(lowest_depth, leaf.bracket_depth)
2169 if leaf.bracket_depth == lowest_depth and is_vararg(
2170 leaf, within=VARARGS_PARENTS
2172 trailing_comma_safe = trailing_comma_safe and py36
2173 leaf_priority = bt.delimiters.get(id(leaf))
2174 if leaf_priority == delimiter_priority:
2177 current_line = Line(depth=line.depth, inside_brackets=line.inside_brackets)
2181 and delimiter_priority == COMMA_PRIORITY
2182 and current_line.leaves[-1].type != token.COMMA
2183 and current_line.leaves[-1].type != STANDALONE_COMMENT
2185 current_line.append(Leaf(token.COMMA, ","))
2189 @dont_increase_indentation
2190 def standalone_comment_split(line: Line, py36: bool = False) -> Iterator[Line]:
2191 """Split standalone comments from the rest of the line."""
2192 if not line.contains_standalone_comments(0):
2193 raise CannotSplit("Line does not have any standalone comments")
2195 current_line = Line(depth=line.depth, inside_brackets=line.inside_brackets)
2197 def append_to_line(leaf: Leaf) -> Iterator[Line]:
2198 """Append `leaf` to current line or to new line if appending impossible."""
2199 nonlocal current_line
2201 current_line.append_safe(leaf, preformatted=True)
2202 except ValueError as ve:
2205 current_line = Line(depth=line.depth, inside_brackets=line.inside_brackets)
2206 current_line.append(leaf)
2208 for index, leaf in enumerate(line.leaves):
2209 yield from append_to_line(leaf)
2211 for comment_after in line.comments_after(leaf, index):
2212 yield from append_to_line(comment_after)
2218 def is_import(leaf: Leaf) -> bool:
2219 """Return True if the given leaf starts an import statement."""
2226 (v == "import" and p and p.type == syms.import_name)
2227 or (v == "from" and p and p.type == syms.import_from)
2232 def normalize_prefix(leaf: Leaf, *, inside_brackets: bool) -> None:
2233 """Leave existing extra newlines if not `inside_brackets`. Remove everything
2236 Note: don't use backslashes for formatting or you'll lose your voting rights.
2238 if not inside_brackets:
2239 spl = leaf.prefix.split("#")
2240 if "\\" not in spl[0]:
2241 nl_count = spl[-1].count("\n")
2244 leaf.prefix = "\n" * nl_count
2250 def normalize_string_prefix(leaf: Leaf, remove_u_prefix: bool = False) -> None:
2251 """Make all string prefixes lowercase.
2253 If remove_u_prefix is given, also removes any u prefix from the string.
2255 Note: Mutates its argument.
2257 match = re.match(r"^([furbFURB]*)(.*)$", leaf.value, re.DOTALL)
2258 assert match is not None, f"failed to match string {leaf.value!r}"
2259 orig_prefix = match.group(1)
2260 new_prefix = orig_prefix.lower()
2262 new_prefix = new_prefix.replace("u", "")
2263 leaf.value = f"{new_prefix}{match.group(2)}"
2266 def normalize_string_quotes(leaf: Leaf) -> None:
2267 """Prefer double quotes but only if it doesn't cause more escaping.
2269 Adds or removes backslashes as appropriate. Doesn't parse and fix
2270 strings nested in f-strings (yet).
2272 Note: Mutates its argument.
2274 value = leaf.value.lstrip("furbFURB")
2275 if value[:3] == '"""':
2278 elif value[:3] == "'''":
2281 elif value[0] == '"':
2287 first_quote_pos = leaf.value.find(orig_quote)
2288 if first_quote_pos == -1:
2289 return # There's an internal error
2291 prefix = leaf.value[:first_quote_pos]
2292 unescaped_new_quote = re.compile(rf"(([^\\]|^)(\\\\)*){new_quote}")
2293 escaped_new_quote = re.compile(rf"([^\\]|^)\\(\\\\)*{new_quote}")
2294 escaped_orig_quote = re.compile(rf"([^\\]|^)\\(\\\\)*{orig_quote}")
2295 body = leaf.value[first_quote_pos + len(orig_quote) : -len(orig_quote)]
2296 if "r" in prefix.casefold():
2297 if unescaped_new_quote.search(body):
2298 # There's at least one unescaped new_quote in this raw string
2299 # so converting is impossible
2302 # Do not introduce or remove backslashes in raw strings
2305 # remove unnecessary quotes
2306 new_body = sub_twice(escaped_new_quote, rf"\1\2{new_quote}", body)
2307 if body != new_body:
2308 # Consider the string without unnecessary quotes as the original
2310 leaf.value = f"{prefix}{orig_quote}{body}{orig_quote}"
2311 new_body = sub_twice(escaped_orig_quote, rf"\1\2{orig_quote}", new_body)
2312 new_body = sub_twice(unescaped_new_quote, rf"\1\\{new_quote}", new_body)
2313 if new_quote == '"""' and new_body[-1] == '"':
2315 new_body = new_body[:-1] + '\\"'
2316 orig_escape_count = body.count("\\")
2317 new_escape_count = new_body.count("\\")
2318 if new_escape_count > orig_escape_count:
2319 return # Do not introduce more escaping
2321 if new_escape_count == orig_escape_count and orig_quote == '"':
2322 return # Prefer double quotes
2324 leaf.value = f"{prefix}{new_quote}{new_body}{new_quote}"
2327 def normalize_invisible_parens(node: Node, parens_after: Set[str]) -> None:
2328 """Make existing optional parentheses invisible or create new ones.
2330 `parens_after` is a set of string leaf values immeditely after which parens
2333 Standardizes on visible parentheses for single-element tuples, and keeps
2334 existing visible parentheses for other tuples and generator expressions.
2337 for child in list(node.children):
2339 if child.type == syms.atom:
2340 maybe_make_parens_invisible_in_atom(child)
2341 elif is_one_tuple(child):
2342 # wrap child in visible parentheses
2343 lpar = Leaf(token.LPAR, "(")
2344 rpar = Leaf(token.RPAR, ")")
2345 index = child.remove() or 0
2346 node.insert_child(index, Node(syms.atom, [lpar, child, rpar]))
2347 elif not (isinstance(child, Leaf) and is_multiline_string(child)):
2348 # wrap child in invisible parentheses
2349 lpar = Leaf(token.LPAR, "")
2350 rpar = Leaf(token.RPAR, "")
2351 index = child.remove() or 0
2352 node.insert_child(index, Node(syms.atom, [lpar, child, rpar]))
2354 check_lpar = isinstance(child, Leaf) and child.value in parens_after
2357 def maybe_make_parens_invisible_in_atom(node: LN) -> bool:
2358 """If it's safe, make the parens in the atom `node` invisible, recusively."""
2360 node.type != syms.atom
2361 or is_empty_tuple(node)
2362 or is_one_tuple(node)
2364 or max_delimiter_priority_in_atom(node) >= COMMA_PRIORITY
2368 first = node.children[0]
2369 last = node.children[-1]
2370 if first.type == token.LPAR and last.type == token.RPAR:
2371 # make parentheses invisible
2372 first.value = "" # type: ignore
2373 last.value = "" # type: ignore
2374 if len(node.children) > 1:
2375 maybe_make_parens_invisible_in_atom(node.children[1])
2381 def is_empty_tuple(node: LN) -> bool:
2382 """Return True if `node` holds an empty tuple."""
2384 node.type == syms.atom
2385 and len(node.children) == 2
2386 and node.children[0].type == token.LPAR
2387 and node.children[1].type == token.RPAR
2391 def is_one_tuple(node: LN) -> bool:
2392 """Return True if `node` holds a tuple with one element, with or without parens."""
2393 if node.type == syms.atom:
2394 if len(node.children) != 3:
2397 lpar, gexp, rpar = node.children
2399 lpar.type == token.LPAR
2400 and gexp.type == syms.testlist_gexp
2401 and rpar.type == token.RPAR
2405 return len(gexp.children) == 2 and gexp.children[1].type == token.COMMA
2408 node.type in IMPLICIT_TUPLE
2409 and len(node.children) == 2
2410 and node.children[1].type == token.COMMA
2414 def is_yield(node: LN) -> bool:
2415 """Return True if `node` holds a `yield` or `yield from` expression."""
2416 if node.type == syms.yield_expr:
2419 if node.type == token.NAME and node.value == "yield": # type: ignore
2422 if node.type != syms.atom:
2425 if len(node.children) != 3:
2428 lpar, expr, rpar = node.children
2429 if lpar.type == token.LPAR and rpar.type == token.RPAR:
2430 return is_yield(expr)
2435 def is_vararg(leaf: Leaf, within: Set[NodeType]) -> bool:
2436 """Return True if `leaf` is a star or double star in a vararg or kwarg.
2438 If `within` includes VARARGS_PARENTS, this applies to function signatures.
2439 If `within` includes UNPACKING_PARENTS, it applies to right hand-side
2440 extended iterable unpacking (PEP 3132) and additional unpacking
2441 generalizations (PEP 448).
2443 if leaf.type not in STARS or not leaf.parent:
2447 if p.type == syms.star_expr:
2448 # Star expressions are also used as assignment targets in extended
2449 # iterable unpacking (PEP 3132). See what its parent is instead.
2455 return p.type in within
2458 def is_multiline_string(leaf: Leaf) -> bool:
2459 """Return True if `leaf` is a multiline string that actually spans many lines."""
2460 value = leaf.value.lstrip("furbFURB")
2461 return value[:3] in {'"""', "'''"} and "\n" in value
2464 def is_stub_suite(node: Node) -> bool:
2465 """Return True if `node` is a suite with a stub body."""
2467 len(node.children) != 4
2468 or node.children[0].type != token.NEWLINE
2469 or node.children[1].type != token.INDENT
2470 or node.children[3].type != token.DEDENT
2474 return is_stub_body(node.children[2])
2477 def is_stub_body(node: LN) -> bool:
2478 """Return True if `node` is a simple statement containing an ellipsis."""
2479 if not isinstance(node, Node) or node.type != syms.simple_stmt:
2482 if len(node.children) != 2:
2485 child = node.children[0]
2487 child.type == syms.atom
2488 and len(child.children) == 3
2489 and all(leaf == Leaf(token.DOT, ".") for leaf in child.children)
2493 def max_delimiter_priority_in_atom(node: LN) -> int:
2494 """Return maximum delimiter priority inside `node`.
2496 This is specific to atoms with contents contained in a pair of parentheses.
2497 If `node` isn't an atom or there are no enclosing parentheses, returns 0.
2499 if node.type != syms.atom:
2502 first = node.children[0]
2503 last = node.children[-1]
2504 if not (first.type == token.LPAR and last.type == token.RPAR):
2507 bt = BracketTracker()
2508 for c in node.children[1:-1]:
2509 if isinstance(c, Leaf):
2512 for leaf in c.leaves():
2515 return bt.max_delimiter_priority()
2521 def ensure_visible(leaf: Leaf) -> None:
2522 """Make sure parentheses are visible.
2524 They could be invisible as part of some statements (see
2525 :func:`normalize_invible_parens` and :func:`visit_import_from`).
2527 if leaf.type == token.LPAR:
2529 elif leaf.type == token.RPAR:
2533 def should_explode(line: Line, opening_bracket: Leaf) -> bool:
2534 """Should `line` immediately be split with `delimiter_split()` after RHS?"""
2536 opening_bracket.parent
2537 and opening_bracket.parent.type in {syms.atom, syms.import_from}
2538 and opening_bracket.value in "[{("
2543 last_leaf = line.leaves[-1]
2544 exclude = {id(last_leaf)} if last_leaf.type == token.COMMA else set()
2545 max_priority = line.bracket_tracker.max_delimiter_priority(exclude=exclude)
2546 except (IndexError, ValueError):
2549 return max_priority == COMMA_PRIORITY
2552 def is_python36(node: Node) -> bool:
2553 """Return True if the current file is using Python 3.6+ features.
2555 Currently looking for:
2557 - trailing commas after * or ** in function signatures and calls.
2559 for n in node.pre_order():
2560 if n.type == token.STRING:
2561 value_head = n.value[:2] # type: ignore
2562 if value_head in {'f"', 'F"', "f'", "F'", "rf", "fr", "RF", "FR"}:
2566 n.type in {syms.typedargslist, syms.arglist}
2568 and n.children[-1].type == token.COMMA
2570 for ch in n.children:
2571 if ch.type in STARS:
2574 if ch.type == syms.argument:
2575 for argch in ch.children:
2576 if argch.type in STARS:
2582 def generate_trailers_to_omit(line: Line, line_length: int) -> Iterator[Set[LeafID]]:
2583 """Generate sets of closing bracket IDs that should be omitted in a RHS.
2585 Brackets can be omitted if the entire trailer up to and including
2586 a preceding closing bracket fits in one line.
2588 Yielded sets are cumulative (contain results of previous yields, too). First
2592 omit: Set[LeafID] = set()
2595 length = 4 * line.depth
2596 opening_bracket = None
2597 closing_bracket = None
2598 optional_brackets: Set[LeafID] = set()
2599 inner_brackets: Set[LeafID] = set()
2600 for index, leaf, leaf_length in enumerate_with_length(line, reversed=True):
2601 length += leaf_length
2602 if length > line_length:
2605 optional_brackets.discard(id(leaf))
2607 if leaf is opening_bracket:
2608 opening_bracket = None
2609 elif leaf.type in CLOSING_BRACKETS:
2610 inner_brackets.add(id(leaf))
2611 elif leaf.type in CLOSING_BRACKETS:
2613 optional_brackets.add(id(opening_bracket))
2616 if index > 0 and line.leaves[index - 1].type in OPENING_BRACKETS:
2617 # Empty brackets would fail a split so treat them as "inner"
2618 # brackets (e.g. only add them to the `omit` set if another
2619 # pair of brackets was good enough.
2620 inner_brackets.add(id(leaf))
2623 opening_bracket = leaf.opening_bracket
2625 omit.add(id(closing_bracket))
2626 omit.update(inner_brackets)
2627 inner_brackets.clear()
2629 closing_bracket = leaf
2632 def get_future_imports(node: Node) -> Set[str]:
2633 """Return a set of __future__ imports in the file."""
2635 for child in node.children:
2636 if child.type != syms.simple_stmt:
2638 first_child = child.children[0]
2639 if isinstance(first_child, Leaf):
2640 # Continue looking if we see a docstring; otherwise stop.
2642 len(child.children) == 2
2643 and first_child.type == token.STRING
2644 and child.children[1].type == token.NEWLINE
2649 elif first_child.type == syms.import_from:
2650 module_name = first_child.children[1]
2651 if not isinstance(module_name, Leaf) or module_name.value != "__future__":
2653 for import_from_child in first_child.children[3:]:
2654 if isinstance(import_from_child, Leaf):
2655 if import_from_child.type == token.NAME:
2656 imports.add(import_from_child.value)
2658 assert import_from_child.type == syms.import_as_names
2659 for leaf in import_from_child.children:
2660 if isinstance(leaf, Leaf) and leaf.type == token.NAME:
2661 imports.add(leaf.value)
2667 PYTHON_EXTENSIONS = {".py", ".pyi"}
2668 BLACKLISTED_DIRECTORIES = {
2681 def gen_python_files_in_dir(path: Path) -> Iterator[Path]:
2682 """Generate all files under `path` which aren't under BLACKLISTED_DIRECTORIES
2683 and have one of the PYTHON_EXTENSIONS.
2685 for child in path.iterdir():
2687 if child.name in BLACKLISTED_DIRECTORIES:
2690 yield from gen_python_files_in_dir(child)
2692 elif child.is_file() and child.suffix in PYTHON_EXTENSIONS:
2698 """Provides a reformatting counter. Can be rendered with `str(report)`."""
2701 change_count: int = 0
2703 failure_count: int = 0
2705 def done(self, src: Path, changed: Changed) -> None:
2706 """Increment the counter for successful reformatting. Write out a message."""
2707 if changed is Changed.YES:
2708 reformatted = "would reformat" if self.check else "reformatted"
2710 out(f"{reformatted} {src}")
2711 self.change_count += 1
2714 if changed is Changed.NO:
2715 msg = f"{src} already well formatted, good job."
2717 msg = f"{src} wasn't modified on disk since last run."
2718 out(msg, bold=False)
2719 self.same_count += 1
2721 def failed(self, src: Path, message: str) -> None:
2722 """Increment the counter for failed reformatting. Write out a message."""
2723 err(f"error: cannot format {src}: {message}")
2724 self.failure_count += 1
2727 def return_code(self) -> int:
2728 """Return the exit code that the app should use.
2730 This considers the current state of changed files and failures:
2731 - if there were any failures, return 123;
2732 - if any files were changed and --check is being used, return 1;
2733 - otherwise return 0.
2735 # According to http://tldp.org/LDP/abs/html/exitcodes.html starting with
2736 # 126 we have special returncodes reserved by the shell.
2737 if self.failure_count:
2740 elif self.change_count and self.check:
2745 def __str__(self) -> str:
2746 """Render a color report of the current state.
2748 Use `click.unstyle` to remove colors.
2751 reformatted = "would be reformatted"
2752 unchanged = "would be left unchanged"
2753 failed = "would fail to reformat"
2755 reformatted = "reformatted"
2756 unchanged = "left unchanged"
2757 failed = "failed to reformat"
2759 if self.change_count:
2760 s = "s" if self.change_count > 1 else ""
2762 click.style(f"{self.change_count} file{s} {reformatted}", bold=True)
2765 s = "s" if self.same_count > 1 else ""
2766 report.append(f"{self.same_count} file{s} {unchanged}")
2767 if self.failure_count:
2768 s = "s" if self.failure_count > 1 else ""
2770 click.style(f"{self.failure_count} file{s} {failed}", fg="red")
2772 return ", ".join(report) + "."
2775 def assert_equivalent(src: str, dst: str) -> None:
2776 """Raise AssertionError if `src` and `dst` aren't equivalent."""
2781 def _v(node: ast.AST, depth: int = 0) -> Iterator[str]:
2782 """Simple visitor generating strings to compare ASTs by content."""
2783 yield f"{' ' * depth}{node.__class__.__name__}("
2785 for field in sorted(node._fields):
2787 value = getattr(node, field)
2788 except AttributeError:
2791 yield f"{' ' * (depth+1)}{field}="
2793 if isinstance(value, list):
2795 if isinstance(item, ast.AST):
2796 yield from _v(item, depth + 2)
2798 elif isinstance(value, ast.AST):
2799 yield from _v(value, depth + 2)
2802 yield f"{' ' * (depth+2)}{value!r}, # {value.__class__.__name__}"
2804 yield f"{' ' * depth}) # /{node.__class__.__name__}"
2807 src_ast = ast.parse(src)
2808 except Exception as exc:
2809 major, minor = sys.version_info[:2]
2810 raise AssertionError(
2811 f"cannot use --safe with this file; failed to parse source file "
2812 f"with Python {major}.{minor}'s builtin AST. Re-run with --fast "
2813 f"or stop using deprecated Python 2 syntax. AST error message: {exc}"
2817 dst_ast = ast.parse(dst)
2818 except Exception as exc:
2819 log = dump_to_file("".join(traceback.format_tb(exc.__traceback__)), dst)
2820 raise AssertionError(
2821 f"INTERNAL ERROR: Black produced invalid code: {exc}. "
2822 f"Please report a bug on https://github.com/ambv/black/issues. "
2823 f"This invalid output might be helpful: {log}"
2826 src_ast_str = "\n".join(_v(src_ast))
2827 dst_ast_str = "\n".join(_v(dst_ast))
2828 if src_ast_str != dst_ast_str:
2829 log = dump_to_file(diff(src_ast_str, dst_ast_str, "src", "dst"))
2830 raise AssertionError(
2831 f"INTERNAL ERROR: Black produced code that is not equivalent to "
2833 f"Please report a bug on https://github.com/ambv/black/issues. "
2834 f"This diff might be helpful: {log}"
2838 def assert_stable(src: str, dst: str, line_length: int, is_pyi: bool = False) -> None:
2839 """Raise AssertionError if `dst` reformats differently the second time."""
2840 newdst = format_str(dst, line_length=line_length, is_pyi=is_pyi)
2843 diff(src, dst, "source", "first pass"),
2844 diff(dst, newdst, "first pass", "second pass"),
2846 raise AssertionError(
2847 f"INTERNAL ERROR: Black produced different code on the second pass "
2848 f"of the formatter. "
2849 f"Please report a bug on https://github.com/ambv/black/issues. "
2850 f"This diff might be helpful: {log}"
2854 def dump_to_file(*output: str) -> str:
2855 """Dump `output` to a temporary file. Return path to the file."""
2858 with tempfile.NamedTemporaryFile(
2859 mode="w", prefix="blk_", suffix=".log", delete=False, encoding="utf8"
2861 for lines in output:
2863 if lines and lines[-1] != "\n":
2868 def diff(a: str, b: str, a_name: str, b_name: str) -> str:
2869 """Return a unified diff string between strings `a` and `b`."""
2872 a_lines = [line + "\n" for line in a.split("\n")]
2873 b_lines = [line + "\n" for line in b.split("\n")]
2875 difflib.unified_diff(a_lines, b_lines, fromfile=a_name, tofile=b_name, n=5)
2879 def cancel(tasks: Iterable[asyncio.Task]) -> None:
2880 """asyncio signal handler that cancels all `tasks` and reports to stderr."""
2886 def shutdown(loop: BaseEventLoop) -> None:
2887 """Cancel all pending tasks on `loop`, wait for them, and close the loop."""
2889 # This part is borrowed from asyncio/runners.py in Python 3.7b2.
2890 to_cancel = [task for task in asyncio.Task.all_tasks(loop) if not task.done()]
2894 for task in to_cancel:
2896 loop.run_until_complete(
2897 asyncio.gather(*to_cancel, loop=loop, return_exceptions=True)
2900 # `concurrent.futures.Future` objects cannot be cancelled once they
2901 # are already running. There might be some when the `shutdown()` happened.
2902 # Silence their logger's spew about the event loop being closed.
2903 cf_logger = logging.getLogger("concurrent.futures")
2904 cf_logger.setLevel(logging.CRITICAL)
2908 def sub_twice(regex: Pattern[str], replacement: str, original: str) -> str:
2909 """Replace `regex` with `replacement` twice on `original`.
2911 This is used by string normalization to perform replaces on
2912 overlapping matches.
2914 return regex.sub(replacement, regex.sub(replacement, original))
2917 def enumerate_reversed(sequence: Sequence[T]) -> Iterator[Tuple[Index, T]]:
2918 """Like `reversed(enumerate(sequence))` if that were possible."""
2919 index = len(sequence) - 1
2920 for element in reversed(sequence):
2921 yield (index, element)
2925 def enumerate_with_length(
2926 line: Line, reversed: bool = False
2927 ) -> Iterator[Tuple[Index, Leaf, int]]:
2928 """Return an enumeration of leaves with their length.
2930 Stops prematurely on multiline strings and standalone comments.
2933 Callable[[Sequence[Leaf]], Iterator[Tuple[Index, Leaf]]],
2934 enumerate_reversed if reversed else enumerate,
2936 for index, leaf in op(line.leaves):
2937 length = len(leaf.prefix) + len(leaf.value)
2938 if "\n" in leaf.value:
2939 return # Multiline strings, we can't continue.
2941 comment: Optional[Leaf]
2942 for comment in line.comments_after(leaf, index):
2943 if "\n" in comment.prefix:
2944 return # Oops, standalone comment!
2946 length += len(comment.value)
2948 yield index, leaf, length
2951 def is_line_short_enough(line: Line, *, line_length: int, line_str: str = "") -> bool:
2952 """Return True if `line` is no longer than `line_length`.
2954 Uses the provided `line_str` rendering, if any, otherwise computes a new one.
2957 line_str = str(line).strip("\n")
2959 len(line_str) <= line_length
2960 and "\n" not in line_str # multiline strings
2961 and not line.contains_standalone_comments()
2965 def can_omit_invisible_parens(line: Line, line_length: int) -> bool:
2966 """Does `line` have a shape safe to reformat without optional parens around it?
2968 Returns True for only a subset of potentially nice looking formattings but
2969 the point is to not return false positives that end up producing lines that
2972 bt = line.bracket_tracker
2973 if not bt.delimiters:
2974 # Without delimiters the optional parentheses are useless.
2977 max_priority = bt.max_delimiter_priority()
2978 if bt.delimiter_count_with_priority(max_priority) > 1:
2979 # With more than one delimiter of a kind the optional parentheses read better.
2982 if max_priority == DOT_PRIORITY:
2983 # A single stranded method call doesn't require optional parentheses.
2986 assert len(line.leaves) >= 2, "Stranded delimiter"
2988 first = line.leaves[0]
2989 second = line.leaves[1]
2990 penultimate = line.leaves[-2]
2991 last = line.leaves[-1]
2993 # With a single delimiter, omit if the expression starts or ends with
2995 if first.type in OPENING_BRACKETS and second.type not in CLOSING_BRACKETS:
2997 length = 4 * line.depth
2998 for _index, leaf, leaf_length in enumerate_with_length(line):
2999 if leaf.type in CLOSING_BRACKETS and leaf.opening_bracket is first:
3002 length += leaf_length
3003 if length > line_length:
3006 if leaf.type in OPENING_BRACKETS:
3007 # There are brackets we can further split on.
3011 # checked the entire string and line length wasn't exceeded
3012 if len(line.leaves) == _index + 1:
3015 # Note: we are not returning False here because a line might have *both*
3016 # a leading opening bracket and a trailing closing bracket. If the
3017 # opening bracket doesn't match our rule, maybe the closing will.
3020 last.type == token.RPAR
3021 or last.type == token.RBRACE
3023 # don't use indexing for omitting optional parentheses;
3025 last.type == token.RSQB
3027 and last.parent.type != syms.trailer
3030 if penultimate.type in OPENING_BRACKETS:
3031 # Empty brackets don't help.
3034 if is_multiline_string(first):
3035 # Additional wrapping of a multiline string in this situation is
3039 length = 4 * line.depth
3040 seen_other_brackets = False
3041 for _index, leaf, leaf_length in enumerate_with_length(line):
3042 length += leaf_length
3043 if leaf is last.opening_bracket:
3044 if seen_other_brackets or length <= line_length:
3047 elif leaf.type in OPENING_BRACKETS:
3048 # There are brackets we can further split on.
3049 seen_other_brackets = True
3054 CACHE_DIR = Path(user_cache_dir("black", version=__version__))
3057 def get_cache_file(line_length: int) -> Path:
3058 return CACHE_DIR / f"cache.{line_length}.pickle"
3061 def read_cache(line_length: int) -> Cache:
3062 """Read the cache if it exists and is well formed.
3064 If it is not well formed, the call to write_cache later should resolve the issue.
3066 cache_file = get_cache_file(line_length)
3067 if not cache_file.exists():
3070 with cache_file.open("rb") as fobj:
3072 cache: Cache = pickle.load(fobj)
3073 except pickle.UnpicklingError:
3079 def get_cache_info(path: Path) -> CacheInfo:
3080 """Return the information used to check if a file is already formatted or not."""
3082 return stat.st_mtime, stat.st_size
3086 cache: Cache, sources: Iterable[Path]
3087 ) -> Tuple[List[Path], List[Path]]:
3088 """Split a list of paths into two.
3090 The first list contains paths of files that modified on disk or are not in the
3091 cache. The other list contains paths to non-modified files.
3096 if cache.get(src) != get_cache_info(src):
3103 def write_cache(cache: Cache, sources: List[Path], line_length: int) -> None:
3104 """Update the cache file."""
3105 cache_file = get_cache_file(line_length)
3107 if not CACHE_DIR.exists():
3108 CACHE_DIR.mkdir(parents=True)
3109 new_cache = {**cache, **{src.resolve(): get_cache_info(src) for src in sources}}
3110 with cache_file.open("wb") as fobj:
3111 pickle.dump(new_cache, fobj, protocol=pickle.HIGHEST_PROTOCOL)
3116 if __name__ == "__main__":