X-Git-Url: https://git.madduck.net/etc/vim.git/blobdiff_plain/1dbe77cd917de67f721df48741d148ea7f642317..44e9cd4a03df470adb24b958ae4b9e47ee890315:/black.py?ds=inline diff --git a/black.py b/black.py index ab2394e..d36db68 100644 --- a/black.py +++ b/black.py @@ -47,10 +47,10 @@ from blib2to3.pgen2 import driver, token from blib2to3.pgen2.parse import ParseError -__version__ = "18.6b4" +__version__ = "18.9b0" DEFAULT_LINE_LENGTH = 88 DEFAULT_EXCLUDES = ( - r"/(\.git|\.hg|\.mypy_cache|\.tox|\.venv|_build|buck-out|build|dist)/" + r"/(\.git|\.hg|\.mypy_cache|\.nox|\.tox|\.venv|_build|buck-out|build|dist)/" ) DEFAULT_INCLUDES = r"\.pyi?$" CACHE_DIR = Path(user_cache_dir("black", version=__version__)) @@ -79,15 +79,15 @@ syms = pygram.python_symbols class NothingChanged(UserWarning): - """Raised by :func:`format_file` when reformatted code is the same as source.""" + """Raised when reformatted code is the same as source.""" class CannotSplit(Exception): - """A readable split that fits the allotted line length is impossible. + """A readable split that fits the allotted line length is impossible.""" - Raised by :func:`left_hand_split`, :func:`right_hand_split`, and - :func:`delimiter_split`. - """ + +class InvalidInput(ValueError): + """Raised when input source code fails all parse attempts.""" class WriteBack(Enum): @@ -115,10 +115,16 @@ class FileMode(Flag): PYTHON36 = 1 PYI = 2 NO_STRING_NORMALIZATION = 4 + NO_NUMERIC_UNDERSCORE_NORMALIZATION = 8 @classmethod def from_configuration( - cls, *, py36: bool, pyi: bool, skip_string_normalization: bool + cls, + *, + py36: bool, + pyi: bool, + skip_string_normalization: bool, + skip_numeric_underscore_normalization: bool, ) -> "FileMode": mode = cls.AUTO_DETECT if py36: @@ -127,6 +133,8 @@ class FileMode(Flag): mode |= cls.PYI if skip_string_normalization: mode |= cls.NO_STRING_NORMALIZATION + if skip_numeric_underscore_normalization: + mode |= cls.NO_NUMERIC_UNDERSCORE_NORMALIZATION return mode @@ -170,7 +178,7 @@ def read_pyproject_toml( "--line-length", type=int, default=DEFAULT_LINE_LENGTH, - help="How many character per line to allow.", + help="How many characters per line to allow.", show_default=True, ) @click.option( @@ -196,6 +204,12 @@ def read_pyproject_toml( is_flag=True, help="Don't normalize string quotes or prefixes.", ) +@click.option( + "-N", + "--skip-numeric-underscore-normalization", + is_flag=True, + help="Don't normalize underscores in numeric literals.", +) @click.option( "--check", is_flag=True, @@ -286,6 +300,7 @@ def main( pyi: bool, py36: bool, skip_string_normalization: bool, + skip_numeric_underscore_normalization: bool, quiet: bool, verbose: bool, include: str, @@ -296,7 +311,10 @@ def main( """The uncompromising code formatter.""" write_back = WriteBack.from_configuration(check=check, diff=diff) mode = FileMode.from_configuration( - py36=py36, pyi=pyi, skip_string_normalization=skip_string_normalization + py36=py36, + pyi=pyi, + skip_string_normalization=skip_string_normalization, + skip_numeric_underscore_normalization=skip_numeric_underscore_normalization, ) if config and verbose: out(f"Using configuration from {config}.", bold=False, fg="blue") @@ -431,52 +449,54 @@ async def schedule_formatting( sources, cached = filter_cached(cache, sources) for src in sorted(cached): report.done(src, Changed.CACHED) + if not sources: + return + cancelled = [] sources_to_cache = [] - if sources: - lock = None - if write_back == WriteBack.DIFF: - # For diff output, we need locks to ensure we don't interleave output - # from different processes. - manager = Manager() - lock = manager.Lock() - tasks = { - loop.run_in_executor( - executor, - format_file_in_place, - src, - line_length, - fast, - write_back, - mode, - lock, - ): src - for src in sorted(sources) - } - pending: Iterable[asyncio.Task] = tasks.keys() - try: - loop.add_signal_handler(signal.SIGINT, cancel, pending) - loop.add_signal_handler(signal.SIGTERM, cancel, pending) - except NotImplementedError: - # There are no good alternatives for these on Windows - pass - while pending: - done, _ = await asyncio.wait(pending, return_when=asyncio.FIRST_COMPLETED) - for task in done: - src = tasks.pop(task) - if task.cancelled(): - cancelled.append(task) - elif task.exception(): - report.failed(src, str(task.exception())) - else: - changed = Changed.YES if task.result() else Changed.NO - # If the file was written back or was successfully checked as - # well-formatted, store this information in the cache. - if write_back is WriteBack.YES or ( - write_back is WriteBack.CHECK and changed is Changed.NO - ): - sources_to_cache.append(src) - report.done(src, changed) + lock = None + if write_back == WriteBack.DIFF: + # For diff output, we need locks to ensure we don't interleave output + # from different processes. + manager = Manager() + lock = manager.Lock() + tasks = { + loop.run_in_executor( + executor, + format_file_in_place, + src, + line_length, + fast, + write_back, + mode, + lock, + ): src + for src in sorted(sources) + } + pending: Iterable[asyncio.Task] = tasks.keys() + try: + loop.add_signal_handler(signal.SIGINT, cancel, pending) + loop.add_signal_handler(signal.SIGTERM, cancel, pending) + except NotImplementedError: + # There are no good alternatives for these on Windows. + pass + while pending: + done, _ = await asyncio.wait(pending, return_when=asyncio.FIRST_COMPLETED) + for task in done: + src = tasks.pop(task) + if task.cancelled(): + cancelled.append(task) + elif task.exception(): + report.failed(src, str(task.exception())) + else: + changed = Changed.YES if task.result() else Changed.NO + # If the file was written back or was successfully checked as + # well-formatted, store this information in the cache. + if write_back is WriteBack.YES or ( + write_back is WriteBack.CHECK and changed is Changed.NO + ): + sources_to_cache.append(src) + report.done(src, changed) if cancelled: await asyncio.gather(*cancelled, loop=loop, return_exceptions=True) if sources_to_cache: @@ -605,7 +625,7 @@ def format_str( `line_length` determines how many characters per line are allowed. """ - src_node = lib2to3_parse(src_contents) + src_node = lib2to3_parse(src_contents.lstrip()) dst_contents = "" future_imports = get_future_imports(src_node) is_pyi = bool(mode & FileMode.PYI) @@ -616,7 +636,8 @@ def format_str( remove_u_prefix=py36 or "unicode_literals" in future_imports, is_pyi=is_pyi, normalize_strings=normalize_strings, - allow_underscores=py36, + allow_underscores=py36 + and not bool(mode & FileMode.NO_NUMERIC_UNDERSCORE_NORMALIZATION), ) elt = EmptyLineTracker(is_pyi=is_pyi) empty_line = Line() @@ -674,7 +695,7 @@ def lib2to3_parse(src_txt: str) -> Node: faulty_line = lines[lineno - 1] except IndexError: faulty_line = "" - exc = ValueError(f"Cannot parse: {lineno}:{column}: {faulty_line}") + exc = InvalidInput(f"Cannot parse: {lineno}:{column}: {faulty_line}") else: raise exc from None @@ -875,8 +896,8 @@ class BracketTracker: bracket_match: Dict[Tuple[Depth, NodeType], Leaf] = Factory(dict) delimiters: Dict[LeafID, Priority] = Factory(dict) previous: Optional[Leaf] = None - _for_loop_variable: int = 0 - _lambda_arguments: int = 0 + _for_loop_depths: List[int] = Factory(list) + _lambda_argument_depths: List[int] = Factory(list) def mark(self, leaf: Leaf) -> None: """Mark `leaf` with bracket-related metadata. Keep track of delimiters. @@ -949,16 +970,21 @@ class BracketTracker: """ if leaf.type == token.NAME and leaf.value == "for": self.depth += 1 - self._for_loop_variable += 1 + self._for_loop_depths.append(self.depth) return True return False def maybe_decrement_after_for_loop_variable(self, leaf: Leaf) -> bool: """See `maybe_increment_for_loop_variable` above for explanation.""" - if self._for_loop_variable and leaf.type == token.NAME and leaf.value == "in": + if ( + self._for_loop_depths + and self._for_loop_depths[-1] == self.depth + and leaf.type == token.NAME + and leaf.value == "in" + ): self.depth -= 1 - self._for_loop_variable -= 1 + self._for_loop_depths.pop() return True return False @@ -971,16 +997,20 @@ class BracketTracker: """ if leaf.type == token.NAME and leaf.value == "lambda": self.depth += 1 - self._lambda_arguments += 1 + self._lambda_argument_depths.append(self.depth) return True return False def maybe_decrement_after_lambda_arguments(self, leaf: Leaf) -> bool: """See `maybe_increment_lambda_arguments` above for explanation.""" - if self._lambda_arguments and leaf.type == token.COLON: + if ( + self._lambda_argument_depths + and self._lambda_argument_depths[-1] == self.depth + and leaf.type == token.COLON + ): self.depth -= 1 - self._lambda_arguments -= 1 + self._lambda_argument_depths.pop() return True return False @@ -1162,7 +1192,7 @@ class Line: self.remove_trailing_comma() return True - # Otheriwsse, if the trailing one is the only one, we might mistakenly + # Otherwise, if the trailing one is the only one, we might mistakenly # change a tuple into a different type by removing the comma. depth = closing.bracket_depth + 1 commas = 0 @@ -1375,7 +1405,7 @@ class EmptyLineTracker: newlines = 1 elif current_line.is_class or self.previous_line.is_class: if current_line.is_stub_class and self.previous_line.is_stub_class: - # No blank line between classes with an emty body + # No blank line between classes with an empty body newlines = 0 else: newlines = 1 @@ -1878,7 +1908,7 @@ def is_split_after_delimiter(leaf: Leaf, previous: Leaf = None) -> int: def is_split_before_delimiter(leaf: Leaf, previous: Leaf = None) -> int: - """Return the priority of the `leaf` delimiter, given a line before after it. + """Return the priority of the `leaf` delimiter, given a line break before it. The delimiter priorities returned here are from those delimiters that would cause a line break before themselves. @@ -1915,15 +1945,20 @@ def is_split_before_delimiter(leaf: Leaf, previous: Leaf = None) -> int: ): return STRING_PRIORITY - if leaf.type != token.NAME: + if leaf.type not in {token.NAME, token.ASYNC}: return 0 if ( leaf.value == "for" and leaf.parent and leaf.parent.type in {syms.comp_for, syms.old_comp_for} + or leaf.type == token.ASYNC ): - return COMPREHENSION_PRIORITY + if ( + not isinstance(leaf.prev_sibling, Leaf) + or leaf.prev_sibling.value != "async" + ): + return COMPREHENSION_PRIORITY if ( leaf.value == "if" @@ -1997,6 +2032,16 @@ def generate_comments(leaf: LN) -> Iterator[Leaf]: @dataclass class ProtoComment: + """Describes a piece of syntax that is a comment. + + It's not a :class:`blib2to3.pytree.Leaf` so that: + + * it can be cached (`Leaf` objects should not be reused more than once as + they store their lineno, column, prefix, and parent information); + * `newlines` and `consumed` fields are kept separate from the `value`. This + simplifies handling of special marker comments like ``# fmt: off/on``. + """ + type: int # token.COMMENT or STANDALONE_COMMENT value: str # content of the comment newlines: int # how many newlines before the comment @@ -2005,6 +2050,7 @@ class ProtoComment: @lru_cache(maxsize=4096) def list_comments(prefix: str, *, is_endmarker: bool) -> List[ProtoComment]: + """Return a list of :class:`ProtoComment` objects parsed from the given `prefix`.""" result: List[ProtoComment] = [] if not prefix or "#" not in prefix: return result @@ -2036,8 +2082,8 @@ def list_comments(prefix: str, *, is_endmarker: bool) -> List[ProtoComment]: def make_comment(content: str) -> str: """Return a consistently formatted comment from the given `content` string. - All comments (except for "##", "#!", "#:") should have a single space between - the hash sign and the content. + All comments (except for "##", "#!", "#:", '#'", "#%%") should have a single + space between the hash sign and the content. If `content` didn't start with a hash sign, one is provided. """ @@ -2047,7 +2093,7 @@ def make_comment(content: str) -> str: if content[0] == "#": content = content[1:] - if content and content[0] not in " !:#": + if content and content[0] not in " !:#'%": content = " " + content return "#" + content @@ -2128,9 +2174,6 @@ def left_hand_split(line: Line, py36: bool = False) -> Iterator[Line]: Prefer RHS otherwise. This is why this function is not symmetrical with :func:`right_hand_split` which also handles optional parentheses. """ - head = Line(depth=line.depth) - body = Line(depth=line.depth + 1, inside_brackets=True) - tail = Line(depth=line.depth) tail_leaves: List[Leaf] = [] body_leaves: List[Leaf] = [] head_leaves: List[Leaf] = [] @@ -2148,15 +2191,12 @@ def left_hand_split(line: Line, py36: bool = False) -> Iterator[Line]: if leaf.type in OPENING_BRACKETS: matching_bracket = leaf current_leaves = body_leaves - # Since body is a new indent level, remove spurious leading whitespace. - if body_leaves: - normalize_prefix(body_leaves[0], inside_brackets=True) - # Build the new lines. - for result, leaves in (head, head_leaves), (body, body_leaves), (tail, tail_leaves): - for leaf in leaves: - result.append(leaf, preformatted=True) - for comment_after in line.comments_after(leaf): - result.append(comment_after, preformatted=True) + if not matching_bracket: + raise CannotSplit("No brackets found") + + head = bracket_split_build_line(head_leaves, line, matching_bracket) + body = bracket_split_build_line(body_leaves, line, matching_bracket, is_body=True) + tail = bracket_split_build_line(tail_leaves, line, matching_bracket) bracket_split_succeeded_or_raise(head, body, tail) for result in (head, body, tail): if result: @@ -2174,9 +2214,6 @@ def right_hand_split( Note: running this function modifies `bracket_depth` on the leaves of `line`. """ - head = Line(depth=line.depth) - body = Line(depth=line.depth + 1, inside_brackets=True) - tail = Line(depth=line.depth) tail_leaves: List[Leaf] = [] body_leaves: List[Leaf] = [] head_leaves: List[Leaf] = [] @@ -2193,25 +2230,18 @@ def right_hand_split( opening_bracket = leaf.opening_bracket closing_bracket = leaf current_leaves = body_leaves - tail_leaves.reverse() - body_leaves.reverse() - head_leaves.reverse() - # Since body is a new indent level, remove spurious leading whitespace. - if body_leaves: - normalize_prefix(body_leaves[0], inside_brackets=True) - if not head_leaves: - # No `head` means the split failed. Either `tail` has all content or + if not (opening_bracket and closing_bracket and head_leaves): + # If there is no opening or closing_bracket that means the split failed and + # all content is in the tail. Otherwise, if `head_leaves` are empty, it means # the matching `opening_bracket` wasn't available on `line` anymore. raise CannotSplit("No brackets found") - # Build the new lines. - for result, leaves in (head, head_leaves), (body, body_leaves), (tail, tail_leaves): - for leaf in leaves: - result.append(leaf, preformatted=True) - for comment_after in line.comments_after(leaf): - result.append(comment_after, preformatted=True) - assert opening_bracket and closing_bracket - body.should_explode = should_explode(body, opening_bracket) + tail_leaves.reverse() + body_leaves.reverse() + head_leaves.reverse() + head = bracket_split_build_line(head_leaves, line, opening_bracket) + body = bracket_split_build_line(body_leaves, line, opening_bracket, is_body=True) + tail = bracket_split_build_line(tail_leaves, line, opening_bracket) bracket_split_succeeded_or_raise(head, body, tail) if ( # the body shouldn't be exploded @@ -2285,6 +2315,35 @@ def bracket_split_succeeded_or_raise(head: Line, body: Line, tail: Line) -> None ) +def bracket_split_build_line( + leaves: List[Leaf], original: Line, opening_bracket: Leaf, *, is_body: bool = False +) -> Line: + """Return a new line with given `leaves` and respective comments from `original`. + + If `is_body` is True, the result line is one-indented inside brackets and as such + has its first leaf's prefix normalized and a trailing comma added when expected. + """ + result = Line(depth=original.depth) + if is_body: + result.inside_brackets = True + result.depth += 1 + if leaves: + # Since body is a new indent level, remove spurious leading whitespace. + normalize_prefix(leaves[0], inside_brackets=True) + # Ensure a trailing comma when expected. + if original.is_import: + if leaves[-1].type != token.COMMA: + leaves.append(Leaf(token.COMMA, ",")) + # Populate the line + for leaf in leaves: + result.append(leaf, preformatted=True) + for comment_after in original.comments_after(leaf): + result.append(comment_after, preformatted=True) + if is_body: + result.should_explode = should_explode(result, opening_bracket) + return result + + def dont_increase_indentation(split_func: SplitFunc) -> SplitFunc: """Normalize prefix of the first leaf in every line returned by `split_func`. @@ -2509,30 +2568,36 @@ def normalize_string_quotes(leaf: Leaf) -> None: def normalize_numeric_literal(leaf: Leaf, allow_underscores: bool) -> None: - """Normalizes numeric (float, int, and complex) literals.""" - # We want all letters (e in exponents, j in complex literals, a-f - # in hex literals) to be lowercase. + """Normalizes numeric (float, int, and complex) literals. + + All letters used in the representation are normalized to lowercase (except + in Python 2 long literals), and long number literals are split using underscores. + """ text = leaf.value.lower() - if text.startswith(("0o", "0x", "0b")): - # Leave octal, hex, and binary literals alone for now. + if text.startswith(("0o", "0b")): + # Leave octal and binary literals alone. pass + elif text.startswith("0x"): + # Change hex literals to upper case. + before, after = text[:2], text[2:] + text = f"{before}{after.upper()}" elif "e" in text: before, after = text.split("e") + sign = "" if after.startswith("-"): after = after[1:] sign = "-" elif after.startswith("+"): after = after[1:] - sign = "" - else: - sign = "" before = format_float_or_int_string(before, allow_underscores) after = format_int_string(after, allow_underscores) text = f"{before}e{sign}{after}" - # Complex numbers and Python 2 longs - elif "j" in text or "l" in text: + elif text.endswith(("j", "l")): number = text[:-1] suffix = text[-1] + # Capitalize in "2L" because "l" looks too similar to "1". + if suffix == "l": + suffix = "L" text = f"{format_float_or_int_string(number, allow_underscores)}{suffix}" else: text = format_float_or_int_string(text, allow_underscores) @@ -2543,24 +2608,40 @@ def format_float_or_int_string(text: str, allow_underscores: bool) -> str: """Formats a float string like "1.0".""" if "." not in text: return format_int_string(text, allow_underscores) + before, after = text.split(".") before = format_int_string(before, allow_underscores) if before else "0" - after = format_int_string(after, allow_underscores) if after else "0" + if after: + after = format_int_string(after, allow_underscores, count_from_end=False) + else: + after = "0" return f"{before}.{after}" -def format_int_string(text: str, allow_underscores: bool) -> str: +def format_int_string( + text: str, allow_underscores: bool, count_from_end: bool = True +) -> str: """Normalizes underscores in a string to e.g. 1_000_000. - Input must be a string consisting only of digits and underscores. + Input must be a string of digits and optional underscores. + If count_from_end is False, we add underscores after groups of three digits + counting from the beginning instead of the end of the strings. This is used + for the fractional part of float literals. """ if not allow_underscores: return text + text = text.replace("_", "") - if len(text) <= 6: - # No underscores for numbers <= 6 digits long. + if len(text) <= 5: + # No underscores for numbers <= 5 digits long. return text - return format(int(text), "3_") + + if count_from_end: + # Avoid removing leading zeros, which are important if we're formatting + # part of a number like "0.001". + return format(int("1" + text), "3_")[1:].lstrip("_") + else: + return "_".join(text[i : i + 3] for i in range(0, len(text), 3)) def normalize_invisible_parens(node: Node, parens_after: Set[str]) -> None: @@ -2581,7 +2662,11 @@ def normalize_invisible_parens(node: Node, parens_after: Set[str]) -> None: for index, child in enumerate(list(node.children)): if check_lpar: if child.type == syms.atom: - maybe_make_parens_invisible_in_atom(child) + if maybe_make_parens_invisible_in_atom(child): + lpar = Leaf(token.LPAR, "") + rpar = Leaf(token.RPAR, "") + index = child.remove() or 0 + node.insert_child(index, Node(syms.atom, [lpar, child, rpar])) elif is_one_tuple(child): # wrap child in visible parentheses lpar = Leaf(token.LPAR, "(") @@ -2689,7 +2774,11 @@ def generate_ignored_nodes(leaf: Leaf) -> Iterator[LN]: def maybe_make_parens_invisible_in_atom(node: LN) -> bool: - """If it's safe, make the parens in the atom `node` invisible, recursively.""" + """If it's safe, make the parens in the atom `node` invisible, recursively. + + Returns whether the node should itself be wrapped in invisible parentheses. + + """ if ( node.type != syms.atom or is_empty_tuple(node) @@ -2707,9 +2796,9 @@ def maybe_make_parens_invisible_in_atom(node: LN) -> bool: last.value = "" # type: ignore if len(node.children) > 1: maybe_make_parens_invisible_in_atom(node.children[1]) - return True + return False - return False + return True def is_empty_tuple(node: LN) -> bool: @@ -2887,7 +2976,8 @@ def is_python36(node: Node) -> bool: """Return True if the current file is using Python 3.6+ features. Currently looking for: - - f-strings; and + - f-strings; + - underscores in numeric literals; and - trailing commas after * or ** in function signatures and calls. """ for n in node.pre_order(): @@ -2896,6 +2986,10 @@ def is_python36(node: Node) -> bool: if value_head in {'f"', 'F"', "f'", "F'", "rf", "fr", "RF", "FR"}: return True + elif n.type == token.NUMBER: + if "_" in n.value: # type: ignore + return True + elif ( n.type in {syms.typedargslist, syms.arglist} and n.children @@ -2929,7 +3023,6 @@ def generate_trailers_to_omit(line: Line, line_length: int) -> Iterator[Set[Leaf length = 4 * line.depth opening_bracket = None closing_bracket = None - optional_brackets: Set[LeafID] = set() inner_brackets: Set[LeafID] = set() for index, leaf, leaf_length in enumerate_with_length(line, reversed=True): length += leaf_length @@ -2940,17 +3033,12 @@ def generate_trailers_to_omit(line: Line, line_length: int) -> Iterator[Set[Leaf if leaf.type == STANDALONE_COMMENT or has_inline_comment: break - optional_brackets.discard(id(leaf)) if opening_bracket: if leaf is opening_bracket: opening_bracket = None elif leaf.type in CLOSING_BRACKETS: inner_brackets.add(id(leaf)) elif leaf.type in CLOSING_BRACKETS: - if not leaf.value: - optional_brackets.add(id(opening_bracket)) - continue - if index > 0 and line.leaves[index - 1].type in OPENING_BRACKETS: # Empty brackets would fail a split so treat them as "inner" # brackets (e.g. only add them to the `omit` set if another @@ -2958,13 +3046,15 @@ def generate_trailers_to_omit(line: Line, line_length: int) -> Iterator[Set[Leaf inner_brackets.add(id(leaf)) continue - opening_bracket = leaf.opening_bracket if closing_bracket: omit.add(id(closing_bracket)) omit.update(inner_brackets) inner_brackets.clear() yield omit - closing_bracket = leaf + + if leaf.value: + opening_bracket = leaf.opening_bracket + closing_bracket = leaf def get_future_imports(node: Node) -> Set[str]: @@ -3129,7 +3219,7 @@ class Report: - otherwise return 0. """ # According to http://tldp.org/LDP/abs/html/exitcodes.html starting with - # 126 we have special returncodes reserved by the shell. + # 126 we have special return codes reserved by the shell. if self.failure_count: return 123