]> git.madduck.net Git - etc/vim.git/blobdiff - black.py

madduck's git repository

Every one of the projects in this repository is available at the canonical URL git://git.madduck.net/madduck/pub/<projectpath> — see each project's metadata for the exact URL.

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.

SSH access, as well as push access can be individually arranged.

If you use my repositories frequently, consider adding the following snippet to ~/.gitconfig and using the third clone URL listed for each project:

[url "git://git.madduck.net/madduck/"]
  insteadOf = madduck:

set entry to black (#553)
[etc/vim.git] / black.py
index 8d53194680309ce664af0e0a77105c31c2343824..4d8586581c131b20f3c44ad1430a72573bf9fe9a 100644 (file)
--- a/black.py
+++ b/black.py
@@ -47,10 +47,10 @@ from blib2to3.pgen2 import driver, token
 from blib2to3.pgen2.parse import ParseError
 
 
 from blib2to3.pgen2.parse import ParseError
 
 
-__version__ = "18.6b4"
+__version__ = "18.9b0"
 DEFAULT_LINE_LENGTH = 88
 DEFAULT_EXCLUDES = (
 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?$"
 CACHE_DIR = Path(user_cache_dir("black", version=__version__))
 )
 DEFAULT_INCLUDES = r"\.pyi?$"
 CACHE_DIR = Path(user_cache_dir("black", version=__version__))
@@ -79,15 +79,15 @@ syms = pygram.python_symbols
 
 
 class NothingChanged(UserWarning):
 
 
 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):
 
 
 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):
 
 
 class WriteBack(Enum):
@@ -115,10 +115,16 @@ class FileMode(Flag):
     PYTHON36 = 1
     PYI = 2
     NO_STRING_NORMALIZATION = 4
     PYTHON36 = 1
     PYI = 2
     NO_STRING_NORMALIZATION = 4
+    NO_NUMERIC_UNDERSCORE_NORMALIZATION = 8
 
     @classmethod
     def from_configuration(
 
     @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:
     ) -> "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
             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
 
 
         return mode
 
 
@@ -196,6 +204,12 @@ def read_pyproject_toml(
     is_flag=True,
     help="Don't normalize string quotes or prefixes.",
 )
     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,
 @click.option(
     "--check",
     is_flag=True,
@@ -286,6 +300,7 @@ def main(
     pyi: bool,
     py36: bool,
     skip_string_normalization: bool,
     pyi: bool,
     py36: bool,
     skip_string_normalization: bool,
+    skip_numeric_underscore_normalization: bool,
     quiet: bool,
     verbose: bool,
     include: str,
     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(
     """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")
     )
     if config and verbose:
         out(f"Using configuration from {config}.", bold=False, fg="blue")
@@ -607,7 +625,7 @@ def format_str(
 
     `line_length` determines how many characters per line are allowed.
     """
 
     `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)
     dst_contents = ""
     future_imports = get_future_imports(src_node)
     is_pyi = bool(mode & FileMode.PYI)
@@ -618,7 +636,8 @@ def format_str(
         remove_u_prefix=py36 or "unicode_literals" in future_imports,
         is_pyi=is_pyi,
         normalize_strings=normalize_strings,
         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()
     )
     elt = EmptyLineTracker(is_pyi=is_pyi)
     empty_line = Line()
@@ -660,7 +679,6 @@ GRAMMARS = [
 
 def lib2to3_parse(src_txt: str) -> Node:
     """Given a string with source, return the lib2to3 Node."""
 
 def lib2to3_parse(src_txt: str) -> Node:
     """Given a string with source, return the lib2to3 Node."""
-    grammar = pygram.python_grammar_no_print_statement
     if src_txt[-1:] != "\n":
         src_txt += "\n"
     for grammar in GRAMMARS:
     if src_txt[-1:] != "\n":
         src_txt += "\n"
     for grammar in GRAMMARS:
@@ -676,7 +694,7 @@ def lib2to3_parse(src_txt: str) -> Node:
                 faulty_line = lines[lineno - 1]
             except IndexError:
                 faulty_line = "<line number missing in source>"
                 faulty_line = lines[lineno - 1]
             except IndexError:
                 faulty_line = "<line number missing in source>"
-            exc = ValueError(f"Cannot parse: {lineno}:{column}: {faulty_line}")
+            exc = InvalidInput(f"Cannot parse: {lineno}:{column}: {faulty_line}")
     else:
         raise exc from None
 
     else:
         raise exc from None
 
@@ -2013,6 +2031,16 @@ def generate_comments(leaf: LN) -> Iterator[Leaf]:
 
 @dataclass
 class ProtoComment:
 
 @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
     type: int  # token.COMMENT or STANDALONE_COMMENT
     value: str  # content of the comment
     newlines: int  # how many newlines before the comment
@@ -2021,6 +2049,7 @@ class ProtoComment:
 
 @lru_cache(maxsize=4096)
 def list_comments(prefix: str, *, is_endmarker: bool) -> List[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
     result: List[ProtoComment] = []
     if not prefix or "#" not in prefix:
         return result
@@ -2052,8 +2081,8 @@ 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.
 
 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.
     """
 
     If `content` didn't start with a hash sign, one is provided.
     """
@@ -2063,7 +2092,7 @@ def make_comment(content: str) -> str:
 
     if content[0] == "#":
         content = content[1:]
 
     if content[0] == "#":
         content = content[1:]
-    if content and content[0] not in " !:#":
+    if content and content[0] not in " !:#'%":
         content = " " + content
     return "#" + content
 
         content = " " + content
     return "#" + content
 
@@ -2126,7 +2155,7 @@ def split_line(
                 result.extend(
                     split_line(l, line_length=line_length, inner=True, py36=py36)
                 )
                 result.extend(
                     split_line(l, line_length=line_length, inner=True, py36=py36)
                 )
-        except CannotSplit as cs:
+        except CannotSplit:
             continue
 
         else:
             continue
 
         else:
@@ -2144,9 +2173,6 @@ 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.
     """
     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] = []
     tail_leaves: List[Leaf] = []
     body_leaves: List[Leaf] = []
     head_leaves: List[Leaf] = []
@@ -2164,15 +2190,12 @@ def left_hand_split(line: Line, py36: bool = False) -> Iterator[Line]:
             if leaf.type in OPENING_BRACKETS:
                 matching_bracket = leaf
                 current_leaves = body_leaves
             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:
     bracket_split_succeeded_or_raise(head, body, tail)
     for result in (head, body, tail):
         if result:
@@ -2190,9 +2213,6 @@ def right_hand_split(
 
     Note: running this function modifies `bracket_depth` on the leaves of `line`.
     """
 
     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] = []
     tail_leaves: List[Leaf] = []
     body_leaves: List[Leaf] = []
     head_leaves: List[Leaf] = []
@@ -2209,25 +2229,18 @@ def right_hand_split(
                 opening_bracket = leaf.opening_bracket
                 closing_bracket = leaf
                 current_leaves = body_leaves
                 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")
 
         # 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
     bracket_split_succeeded_or_raise(head, body, tail)
     if (
         # the body shouldn't be exploded
@@ -2301,6 +2314,35 @@ 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`.
 
 def dont_increase_indentation(split_func: SplitFunc) -> SplitFunc:
     """Normalize prefix of the first leaf in every line returned by `split_func`.
 
@@ -2347,7 +2389,7 @@ def delimiter_split(line: Line, py36: bool = False) -> Iterator[Line]:
         nonlocal current_line
         try:
             current_line.append_safe(leaf, preformatted=True)
         nonlocal current_line
         try:
             current_line.append_safe(leaf, preformatted=True)
-        except ValueError as ve:
+        except ValueError:
             yield current_line
 
             current_line = Line(depth=line.depth, inside_brackets=line.inside_brackets)
             yield current_line
 
             current_line = Line(depth=line.depth, inside_brackets=line.inside_brackets)
@@ -2531,9 +2573,13 @@ def normalize_numeric_literal(leaf: Leaf, allow_underscores: bool) -> None:
     in Python 2 long literals), and long number literals are split using underscores.
     """
     text = leaf.value.lower()
     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.
+    if text.startswith(("0o", "0b")):
+        # Leave octal and binary literals alone.
         pass
         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 = ""
     elif "e" in text:
         before, after = text.split("e")
         sign = ""
@@ -2585,8 +2631,8 @@ def format_int_string(
         return text
 
     text = text.replace("_", "")
         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
 
     if count_from_end:
         return text
 
     if count_from_end:
@@ -3616,6 +3662,10 @@ def patch_click() -> None:
             module._verify_python3_env = lambda: None
 
 
             module._verify_python3_env = lambda: None
 
 
-if __name__ == "__main__":
+def patched_main() -> None:
     patch_click()
     main()
     patch_click()
     main()
+
+
+if __name__ == "__main__":
+    patched_main()