]> 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:

Re-use indexes of current iteration in `comments_after()`
[etc/vim.git] / black.py
index 36380e6ae847b2314b4ce0e8d2c816071cdc418b..43b9bd17f6f2dcd3b52648908e54239620ddd4b3 100644 (file)
--- a/black.py
+++ b/black.py
@@ -409,9 +409,10 @@ def format_str(src_contents: str, line_length: int) -> FileContent:
     """
     src_node = lib2to3_parse(src_contents)
     dst_contents = ""
-    lines = LineGenerator()
-    elt = EmptyLineTracker()
+    future_imports = get_future_imports(src_node)
     py36 = is_python36(src_node)
+    lines = LineGenerator(remove_u_prefix=py36 or "unicode_literals" in future_imports)
+    elt = EmptyLineTracker()
     empty_line = Line()
     after = 0
     for current_line in lines.visit(src_node):
@@ -599,6 +600,22 @@ TEST_DESCENDANTS = {
     syms.term,
     syms.power,
 }
+ASSIGNMENTS = {
+    "=",
+    "+=",
+    "-=",
+    "*=",
+    "@=",
+    "/=",
+    "%=",
+    "&=",
+    "|=",
+    "^=",
+    "<<=",
+    ">>=",
+    "**=",
+    "//=",
+}
 COMPREHENSION_PRIORITY = 20
 COMMA_PRIORITY = 18
 TERNARY_PRIORITY = 16
@@ -631,8 +648,8 @@ class BracketTracker:
     bracket_match: Dict[Tuple[Depth, NodeType], Leaf] = Factory(dict)
     delimiters: Dict[LeafID, Priority] = Factory(dict)
     previous: Optional[Leaf] = None
-    _for_loop_variable: bool = False
-    _lambda_arguments: bool = False
+    _for_loop_variable: int = 0
+    _lambda_arguments: int = 0
 
     def mark(self, leaf: Leaf) -> None:
         """Mark `leaf` with bracket-related metadata. Keep track of delimiters.
@@ -694,7 +711,7 @@ class BracketTracker:
         """
         if leaf.type == token.NAME and leaf.value == "for":
             self.depth += 1
-            self._for_loop_variable = True
+            self._for_loop_variable += 1
             return True
 
         return False
@@ -703,7 +720,7 @@ class BracketTracker:
         """See `maybe_increment_for_loop_variable` above for explanation."""
         if self._for_loop_variable and leaf.type == token.NAME and leaf.value == "in":
             self.depth -= 1
-            self._for_loop_variable = False
+            self._for_loop_variable -= 1
             return True
 
         return False
@@ -716,7 +733,7 @@ class BracketTracker:
         """
         if leaf.type == token.NAME and leaf.value == "lambda":
             self.depth += 1
-            self._lambda_arguments = True
+            self._lambda_arguments += 1
             return True
 
         return False
@@ -725,7 +742,7 @@ class BracketTracker:
         """See `maybe_increment_lambda_arguments` above for explanation."""
         if self._lambda_arguments and leaf.type == token.COLON:
             self.depth -= 1
-            self._lambda_arguments = False
+            self._lambda_arguments -= 1
             return True
 
         return False
@@ -958,17 +975,21 @@ class Line:
             self.comments.append((after, comment))
             return True
 
-    def comments_after(self, leaf: Leaf) -> Iterator[Leaf]:
-        """Generate comments that should appear directly after `leaf`."""
-        for _leaf_index, _leaf in enumerate(self.leaves):
-            if leaf is _leaf:
-                break
+    def comments_after(self, leaf: Leaf, _index: int = -1) -> Iterator[Leaf]:
+        """Generate comments that should appear directly after `leaf`.
 
-        else:
-            return
+        Provide a non-negative leaf `_index` to speed up the function.
+        """
+        if _index == -1:
+            for _index, _leaf in enumerate(self.leaves):
+                if leaf is _leaf:
+                    break
+
+            else:
+                return
 
         for index, comment_after in self.comments:
-            if _leaf_index == index:
+            if _index == index:
                 yield comment_after
 
     def remove_trailing_comma(self) -> None:
@@ -994,8 +1015,9 @@ class Line:
             and subscript_start.type == syms.subscriptlist
         ):
             subscript_start = child_towards(subscript_start, leaf)
-        return subscript_start is not None and any(
-            n.type in TEST_DESCENDANTS for n in subscript_start.pre_order()
+        return (
+            subscript_start is not None
+            and any(n.type in TEST_DESCENDANTS for n in subscript_start.pre_order())
         )
 
     def __str__(self) -> str:
@@ -1154,6 +1176,7 @@ class LineGenerator(Visitor[Line]):
     in ways that will no longer stringify to valid Python code on the tree.
     """
     current_line: Line = Factory(Line)
+    remove_u_prefix: bool = False
 
     def line(self, indent: int = 0, type: Type[Line] = Line) -> Iterator[Line]:
         """Generate a line.
@@ -1221,6 +1244,7 @@ class LineGenerator(Visitor[Line]):
             else:
                 normalize_prefix(node, inside_brackets=any_open_brackets)
                 if node.type == token.STRING:
+                    normalize_string_prefix(node, remove_u_prefix=self.remove_u_prefix)
                     normalize_string_quotes(node)
                 if node.type not in WHITESPACE:
                     self.current_line.append(node)
@@ -1252,14 +1276,13 @@ class LineGenerator(Visitor[Line]):
         """Visit a statement.
 
         This implementation is shared for `if`, `while`, `for`, `try`, `except`,
-        `def`, `with`, `class`, and `assert`.
+        `def`, `with`, `class`, `assert` and assignments.
 
         The relevant Python language `keywords` for a given statement will be
         NAME leaves within it. This methods puts those on a separate line.
 
-        `parens` holds pairs of nodes where invisible parentheses should be put.
-        Keys hold nodes after which opening parentheses should be put, values
-        hold nodes before which closing parentheses should be put.
+        `parens` holds a set of string leaf values immeditely after which
+        invisible parens should be put.
         """
         normalize_invisible_parens(node, parens_after=parens)
         for child in node.children:
@@ -1359,7 +1382,9 @@ class LineGenerator(Visitor[Line]):
         v = self.visit_stmt
         Ø: Set[str] = set()
         self.visit_assert_stmt = partial(v, keywords={"assert"}, parens={"assert", ","})
-        self.visit_if_stmt = partial(v, keywords={"if", "else", "elif"}, parens={"if"})
+        self.visit_if_stmt = partial(
+            v, keywords={"if", "else", "elif"}, parens={"if", "elif"}
+        )
         self.visit_while_stmt = partial(v, keywords={"while", "else"}, parens={"while"})
         self.visit_for_stmt = partial(v, keywords={"for", "else"}, parens={"for", "in"})
         self.visit_try_stmt = partial(
@@ -1369,6 +1394,8 @@ class LineGenerator(Visitor[Line]):
         self.visit_with_stmt = partial(v, keywords={"with"}, parens=Ø)
         self.visit_funcdef = partial(v, keywords={"def"}, parens=Ø)
         self.visit_classdef = partial(v, keywords={"class"}, parens=Ø)
+        self.visit_expr_stmt = partial(v, keywords=Ø, parens=ASSIGNMENTS)
+        self.visit_return_stmt = partial(v, keywords={"return"}, parens={"return"})
         self.visit_async_funcdef = self.visit_async_stmt
         self.visit_decorated = self.visit_decorators
 
@@ -2032,10 +2059,10 @@ def delimiter_split(line: Line, py36: bool = False) -> Iterator[Line]:
             current_line = Line(depth=line.depth, inside_brackets=line.inside_brackets)
             current_line.append(leaf)
 
-    for leaf in line.leaves:
+    for index, leaf in enumerate(line.leaves):
         yield from append_to_line(leaf)
 
-        for comment_after in line.comments_after(leaf):
+        for comment_after in line.comments_after(leaf, index):
             yield from append_to_line(comment_after)
 
         lowest_depth = min(lowest_depth, leaf.bracket_depth)
@@ -2079,10 +2106,10 @@ def standalone_comment_split(line: Line, py36: bool = False) -> Iterator[Line]:
             current_line = Line(depth=line.depth, inside_brackets=line.inside_brackets)
             current_line.append(leaf)
 
-    for leaf in line.leaves:
+    for index, leaf in enumerate(line.leaves):
         yield from append_to_line(leaf)
 
-        for comment_after in line.comments_after(leaf):
+        for comment_after in line.comments_after(leaf, index):
             yield from append_to_line(comment_after)
 
     if current_line:
@@ -2141,6 +2168,22 @@ def normalize_prefix(leaf: Leaf, *, inside_brackets: bool) -> None:
     leaf.prefix = ""
 
 
+def normalize_string_prefix(leaf: Leaf, remove_u_prefix: bool = False) -> None:
+    """Make all string prefixes lowercase.
+
+    If remove_u_prefix is given, also removes any u prefix from the string.
+
+    Note: Mutates its argument.
+    """
+    match = re.match(r"^([furbFURB]*)(.*)$", leaf.value, re.DOTALL)
+    assert match is not None, f"failed to match string {leaf.value!r}"
+    orig_prefix = match.group(1)
+    new_prefix = orig_prefix.lower()
+    if remove_u_prefix:
+        new_prefix = new_prefix.replace("u", "")
+    leaf.value = f"{new_prefix}{match.group(2)}"
+
+
 def normalize_string_quotes(leaf: Leaf) -> None:
     """Prefer double quotes but only if it doesn't cause more escaping.
 
@@ -2205,6 +2248,9 @@ def normalize_string_quotes(leaf: Leaf) -> None:
 def normalize_invisible_parens(node: Node, parens_after: Set[str]) -> None:
     """Make existing optional parentheses invisible or create new ones.
 
+    `parens_after` is a set of string leaf values immeditely after which parens
+    should be put.
+
     Standardizes on visible parentheses for single-element tuples, and keeps
     existing visible parentheses for other tuples and generator expressions.
     """
@@ -2235,6 +2281,7 @@ def maybe_make_parens_invisible_in_atom(node: LN) -> bool:
         node.type != syms.atom
         or is_empty_tuple(node)
         or is_one_tuple(node)
+        or is_yield(node)
         or max_delimiter_priority_in_atom(node) >= COMMA_PRIORITY
     ):
         return False
@@ -2285,12 +2332,33 @@ def is_one_tuple(node: LN) -> bool:
     )
 
 
+def is_yield(node: LN) -> bool:
+    """Return True if `node` holds a `yield` or `yield from` expression."""
+    if node.type == syms.yield_expr:
+        return True
+
+    if node.type == token.NAME and node.value == "yield":  # type: ignore
+        return True
+
+    if node.type != syms.atom:
+        return False
+
+    if len(node.children) != 3:
+        return False
+
+    lpar, expr, rpar = node.children
+    if lpar.type == token.LPAR and rpar.type == token.RPAR:
+        return is_yield(expr)
+
+    return False
+
+
 def is_vararg(leaf: Leaf, within: Set[NodeType]) -> bool:
     """Return True if `leaf` is a star or double star in a vararg or kwarg.
 
     If `within` includes VARARGS_PARENTS, this applies to function signatures.
-    If `within` includes COLLECTION_LIBERALS_PARENTS, it applies to right
-    hand-side extended iterable unpacking (PEP 3132) and additional unpacking
+    If `within` includes UNPACKING_PARENTS, it applies to right hand-side
+    extended iterable unpacking (PEP 3132) and additional unpacking
     generalizations (PEP 448).
     """
     if leaf.type not in STARS or not leaf.parent:
@@ -2378,6 +2446,41 @@ def is_python36(node: Node) -> bool:
     return False
 
 
+def get_future_imports(node: Node) -> Set[str]:
+    """Return a set of __future__ imports in the file."""
+    imports = set()
+    for child in node.children:
+        if child.type != syms.simple_stmt:
+            break
+        first_child = child.children[0]
+        if isinstance(first_child, Leaf):
+            # Continue looking if we see a docstring; otherwise stop.
+            if (
+                len(child.children) == 2
+                and first_child.type == token.STRING
+                and child.children[1].type == token.NEWLINE
+            ):
+                continue
+            else:
+                break
+        elif first_child.type == syms.import_from:
+            module_name = first_child.children[1]
+            if not isinstance(module_name, Leaf) or module_name.value != "__future__":
+                break
+            for import_from_child in first_child.children[3:]:
+                if isinstance(import_from_child, Leaf):
+                    if import_from_child.type == token.NAME:
+                        imports.add(import_from_child.value)
+                else:
+                    assert import_from_child.type == syms.import_as_names
+                    for leaf in import_from_child.children:
+                        if isinstance(leaf, Leaf) and leaf.type == token.NAME:
+                            imports.add(leaf.value)
+        else:
+            break
+    return imports
+
+
 PYTHON_EXTENSIONS = {".py"}
 BLACKLISTED_DIRECTORIES = {
     "build", "buck-out", "dist", "_build", ".git", ".hg", ".mypy_cache", ".tox", ".venv"
@@ -2395,7 +2498,7 @@ def gen_python_files_in_dir(path: Path) -> Iterator[Path]:
 
             yield from gen_python_files_in_dir(child)
 
-        elif child.suffix in PYTHON_EXTENSIONS:
+        elif child.is_file() and child.suffix in PYTHON_EXTENSIONS:
             yield child