]> 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:

Add flake8 to CI, too
[etc/vim.git] / black.py
index 774d91dc2a7f64ec2dbc252a55e0fe051d5c429c..48ba0a266cc7c5354f87ec30c13600aed8528599 100644 (file)
--- a/black.py
+++ b/black.py
@@ -7,6 +7,7 @@ import keyword
 import os
 from pathlib import Path
 import tokenize
 import os
 from pathlib import Path
 import tokenize
+import sys
 from typing import (
     Dict, Generic, Iterable, Iterator, List, Optional, Set, Tuple, TypeVar, Union
 )
 from typing import (
     Dict, Generic, Iterable, Iterator, List, Optional, Set, Tuple, TypeVar, Union
 )
@@ -20,7 +21,7 @@ from blib2to3 import pygram, pytree
 from blib2to3.pgen2 import driver, token
 from blib2to3.pgen2.parse import ParseError
 
 from blib2to3.pgen2 import driver, token
 from blib2to3.pgen2.parse import ParseError
 
-__version__ = "18.3a1"
+__version__ = "18.3a2"
 DEFAULT_LINE_LENGTH = 88
 # types
 syms = pygram.python_symbols
 DEFAULT_LINE_LENGTH = 88
 # types
 syms = pygram.python_symbols
@@ -192,6 +193,7 @@ def format_str(src_contents: str, line_length: int) -> FileContent:
     comments: List[Line] = []
     lines = LineGenerator()
     elt = EmptyLineTracker()
     comments: List[Line] = []
     lines = LineGenerator()
     elt = EmptyLineTracker()
+    py36 = is_python36(src_node)
     empty_line = Line()
     after = 0
     for current_line in lines.visit(src_node):
     empty_line = Line()
     after = 0
     for current_line in lines.visit(src_node):
@@ -204,7 +206,7 @@ def format_str(src_contents: str, line_length: int) -> FileContent:
             for comment in comments:
                 dst_contents += str(comment)
             comments = []
             for comment in comments:
                 dst_contents += str(comment)
             comments = []
-            for line in split_line(current_line, line_length=line_length):
+            for line in split_line(current_line, line_length=line_length, py36=py36):
                 dst_contents += str(line)
         else:
             comments.append(current_line)
                 dst_contents += str(line)
         else:
             comments.append(current_line)
@@ -345,8 +347,8 @@ class BracketTracker:
         if leaf.type in CLOSING_BRACKETS:
             self.depth -= 1
             opening_bracket = self.bracket_match.pop((self.depth, leaf.type))
         if leaf.type in CLOSING_BRACKETS:
             self.depth -= 1
             opening_bracket = self.bracket_match.pop((self.depth, leaf.type))
-            leaf.opening_bracket = opening_bracket  # type: ignore
-        leaf.bracket_depth = self.depth  # type: ignore
+            leaf.opening_bracket = opening_bracket
+        leaf.bracket_depth = self.depth
         if self.depth == 0:
             delim = is_delimiter(leaf)
             if delim:
         if self.depth == 0:
             delim = is_delimiter(leaf)
             if delim:
@@ -489,9 +491,9 @@ class Line:
         # For parens let's check if it's safe to remove the comma.  If the
         # trailing one is the only one, we might mistakenly change a tuple
         # into a different type by removing the comma.
         # For parens let's check if it's safe to remove the comma.  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  # type: ignore
+        depth = closing.bracket_depth + 1
         commas = 0
         commas = 0
-        opening = closing.opening_bracket  # type: ignore
+        opening = closing.opening_bracket
         for _opening_index, leaf in enumerate(self.leaves):
             if leaf is opening:
                 break
         for _opening_index, leaf in enumerate(self.leaves):
             if leaf is opening:
                 break
@@ -503,7 +505,7 @@ class Line:
             if leaf is closing:
                 break
 
             if leaf is closing:
                 break
 
-            bracket_depth = leaf.bracket_depth  # type: ignore
+            bracket_depth = leaf.bracket_depth
             if bracket_depth == depth and leaf.type == token.COMMA:
                 commas += 1
         if commas > 1:
             if bracket_depth == depth and leaf.type == token.COMMA:
                 commas += 1
         if commas > 1:
@@ -802,9 +804,10 @@ BRACKET = {token.LPAR: token.RPAR, token.LSQB: token.RSQB, token.LBRACE: token.R
 OPENING_BRACKETS = set(BRACKET.keys())
 CLOSING_BRACKETS = set(BRACKET.values())
 BRACKETS = OPENING_BRACKETS | CLOSING_BRACKETS
 OPENING_BRACKETS = set(BRACKET.keys())
 CLOSING_BRACKETS = set(BRACKET.values())
 BRACKETS = OPENING_BRACKETS | CLOSING_BRACKETS
+ALWAYS_NO_SPACE = CLOSING_BRACKETS | {token.COMMA, token.COLON, STANDALONE_COMMENT}
 
 
 
 
-def whitespace(leaf: Leaf) -> str:
+def whitespace(leaf: Leaf) -> str:  # noqa C901
     """Return whitespace prefix if needed for the given `leaf`."""
     NO = ''
     SPACE = ' '
     """Return whitespace prefix if needed for the given `leaf`."""
     NO = ''
     SPACE = ' '
@@ -812,24 +815,12 @@ def whitespace(leaf: Leaf) -> str:
     t = leaf.type
     p = leaf.parent
     v = leaf.value
     t = leaf.type
     p = leaf.parent
     v = leaf.value
-    if t == token.COLON:
-        return NO
-
-    if t == token.COMMA:
-        return NO
-
-    if t == token.RPAR:
+    if t in ALWAYS_NO_SPACE:
         return NO
 
     if t == token.COMMENT:
         return DOUBLESPACE
 
         return NO
 
     if t == token.COMMENT:
         return DOUBLESPACE
 
-    if t == STANDALONE_COMMENT:
-        return NO
-
-    if t in CLOSING_BRACKETS:
-        return NO
-
     assert p is not None, f"INTERNAL ERROR: hand-made leaf without parent: {leaf!r}"
     prev = leaf.prev_sibling
     if not prev:
     assert p is not None, f"INTERNAL ERROR: hand-made leaf without parent: {leaf!r}"
     prev = leaf.prev_sibling
     if not prev:
@@ -861,7 +852,7 @@ def whitespace(leaf: Leaf) -> str:
             if prevp.parent and prevp.parent.type == syms.subscript:
                 return NO
 
             if prevp.parent and prevp.parent.type == syms.subscript:
                 return NO
 
-        elif prevp.parent and prevp.parent.type == syms.factor:
+        elif prevp.parent and prevp.parent.type in {syms.factor, syms.star_expr}:
             return NO
 
     elif prev.type in OPENING_BRACKETS:
             return NO
 
     elif prev.type in OPENING_BRACKETS:
@@ -958,7 +949,14 @@ def whitespace(leaf: Leaf) -> str:
 
     elif p.type == syms.subscript:
         # indexing
 
     elif p.type == syms.subscript:
         # indexing
-        if not prev or prev.type == token.COLON:
+        if not prev:
+            assert p.parent is not None, "subscripts are always parented"
+            if p.parent.type == syms.subscriptlist:
+                return SPACE
+
+            return NO
+
+        elif prev.type == token.COLON:
             return NO
 
     elif p.type == syms.atom:
             return NO
 
     elif p.type == syms.atom:
@@ -983,7 +981,7 @@ def whitespace(leaf: Leaf) -> str:
         if prev.type == token.DOUBLESTAR:
             return NO
 
         if prev.type == token.DOUBLESTAR:
             return NO
 
-    elif p.type == syms.factor or p.type == syms.star_expr:
+    elif p.type in {syms.factor, syms.star_expr}:
         # unary ops
         if not prev:
             prevp = preceding_leaf(p)
         # unary ops
         if not prev:
             prevp = preceding_leaf(p)
@@ -1108,13 +1106,18 @@ def generate_comments(leaf: Leaf) -> Iterator[Leaf]:
         yield Leaf(STANDALONE_COMMENT, line)
 
 
         yield Leaf(STANDALONE_COMMENT, line)
 
 
-def split_line(line: Line, line_length: int, inner: bool = False) -> Iterator[Line]:
+def split_line(
+    line: Line, line_length: int, inner: bool = False, py36: bool = False
+) -> Iterator[Line]:
     """Splits a `line` into potentially many lines.
 
     They should fit in the allotted `line_length` but might not be able to.
     `inner` signifies that there were a pair of brackets somewhere around the
     current `line`, possibly transitively. This means we can fallback to splitting
     by delimiters if the LHS/RHS don't yield any results.
     """Splits a `line` into potentially many lines.
 
     They should fit in the allotted `line_length` but might not be able to.
     `inner` signifies that there were a pair of brackets somewhere around the
     current `line`, possibly transitively. This means we can fallback to splitting
     by delimiters if the LHS/RHS don't yield any results.
+
+    If `py36` is True, splitting may generate syntax that is only compatible
+    with Python 3.6 and later.
     """
     line_str = str(line).strip('\n')
     if len(line_str) <= line_length and '\n' not in line_str:
     """
     line_str = str(line).strip('\n')
     if len(line_str) <= line_length and '\n' not in line_str:
@@ -1137,11 +1140,13 @@ def split_line(line: Line, line_length: int, inner: bool = False) -> Iterator[Li
         # split altogether.
         result: List[Line] = []
         try:
         # split altogether.
         result: List[Line] = []
         try:
-            for l in split_func(line):
+            for l in split_func(line, py36=py36):
                 if str(l).strip('\n') == line_str:
                     raise CannotSplit("Split function returned an unchanged result")
 
                 if str(l).strip('\n') == line_str:
                     raise CannotSplit("Split function returned an unchanged result")
 
-                result.extend(split_line(l, line_length=line_length, inner=True))
+                result.extend(
+                    split_line(l, line_length=line_length, inner=True, py36=py36)
+                )
         except CannotSplit as cs:
             continue
 
         except CannotSplit as cs:
             continue
 
@@ -1153,7 +1158,7 @@ def split_line(line: Line, line_length: int, inner: bool = False) -> Iterator[Li
         yield line
 
 
         yield line
 
 
-def left_hand_split(line: Line) -> Iterator[Line]:
+def left_hand_split(line: Line, py36: bool = False) -> Iterator[Line]:
     """Split line into many lines, starting with the first matching bracket pair.
 
     Note: this usually looks weird, only use this for function definitions.
     """Split line into many lines, starting with the first matching bracket pair.
 
     Note: this usually looks weird, only use this for function definitions.
@@ -1171,7 +1176,7 @@ def left_hand_split(line: Line) -> Iterator[Line]:
         if (
             current_leaves is body_leaves and
             leaf.type in CLOSING_BRACKETS and
         if (
             current_leaves is body_leaves and
             leaf.type in CLOSING_BRACKETS and
-            leaf.opening_bracket is matching_bracket  # type: ignore
+            leaf.opening_bracket is matching_bracket
         ):
             current_leaves = tail_leaves
         current_leaves.append(leaf)
         ):
             current_leaves = tail_leaves
         current_leaves.append(leaf)
@@ -1208,7 +1213,7 @@ def left_hand_split(line: Line) -> Iterator[Line]:
             yield result
 
 
             yield result
 
 
-def right_hand_split(line: Line) -> Iterator[Line]:
+def right_hand_split(line: Line, py36: bool = False) -> Iterator[Line]:
     """Split line into many lines, starting with the last matching bracket pair."""
     head = Line(depth=line.depth)
     body = Line(depth=line.depth + 1, inside_brackets=True)
     """Split line into many lines, starting with the last matching bracket pair."""
     head = Line(depth=line.depth)
     body = Line(depth=line.depth + 1, inside_brackets=True)
@@ -1225,7 +1230,7 @@ def right_hand_split(line: Line) -> Iterator[Line]:
         current_leaves.append(leaf)
         if current_leaves is tail_leaves:
             if leaf.type in CLOSING_BRACKETS:
         current_leaves.append(leaf)
         if current_leaves is tail_leaves:
             if leaf.type in CLOSING_BRACKETS:
-                opening_bracket = leaf.opening_bracket  # type: ignore
+                opening_bracket = leaf.opening_bracket
                 current_leaves = body_leaves
     tail_leaves.reverse()
     body_leaves.reverse()
                 current_leaves = body_leaves
     tail_leaves.reverse()
     body_leaves.reverse()
@@ -1259,10 +1264,12 @@ def right_hand_split(line: Line) -> Iterator[Line]:
             yield result
 
 
             yield result
 
 
-def delimiter_split(line: Line) -> Iterator[Line]:
+def delimiter_split(line: Line, py36: bool = False) -> Iterator[Line]:
     """Split according to delimiters of the highest priority.
 
     This kind of split doesn't increase indentation.
     """Split according to delimiters of the highest priority.
 
     This kind of split doesn't increase indentation.
+    If `py36` is True, the split will add trailing commas also in function
+    signatures that contain * and **.
     """
     try:
         last_leaf = line.leaves[-1]
     """
     try:
         last_leaf = line.leaves[-1]
@@ -1276,11 +1283,20 @@ def delimiter_split(line: Line) -> Iterator[Line]:
         raise CannotSplit("No delimiters found")
 
     current_line = Line(depth=line.depth, inside_brackets=line.inside_brackets)
         raise CannotSplit("No delimiters found")
 
     current_line = Line(depth=line.depth, inside_brackets=line.inside_brackets)
+    lowest_depth = sys.maxsize
+    trailing_comma_safe = True
     for leaf in line.leaves:
         current_line.append(leaf, preformatted=True)
         comment_after = line.comments.get(id(leaf))
         if comment_after:
             current_line.append(comment_after, preformatted=True)
     for leaf in line.leaves:
         current_line.append(leaf, preformatted=True)
         comment_after = line.comments.get(id(leaf))
         if comment_after:
             current_line.append(comment_after, preformatted=True)
+        lowest_depth = min(lowest_depth, leaf.bracket_depth)
+        if (
+            leaf.bracket_depth == lowest_depth and
+            leaf.type == token.STAR or
+            leaf.type == token.DOUBLESTAR
+        ):
+            trailing_comma_safe = trailing_comma_safe and py36
         leaf_priority = delimiters.get(id(leaf))
         if leaf_priority == delimiter_priority:
             normalize_prefix(current_line.leaves[0])
         leaf_priority = delimiters.get(id(leaf))
         if leaf_priority == delimiter_priority:
             normalize_prefix(current_line.leaves[0])
@@ -1290,7 +1306,8 @@ def delimiter_split(line: Line) -> Iterator[Line]:
     if current_line:
         if (
             delimiter_priority == COMMA_PRIORITY and
     if current_line:
         if (
             delimiter_priority == COMMA_PRIORITY and
-            current_line.leaves[-1].type != token.COMMA
+            current_line.leaves[-1].type != token.COMMA and
+            trailing_comma_safe
         ):
             current_line.append(Leaf(token.COMMA, ','))
         normalize_prefix(current_line.leaves[0])
         ):
             current_line.append(Leaf(token.COMMA, ','))
         normalize_prefix(current_line.leaves[0])
@@ -1325,6 +1342,31 @@ def normalize_prefix(leaf: Leaf) -> None:
     leaf.prefix = ''
 
 
     leaf.prefix = ''
 
 
+def is_python36(node: Node) -> bool:
+    """Returns True if the current file is using Python 3.6+ features.
+
+    Currently looking for:
+    - f-strings; and
+    - trailing commas after * or ** in function signatures.
+    """
+    for n in node.pre_order():
+        if n.type == token.STRING:
+            value_head = n.value[:2]  # type: ignore
+            if value_head in {'f"', 'F"', "f'", "F'", 'rf', 'fr', 'RF', 'FR'}:
+                return True
+
+        elif (
+            n.type == syms.typedargslist and
+            n.children and
+            n.children[-1].type == token.COMMA
+        ):
+            for ch in n.children:
+                if ch.type == token.STAR or ch.type == token.DOUBLESTAR:
+                    return True
+
+    return False
+
+
 PYTHON_EXTENSIONS = {'.py'}
 BLACKLISTED_DIRECTORIES = {
     'build', 'buck-out', 'dist', '_build', '.git', '.hg', '.mypy_cache', '.tox', '.venv'
 PYTHON_EXTENSIONS = {'.py'}
 BLACKLISTED_DIRECTORIES = {
     'build', 'buck-out', 'dist', '_build', '.git', '.hg', '.mypy_cache', '.tox', '.venv'