]> 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:

Only use trailing commas in function signatures when it's safe
[etc/vim.git] / black.py
index e2c427cb536c752b6e5d0e00c07ca2b2a2a6a23e..0a9d3eae46fa1a548f1eb3c9674fc7be3ba6812b 100644 (file)
--- a/black.py
+++ b/black.py
@@ -7,6 +7,7 @@ import keyword
 import os
 from pathlib import Path
 import tokenize
+import sys
 from typing import (
     Dict, Generic, Iterable, Iterator, List, Optional, Set, Tuple, TypeVar, Union
 )
@@ -192,6 +193,7 @@ def format_str(src_contents: str, line_length: int) -> FileContent:
     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):
@@ -204,7 +206,7 @@ def format_str(src_contents: str, line_length: int) -> FileContent:
             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)
@@ -392,6 +394,8 @@ class Line:
     comments: Dict[LeafID, Leaf] = attrib(default=Factory(dict))
     bracket_tracker: BracketTracker = attrib(default=Factory(BracketTracker))
     inside_brackets: bool = attrib(default=False)
+    has_for: bool = attrib(default=False)
+    _for_loop_variable: bool = attrib(default=False, init=False)
 
     def append(self, leaf: Leaf, preformatted: bool = False) -> None:
         has_value = leaf.value.strip()
@@ -403,8 +407,10 @@ class Line:
             # imports, for which we only preserve newlines.
             leaf.prefix += whitespace(leaf)
         if self.inside_brackets or not preformatted:
+            self.maybe_decrement_after_for_loop_variable(leaf)
             self.bracket_tracker.mark(leaf)
             self.maybe_remove_trailing_comma(leaf)
+            self.maybe_increment_for_loop_variable(leaf)
             if self.maybe_adapt_standalone_comment(leaf):
                 return
 
@@ -508,6 +514,29 @@ class Line:
 
         return False
 
+    def maybe_increment_for_loop_variable(self, leaf: Leaf) -> bool:
+        """In a for loop, or comprehension, the variables are often unpacks.
+
+        To avoid splitting on the comma in this situation, we will increase
+        the depth of tokens between `for` and `in`.
+        """
+        if leaf.type == token.NAME and leaf.value == 'for':
+            self.has_for = True
+            self.bracket_tracker.depth += 1
+            self._for_loop_variable = True
+            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':
+            self.bracket_tracker.depth -= 1
+            self._for_loop_variable = False
+            return True
+
+        return False
+
     def maybe_adapt_standalone_comment(self, comment: Leaf) -> bool:
         """Hack a standalone comment to act as a trailing comment for line splitting.
 
@@ -1081,13 +1110,18 @@ def generate_comments(leaf: Leaf) -> Iterator[Leaf]:
         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.
+
+    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:
@@ -1110,11 +1144,13 @@ def split_line(line: Line, line_length: int, inner: bool = False) -> Iterator[Li
         # 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")
 
-                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
 
@@ -1126,7 +1162,7 @@ def split_line(line: Line, line_length: int, inner: bool = False) -> Iterator[Li
         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.
@@ -1181,7 +1217,7 @@ def left_hand_split(line: Line) -> Iterator[Line]:
             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)
@@ -1232,10 +1268,12 @@ def right_hand_split(line: Line) -> Iterator[Line]:
             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.
+    If `py36` is True, the split will add trailing commas also in function
+    signatures that contain * and **.
     """
     try:
         last_leaf = line.leaves[-1]
@@ -1249,11 +1287,20 @@ def delimiter_split(line: Line) -> Iterator[Line]:
         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)
+        lowest_depth = min(lowest_depth, leaf.bracket_depth)
+        if (
+            leaf.bracket_depth == lowest_depth and  # type: ignore
+            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])
@@ -1263,7 +1310,8 @@ def delimiter_split(line: Line) -> Iterator[Line]:
     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])
@@ -1298,6 +1346,31 @@ def normalize_prefix(leaf: Leaf) -> None:
     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:
+            assert isinstance(n, Leaf)
+            if n.value[:2] 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'