[blib2to3] Support non-ASCII identifiers

[etc/vim.git] / black.py

diff --git a/black.py b/black.py
index 4995b53d51b0b4b4cc4270b9595a22abd94ead17..9b144edb579d923d7479bdae8565f5df7507fd81 100644 (file)
--- a/black.py
+++ b/black.py
@@ -10,6 +10,7 @@ import logging
 from multiprocessing import Manager
 import os
 from pathlib import Path
 from multiprocessing import Manager
 import os
 from pathlib import Path

+import re

 import tokenize
 import signal
 import sys
 import tokenize
 import signal
 import sys


@@ -561,12 +562,13 @@ class BracketTracker:
             leaf.opening_bracket = opening_bracket
         leaf.bracket_depth = self.depth
         if self.depth == 0:
             leaf.opening_bracket = opening_bracket
         leaf.bracket_depth = self.depth
         if self.depth == 0:

-            after_delim = is_split_after_delimiter(leaf, self.previous)
-            before_delim = is_split_before_delimiter(leaf, self.previous)
-            if after_delim > before_delim:
-                self.delimiters[id(leaf)] = after_delim
-            elif before_delim > after_delim and self.previous is not None:
-                self.delimiters[id(self.previous)] = before_delim
+            delim = is_split_before_delimiter(leaf, self.previous)
+            if delim and self.previous is not None:
+                self.delimiters[id(self.previous)] = delim
+            else:
+                delim = is_split_after_delimiter(leaf, self.previous)
+                if delim:
+                    self.delimiters[id(leaf)] = delim

         if leaf.type in OPENING_BRACKETS:
             self.bracket_match[self.depth, BRACKET[leaf.type]] = leaf
             self.depth += 1
         if leaf.type in OPENING_BRACKETS:
             self.bracket_match[self.depth, BRACKET[leaf.type]] = leaf
             self.depth += 1


@@ -1438,13 +1440,6 @@ def is_split_after_delimiter(leaf: Leaf, previous: Leaf = None) -> int:
     if leaf.type == token.COMMA:
         return COMMA_PRIORITY
 
     if leaf.type == token.COMMA:
         return COMMA_PRIORITY
 

-    if (
-        leaf.type in VARARGS
-        and leaf.parent
-        and leaf.parent.type in {syms.argument, syms.typedargslist}
-    ):
-        return MATH_PRIORITY
-

     return 0
 
 
     return 0
 
 


@@ -1456,6 +1451,15 @@ def is_split_before_delimiter(leaf: Leaf, previous: Leaf = None) -> int:
 
     Higher numbers are higher priority.
     """
 
     Higher numbers are higher priority.
     """

+    if (
+        leaf.type in VARARGS
+        and leaf.parent
+        and leaf.parent.type in {syms.argument, syms.typedargslist}
+    ):
+        # * and ** might also be MATH_OPERATORS but in this case they are not.
+        # Don't treat them as a delimiter.
+        return 0
+

     if (
         leaf.type in MATH_OPERATORS
         and leaf.parent
     if (
         leaf.type in MATH_OPERATORS
         and leaf.parent


@@ -1919,8 +1923,10 @@ def normalize_string_quotes(leaf: Leaf) -> None:
 
     prefix = leaf.value[:first_quote_pos]
     body = leaf.value[first_quote_pos + len(orig_quote):-len(orig_quote)]
 
     prefix = leaf.value[:first_quote_pos]
     body = leaf.value[first_quote_pos + len(orig_quote):-len(orig_quote)]

+    unescaped_new_quote = re.compile(r"(([^\\]|^)(\\\\)*)" + new_quote)
+    escaped_orig_quote = re.compile(r"\\(\\\\)*" + orig_quote)

     if "r" in prefix.casefold():
     if "r" in prefix.casefold():

-        if body.count(new_quote) != body.count(f"\\{new_quote}"):
+        if unescaped_new_quote.search(body):

             # There's at least one unescaped new_quote in this raw string
             # so converting is impossible
             return
             # There's at least one unescaped new_quote in this raw string
             # so converting is impossible
             return


@@ -1928,9 +1934,8 @@ def normalize_string_quotes(leaf: Leaf) -> None:
         # Do not introduce or remove backslashes in raw strings
         new_body = body
     else:
         # Do not introduce or remove backslashes in raw strings
         new_body = body
     else:

-        new_body = body.replace(f"\\{orig_quote}", orig_quote).replace(
-            new_quote, f"\\{new_quote}"
-        )
+        new_body = escaped_orig_quote.sub(f"\\1{orig_quote}", body)
+        new_body = unescaped_new_quote.sub(f"\\1\\\\{new_quote}", new_body)

     if new_quote == '"""' and new_body[-1] == '"':
         # edge case:
         new_body = new_body[:-1] + '\\"'
     if new_quote == '"""' and new_body[-1] == '"':
         # edge case:
         new_body = new_body[:-1] + '\\"'