Preserve prefixes outside line ranges

Author: cyphercodes

CHANGES.md

@@ -27,6 +27,8 @@
   `from x import (  # fmt: skip`) when a standalone comment is among the bracket's
   contents: the whole statement is now preserved instead of being reformatted (and
   previously crashing) (#5161)
+- Preserve comments and blank lines outside requested ranges when formatting with
+  `--line-ranges` (#5175)
 
 ### Preview style
 

src/black/comments.py

@@ -55,9 +55,12 @@ class ProtoComment:
     consumed: int  # how many characters of the original leaf's prefix did we consume
     form_feed: bool  # is there a form feed before the comment
     leading_whitespace: str  # leading whitespace before the comment, if any
+    original_value: str  # original comment text, before normalization
 
 
-def generate_comments(leaf: LN, mode: Mode) -> Iterator[Leaf]:
+def generate_comments(
+    leaf: LN, mode: Mode, *, preserve_comment_formatting: bool = False
+) -> Iterator[Leaf]:
     """Clean the prefix of the `leaf` and generate comments from it, if any.
 
     Comments in lib2to3 are shoved into the whitespace prefix.  This happens
@@ -82,7 +85,10 @@ def generate_comments(leaf: LN, mode: Mode) -> Iterator[Leaf]:
     ):
         total_consumed = pc.consumed
         prefix = make_simple_prefix(pc.newlines, pc.form_feed)
-        yield Leaf(pc.type, pc.value, prefix=prefix)
+        value = pc.value
+        if preserve_comment_formatting:
+            value = pc.leading_whitespace + pc.original_value
+        yield Leaf(pc.type, value, prefix=prefix)
     normalize_trailing_prefix(leaf, total_consumed)
 
 
@@ -127,6 +133,7 @@ def list_comments(prefix: str, *, is_endmarker: bool, mode: Mode) -> list[ProtoC
                 consumed=consumed,
                 form_feed=form_feed,
                 leading_whitespace=whitespace,
+                original_value=line.rstrip(),
             )
         )
         form_feed = False

src/black/linegen.py

@@ -148,7 +148,20 @@ def visit_default(self, node: LN) -> Iterator[Line]:
         """Default `visit_*()` implementation. Recurses to children of `node`."""
         if isinstance(node, Leaf):
             any_open_brackets = self.current_line.bracket_tracker.any_open_brackets()
-            for comment in generate_comments(node, mode=self.mode):
+            preserve_comment_formatting = (
+                node.type == STANDALONE_COMMENT
+                and node.fmt_pass_converted_first_leaf is not None
+                and node.line_ranges_converted
+            )
+            for comment in generate_comments(
+                node,
+                mode=self.mode,
+                preserve_comment_formatting=preserve_comment_formatting,
+            ):
+                if preserve_comment_formatting:
+                    indent = "    " * self.current_line.depth
+                    if indent and comment.value.startswith(indent):
+                        comment.value = comment.value[len(indent) :]
                 if any_open_brackets:
                     # any comment within brackets is subject to splitting
                     self.current_line.append(comment)

src/black/lines.py

@@ -981,7 +981,8 @@ def _maybe_empty_lines(self, current_line: Line) -> tuple[int, int]:
             # Consume the first leaf's extra newlines.
             first_leaf = current_line.leaves[0]
             before = first_leaf.prefix.count("\n")
-            before = min(before, max_allowed)
+            if not current_line.is_fmt_pass_converted():
+                before = min(before, max_allowed)
             first_leaf.prefix = ""
         else:
             before = 0

src/black/ranges.py

@@ -362,6 +362,7 @@ def _convert_node_to_standalone_comment(node: LN) -> None:
                 value,
                 prefix=prefix,
                 fmt_pass_converted_first_leaf=first,
+                line_ranges_converted=True,
             ),
         )
 
@@ -392,6 +393,7 @@ def _convert_nodes_to_standalone_comment(nodes: Sequence[LN], *, newline: Leaf)
                 value,
                 prefix=prefix,
                 fmt_pass_converted_first_leaf=first,
+                line_ranges_converted=True,
             ),
         )
 

src/blib2to3/pytree.py

@@ -381,6 +381,8 @@ class Leaf(Base):
     # code, and `fmt_pass_converted_first_leaf` points to the first Leaf in the
     # converted code.
     fmt_pass_converted_first_leaf: Optional["Leaf"] = None
+    # True when this Leaf is converted from unchanged code for --line-ranges.
+    line_ranges_converted: bool = False
 
     def __init__(
         self,
@@ -391,6 +393,7 @@ def __init__(
         fixers_applied: list[Any] = [],
         opening_bracket: Optional["Leaf"] = None,
         fmt_pass_converted_first_leaf: Optional["Leaf"] = None,
+        line_ranges_converted: bool = False,
     ) -> None:
         """
         Initializer.
@@ -410,6 +413,7 @@ def __init__(
         self.children = []
         self.opening_bracket = opening_bracket
         self.fmt_pass_converted_first_leaf = fmt_pass_converted_first_leaf
+        self.line_ranges_converted = line_ranges_converted
 
     def __repr__(self) -> str:
         """Return a canonical string representation."""

tests/data/cases/line_ranges_preserve_outside_prefix.py

@@ -0,0 +1,46 @@
+# flags: --line-ranges=2-10
+print("format me"  )
+ # format whitespace
+  # format whitespace
+   # format whitespace
+print( "format me" )
+   # format whitespace
+  # format whitespace
+ # format whitespace
+print(  "format me")
+
+
+
+print("don't format me"  )
+ # don't format whitespace
+  # don't format whitespace
+   # don't format whitespace
+print( "don't format me" )
+   # don't format whitespace
+  # don't format whitespace
+ # don't format whitespace
+print(  "don't format me")
+
+# output
+# flags: --line-ranges=2-10
+print("format me")
+# format whitespace
+# format whitespace
+# format whitespace
+print("format me")
+# format whitespace
+# format whitespace
+# format whitespace
+print("format me")
+
+
+
+print("don't format me"  )
+ # don't format whitespace
+  # don't format whitespace
+   # don't format whitespace
+print( "don't format me" )
+   # don't format whitespace
+  # don't format whitespace
+ # don't format whitespace
+print(  "don't format me")