dont skip formatting #%% (#2919)

Fixes #2588

diff --git a/CHANGES.md b/CHANGES.md
index bb3ccb9ed9f89c7b1c56ef02f09d37c073915a5a..d0faf7cecb9af5455e5910253ca888b3f747f299 100644 (file)
--- a/CHANGES.md
+++ b/CHANGES.md
@@ -14,6 +14,8 @@
 
 <!-- Changes that affect Black's preview style -->
 
 
 <!-- Changes that affect Black's preview style -->
 

+- Code cell separators `#%%` are now standardised to `# %%` (#2919)
+

 ### _Blackd_
 
 <!-- Changes to blackd -->
 ### _Blackd_
 
 <!-- Changes to blackd -->

diff --git a/src/black/__init__.py b/src/black/__init__.py
index c4ec99b441f86f063a660e53f03f7a08df40a685..51e31e9598b29bd250fbba49b48ad823e5328eeb 100644 (file)
--- a/src/black/__init__.py
+++ b/src/black/__init__.py
@@ -1166,7 +1166,7 @@ def _format_str_once(src_contents: str, *, mode: Mode) -> str:
     else:
         versions = detect_target_versions(src_node, future_imports=future_imports)
 
     else:
         versions = detect_target_versions(src_node, future_imports=future_imports)
 

-    normalize_fmt_off(src_node)
+    normalize_fmt_off(src_node, preview=mode.preview)

     lines = LineGenerator(mode=mode)
     elt = EmptyLineTracker(is_pyi=mode.is_pyi)
     empty_line = Line(mode=mode)
     lines = LineGenerator(mode=mode)
     elt = EmptyLineTracker(is_pyi=mode.is_pyi)
     empty_line = Line(mode=mode)

diff --git a/src/black/comments.py b/src/black/comments.py
index 28b9117101d7aa0eb19ce64190b6c2a83b83a4fd..455326469f0f3b9c42764b0867fda22d121914dc 100644 (file)
--- a/src/black/comments.py
+++ b/src/black/comments.py
@@ -23,6 +23,8 @@ FMT_SKIP: Final = {"# fmt: skip", "# fmt:skip"}
 FMT_PASS: Final = {*FMT_OFF, *FMT_SKIP}
 FMT_ON: Final = {"# fmt: on", "# fmt:on", "# yapf: enable"}
 
 FMT_PASS: Final = {*FMT_OFF, *FMT_SKIP}
 FMT_ON: Final = {"# fmt: on", "# fmt:on", "# yapf: enable"}
 

+COMMENT_EXCEPTIONS = {True: " !:#'", False: " !:#'%"}
+

 
 @dataclass
 class ProtoComment:
 
 @dataclass
 class ProtoComment:


@@ -42,7 +44,7 @@ class ProtoComment:
     consumed: int  # how many characters of the original leaf's prefix did we consume
 
 
     consumed: int  # how many characters of the original leaf's prefix did we consume
 
 

-def generate_comments(leaf: LN) -> Iterator[Leaf]:
+def generate_comments(leaf: LN, *, preview: bool) -> 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
     """Clean the prefix of the `leaf` and generate comments from it, if any.
 
     Comments in lib2to3 are shoved into the whitespace prefix.  This happens


@@ -61,12 +63,16 @@ def generate_comments(leaf: LN) -> Iterator[Leaf]:
     Inline comments are emitted as regular token.COMMENT leaves.  Standalone
     are emitted with a fake STANDALONE_COMMENT token identifier.
     """
     Inline comments are emitted as regular token.COMMENT leaves.  Standalone
     are emitted with a fake STANDALONE_COMMENT token identifier.
     """

-    for pc in list_comments(leaf.prefix, is_endmarker=leaf.type == token.ENDMARKER):
+    for pc in list_comments(
+        leaf.prefix, is_endmarker=leaf.type == token.ENDMARKER, preview=preview
+    ):

         yield Leaf(pc.type, pc.value, prefix="\n" * pc.newlines)
 
 
 @lru_cache(maxsize=4096)
         yield Leaf(pc.type, pc.value, prefix="\n" * pc.newlines)
 
 
 @lru_cache(maxsize=4096)

-def list_comments(prefix: str, *, is_endmarker: bool) -> List[ProtoComment]:
+def list_comments(
+    prefix: str, *, is_endmarker: bool, preview: bool
+) -> List[ProtoComment]:

     """Return a list of :class:`ProtoComment` objects parsed from the given `prefix`."""
     result: List[ProtoComment] = []
     if not prefix or "#" not in prefix:
     """Return a list of :class:`ProtoComment` objects parsed from the given `prefix`."""
     result: List[ProtoComment] = []
     if not prefix or "#" not in prefix:


@@ -92,7 +98,7 @@ def list_comments(prefix: str, *, is_endmarker: bool) -> List[ProtoComment]:
             comment_type = token.COMMENT  # simple trailing comment
         else:
             comment_type = STANDALONE_COMMENT
             comment_type = token.COMMENT  # simple trailing comment
         else:
             comment_type = STANDALONE_COMMENT

-        comment = make_comment(line)
+        comment = make_comment(line, preview=preview)

         result.append(
             ProtoComment(
                 type=comment_type, value=comment, newlines=nlines, consumed=consumed
         result.append(
             ProtoComment(
                 type=comment_type, value=comment, newlines=nlines, consumed=consumed


@@ -102,10 +108,10 @@ def list_comments(prefix: str, *, is_endmarker: bool) -> List[ProtoComment]:
     return result
 
 
     return result
 
 

-def make_comment(content: str) -> str:
+def make_comment(content: str, *, preview: bool) -> str:

     """Return a consistently formatted comment from the given `content` string.
 
     """Return a consistently formatted comment from the given `content` string.
 

-    All comments (except for "##", "#!", "#:", '#'", "#%%") should have a single
+    All comments (except for "##", "#!", "#:", '#'") should have a single

     space between the hash sign and the content.
 
     If `content` didn't start with a hash sign, one is provided.
     space between the hash sign and the content.
 
     If `content` didn't start with a hash sign, one is provided.


@@ -123,26 +129,26 @@ def make_comment(content: str) -> str:
         and not content.lstrip().startswith("type:")
     ):
         content = " " + content[1:]  # Replace NBSP by a simple space
         and not content.lstrip().startswith("type:")
     ):
         content = " " + content[1:]  # Replace NBSP by a simple space

-    if content and content[0] not in " !:#'%":
+    if content and content[0] not in COMMENT_EXCEPTIONS[preview]:

         content = " " + content
     return "#" + content
 
 
         content = " " + content
     return "#" + content
 
 

-def normalize_fmt_off(node: Node) -> None:
+def normalize_fmt_off(node: Node, *, preview: bool) -> None:

     """Convert content between `# fmt: off`/`# fmt: on` into standalone comments."""
     try_again = True
     while try_again:
     """Convert content between `# fmt: off`/`# fmt: on` into standalone comments."""
     try_again = True
     while try_again:

-        try_again = convert_one_fmt_off_pair(node)
+        try_again = convert_one_fmt_off_pair(node, preview=preview)

 
 
 
 

-def convert_one_fmt_off_pair(node: Node) -> bool:
+def convert_one_fmt_off_pair(node: Node, *, preview: bool) -> bool:

     """Convert content of a single `# fmt: off`/`# fmt: on` into a standalone comment.
 
     Returns True if a pair was converted.
     """
     for leaf in node.leaves():
         previous_consumed = 0
     """Convert content of a single `# fmt: off`/`# fmt: on` into a standalone comment.
 
     Returns True if a pair was converted.
     """
     for leaf in node.leaves():
         previous_consumed = 0

-        for comment in list_comments(leaf.prefix, is_endmarker=False):
+        for comment in list_comments(leaf.prefix, is_endmarker=False, preview=preview):

             if comment.value not in FMT_PASS:
                 previous_consumed = comment.consumed
                 continue
             if comment.value not in FMT_PASS:
                 previous_consumed = comment.consumed
                 continue


@@ -157,7 +163,7 @@ def convert_one_fmt_off_pair(node: Node) -> bool:
                     if comment.value in FMT_SKIP and prev.type in WHITESPACE:
                         continue
 
                     if comment.value in FMT_SKIP and prev.type in WHITESPACE:
                         continue
 

-            ignored_nodes = list(generate_ignored_nodes(leaf, comment))
+            ignored_nodes = list(generate_ignored_nodes(leaf, comment, preview=preview))

             if not ignored_nodes:
                 continue
 
             if not ignored_nodes:
                 continue
 


@@ -197,7 +203,9 @@ def convert_one_fmt_off_pair(node: Node) -> bool:
     return False
 
 
     return False
 
 

-def generate_ignored_nodes(leaf: Leaf, comment: ProtoComment) -> Iterator[LN]:
+def generate_ignored_nodes(
+    leaf: Leaf, comment: ProtoComment, *, preview: bool
+) -> Iterator[LN]:

     """Starting from the container of `leaf`, generate all leaves until `# fmt: on`.
 
     If comment is skip, returns leaf only.
     """Starting from the container of `leaf`, generate all leaves until `# fmt: on`.
 
     If comment is skip, returns leaf only.


@@ -221,13 +229,13 @@ def generate_ignored_nodes(leaf: Leaf, comment: ProtoComment) -> Iterator[LN]:
             yield leaf.parent
         return
     while container is not None and container.type != token.ENDMARKER:
             yield leaf.parent
         return
     while container is not None and container.type != token.ENDMARKER:

-        if is_fmt_on(container):
+        if is_fmt_on(container, preview=preview):

             return
 
         # fix for fmt: on in children
             return
 
         # fix for fmt: on in children

-        if contains_fmt_on_at_column(container, leaf.column):
+        if contains_fmt_on_at_column(container, leaf.column, preview=preview):

             for child in container.children:
             for child in container.children:

-                if contains_fmt_on_at_column(child, leaf.column):
+                if contains_fmt_on_at_column(child, leaf.column, preview=preview):

                     return
                 yield child
         else:
                     return
                 yield child
         else:


@@ -235,12 +243,12 @@ def generate_ignored_nodes(leaf: Leaf, comment: ProtoComment) -> Iterator[LN]:
             container = container.next_sibling
 
 
             container = container.next_sibling
 
 

-def is_fmt_on(container: LN) -> bool:
+def is_fmt_on(container: LN, preview: bool) -> bool:

     """Determine whether formatting is switched on within a container.
     Determined by whether the last `# fmt:` comment is `on` or `off`.
     """
     fmt_on = False
     """Determine whether formatting is switched on within a container.
     Determined by whether the last `# fmt:` comment is `on` or `off`.
     """
     fmt_on = False

-    for comment in list_comments(container.prefix, is_endmarker=False):
+    for comment in list_comments(container.prefix, is_endmarker=False, preview=preview):

         if comment.value in FMT_ON:
             fmt_on = True
         elif comment.value in FMT_OFF:
         if comment.value in FMT_ON:
             fmt_on = True
         elif comment.value in FMT_OFF:


@@ -248,7 +256,7 @@ def is_fmt_on(container: LN) -> bool:
     return fmt_on
 
 
     return fmt_on
 
 

-def contains_fmt_on_at_column(container: LN, column: int) -> bool:
+def contains_fmt_on_at_column(container: LN, column: int, *, preview: bool) -> bool:

     """Determine if children at a given column have formatting switched on."""
     for child in container.children:
         if (
     """Determine if children at a given column have formatting switched on."""
     for child in container.children:
         if (


@@ -257,7 +265,7 @@ def contains_fmt_on_at_column(container: LN, column: int) -> bool:
             or isinstance(child, Leaf)
             and child.column == column
         ):
             or isinstance(child, Leaf)
             and child.column == column
         ):

-            if is_fmt_on(child):
+            if is_fmt_on(child, preview=preview):

                 return True
 
     return False
                 return True
 
     return False

diff --git a/src/black/linegen.py b/src/black/linegen.py
index 4dc242a1dfe05b9f8ef7ff53b26333073469cf57..79475a83f0e363c5f9f223278c5a3ebdf5f7fad6 100644 (file)
--- a/src/black/linegen.py
+++ b/src/black/linegen.py
@@ -72,7 +72,7 @@ class LineGenerator(Visitor[Line]):
         """Default `visit_*()` implementation. Recurses to children of `node`."""
         if isinstance(node, Leaf):
             any_open_brackets = self.current_line.bracket_tracker.any_open_brackets()
         """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):
+            for comment in generate_comments(node, preview=self.mode.preview):

                 if any_open_brackets:
                     # any comment within brackets is subject to splitting
                     self.current_line.append(comment)
                 if any_open_brackets:
                     # any comment within brackets is subject to splitting
                     self.current_line.append(comment)


@@ -132,7 +132,7 @@ class LineGenerator(Visitor[Line]):
         `parens` holds a set of string leaf values immediately after which
         invisible parens should be put.
         """
         `parens` holds a set of string leaf values immediately after which
         invisible parens should be put.
         """

-        normalize_invisible_parens(node, parens_after=parens)
+        normalize_invisible_parens(node, parens_after=parens, preview=self.mode.preview)

         for child in node.children:
             if is_name_token(child) and child.value in keywords:
                 yield from self.line()
         for child in node.children:
             if is_name_token(child) and child.value in keywords:
                 yield from self.line()


@@ -141,7 +141,7 @@ class LineGenerator(Visitor[Line]):
 
     def visit_match_case(self, node: Node) -> Iterator[Line]:
         """Visit either a match or case statement."""
 
     def visit_match_case(self, node: Node) -> Iterator[Line]:
         """Visit either a match or case statement."""

-        normalize_invisible_parens(node, parens_after=set())
+        normalize_invisible_parens(node, parens_after=set(), preview=self.mode.preview)

 
         yield from self.line()
         for child in node.children:
 
         yield from self.line()
         for child in node.children:


@@ -802,7 +802,9 @@ def normalize_prefix(leaf: Leaf, *, inside_brackets: bool) -> None:
     leaf.prefix = ""
 
 
     leaf.prefix = ""
 
 

-def normalize_invisible_parens(node: Node, parens_after: Set[str]) -> None:
+def normalize_invisible_parens(
+    node: Node, parens_after: Set[str], *, preview: bool
+) -> None:

     """Make existing optional parentheses invisible or create new ones.
 
     `parens_after` is a set of string leaf values immediately after which parens
     """Make existing optional parentheses invisible or create new ones.
 
     `parens_after` is a set of string leaf values immediately after which parens


@@ -811,7 +813,7 @@ def normalize_invisible_parens(node: Node, parens_after: Set[str]) -> None:
     Standardizes on visible parentheses for single-element tuples, and keeps
     existing visible parentheses for other tuples and generator expressions.
     """
     Standardizes on visible parentheses for single-element tuples, and keeps
     existing visible parentheses for other tuples and generator expressions.
     """

-    for pc in list_comments(node.prefix, is_endmarker=False):
+    for pc in list_comments(node.prefix, is_endmarker=False, preview=preview):

         if pc.value in FMT_OFF:
             # This `node` has a prefix with `# fmt: off`, don't mess with parens.
             return
         if pc.value in FMT_OFF:
             # This `node` has a prefix with `# fmt: off`, don't mess with parens.
             return


@@ -820,7 +822,9 @@ def normalize_invisible_parens(node: Node, parens_after: Set[str]) -> None:
         # Fixes a bug where invisible parens are not properly stripped from
         # assignment statements that contain type annotations.
         if isinstance(child, Node) and child.type == syms.annassign:
         # Fixes a bug where invisible parens are not properly stripped from
         # assignment statements that contain type annotations.
         if isinstance(child, Node) and child.type == syms.annassign:

-            normalize_invisible_parens(child, parens_after=parens_after)
+            normalize_invisible_parens(
+                child, parens_after=parens_after, preview=preview
+            )

 
         # Add parentheses around long tuple unpacking in assignments.
         if (
 
         # Add parentheses around long tuple unpacking in assignments.
         if (

diff --git a/tests/data/comments8.py b/tests/data/comments8.py
new file mode 100644 (file)
index 0000000..a2030c2
--- /dev/null
+++ b/tests/data/comments8.py
@@ -0,0 +1,15 @@
+# The percent-percent comments are Spyder IDE cells.
+# Both `#%%`` and `# %%` are accepted, so `black` standardises
+# to the latter.
+
+#%%
+# %%
+
+# output
+
+# The percent-percent comments are Spyder IDE cells.
+# Both `#%%`` and `# %%` are accepted, so `black` standardises
+# to the latter.
+
+# %%
+# %%

diff --git a/tests/test_format.py b/tests/test_format.py
index 269bbacd249ca0dcb836282dab8bb8e3a26678b6..667d5c110fad03ba775fcd0f29736440da62f71d 100644 (file)
--- a/tests/test_format.py
+++ b/tests/test_format.py
@@ -75,6 +75,7 @@ PREVIEW_CASES: List[str] = [
     # string processing
     "cantfit",
     "comments7",
     # string processing
     "cantfit",
     "comments7",

+    "comments8",

     "long_strings",
     "long_strings__edge_case",
     "long_strings__regression",
     "long_strings",
     "long_strings__edge_case",
     "long_strings__regression",