Skip to content

gh-153568: Factor shared prefixes in expression rules #157437

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
pablogsal wants to merge 1 commit into
base: main
Choose a base branch
from
+1,617 −1,769
Open

gh-153568: Factor shared prefixes in expression rules #157437

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
61 changes: 27 additions & 34 deletions Grammar/python.gram
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -708,10 +708,10 @@ type_param_starred_default[expr_ty]: '=' e=star_expression {
# -----------

expressions[expr_ty]:
| a=expression b=(',' c=expression { c })+ [','] {
_PyAST_Tuple(CHECK(asdl_expr_seq*, _PyPegen_seq_insert_in_front(p, a, b)), Load, EXTRA) }
| a=expression ',' { _PyAST_Tuple(CHECK(asdl_expr_seq*, _PyPegen_singleton_seq(p, a)), Load, EXTRA) }
| expression
| a=expression b=[(',' c=expression { c })+] trailing=[','] {
(b ? b : trailing) ? _PyAST_Tuple(
b ? CHECK(asdl_expr_seq*, _PyPegen_seq_insert_in_front(p, a, b)) :
CHECK(asdl_expr_seq*, _PyPegen_singleton_seq(p, a)), Load, EXTRA) : a }

expression[expr_ty] (memo):
| invalid_if_expression
Expand All @@ -729,10 +729,10 @@ yield_expr[expr_ty]:
| 'yield' a=[star_expressions] { _PyAST_Yield(a, EXTRA) }

star_expressions[expr_ty]:
| a=star_expression b=(',' c=star_expression { c })+ [','] {
_PyAST_Tuple(CHECK(asdl_expr_seq*, _PyPegen_seq_insert_in_front(p, a, b)), Load, EXTRA) }
| a=star_expression ',' { _PyAST_Tuple(CHECK(asdl_expr_seq*, _PyPegen_singleton_seq(p, a)), Load, EXTRA) }
| star_expression
| a=star_expression b=[(',' c=star_expression { c })+] trailing=[','] {
(b ? b : trailing) ? _PyAST_Tuple(
b ? CHECK(asdl_expr_seq*, _PyPegen_seq_insert_in_front(p, a, b)) :
CHECK(asdl_expr_seq*, _PyPegen_singleton_seq(p, a)), Load, EXTRA) : a }

star_expression[expr_ty] (memo):
| '*' a=bitwise_or { _PyAST_Starred(a, Load, EXTRA) }
Expand Down Expand Up @@ -761,18 +761,14 @@ named_expression[expr_ty]:
| expression !':='

disjunction[expr_ty] (memo):
| a=conjunction b=('or' c=conjunction { c })+ { _PyAST_BoolOp(
Or,
CHECK(asdl_expr_seq*, _PyPegen_seq_insert_in_front(p, a, b)),
EXTRA) }
| conjunction
| a=conjunction b=[('or' c=conjunction { c })+] {
b ? _PyAST_BoolOp(Or,
CHECK(asdl_expr_seq*, _PyPegen_seq_insert_in_front(p, a, b)), EXTRA) : a }

conjunction[expr_ty] (memo):
| a=inversion b=('and' c=inversion { c })+ { _PyAST_BoolOp(
And,
CHECK(asdl_expr_seq*, _PyPegen_seq_insert_in_front(p, a, b)),
EXTRA) }
| inversion
| a=inversion b=[('and' c=inversion { c })+] {
b ? _PyAST_BoolOp(And,
CHECK(asdl_expr_seq*, _PyPegen_seq_insert_in_front(p, a, b)), EXTRA) : a }

inversion[expr_ty] (memo):
| 'not' a=inversion { _PyAST_UnaryOp(Not, a, EXTRA) }
Expand All @@ -782,13 +778,10 @@ inversion[expr_ty] (memo):
# --------------------

comparison[expr_ty]:
| a=bitwise_or b=compare_op_bitwise_or_pair+ {
_PyAST_Compare(
a,
| a=bitwise_or b=[compare_op_bitwise_or_pair+] {
b ? _PyAST_Compare(a,
CHECK(asdl_int_seq*, _PyPegen_get_cmpops(p, b)),
CHECK(asdl_expr_seq*, _PyPegen_get_exprs(p, b)),
EXTRA) }
| bitwise_or
CHECK(asdl_expr_seq*, _PyPegen_get_exprs(p, b)), EXTRA) : a }

compare_op_bitwise_or_pair[CmpopExprPair*]:
| eq_bitwise_or
Expand Down Expand Up @@ -832,24 +825,24 @@ bitwise_and[expr_ty]:
| shift_expr

shift_expr[expr_ty]:
| a=shift_expr '<<' b=sum { _PyAST_BinOp(a, LShift, b, EXTRA) }
| a=shift_expr '>>' b=sum { _PyAST_BinOp(a, RShift, b, EXTRA) }
| a=shift_expr op[Token*]=('<<' | '>>') b=sum {
_PyAST_BinOp(a, op->type == LEFTSHIFT ? LShift : RShift, b, EXTRA) }
| sum

# Arithmetic operators
# --------------------

sum[expr_ty]:
| a=sum '+' b=term { _PyAST_BinOp(a, Add, b, EXTRA) }
| a=sum '-' b=term { _PyAST_BinOp(a, Sub, b, EXTRA) }
| a=sum op[Token*]=('+' | '-') b=term {
_PyAST_BinOp(a, op->type == PLUS ? Add : Sub, b, EXTRA) }
| invalid_arithmetic
| term

term[expr_ty]:
| a=term '*' b=factor { _PyAST_BinOp(a, Mult, b, EXTRA) }
| a=term '/' b=factor { _PyAST_BinOp(a, Div, b, EXTRA) }
| a=term '//' b=factor { _PyAST_BinOp(a, FloorDiv, b, EXTRA) }
| a=term '%' b=factor { _PyAST_BinOp(a, Mod, b, EXTRA) }
| a=term op[Token*]=('*' | '/' | '//' | '%') b=factor {
_PyAST_BinOp(a, op->type == STAR ? Mult :
op->type == SLASH ? Div :
op->type == DOUBLESLASH ? FloorDiv : Mod, b, EXTRA) }
| a=term '@' b=factor { CHECK_VERSION(expr_ty, 5, "The '@' operator is", _PyAST_BinOp(a, MatMult, b, EXTRA)) }
| factor

Expand All @@ -861,8 +854,8 @@ factor[expr_ty] (memo):
| invalid_factor

power[expr_ty]:
| a=await_primary '**' b=factor { _PyAST_BinOp(a, Pow, b, EXTRA) }
| await_primary
| a=await_primary b=['**' c=factor { c }] {
b ? _PyAST_BinOp(a, Pow, b, EXTRA) : a }

# Primary elements
# ----------------
Expand Down
44 changes: 44 additions & 0 deletions Lib/test/test_ast/test_ast.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -696,6 +696,50 @@ def test_issue18374_binop_col_offset(self):
self.assertEqual(grandchild_binop.end_col_offset, 3)
self.assertEqual(grandchild_binop.end_lineno, 1)

def test_mixed_binop_associativity_and_locations(self):
cases = [
('<<', '>>', ast.LShift, ast.RShift),
('+', '-', ast.Add, ast.Sub),
('-', '+', ast.Sub, ast.Add),
('*', '/', ast.Mult, ast.Div),
('/', '//', ast.Div, ast.FloorDiv),
('//', '%', ast.FloorDiv, ast.Mod),
('%', '@', ast.Mod, ast.MatMult),
('@', '*', ast.MatMult, ast.Mult),
]
for first, second, first_op, second_op in cases:
source = f'(a {first} b\n {second} c)'
with self.subTest(source=source):
node = ast.parse(source, mode='eval').body
self.assertIsInstance(node, ast.BinOp)
self.assertIsInstance(node.op, second_op)
self.assertIsInstance(node.left, ast.BinOp)
self.assertIsInstance(node.left.op, first_op)
self.assertEqual(ast.get_source_segment(source, node), source[1:-1])
self.assertEqual(ast.get_source_segment(source, node.left),
f'a {first} b')

def test_expression_tail_locations(self):
cases = [
('a', ast.Name),
('a,', ast.Tuple),
('a, b', ast.Tuple),
('a, b,', ast.Tuple),
('a or b or c', ast.BoolOp),
('a and b and c', ast.BoolOp),
('a < b <= c', ast.Compare),
('a ** b ** c', ast.BinOp),
]
for expression, node_type in cases:
for mode in ('eval', 'exec'):
with self.subTest(expression=expression, mode=mode):
source = expression + ' # comment\n'
tree = ast.parse(source, mode=mode)
node = tree.body if mode == 'eval' else tree.body[0].value
self.assertIsInstance(node, node_type)
self.assertEqual(ast.get_source_segment(source, node),
expression)

def test_issue39579_dotted_name_end_col_offset(self):
tree = ast.parse('@a.b.c\ndef f(): pass')
attr_b = tree.body[0].decorator_list[0].value
Expand Down
2 changes: 2 additions & 0 deletions Misc/NEWS.d/next/Core_and_Builtins/2026-09-13-15-00-00.gh-issue-153568.GpF4tr.rst
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,2 @@
Speed up parsing expressions by factoring shared prefixes and optional tails
in the grammar.
Loading
Loading