Spaces:
Runtime error
Runtime error
File size: 61,298 Bytes
6eefbd7 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260 1261 1262 1263 1264 1265 1266 1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389 1390 1391 1392 1393 1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423 1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 1442 1443 1444 1445 1446 1447 1448 1449 1450 1451 1452 1453 1454 1455 1456 1457 1458 1459 1460 1461 1462 1463 1464 1465 1466 1467 1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 1479 1480 1481 1482 1483 1484 1485 1486 1487 1488 1489 1490 1491 1492 1493 1494 1495 1496 1497 1498 1499 1500 1501 1502 1503 1504 1505 1506 1507 1508 1509 1510 1511 1512 1513 1514 1515 1516 1517 1518 1519 1520 1521 1522 1523 1524 1525 1526 1527 1528 1529 1530 1531 1532 1533 1534 1535 1536 1537 1538 1539 1540 1541 1542 1543 1544 1545 1546 1547 1548 1549 1550 1551 1552 1553 1554 1555 1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566 1567 1568 1569 1570 1571 1572 1573 |
"""
Generating lines of code.
"""
import sys
from dataclasses import replace
from enum import Enum, auto
from functools import partial, wraps
from typing import Collection, Iterator, List, Optional, Set, Union, cast
from black.brackets import (
COMMA_PRIORITY,
DOT_PRIORITY,
get_leaves_inside_matching_brackets,
max_delimiter_priority_in_atom,
)
from black.comments import FMT_OFF, generate_comments, list_comments
from black.lines import (
Line,
RHSResult,
append_leaves,
can_be_split,
can_omit_invisible_parens,
is_line_short_enough,
line_to_string,
)
from black.mode import Feature, Mode, Preview
from black.nodes import (
ASSIGNMENTS,
BRACKETS,
CLOSING_BRACKETS,
OPENING_BRACKETS,
RARROW,
STANDALONE_COMMENT,
STATEMENT,
WHITESPACE,
Visitor,
ensure_visible,
is_arith_like,
is_async_stmt_or_funcdef,
is_atom_with_invisible_parens,
is_docstring,
is_empty_tuple,
is_lpar_token,
is_multiline_string,
is_name_token,
is_one_sequence_between,
is_one_tuple,
is_rpar_token,
is_stub_body,
is_stub_suite,
is_tuple_containing_walrus,
is_type_ignore_comment_string,
is_vararg,
is_walrus_assignment,
is_yield,
syms,
wrap_in_parentheses,
)
from black.numerics import normalize_numeric_literal
from black.strings import (
fix_docstring,
get_string_prefix,
normalize_string_prefix,
normalize_string_quotes,
normalize_unicode_escape_sequences,
)
from black.trans import (
CannotTransform,
StringMerger,
StringParenStripper,
StringParenWrapper,
StringSplitter,
Transformer,
hug_power_op,
)
from blib2to3.pgen2 import token
from blib2to3.pytree import Leaf, Node
# types
LeafID = int
LN = Union[Leaf, Node]
class CannotSplit(CannotTransform):
"""A readable split that fits the allotted line length is impossible."""
# This isn't a dataclass because @dataclass + Generic breaks mypyc.
# See also https://github.com/mypyc/mypyc/issues/827.
class LineGenerator(Visitor[Line]):
"""Generates reformatted Line objects. Empty lines are not emitted.
Note: destroys the tree it's visiting by mutating prefixes of its leaves
in ways that will no longer stringify to valid Python code on the tree.
"""
def __init__(self, mode: Mode, features: Collection[Feature]) -> None:
self.mode = mode
self.features = features
self.current_line: Line
self.__post_init__()
def line(self, indent: int = 0) -> Iterator[Line]:
"""Generate a line.
If the line is empty, only emit if it makes sense.
If the line is too long, split it first and then generate.
If any lines were generated, set up a new current_line.
"""
if not self.current_line:
self.current_line.depth += indent
return # Line is empty, don't emit. Creating a new one unnecessary.
if (
Preview.improved_async_statements_handling in self.mode
and len(self.current_line.leaves) == 1
and is_async_stmt_or_funcdef(self.current_line.leaves[0])
):
# Special case for async def/for/with statements. `visit_async_stmt`
# adds an `ASYNC` leaf then visits the child def/for/with statement
# nodes. Line yields from those nodes shouldn't treat the former
# `ASYNC` leaf as a complete line.
return
complete_line = self.current_line
self.current_line = Line(mode=self.mode, depth=complete_line.depth + indent)
yield complete_line
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):
if any_open_brackets:
# any comment within brackets is subject to splitting
self.current_line.append(comment)
elif comment.type == token.COMMENT:
# regular trailing comment
self.current_line.append(comment)
yield from self.line()
else:
# regular standalone comment
yield from self.line()
self.current_line.append(comment)
yield from self.line()
normalize_prefix(node, inside_brackets=any_open_brackets)
if self.mode.string_normalization and node.type == token.STRING:
node.value = normalize_string_prefix(node.value)
node.value = normalize_string_quotes(node.value)
if node.type == token.NUMBER:
normalize_numeric_literal(node)
if node.type not in WHITESPACE:
self.current_line.append(node)
yield from super().visit_default(node)
def visit_test(self, node: Node) -> Iterator[Line]:
"""Visit an `x if y else z` test"""
if Preview.parenthesize_conditional_expressions in self.mode:
already_parenthesized = (
node.prev_sibling and node.prev_sibling.type == token.LPAR
)
if not already_parenthesized:
lpar = Leaf(token.LPAR, "")
rpar = Leaf(token.RPAR, "")
node.insert_child(0, lpar)
node.append_child(rpar)
yield from self.visit_default(node)
def visit_INDENT(self, node: Leaf) -> Iterator[Line]:
"""Increase indentation level, maybe yield a line."""
# In blib2to3 INDENT never holds comments.
yield from self.line(+1)
yield from self.visit_default(node)
def visit_DEDENT(self, node: Leaf) -> Iterator[Line]:
"""Decrease indentation level, maybe yield a line."""
# The current line might still wait for trailing comments. At DEDENT time
# there won't be any (they would be prefixes on the preceding NEWLINE).
# Emit the line then.
yield from self.line()
# While DEDENT has no value, its prefix may contain standalone comments
# that belong to the current indentation level. Get 'em.
yield from self.visit_default(node)
# Finally, emit the dedent.
yield from self.line(-1)
def visit_stmt(
self, node: Node, keywords: Set[str], parens: Set[str]
) -> Iterator[Line]:
"""Visit a statement.
This implementation is shared for `if`, `while`, `for`, `try`, `except`,
`def`, `with`, `class`, `assert`, and assignments.
The relevant Python language `keywords` for a given statement will be
NAME leaves within it. This methods puts those on a separate line.
`parens` holds a set of string leaf values immediately after which
invisible parens should be put.
"""
normalize_invisible_parens(
node, parens_after=parens, mode=self.mode, features=self.features
)
for child in node.children:
if is_name_token(child) and child.value in keywords:
yield from self.line()
yield from self.visit(child)
def visit_typeparams(self, node: Node) -> Iterator[Line]:
yield from self.visit_default(node)
node.children[0].prefix = ""
def visit_typevartuple(self, node: Node) -> Iterator[Line]:
yield from self.visit_default(node)
node.children[1].prefix = ""
def visit_paramspec(self, node: Node) -> Iterator[Line]:
yield from self.visit_default(node)
node.children[1].prefix = ""
def visit_dictsetmaker(self, node: Node) -> Iterator[Line]:
if Preview.wrap_long_dict_values_in_parens in self.mode:
for i, child in enumerate(node.children):
if i == 0:
continue
if node.children[i - 1].type == token.COLON:
if child.type == syms.atom and child.children[0].type == token.LPAR:
if maybe_make_parens_invisible_in_atom(
child,
parent=node,
remove_brackets_around_comma=False,
):
wrap_in_parentheses(node, child, visible=False)
else:
wrap_in_parentheses(node, child, visible=False)
yield from self.visit_default(node)
def visit_funcdef(self, node: Node) -> Iterator[Line]:
"""Visit function definition."""
yield from self.line()
# Remove redundant brackets around return type annotation.
is_return_annotation = False
for child in node.children:
if child.type == token.RARROW:
is_return_annotation = True
elif is_return_annotation:
if child.type == syms.atom and child.children[0].type == token.LPAR:
if maybe_make_parens_invisible_in_atom(
child,
parent=node,
remove_brackets_around_comma=False,
):
wrap_in_parentheses(node, child, visible=False)
else:
wrap_in_parentheses(node, child, visible=False)
is_return_annotation = False
for child in node.children:
yield from self.visit(child)
def visit_match_case(self, node: Node) -> Iterator[Line]:
"""Visit either a match or case statement."""
normalize_invisible_parens(
node, parens_after=set(), mode=self.mode, features=self.features
)
yield from self.line()
for child in node.children:
yield from self.visit(child)
def visit_suite(self, node: Node) -> Iterator[Line]:
"""Visit a suite."""
if self.mode.is_pyi and is_stub_suite(node):
yield from self.visit(node.children[2])
else:
yield from self.visit_default(node)
def visit_simple_stmt(self, node: Node) -> Iterator[Line]:
"""Visit a statement without nested statements."""
prev_type: Optional[int] = None
for child in node.children:
if (prev_type is None or prev_type == token.SEMI) and is_arith_like(child):
wrap_in_parentheses(node, child, visible=False)
prev_type = child.type
is_suite_like = node.parent and node.parent.type in STATEMENT
if is_suite_like:
if self.mode.is_pyi and is_stub_body(node):
yield from self.visit_default(node)
else:
yield from self.line(+1)
yield from self.visit_default(node)
yield from self.line(-1)
else:
if (
not self.mode.is_pyi
or not node.parent
or not is_stub_suite(node.parent)
):
yield from self.line()
yield from self.visit_default(node)
def visit_async_stmt(self, node: Node) -> Iterator[Line]:
"""Visit `async def`, `async for`, `async with`."""
yield from self.line()
children = iter(node.children)
for child in children:
yield from self.visit(child)
if child.type == token.ASYNC or child.type == STANDALONE_COMMENT:
# STANDALONE_COMMENT happens when `# fmt: skip` is applied on the async
# line.
break
internal_stmt = next(children)
if Preview.improved_async_statements_handling in self.mode:
yield from self.visit(internal_stmt)
else:
for child in internal_stmt.children:
yield from self.visit(child)
def visit_decorators(self, node: Node) -> Iterator[Line]:
"""Visit decorators."""
for child in node.children:
yield from self.line()
yield from self.visit(child)
def visit_power(self, node: Node) -> Iterator[Line]:
for idx, leaf in enumerate(node.children[:-1]):
next_leaf = node.children[idx + 1]
if not isinstance(leaf, Leaf):
continue
value = leaf.value.lower()
if (
leaf.type == token.NUMBER
and next_leaf.type == syms.trailer
# Ensure that we are in an attribute trailer
and next_leaf.children[0].type == token.DOT
# It shouldn't wrap hexadecimal, binary and octal literals
and not value.startswith(("0x", "0b", "0o"))
# It shouldn't wrap complex literals
and "j" not in value
):
wrap_in_parentheses(node, leaf)
remove_await_parens(node)
yield from self.visit_default(node)
def visit_SEMI(self, leaf: Leaf) -> Iterator[Line]:
"""Remove a semicolon and put the other statement on a separate line."""
yield from self.line()
def visit_ENDMARKER(self, leaf: Leaf) -> Iterator[Line]:
"""End of file. Process outstanding comments and end with a newline."""
yield from self.visit_default(leaf)
yield from self.line()
def visit_STANDALONE_COMMENT(self, leaf: Leaf) -> Iterator[Line]:
if not self.current_line.bracket_tracker.any_open_brackets():
yield from self.line()
yield from self.visit_default(leaf)
def visit_factor(self, node: Node) -> Iterator[Line]:
"""Force parentheses between a unary op and a binary power:
-2 ** 8 -> -(2 ** 8)
"""
_operator, operand = node.children
if (
operand.type == syms.power
and len(operand.children) == 3
and operand.children[1].type == token.DOUBLESTAR
):
lpar = Leaf(token.LPAR, "(")
rpar = Leaf(token.RPAR, ")")
index = operand.remove() or 0
node.insert_child(index, Node(syms.atom, [lpar, operand, rpar]))
yield from self.visit_default(node)
def visit_STRING(self, leaf: Leaf) -> Iterator[Line]:
if Preview.hex_codes_in_unicode_sequences in self.mode:
normalize_unicode_escape_sequences(leaf)
if is_docstring(leaf) and "\\\n" not in leaf.value:
# We're ignoring docstrings with backslash newline escapes because changing
# indentation of those changes the AST representation of the code.
if self.mode.string_normalization:
docstring = normalize_string_prefix(leaf.value)
# visit_default() does handle string normalization for us, but
# since this method acts differently depending on quote style (ex.
# see padding logic below), there's a possibility for unstable
# formatting as visit_default() is called *after*. To avoid a
# situation where this function formats a docstring differently on
# the second pass, normalize it early.
docstring = normalize_string_quotes(docstring)
else:
docstring = leaf.value
prefix = get_string_prefix(docstring)
docstring = docstring[len(prefix) :] # Remove the prefix
quote_char = docstring[0]
# A natural way to remove the outer quotes is to do:
# docstring = docstring.strip(quote_char)
# but that breaks on """""x""" (which is '""x').
# So we actually need to remove the first character and the next two
# characters but only if they are the same as the first.
quote_len = 1 if docstring[1] != quote_char else 3
docstring = docstring[quote_len:-quote_len]
docstring_started_empty = not docstring
indent = " " * 4 * self.current_line.depth
if is_multiline_string(leaf):
docstring = fix_docstring(docstring, indent)
else:
docstring = docstring.strip()
has_trailing_backslash = False
if docstring:
# Add some padding if the docstring starts / ends with a quote mark.
if docstring[0] == quote_char:
docstring = " " + docstring
if docstring[-1] == quote_char:
docstring += " "
if docstring[-1] == "\\":
backslash_count = len(docstring) - len(docstring.rstrip("\\"))
if backslash_count % 2:
# Odd number of tailing backslashes, add some padding to
# avoid escaping the closing string quote.
docstring += " "
has_trailing_backslash = True
elif not docstring_started_empty:
docstring = " "
# We could enforce triple quotes at this point.
quote = quote_char * quote_len
# It's invalid to put closing single-character quotes on a new line.
if self.mode and quote_len == 3:
# We need to find the length of the last line of the docstring
# to find if we can add the closing quotes to the line without
# exceeding the maximum line length.
# If docstring is one line, we don't put the closing quotes on a
# separate line because it looks ugly (#3320).
lines = docstring.splitlines()
last_line_length = len(lines[-1]) if docstring else 0
# If adding closing quotes would cause the last line to exceed
# the maximum line length then put a line break before the
# closing quotes
if (
len(lines) > 1
and last_line_length + quote_len > self.mode.line_length
and len(indent) + quote_len <= self.mode.line_length
and not has_trailing_backslash
):
leaf.value = prefix + quote + docstring + "\n" + indent + quote
else:
leaf.value = prefix + quote + docstring + quote
else:
leaf.value = prefix + quote + docstring + quote
yield from self.visit_default(leaf)
def __post_init__(self) -> None:
"""You are in a twisty little maze of passages."""
self.current_line = Line(mode=self.mode)
v = self.visit_stmt
Ø: Set[str] = set()
self.visit_assert_stmt = partial(v, keywords={"assert"}, parens={"assert", ","})
self.visit_if_stmt = partial(
v, keywords={"if", "else", "elif"}, parens={"if", "elif"}
)
self.visit_while_stmt = partial(v, keywords={"while", "else"}, parens={"while"})
self.visit_for_stmt = partial(v, keywords={"for", "else"}, parens={"for", "in"})
self.visit_try_stmt = partial(
v, keywords={"try", "except", "else", "finally"}, parens=Ø
)
self.visit_except_clause = partial(v, keywords={"except"}, parens={"except"})
self.visit_with_stmt = partial(v, keywords={"with"}, parens={"with"})
self.visit_classdef = partial(v, keywords={"class"}, parens=Ø)
self.visit_expr_stmt = partial(v, keywords=Ø, parens=ASSIGNMENTS)
self.visit_return_stmt = partial(v, keywords={"return"}, parens={"return"})
self.visit_import_from = partial(v, keywords=Ø, parens={"import"})
self.visit_del_stmt = partial(v, keywords=Ø, parens={"del"})
self.visit_async_funcdef = self.visit_async_stmt
self.visit_decorated = self.visit_decorators
# PEP 634
self.visit_match_stmt = self.visit_match_case
self.visit_case_block = self.visit_match_case
def transform_line(
line: Line, mode: Mode, features: Collection[Feature] = ()
) -> Iterator[Line]:
"""Transform a `line`, potentially splitting it into many lines.
They should fit in the allotted `line_length` but might not be able to.
`features` are syntactical features that may be used in the output.
"""
if line.is_comment:
yield line
return
line_str = line_to_string(line)
ll = mode.line_length
sn = mode.string_normalization
string_merge = StringMerger(ll, sn)
string_paren_strip = StringParenStripper(ll, sn)
string_split = StringSplitter(ll, sn)
string_paren_wrap = StringParenWrapper(ll, sn)
transformers: List[Transformer]
if (
not line.contains_uncollapsable_type_comments()
and not line.should_split_rhs
and not line.magic_trailing_comma
and (
is_line_short_enough(line, mode=mode, line_str=line_str)
or line.contains_unsplittable_type_ignore()
)
and not (line.inside_brackets and line.contains_standalone_comments())
):
# Only apply basic string preprocessing, since lines shouldn't be split here.
if Preview.string_processing in mode:
transformers = [string_merge, string_paren_strip]
else:
transformers = []
elif line.is_def:
transformers = [left_hand_split]
else:
def _rhs(
self: object, line: Line, features: Collection[Feature], mode: Mode
) -> Iterator[Line]:
"""Wraps calls to `right_hand_split`.
The calls increasingly `omit` right-hand trailers (bracket pairs with
content), meaning the trailers get glued together to split on another
bracket pair instead.
"""
for omit in generate_trailers_to_omit(line, mode.line_length):
lines = list(right_hand_split(line, mode, features, omit=omit))
# Note: this check is only able to figure out if the first line of the
# *current* transformation fits in the line length. This is true only
# for simple cases. All others require running more transforms via
# `transform_line()`. This check doesn't know if those would succeed.
if is_line_short_enough(lines[0], mode=mode):
yield from lines
return
# All splits failed, best effort split with no omits.
# This mostly happens to multiline strings that are by definition
# reported as not fitting a single line, as well as lines that contain
# trailing commas (those have to be exploded).
yield from right_hand_split(line, mode, features=features)
# HACK: nested functions (like _rhs) compiled by mypyc don't retain their
# __name__ attribute which is needed in `run_transformer` further down.
# Unfortunately a nested class breaks mypyc too. So a class must be created
# via type ... https://github.com/mypyc/mypyc/issues/884
rhs = type("rhs", (), {"__call__": _rhs})()
if Preview.string_processing in mode:
if line.inside_brackets:
transformers = [
string_merge,
string_paren_strip,
string_split,
delimiter_split,
standalone_comment_split,
string_paren_wrap,
rhs,
]
else:
transformers = [
string_merge,
string_paren_strip,
string_split,
string_paren_wrap,
rhs,
]
else:
if line.inside_brackets:
transformers = [delimiter_split, standalone_comment_split, rhs]
else:
transformers = [rhs]
# It's always safe to attempt hugging of power operations and pretty much every line
# could match.
transformers.append(hug_power_op)
for transform in transformers:
# We are accumulating lines in `result` because we might want to abort
# mission and return the original line in the end, or attempt a different
# split altogether.
try:
result = run_transformer(line, transform, mode, features, line_str=line_str)
except CannotTransform:
continue
else:
yield from result
break
else:
yield line
class _BracketSplitComponent(Enum):
head = auto()
body = auto()
tail = auto()
def left_hand_split(
line: Line, _features: Collection[Feature], mode: Mode
) -> 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.
Prefer RHS otherwise. This is why this function is not symmetrical with
:func:`right_hand_split` which also handles optional parentheses.
"""
tail_leaves: List[Leaf] = []
body_leaves: List[Leaf] = []
head_leaves: List[Leaf] = []
current_leaves = head_leaves
matching_bracket: Optional[Leaf] = None
for leaf in line.leaves:
if (
current_leaves is body_leaves
and leaf.type in CLOSING_BRACKETS
and leaf.opening_bracket is matching_bracket
and isinstance(matching_bracket, Leaf)
):
ensure_visible(leaf)
ensure_visible(matching_bracket)
current_leaves = tail_leaves if body_leaves else head_leaves
current_leaves.append(leaf)
if current_leaves is head_leaves:
if leaf.type in OPENING_BRACKETS:
matching_bracket = leaf
current_leaves = body_leaves
if not matching_bracket:
raise CannotSplit("No brackets found")
head = bracket_split_build_line(
head_leaves, line, matching_bracket, component=_BracketSplitComponent.head
)
body = bracket_split_build_line(
body_leaves, line, matching_bracket, component=_BracketSplitComponent.body
)
tail = bracket_split_build_line(
tail_leaves, line, matching_bracket, component=_BracketSplitComponent.tail
)
bracket_split_succeeded_or_raise(head, body, tail)
for result in (head, body, tail):
if result:
yield result
def right_hand_split(
line: Line,
mode: Mode,
features: Collection[Feature] = (),
omit: Collection[LeafID] = (),
) -> Iterator[Line]:
"""Split line into many lines, starting with the last matching bracket pair.
If the split was by optional parentheses, attempt splitting without them, too.
`omit` is a collection of closing bracket IDs that shouldn't be considered for
this split.
Note: running this function modifies `bracket_depth` on the leaves of `line`.
"""
rhs_result = _first_right_hand_split(line, omit=omit)
yield from _maybe_split_omitting_optional_parens(
rhs_result, line, mode, features=features, omit=omit
)
def _first_right_hand_split(
line: Line,
omit: Collection[LeafID] = (),
) -> RHSResult:
"""Split the line into head, body, tail starting with the last bracket pair.
Note: this function should not have side effects. It's relied upon by
_maybe_split_omitting_optional_parens to get an opinion whether to prefer
splitting on the right side of an assignment statement.
"""
tail_leaves: List[Leaf] = []
body_leaves: List[Leaf] = []
head_leaves: List[Leaf] = []
current_leaves = tail_leaves
opening_bracket: Optional[Leaf] = None
closing_bracket: Optional[Leaf] = None
for leaf in reversed(line.leaves):
if current_leaves is body_leaves:
if leaf is opening_bracket:
current_leaves = head_leaves if body_leaves else tail_leaves
current_leaves.append(leaf)
if current_leaves is tail_leaves:
if leaf.type in CLOSING_BRACKETS and id(leaf) not in omit:
opening_bracket = leaf.opening_bracket
closing_bracket = leaf
current_leaves = body_leaves
if not (opening_bracket and closing_bracket and head_leaves):
# If there is no opening or closing_bracket that means the split failed and
# all content is in the tail. Otherwise, if `head_leaves` are empty, it means
# the matching `opening_bracket` wasn't available on `line` anymore.
raise CannotSplit("No brackets found")
tail_leaves.reverse()
body_leaves.reverse()
head_leaves.reverse()
head = bracket_split_build_line(
head_leaves, line, opening_bracket, component=_BracketSplitComponent.head
)
body = bracket_split_build_line(
body_leaves, line, opening_bracket, component=_BracketSplitComponent.body
)
tail = bracket_split_build_line(
tail_leaves, line, opening_bracket, component=_BracketSplitComponent.tail
)
bracket_split_succeeded_or_raise(head, body, tail)
return RHSResult(head, body, tail, opening_bracket, closing_bracket)
def _maybe_split_omitting_optional_parens(
rhs: RHSResult,
line: Line,
mode: Mode,
features: Collection[Feature] = (),
omit: Collection[LeafID] = (),
) -> Iterator[Line]:
if (
Feature.FORCE_OPTIONAL_PARENTHESES not in features
# the opening bracket is an optional paren
and rhs.opening_bracket.type == token.LPAR
and not rhs.opening_bracket.value
# the closing bracket is an optional paren
and rhs.closing_bracket.type == token.RPAR
and not rhs.closing_bracket.value
# it's not an import (optional parens are the only thing we can split on
# in this case; attempting a split without them is a waste of time)
and not line.is_import
# there are no standalone comments in the body
and not rhs.body.contains_standalone_comments(0)
# and we can actually remove the parens
and can_omit_invisible_parens(rhs, mode.line_length)
):
omit = {id(rhs.closing_bracket), *omit}
try:
# The RHSResult Omitting Optional Parens.
rhs_oop = _first_right_hand_split(line, omit=omit)
if not (
Preview.prefer_splitting_right_hand_side_of_assignments in line.mode
# the split is right after `=`
and len(rhs.head.leaves) >= 2
and rhs.head.leaves[-2].type == token.EQUAL
# the left side of assignment contains brackets
and any(leaf.type in BRACKETS for leaf in rhs.head.leaves[:-1])
# the left side of assignment is short enough (the -1 is for the ending
# optional paren)
and is_line_short_enough(
rhs.head, mode=replace(mode, line_length=mode.line_length - 1)
)
# the left side of assignment won't explode further because of magic
# trailing comma
and rhs.head.magic_trailing_comma is None
# the split by omitting optional parens isn't preferred by some other
# reason
and not _prefer_split_rhs_oop(rhs_oop, mode)
):
yield from _maybe_split_omitting_optional_parens(
rhs_oop, line, mode, features=features, omit=omit
)
return
except CannotSplit as e:
if not (
can_be_split(rhs.body) or is_line_short_enough(rhs.body, mode=mode)
):
raise CannotSplit(
"Splitting failed, body is still too long and can't be split."
) from e
elif (
rhs.head.contains_multiline_strings()
or rhs.tail.contains_multiline_strings()
):
raise CannotSplit(
"The current optional pair of parentheses is bound to fail to"
" satisfy the splitting algorithm because the head or the tail"
" contains multiline strings which by definition never fit one"
" line."
) from e
ensure_visible(rhs.opening_bracket)
ensure_visible(rhs.closing_bracket)
for result in (rhs.head, rhs.body, rhs.tail):
if result:
yield result
def _prefer_split_rhs_oop(rhs_oop: RHSResult, mode: Mode) -> bool:
"""
Returns whether we should prefer the result from a split omitting optional parens.
"""
has_closing_bracket_after_assign = False
for leaf in reversed(rhs_oop.head.leaves):
if leaf.type == token.EQUAL:
break
if leaf.type in CLOSING_BRACKETS:
has_closing_bracket_after_assign = True
break
return (
# contains matching brackets after the `=` (done by checking there is a
# closing bracket)
has_closing_bracket_after_assign
or (
# the split is actually from inside the optional parens (done by checking
# the first line still contains the `=`)
any(leaf.type == token.EQUAL for leaf in rhs_oop.head.leaves)
# the first line is short enough
and is_line_short_enough(rhs_oop.head, mode=mode)
)
# contains unsplittable type ignore
or rhs_oop.head.contains_unsplittable_type_ignore()
or rhs_oop.body.contains_unsplittable_type_ignore()
or rhs_oop.tail.contains_unsplittable_type_ignore()
)
def bracket_split_succeeded_or_raise(head: Line, body: Line, tail: Line) -> None:
"""Raise :exc:`CannotSplit` if the last left- or right-hand split failed.
Do nothing otherwise.
A left- or right-hand split is based on a pair of brackets. Content before
(and including) the opening bracket is left on one line, content inside the
brackets is put on a separate line, and finally content starting with and
following the closing bracket is put on a separate line.
Those are called `head`, `body`, and `tail`, respectively. If the split
produced the same line (all content in `head`) or ended up with an empty `body`
and the `tail` is just the closing bracket, then it's considered failed.
"""
tail_len = len(str(tail).strip())
if not body:
if tail_len == 0:
raise CannotSplit("Splitting brackets produced the same line")
elif tail_len < 3:
raise CannotSplit(
f"Splitting brackets on an empty body to save {tail_len} characters is"
" not worth it"
)
def bracket_split_build_line(
leaves: List[Leaf],
original: Line,
opening_bracket: Leaf,
*,
component: _BracketSplitComponent,
) -> Line:
"""Return a new line with given `leaves` and respective comments from `original`.
If it's the head component, brackets will be tracked so trailing commas are
respected.
If it's the body component, the result line is one-indented inside brackets and as
such has its first leaf's prefix normalized and a trailing comma added when
expected.
"""
result = Line(mode=original.mode, depth=original.depth)
if component is _BracketSplitComponent.body:
result.inside_brackets = True
result.depth += 1
if leaves:
# Since body is a new indent level, remove spurious leading whitespace.
normalize_prefix(leaves[0], inside_brackets=True)
# Ensure a trailing comma for imports and standalone function arguments, but
# be careful not to add one after any comments or within type annotations.
no_commas = (
original.is_def
and opening_bracket.value == "("
and not any(leaf.type == token.COMMA for leaf in leaves)
# In particular, don't add one within a parenthesized return annotation.
# Unfortunately the indicator we're in a return annotation (RARROW) may
# be defined directly in the parent node, the parent of the parent ...
# and so on depending on how complex the return annotation is.
# This isn't perfect and there's some false negatives but they are in
# contexts were a comma is actually fine.
and not any(
node.prev_sibling.type == RARROW
for node in (
leaves[0].parent,
getattr(leaves[0].parent, "parent", None),
)
if isinstance(node, Node) and isinstance(node.prev_sibling, Leaf)
)
# Except the false negatives above for PEP 604 unions where we
# can't add the comma.
and not (
leaves[0].parent
and leaves[0].parent.next_sibling
and leaves[0].parent.next_sibling.type == token.VBAR
)
)
if original.is_import or no_commas:
for i in range(len(leaves) - 1, -1, -1):
if leaves[i].type == STANDALONE_COMMENT:
continue
if leaves[i].type != token.COMMA:
new_comma = Leaf(token.COMMA, ",")
leaves.insert(i + 1, new_comma)
break
leaves_to_track: Set[LeafID] = set()
if component is _BracketSplitComponent.head:
leaves_to_track = get_leaves_inside_matching_brackets(leaves)
# Populate the line
for leaf in leaves:
result.append(
leaf,
preformatted=True,
track_bracket=id(leaf) in leaves_to_track,
)
for comment_after in original.comments_after(leaf):
result.append(comment_after, preformatted=True)
if component is _BracketSplitComponent.body and should_split_line(
result, opening_bracket
):
result.should_split_rhs = True
return result
def dont_increase_indentation(split_func: Transformer) -> Transformer:
"""Normalize prefix of the first leaf in every line returned by `split_func`.
This is a decorator over relevant split functions.
"""
@wraps(split_func)
def split_wrapper(
line: Line, features: Collection[Feature], mode: Mode
) -> Iterator[Line]:
for split_line in split_func(line, features, mode):
normalize_prefix(split_line.leaves[0], inside_brackets=True)
yield split_line
return split_wrapper
def _get_last_non_comment_leaf(line: Line) -> Optional[int]:
for leaf_idx in range(len(line.leaves) - 1, 0, -1):
if line.leaves[leaf_idx].type != STANDALONE_COMMENT:
return leaf_idx
return None
def _safe_add_trailing_comma(safe: bool, delimiter_priority: int, line: Line) -> Line:
if (
safe
and delimiter_priority == COMMA_PRIORITY
and line.leaves[-1].type != token.COMMA
and line.leaves[-1].type != STANDALONE_COMMENT
):
new_comma = Leaf(token.COMMA, ",")
line.append(new_comma)
return line
@dont_increase_indentation
def delimiter_split(
line: Line, features: Collection[Feature], mode: Mode
) -> Iterator[Line]:
"""Split according to delimiters of the highest priority.
If the appropriate Features are given, the split will add trailing commas
also in function signatures and calls that contain `*` and `**`.
"""
try:
last_leaf = line.leaves[-1]
except IndexError:
raise CannotSplit("Line empty") from None
bt = line.bracket_tracker
try:
delimiter_priority = bt.max_delimiter_priority(exclude={id(last_leaf)})
except ValueError:
raise CannotSplit("No delimiters found") from None
if delimiter_priority == DOT_PRIORITY:
if bt.delimiter_count_with_priority(delimiter_priority) == 1:
raise CannotSplit("Splitting a single attribute from its owner looks wrong")
current_line = Line(
mode=line.mode, depth=line.depth, inside_brackets=line.inside_brackets
)
lowest_depth = sys.maxsize
trailing_comma_safe = True
def append_to_line(leaf: Leaf) -> Iterator[Line]:
"""Append `leaf` to current line or to new line if appending impossible."""
nonlocal current_line
try:
current_line.append_safe(leaf, preformatted=True)
except ValueError:
yield current_line
current_line = Line(
mode=line.mode, depth=line.depth, inside_brackets=line.inside_brackets
)
current_line.append(leaf)
last_non_comment_leaf = _get_last_non_comment_leaf(line)
for leaf_idx, leaf in enumerate(line.leaves):
yield from append_to_line(leaf)
for comment_after in line.comments_after(leaf):
yield from append_to_line(comment_after)
lowest_depth = min(lowest_depth, leaf.bracket_depth)
if leaf.bracket_depth == lowest_depth:
if is_vararg(leaf, within={syms.typedargslist}):
trailing_comma_safe = (
trailing_comma_safe and Feature.TRAILING_COMMA_IN_DEF in features
)
elif is_vararg(leaf, within={syms.arglist, syms.argument}):
trailing_comma_safe = (
trailing_comma_safe and Feature.TRAILING_COMMA_IN_CALL in features
)
if (
Preview.add_trailing_comma_consistently in mode
and last_leaf.type == STANDALONE_COMMENT
and leaf_idx == last_non_comment_leaf
):
current_line = _safe_add_trailing_comma(
trailing_comma_safe, delimiter_priority, current_line
)
leaf_priority = bt.delimiters.get(id(leaf))
if leaf_priority == delimiter_priority:
yield current_line
current_line = Line(
mode=line.mode, depth=line.depth, inside_brackets=line.inside_brackets
)
if current_line:
current_line = _safe_add_trailing_comma(
trailing_comma_safe, delimiter_priority, current_line
)
yield current_line
@dont_increase_indentation
def standalone_comment_split(
line: Line, features: Collection[Feature], mode: Mode
) -> Iterator[Line]:
"""Split standalone comments from the rest of the line."""
if not line.contains_standalone_comments(0):
raise CannotSplit("Line does not have any standalone comments")
current_line = Line(
mode=line.mode, depth=line.depth, inside_brackets=line.inside_brackets
)
def append_to_line(leaf: Leaf) -> Iterator[Line]:
"""Append `leaf` to current line or to new line if appending impossible."""
nonlocal current_line
try:
current_line.append_safe(leaf, preformatted=True)
except ValueError:
yield current_line
current_line = Line(
line.mode, depth=line.depth, inside_brackets=line.inside_brackets
)
current_line.append(leaf)
for leaf in line.leaves:
yield from append_to_line(leaf)
for comment_after in line.comments_after(leaf):
yield from append_to_line(comment_after)
if current_line:
yield current_line
def normalize_prefix(leaf: Leaf, *, inside_brackets: bool) -> None:
"""Leave existing extra newlines if not `inside_brackets`. Remove everything
else.
Note: don't use backslashes for formatting or you'll lose your voting rights.
"""
if not inside_brackets:
spl = leaf.prefix.split("#")
if "\\" not in spl[0]:
nl_count = spl[-1].count("\n")
if len(spl) > 1:
nl_count -= 1
leaf.prefix = "\n" * nl_count
return
leaf.prefix = ""
def normalize_invisible_parens(
node: Node, parens_after: Set[str], *, mode: Mode, features: Collection[Feature]
) -> None:
"""Make existing optional parentheses invisible or create new ones.
`parens_after` is a set of string leaf values immediately after which parens
should be put.
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):
if pc.value in FMT_OFF:
# This `node` has a prefix with `# fmt: off`, don't mess with parens.
return
# The multiple context managers grammar has a different pattern, thus this is
# separate from the for-loop below. This possibly wraps them in invisible parens,
# and later will be removed in remove_with_parens when needed.
if node.type == syms.with_stmt:
_maybe_wrap_cms_in_parens(node, mode, features)
check_lpar = False
for index, child in enumerate(list(node.children)):
# 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, mode=mode, features=features
)
# Add parentheses around long tuple unpacking in assignments.
if (
index == 0
and isinstance(child, Node)
and child.type == syms.testlist_star_expr
):
check_lpar = True
if check_lpar:
if (
child.type == syms.atom
and node.type == syms.for_stmt
and isinstance(child.prev_sibling, Leaf)
and child.prev_sibling.type == token.NAME
and child.prev_sibling.value == "for"
):
if maybe_make_parens_invisible_in_atom(
child,
parent=node,
remove_brackets_around_comma=True,
):
wrap_in_parentheses(node, child, visible=False)
elif isinstance(child, Node) and node.type == syms.with_stmt:
remove_with_parens(child, node)
elif child.type == syms.atom:
if maybe_make_parens_invisible_in_atom(
child,
parent=node,
):
wrap_in_parentheses(node, child, visible=False)
elif is_one_tuple(child):
wrap_in_parentheses(node, child, visible=True)
elif node.type == syms.import_from:
_normalize_import_from(node, child, index)
break
elif (
index == 1
and child.type == token.STAR
and node.type == syms.except_clause
):
# In except* (PEP 654), the star is actually part of
# of the keyword. So we need to skip the insertion of
# invisible parentheses to work more precisely.
continue
elif not (isinstance(child, Leaf) and is_multiline_string(child)):
wrap_in_parentheses(node, child, visible=False)
comma_check = child.type == token.COMMA
check_lpar = isinstance(child, Leaf) and (
child.value in parens_after or comma_check
)
def _normalize_import_from(parent: Node, child: LN, index: int) -> None:
# "import from" nodes store parentheses directly as part of
# the statement
if is_lpar_token(child):
assert is_rpar_token(parent.children[-1])
# make parentheses invisible
child.value = ""
parent.children[-1].value = ""
elif child.type != token.STAR:
# insert invisible parentheses
parent.insert_child(index, Leaf(token.LPAR, ""))
parent.append_child(Leaf(token.RPAR, ""))
def remove_await_parens(node: Node) -> None:
if node.children[0].type == token.AWAIT and len(node.children) > 1:
if (
node.children[1].type == syms.atom
and node.children[1].children[0].type == token.LPAR
):
if maybe_make_parens_invisible_in_atom(
node.children[1],
parent=node,
remove_brackets_around_comma=True,
):
wrap_in_parentheses(node, node.children[1], visible=False)
# Since await is an expression we shouldn't remove
# brackets in cases where this would change
# the AST due to operator precedence.
# Therefore we only aim to remove brackets around
# power nodes that aren't also await expressions themselves.
# https://peps.python.org/pep-0492/#updated-operator-precedence-table
# N.B. We've still removed any redundant nested brackets though :)
opening_bracket = cast(Leaf, node.children[1].children[0])
closing_bracket = cast(Leaf, node.children[1].children[-1])
bracket_contents = node.children[1].children[1]
if isinstance(bracket_contents, Node):
if bracket_contents.type != syms.power:
ensure_visible(opening_bracket)
ensure_visible(closing_bracket)
elif (
bracket_contents.type == syms.power
and bracket_contents.children[0].type == token.AWAIT
):
ensure_visible(opening_bracket)
ensure_visible(closing_bracket)
# If we are in a nested await then recurse down.
remove_await_parens(bracket_contents)
def _maybe_wrap_cms_in_parens(
node: Node, mode: Mode, features: Collection[Feature]
) -> None:
"""When enabled and safe, wrap the multiple context managers in invisible parens.
It is only safe when `features` contain Feature.PARENTHESIZED_CONTEXT_MANAGERS.
"""
if (
Feature.PARENTHESIZED_CONTEXT_MANAGERS not in features
or Preview.wrap_multiple_context_managers_in_parens not in mode
or len(node.children) <= 2
# If it's an atom, it's already wrapped in parens.
or node.children[1].type == syms.atom
):
return
colon_index: Optional[int] = None
for i in range(2, len(node.children)):
if node.children[i].type == token.COLON:
colon_index = i
break
if colon_index is not None:
lpar = Leaf(token.LPAR, "")
rpar = Leaf(token.RPAR, "")
context_managers = node.children[1:colon_index]
for child in context_managers:
child.remove()
# After wrapping, the with_stmt will look like this:
# with_stmt
# NAME 'with'
# atom
# LPAR ''
# testlist_gexp
# ... <-- context_managers
# /testlist_gexp
# RPAR ''
# /atom
# COLON ':'
new_child = Node(
syms.atom, [lpar, Node(syms.testlist_gexp, context_managers), rpar]
)
node.insert_child(1, new_child)
def remove_with_parens(node: Node, parent: Node) -> None:
"""Recursively hide optional parens in `with` statements."""
# Removing all unnecessary parentheses in with statements in one pass is a tad
# complex as different variations of bracketed statements result in pretty
# different parse trees:
#
# with (open("file")) as f: # this is an asexpr_test
# ...
#
# with (open("file") as f): # this is an atom containing an
# ... # asexpr_test
#
# with (open("file")) as f, (open("file")) as f: # this is asexpr_test, COMMA,
# ... # asexpr_test
#
# with (open("file") as f, open("file") as f): # an atom containing a
# ... # testlist_gexp which then
# # contains multiple asexpr_test(s)
if node.type == syms.atom:
if maybe_make_parens_invisible_in_atom(
node,
parent=parent,
remove_brackets_around_comma=True,
):
wrap_in_parentheses(parent, node, visible=False)
if isinstance(node.children[1], Node):
remove_with_parens(node.children[1], node)
elif node.type == syms.testlist_gexp:
for child in node.children:
if isinstance(child, Node):
remove_with_parens(child, node)
elif node.type == syms.asexpr_test and not any(
leaf.type == token.COLONEQUAL for leaf in node.leaves()
):
if maybe_make_parens_invisible_in_atom(
node.children[0],
parent=node,
remove_brackets_around_comma=True,
):
wrap_in_parentheses(node, node.children[0], visible=False)
def maybe_make_parens_invisible_in_atom(
node: LN,
parent: LN,
remove_brackets_around_comma: bool = False,
) -> bool:
"""If it's safe, make the parens in the atom `node` invisible, recursively.
Additionally, remove repeated, adjacent invisible parens from the atom `node`
as they are redundant.
Returns whether the node should itself be wrapped in invisible parentheses.
"""
if (
node.type != syms.atom
or is_empty_tuple(node)
or is_one_tuple(node)
or (is_yield(node) and parent.type != syms.expr_stmt)
or (
# This condition tries to prevent removing non-optional brackets
# around a tuple, however, can be a bit overzealous so we provide
# and option to skip this check for `for` and `with` statements.
not remove_brackets_around_comma
and max_delimiter_priority_in_atom(node) >= COMMA_PRIORITY
)
or is_tuple_containing_walrus(node)
):
return False
if is_walrus_assignment(node):
if parent.type in [
syms.annassign,
syms.expr_stmt,
syms.assert_stmt,
syms.return_stmt,
syms.except_clause,
syms.funcdef,
syms.with_stmt,
# these ones aren't useful to end users, but they do please fuzzers
syms.for_stmt,
syms.del_stmt,
syms.for_stmt,
]:
return False
first = node.children[0]
last = node.children[-1]
if is_lpar_token(first) and is_rpar_token(last):
middle = node.children[1]
# make parentheses invisible
if (
# If the prefix of `middle` includes a type comment with
# ignore annotation, then we do not remove the parentheses
not is_type_ignore_comment_string(middle.prefix.strip())
):
first.value = ""
last.value = ""
maybe_make_parens_invisible_in_atom(
middle,
parent=parent,
remove_brackets_around_comma=remove_brackets_around_comma,
)
if is_atom_with_invisible_parens(middle):
# Strip the invisible parens from `middle` by replacing
# it with the child in-between the invisible parens
middle.replace(middle.children[1])
return False
return True
def should_split_line(line: Line, opening_bracket: Leaf) -> bool:
"""Should `line` be immediately split with `delimiter_split()` after RHS?"""
if not (opening_bracket.parent and opening_bracket.value in "[{("):
return False
# We're essentially checking if the body is delimited by commas and there's more
# than one of them (we're excluding the trailing comma and if the delimiter priority
# is still commas, that means there's more).
exclude = set()
trailing_comma = False
try:
last_leaf = line.leaves[-1]
if last_leaf.type == token.COMMA:
trailing_comma = True
exclude.add(id(last_leaf))
max_priority = line.bracket_tracker.max_delimiter_priority(exclude=exclude)
except (IndexError, ValueError):
return False
return max_priority == COMMA_PRIORITY and (
(line.mode.magic_trailing_comma and trailing_comma)
# always explode imports
or opening_bracket.parent.type in {syms.atom, syms.import_from}
)
def generate_trailers_to_omit(line: Line, line_length: int) -> Iterator[Set[LeafID]]:
"""Generate sets of closing bracket IDs that should be omitted in a RHS.
Brackets can be omitted if the entire trailer up to and including
a preceding closing bracket fits in one line.
Yielded sets are cumulative (contain results of previous yields, too). First
set is empty, unless the line should explode, in which case bracket pairs until
the one that needs to explode are omitted.
"""
omit: Set[LeafID] = set()
if not line.magic_trailing_comma:
yield omit
length = 4 * line.depth
opening_bracket: Optional[Leaf] = None
closing_bracket: Optional[Leaf] = None
inner_brackets: Set[LeafID] = set()
for index, leaf, leaf_length in line.enumerate_with_length(reversed=True):
length += leaf_length
if length > line_length:
break
has_inline_comment = leaf_length > len(leaf.value) + len(leaf.prefix)
if leaf.type == STANDALONE_COMMENT or has_inline_comment:
break
if opening_bracket:
if leaf is opening_bracket:
opening_bracket = None
elif leaf.type in CLOSING_BRACKETS:
prev = line.leaves[index - 1] if index > 0 else None
if (
prev
and prev.type == token.COMMA
and leaf.opening_bracket is not None
and not is_one_sequence_between(
leaf.opening_bracket, leaf, line.leaves
)
):
# Never omit bracket pairs with trailing commas.
# We need to explode on those.
break
inner_brackets.add(id(leaf))
elif leaf.type in CLOSING_BRACKETS:
prev = line.leaves[index - 1] if index > 0 else None
if prev and prev.type in OPENING_BRACKETS:
# Empty brackets would fail a split so treat them as "inner"
# brackets (e.g. only add them to the `omit` set if another
# pair of brackets was good enough.
inner_brackets.add(id(leaf))
continue
if closing_bracket:
omit.add(id(closing_bracket))
omit.update(inner_brackets)
inner_brackets.clear()
yield omit
if (
prev
and prev.type == token.COMMA
and leaf.opening_bracket is not None
and not is_one_sequence_between(leaf.opening_bracket, leaf, line.leaves)
):
# Never omit bracket pairs with trailing commas.
# We need to explode on those.
break
if leaf.value:
opening_bracket = leaf.opening_bracket
closing_bracket = leaf
def run_transformer(
line: Line,
transform: Transformer,
mode: Mode,
features: Collection[Feature],
*,
line_str: str = "",
) -> List[Line]:
if not line_str:
line_str = line_to_string(line)
result: List[Line] = []
for transformed_line in transform(line, features, mode):
if str(transformed_line).strip("\n") == line_str:
raise CannotTransform("Line transformer returned an unchanged result")
result.extend(transform_line(transformed_line, mode=mode, features=features))
features_set = set(features)
if (
Feature.FORCE_OPTIONAL_PARENTHESES in features_set
or transform.__class__.__name__ != "rhs"
or not line.bracket_tracker.invisible
or any(bracket.value for bracket in line.bracket_tracker.invisible)
or line.contains_multiline_strings()
or result[0].contains_uncollapsable_type_comments()
or result[0].contains_unsplittable_type_ignore()
or is_line_short_enough(result[0], mode=mode)
# If any leaves have no parents (which _can_ occur since
# `transform(line)` potentially destroys the line's underlying node
# structure), then we can't proceed. Doing so would cause the below
# call to `append_leaves()` to fail.
or any(leaf.parent is None for leaf in line.leaves)
):
return result
line_copy = line.clone()
append_leaves(line_copy, line, line.leaves)
features_fop = features_set | {Feature.FORCE_OPTIONAL_PARENTHESES}
second_opinion = run_transformer(
line_copy, transform, mode, features_fop, line_str=line_str
)
if all(is_line_short_enough(ln, mode=mode) for ln in second_opinion):
result = second_opinion
return result
|