/usr/lib/python2.7/dist-packages/billiard/pool.py is in python-billiard 3.3.0.22-1.
This file is owned by root:root, with mode 0o644.
The actual contents of the file can be viewed below.
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 1574 1575 1576 1577 1578 1579 1580 1581 1582 1583 1584 1585 1586 1587 1588 1589 1590 1591 1592 1593 1594 1595 1596 1597 1598 1599 1600 1601 1602 1603 1604 1605 1606 1607 1608 1609 1610 1611 1612 1613 1614 1615 1616 1617 1618 1619 1620 1621 1622 1623 1624 1625 1626 1627 1628 1629 1630 1631 1632 1633 1634 1635 1636 1637 1638 1639 1640 1641 1642 1643 1644 1645 1646 1647 1648 1649 1650 1651 1652 1653 1654 1655 1656 1657 1658 1659 1660 1661 1662 1663 1664 1665 1666 1667 1668 1669 1670 1671 1672 1673 1674 1675 1676 1677 1678 1679 1680 1681 1682 1683 1684 1685 1686 1687 1688 1689 1690 1691 1692 1693 1694 1695 1696 1697 1698 1699 1700 1701 1702 1703 1704 1705 1706 1707 1708 1709 1710 1711 1712 1713 1714 1715 1716 1717 1718 1719 1720 1721 1722 1723 1724 1725 1726 1727 1728 1729 1730 1731 1732 1733 1734 1735 1736 1737 1738 1739 1740 1741 1742 1743 1744 1745 1746 1747 1748 1749 1750 1751 1752 1753 1754 1755 1756 1757 1758 1759 1760 1761 1762 1763 1764 1765 1766 1767 1768 1769 1770 1771 1772 1773 1774 1775 1776 1777 1778 1779 1780 1781 1782 1783 1784 1785 1786 1787 1788 1789 1790 1791 1792 1793 1794 1795 1796 1797 1798 1799 1800 1801 1802 1803 1804 1805 1806 1807 1808 1809 1810 1811 1812 1813 1814 1815 1816 1817 1818 1819 1820 1821 1822 1823 1824 1825 1826 1827 1828 1829 1830 1831 1832 1833 1834 1835 1836 1837 1838 1839 1840 1841 1842 1843 1844 1845 1846 1847 1848 1849 1850 1851 1852 1853 1854 1855 1856 1857 1858 1859 1860 1861 1862 1863 1864 1865 1866 1867 1868 1869 1870 1871 1872 1873 1874 1875 1876 1877 1878 1879 1880 1881 1882 1883 1884 1885 1886 1887 1888 1889 1890 1891 1892 1893 1894 1895 1896 1897 1898 1899 1900 1901 1902 1903 1904 1905 1906 1907 1908 1909 1910 1911 1912 1913 1914 1915 1916 1917 1918 1919 1920 1921 1922 1923 1924 1925 1926 1927 1928 1929 1930 1931 1932 1933 1934 1935 1936 1937 1938 1939 1940 1941 1942 1943 1944 1945 1946 1947 1948 1949 1950 1951 1952 1953 1954 1955 1956 1957 1958 1959 | # -*- coding: utf-8 -*-
#
# Module providing the `Pool` class for managing a process pool
#
# multiprocessing/pool.py
#
# Copyright (c) 2006-2008, R Oudkerk
# Licensed to PSF under a Contributor Agreement.
#
from __future__ import absolute_import
#
# Imports
#
import errno
import itertools
import os
import platform
import signal
import sys
import threading
import time
import warnings
from collections import deque
from functools import partial
from . import Event, Process, cpu_count
from . import util
from .common import pickle_loads, reset_signals, restart_state
from .compat import get_errno, send_offset
from .einfo import ExceptionInfo
from .dummy import DummyProcess
from .exceptions import (
CoroStop,
RestartFreqExceeded,
SoftTimeLimitExceeded,
Terminated,
TimeLimitExceeded,
TimeoutError,
WorkerLostError,
)
from .five import Empty, Queue, range, values, reraise, monotonic
from .util import Finalize, debug
PY3 = sys.version_info[0] == 3
if platform.system() == 'Windows': # pragma: no cover
# On Windows os.kill calls TerminateProcess which cannot be
# handled by # any process, so this is needed to terminate the task
# *and its children* (if any).
from ._win import kill_processtree as _kill # noqa
SIGKILL = signal.SIGTERM
else:
from os import kill as _kill # noqa
SIGKILL = signal.SIGKILL
try:
TIMEOUT_MAX = threading.TIMEOUT_MAX
except AttributeError: # pragma: no cover
TIMEOUT_MAX = 1e10 # noqa
if sys.version_info >= (3, 3):
_Semaphore = threading.Semaphore
else:
# Semaphore is a factory function pointing to _Semaphore
_Semaphore = threading._Semaphore # noqa
SIGMAP = dict(
(getattr(signal, n), n) for n in dir(signal) if n.startswith('SIG')
)
#
# Constants representing the state of a pool
#
RUN = 0
CLOSE = 1
TERMINATE = 2
#
# Constants representing the state of a job
#
ACK = 0
READY = 1
TASK = 2
NACK = 3
DEATH = 4
#
# Exit code constants
#
EX_OK = 0
EX_FAILURE = 1
EX_RECYCLE = 0x9B
# Signal used for soft time limits.
SIG_SOFT_TIMEOUT = getattr(signal, "SIGUSR1", None)
#
# Miscellaneous
#
LOST_WORKER_TIMEOUT = 10.0
EX_OK = getattr(os, "EX_OK", 0)
job_counter = itertools.count()
Lock = threading.Lock
def _get_send_offset(connection):
try:
native = connection.send_offset
except AttributeError:
native = None
if native is None:
return partial(send_offset, connection.fileno())
return native
def human_status(status):
if (status or 0) < 0:
try:
return 'signal {0} ({1})'.format(-status, SIGMAP[-status])
except KeyError:
return 'signal {0}'.format(-status)
return 'exitcode {0}'.format(status)
def mapstar(args):
return list(map(*args))
def starmapstar(args):
return list(itertools.starmap(args[0], args[1]))
def error(msg, *args, **kwargs):
if util._logger:
util._logger.error(msg, *args, **kwargs)
def stop_if_not_current(thread, timeout=None):
if thread is not threading.currentThread():
thread.stop(timeout)
class LaxBoundedSemaphore(_Semaphore):
"""Semaphore that checks that # release is <= # acquires,
but ignores if # releases >= value."""
def __init__(self, value=1, verbose=None):
if PY3:
_Semaphore.__init__(self, value)
else:
_Semaphore.__init__(self, value, verbose)
self._initial_value = value
def shrink(self):
self._initial_value -= 1
self.acquire()
if PY3:
def release(self):
cond = self._cond
with cond:
if self._value < self._initial_value:
self._value += 1
cond.notify_all()
def clear(self):
while self._value < self._initial_value:
_Semaphore.release(self)
def grow(self):
with self._cond:
self._initial_value += 1
self._value += 1
self._cond.notify()
else:
def grow(self):
cond = self._Semaphore__cond
with cond:
self._initial_value += 1
self._Semaphore__value += 1
cond.notify()
def release(self): # noqa
cond = self._Semaphore__cond
with cond:
if self._Semaphore__value < self._initial_value:
self._Semaphore__value += 1
cond.notifyAll()
def clear(self): # noqa
while self._Semaphore__value < self._initial_value:
_Semaphore.release(self)
#
# Exceptions
#
class MaybeEncodingError(Exception):
"""Wraps possible unpickleable errors, so they can be
safely sent through the socket."""
def __init__(self, exc, value):
self.exc = repr(exc)
self.value = repr(value)
super(MaybeEncodingError, self).__init__(self.exc, self.value)
def __repr__(self):
return "<MaybeEncodingError: %s>" % str(self)
def __str__(self):
return "Error sending result: '%r'. Reason: '%r'." % (
self.value, self.exc)
class WorkersJoined(Exception):
"""All workers have terminated."""
def soft_timeout_sighandler(signum, frame):
raise SoftTimeLimitExceeded()
#
# Code run by worker processes
#
class Worker(Process):
_controlled_termination = False
_job_terminated = False
def __init__(self, inq, outq, synq=None, initializer=None, initargs=(),
maxtasks=None, sentinel=None, on_exit=None,
sigprotection=True):
assert maxtasks is None or (type(maxtasks) == int and maxtasks > 0)
self.initializer = initializer
self.initargs = initargs
self.maxtasks = maxtasks
self._shutdown = sentinel
self.on_exit = on_exit
self.sigprotection = sigprotection
self.inq, self.outq, self.synq = inq, outq, synq
self._make_shortcuts()
super(Worker, self).__init__()
def __reduce__(self):
return self.__class__, (
self.inq, self.outq, self.synq, self.initializer,
self.initargs, self.maxtasks, self._shutdown,
)
def _make_shortcuts(self):
self.inqW_fd = self.inq._writer.fileno() # inqueue write fd
self.outqR_fd = self.outq._reader.fileno() # outqueue read fd
if self.synq:
self.synqR_fd = self.synq._reader.fileno() # synqueue read fd
self.synqW_fd = self.synq._writer.fileno() # synqueue write fd
self.send_syn_offset = _get_send_offset(self.synq._writer)
else:
self.synqR_fd = self.synqW_fd = self._send_syn_offset = None
self._quick_put = self.inq._writer.send
self._quick_get = self.outq._reader.recv
self.send_job_offset = _get_send_offset(self.inq._writer)
def run(self):
_exit = sys.exit
_exitcode = [None]
def exit(status=None):
_exitcode[0] = status
return _exit()
sys.exit = exit
pid = os.getpid()
self._make_child_methods()
self.after_fork()
self.on_loop_start(pid=pid) # callback on loop start
try:
sys.exit(self.workloop(pid=pid))
except Exception as exc:
error('Pool process %r error: %r', self, exc, exc_info=1)
self._do_exit(pid, _exitcode[0], exc)
finally:
self._do_exit(pid, _exitcode[0], None)
def _do_exit(self, pid, exitcode, exc=None):
if exitcode is None:
exitcode = EX_FAILURE if exc else EX_OK
if self.on_exit is not None:
self.on_exit(pid, exitcode)
if sys.platform != 'win32':
try:
self.outq.put((DEATH, (pid, exitcode)))
time.sleep(1)
finally:
os._exit(exitcode)
else:
os._exit(exitcode)
def on_loop_start(self, pid):
pass
def terminate_controlled(self):
self._controlled_termination = True
self.terminate()
def prepare_result(self, result):
return result
def workloop(self, debug=debug, now=monotonic, pid=None):
pid = pid or os.getpid()
put = self.outq.put
inqW_fd = self.inqW_fd
synqW_fd = self.synqW_fd
maxtasks = self.maxtasks
prepare_result = self.prepare_result
wait_for_job = self.wait_for_job
_wait_for_syn = self.wait_for_syn
def wait_for_syn(jid):
i = 0
while 1:
if i > 60:
error('!!!WAIT FOR ACK TIMEOUT: job:%r fd:%r!!!',
jid, self.synq._reader.fileno(), exc_info=1)
req = _wait_for_syn()
if req:
type_, args = req
if type_ == NACK:
return False
assert type_ == ACK
return True
i += 1
completed = 0
while maxtasks is None or (maxtasks and completed < maxtasks):
req = wait_for_job()
if req:
type_, args_ = req
assert type_ == TASK
job, i, fun, args, kwargs = args_
put((ACK, (job, i, now(), pid, synqW_fd)))
if _wait_for_syn:
confirm = wait_for_syn(job)
if not confirm:
continue # received NACK
try:
result = (True, prepare_result(fun(*args, **kwargs)))
except Exception:
result = (False, ExceptionInfo())
try:
put((READY, (job, i, result, inqW_fd)))
except Exception as exc:
_, _, tb = sys.exc_info()
try:
wrapped = MaybeEncodingError(exc, result[1])
einfo = ExceptionInfo((
MaybeEncodingError, wrapped, tb,
))
put((READY, (job, i, (False, einfo), inqW_fd)))
finally:
del(tb)
completed += 1
debug('worker exiting after %d tasks', completed)
if maxtasks:
return EX_RECYCLE if completed == maxtasks else EX_FAILURE
return EX_OK
def after_fork(self):
if hasattr(self.inq, '_writer'):
self.inq._writer.close()
if hasattr(self.outq, '_reader'):
self.outq._reader.close()
if self.initializer is not None:
self.initializer(*self.initargs)
# Make sure all exiting signals call finally: blocks.
# This is important for the semaphore to be released.
reset_signals(full=self.sigprotection)
# install signal handler for soft timeouts.
if SIG_SOFT_TIMEOUT is not None:
signal.signal(SIG_SOFT_TIMEOUT, soft_timeout_sighandler)
try:
signal.signal(signal.SIGINT, signal.SIG_IGN)
except AttributeError:
pass
def _make_recv_method(self, conn):
get = conn.get
if hasattr(conn, '_reader'):
_poll = conn._reader.poll
if hasattr(conn, 'get_payload') and conn.get_payload:
get_payload = conn.get_payload
def _recv(timeout, loads=pickle_loads):
return True, loads(get_payload())
else:
def _recv(timeout): # noqa
if _poll(timeout):
return True, get()
return False, None
else:
def _recv(timeout): # noqa
try:
return True, get(timeout=timeout)
except Queue.Empty:
return False, None
return _recv
def _make_child_methods(self, loads=pickle_loads):
self.wait_for_job = self._make_protected_receive(self.inq)
self.wait_for_syn = (self._make_protected_receive(self.synq)
if self.synq else None)
def _make_protected_receive(self, conn):
_receive = self._make_recv_method(conn)
should_shutdown = self._shutdown.is_set if self._shutdown else None
def receive(debug=debug):
if should_shutdown and should_shutdown():
debug('worker got sentinel -- exiting')
raise SystemExit(EX_OK)
try:
ready, req = _receive(1.0)
if not ready:
return None
except (EOFError, IOError) as exc:
if get_errno(exc) == errno.EINTR:
return None # interrupted, maybe by gdb
debug('worker got %s -- exiting', type(exc).__name__)
raise SystemExit(EX_FAILURE)
if req is None:
debug('worker got sentinel -- exiting')
raise SystemExit(EX_FAILURE)
return req
return receive
#
# Class representing a process pool
#
class PoolThread(DummyProcess):
def __init__(self, *args, **kwargs):
DummyProcess.__init__(self)
self._state = RUN
self._was_started = False
self.daemon = True
def run(self):
try:
return self.body()
except RestartFreqExceeded as exc:
error("Thread %r crashed: %r", type(self).__name__, exc,
exc_info=1)
_kill(os.getpid(), signal.SIGTERM)
sys.exit()
except Exception as exc:
error("Thread %r crashed: %r", type(self).__name__, exc,
exc_info=1)
os._exit(1)
def start(self, *args, **kwargs):
self._was_started = True
super(PoolThread, self).start(*args, **kwargs)
def on_stop_not_started(self):
pass
def stop(self, timeout=None):
if self._was_started:
self.join(timeout)
return
self.on_stop_not_started()
def terminate(self):
self._state = TERMINATE
def close(self):
self._state = CLOSE
class Supervisor(PoolThread):
def __init__(self, pool):
self.pool = pool
super(Supervisor, self).__init__()
def body(self):
debug('worker handler starting')
time.sleep(0.8)
pool = self.pool
try:
# do a burst at startup to verify that we can start
# our pool processes, and in that time we lower
# the max restart frequency.
prev_state = pool.restart_state
pool.restart_state = restart_state(10 * pool._processes, 1)
for _ in range(10):
if self._state == RUN and pool._state == RUN:
pool._maintain_pool()
time.sleep(0.1)
# Keep maintaing workers until the cache gets drained, unless
# the pool is termianted
pool.restart_state = prev_state
while self._state == RUN and pool._state == RUN:
pool._maintain_pool()
time.sleep(0.8)
except RestartFreqExceeded:
pool.close()
pool.join()
raise
debug('worker handler exiting')
class TaskHandler(PoolThread):
def __init__(self, taskqueue, put, outqueue, pool):
self.taskqueue = taskqueue
self.put = put
self.outqueue = outqueue
self.pool = pool
super(TaskHandler, self).__init__()
def body(self):
taskqueue = self.taskqueue
put = self.put
for taskseq, set_length in iter(taskqueue.get, None):
try:
i = -1
for i, task in enumerate(taskseq):
if self._state:
debug('task handler found thread._state != RUN')
break
try:
put(task)
except IOError:
debug('could not put task on queue')
break
else:
if set_length:
debug('doing set_length()')
set_length(i + 1)
continue
break
except Exception as exc:
error('Task Handler ERROR: %r', exc, exc_info=1)
break
else:
debug('task handler got sentinel')
self.tell_others()
def tell_others(self):
outqueue = self.outqueue
put = self.put
pool = self.pool
try:
# tell result handler to finish when cache is empty
debug('task handler sending sentinel to result handler')
outqueue.put(None)
# tell workers there is no more work
debug('task handler sending sentinel to workers')
for p in pool:
put(None)
except IOError:
debug('task handler got IOError when sending sentinels')
debug('task handler exiting')
def on_stop_not_started(self):
self.tell_others()
class TimeoutHandler(PoolThread):
def __init__(self, processes, cache, t_soft, t_hard):
self.processes = processes
self.cache = cache
self.t_soft = t_soft
self.t_hard = t_hard
self._it = None
super(TimeoutHandler, self).__init__()
def _process_by_pid(self, pid):
return next((
(proc, i) for i, proc in enumerate(self.processes)
if proc.pid == pid
), (None, None))
def on_soft_timeout(self, job):
debug('soft time limit exceeded for %r', job)
process, _index = self._process_by_pid(job._worker_pid)
if not process:
return
# Run timeout callback
job.handle_timeout(soft=True)
try:
_kill(job._worker_pid, SIG_SOFT_TIMEOUT)
except OSError as exc:
if get_errno(exc) != errno.ESRCH:
raise
def on_hard_timeout(self, job):
if job.ready():
return
debug('hard time limit exceeded for %r', job)
# Remove from cache and set return value to an exception
try:
raise TimeLimitExceeded(job._timeout)
except TimeLimitExceeded:
job._set(job._job, (False, ExceptionInfo()))
else: # pragma: no cover
pass
# Remove from _pool
process, _index = self._process_by_pid(job._worker_pid)
# Run timeout callback
job.handle_timeout(soft=False)
if process:
self._trywaitkill(process)
def _trywaitkill(self, worker):
debug('timeout: sending TERM to %s', worker._name)
try:
worker.terminate()
except OSError:
pass
else:
if worker._popen.wait(timeout=0.1):
return
debug('timeout: TERM timed-out, now sending KILL to %s', worker._name)
try:
_kill(worker.pid, SIGKILL)
except OSError:
pass
def handle_timeouts(self):
cache = self.cache
t_hard, t_soft = self.t_hard, self.t_soft
dirty = set()
on_soft_timeout = self.on_soft_timeout
on_hard_timeout = self.on_hard_timeout
def _timed_out(start, timeout):
if not start or not timeout:
return False
if monotonic() >= start + timeout:
return True
# Inner-loop
while self._state == RUN:
# Remove dirty items not in cache anymore
if dirty:
dirty = set(k for k in dirty if k in cache)
for i, job in list(cache.items()):
ack_time = job._time_accepted
soft_timeout = job._soft_timeout
if soft_timeout is None:
soft_timeout = t_soft
hard_timeout = job._timeout
if hard_timeout is None:
hard_timeout = t_hard
if _timed_out(ack_time, hard_timeout):
on_hard_timeout(job)
elif i not in dirty and _timed_out(ack_time, soft_timeout):
on_soft_timeout(job)
dirty.add(i)
yield
def body(self):
while self._state == RUN:
try:
for _ in self.handle_timeouts():
time.sleep(1.0) # don't spin
except CoroStop:
break
debug('timeout handler exiting')
def handle_event(self, *args):
if self._it is None:
self._it = self.handle_timeouts()
try:
next(self._it)
except StopIteration:
self._it = None
class ResultHandler(PoolThread):
def __init__(self, outqueue, get, cache, poll,
join_exited_workers, putlock, restart_state,
check_timeouts, on_job_ready):
self.outqueue = outqueue
self.get = get
self.cache = cache
self.poll = poll
self.join_exited_workers = join_exited_workers
self.putlock = putlock
self.restart_state = restart_state
self._it = None
self._shutdown_complete = False
self.check_timeouts = check_timeouts
self.on_job_ready = on_job_ready
self._make_methods()
super(ResultHandler, self).__init__()
def on_stop_not_started(self):
# used when pool started without result handler thread.
self.finish_at_shutdown(handle_timeouts=True)
def _make_methods(self):
cache = self.cache
putlock = self.putlock
restart_state = self.restart_state
on_job_ready = self.on_job_ready
def on_ack(job, i, time_accepted, pid, synqW_fd):
restart_state.R = 0
try:
cache[job]._ack(i, time_accepted, pid, synqW_fd)
except (KeyError, AttributeError):
# Object gone or doesn't support _ack (e.g. IMAPIterator).
pass
def on_ready(job, i, obj, inqW_fd):
if on_job_ready is not None:
on_job_ready(job, i, obj, inqW_fd)
try:
item = cache[job]
except KeyError:
return
if not item.ready():
if putlock is not None:
putlock.release()
try:
item._set(i, obj)
except KeyError:
pass
def on_death(pid, exitcode):
try:
os.kill(pid, signal.SIGTERM)
except OSError as exc:
if get_errno(exc) != errno.ESRCH:
raise
state_handlers = self.state_handlers = {
ACK: on_ack, READY: on_ready, DEATH: on_death
}
def on_state_change(task):
state, args = task
try:
state_handlers[state](*args)
except KeyError:
debug("Unknown job state: %s (args=%s)", state, args)
self.on_state_change = on_state_change
def _process_result(self, timeout=1.0):
poll = self.poll
on_state_change = self.on_state_change
while 1:
try:
ready, task = poll(timeout)
except (IOError, EOFError) as exc:
debug('result handler got %r -- exiting', exc)
raise CoroStop()
if self._state:
assert self._state == TERMINATE
debug('result handler found thread._state=TERMINATE')
raise CoroStop()
if ready:
if task is None:
debug('result handler got sentinel')
raise CoroStop()
on_state_change(task)
if timeout != 0: # blocking
break
else:
break
yield
def handle_event(self, fileno=None, events=None):
if self._state == RUN:
if self._it is None:
self._it = self._process_result(0) # non-blocking
try:
next(self._it)
except (StopIteration, CoroStop):
self._it = None
def body(self):
debug('result handler starting')
try:
while self._state == RUN:
try:
for _ in self._process_result(1.0): # blocking
pass
except CoroStop:
break
finally:
self.finish_at_shutdown()
def finish_at_shutdown(self, handle_timeouts=False):
self._shutdown_complete = True
get = self.get
outqueue = self.outqueue
cache = self.cache
poll = self.poll
join_exited_workers = self.join_exited_workers
check_timeouts = self.check_timeouts
on_state_change = self.on_state_change
time_terminate = None
while cache and self._state != TERMINATE:
if check_timeouts is not None:
check_timeouts()
try:
ready, task = poll(1.0)
except (IOError, EOFError) as exc:
debug('result handler got %r -- exiting', exc)
return
if ready:
if task is None:
debug('result handler ignoring extra sentinel')
continue
on_state_change(task)
try:
join_exited_workers(shutdown=True)
except WorkersJoined:
now = monotonic()
if not time_terminate:
time_terminate = now
else:
if now - time_terminate > 5.0:
debug('result handler exiting: timed out')
break
debug('result handler: all workers terminated, '
'timeout in %ss',
abs(min(now - time_terminate - 5.0, 0)))
if hasattr(outqueue, '_reader'):
debug('ensuring that outqueue is not full')
# If we don't make room available in outqueue then
# attempts to add the sentinel (None) to outqueue may
# block. There is guaranteed to be no more than 2 sentinels.
try:
for i in range(10):
if not outqueue._reader.poll():
break
get()
except (IOError, EOFError):
pass
debug('result handler exiting: len(cache)=%s, thread._state=%s',
len(cache), self._state)
class Pool(object):
'''
Class which supports an async version of applying functions to arguments.
'''
Worker = Worker
Supervisor = Supervisor
TaskHandler = TaskHandler
TimeoutHandler = TimeoutHandler
ResultHandler = ResultHandler
SoftTimeLimitExceeded = SoftTimeLimitExceeded
def __init__(self, processes=None, initializer=None, initargs=(),
maxtasksperchild=None, timeout=None, soft_timeout=None,
lost_worker_timeout=None,
max_restarts=None, max_restart_freq=1,
on_process_up=None,
on_process_down=None,
on_timeout_set=None,
on_timeout_cancel=None,
threads=True,
semaphore=None,
putlocks=False,
allow_restart=False,
synack=False,
on_process_exit=None,
**kwargs):
self.synack = synack
self._setup_queues()
self._taskqueue = Queue()
self._cache = {}
self._state = RUN
self.timeout = timeout
self.soft_timeout = soft_timeout
self._maxtasksperchild = maxtasksperchild
self._initializer = initializer
self._initargs = initargs
self._on_process_exit = on_process_exit
self.lost_worker_timeout = lost_worker_timeout or LOST_WORKER_TIMEOUT
self.on_process_up = on_process_up
self.on_process_down = on_process_down
self.on_timeout_set = on_timeout_set
self.on_timeout_cancel = on_timeout_cancel
self.threads = threads
self.readers = {}
self.allow_restart = allow_restart
if soft_timeout and SIG_SOFT_TIMEOUT is None:
warnings.warn(UserWarning(
"Soft timeouts are not supported: "
"on this platform: It does not have the SIGUSR1 signal.",
))
soft_timeout = None
self._processes = self.cpu_count() if processes is None else processes
self.max_restarts = max_restarts or round(self._processes * 100)
self.restart_state = restart_state(max_restarts, max_restart_freq or 1)
if initializer is not None and not callable(initializer):
raise TypeError('initializer must be a callable')
if on_process_exit is not None and not callable(on_process_exit):
raise TypeError('on_process_exit must be callable')
self._pool = []
self._poolctrl = {}
self.putlocks = putlocks
self._putlock = semaphore or LaxBoundedSemaphore(self._processes)
for i in range(self._processes):
self._create_worker_process(i)
self._worker_handler = self.Supervisor(self)
if threads:
self._worker_handler.start()
self._task_handler = self.TaskHandler(self._taskqueue,
self._quick_put,
self._outqueue,
self._pool)
if threads:
self._task_handler.start()
# Thread killing timedout jobs.
self._timeout_handler = self.TimeoutHandler(
self._pool, self._cache,
self.soft_timeout, self.timeout,
)
self._timeout_handler_mutex = Lock()
self._timeout_handler_started = False
if self.timeout is not None or self.soft_timeout is not None:
self._start_timeout_handler()
# If running without threads, we need to check for timeouts
# while waiting for unfinished work at shutdown.
self.check_timeouts = None
if not threads:
self.check_timeouts = self._timeout_handler.handle_event
# Thread processing results in the outqueue.
self._result_handler = self.create_result_handler()
self.handle_result_event = self._result_handler.handle_event
if threads:
self._result_handler.start()
self._terminate = Finalize(
self, self._terminate_pool,
args=(self._taskqueue, self._inqueue, self._outqueue,
self._pool, self._worker_handler, self._task_handler,
self._result_handler, self._cache,
self._timeout_handler,
self._help_stuff_finish_args()),
exitpriority=15,
)
def create_result_handler(self, **extra_kwargs):
return self.ResultHandler(
self._outqueue, self._quick_get, self._cache,
self._poll_result, self._join_exited_workers,
self._putlock, self.restart_state, self.check_timeouts,
self.on_job_ready, **extra_kwargs
)
def on_job_ready(self, job, i, obj, inqW_fd):
pass
def _help_stuff_finish_args(self):
return self._inqueue, self._task_handler, self._pool
def cpu_count(self):
try:
return cpu_count()
except NotImplementedError:
return 1
def handle_result_event(self, *args):
return self._result_handler.handle_event(*args)
def _process_register_queues(self, worker, queues):
pass
def _process_by_pid(self, pid):
return next((
(proc, i) for i, proc in enumerate(self._pool)
if proc.pid == pid
), (None, None))
def get_process_queues(self):
return self._inqueue, self._outqueue, None
def _create_worker_process(self, i):
sentinel = Event() if self.allow_restart else None
inq, outq, synq = self.get_process_queues()
w = self.Worker(
inq, outq, synq, self._initializer, self._initargs,
self._maxtasksperchild, sentinel, self._on_process_exit,
# Need to handle all signals if using the ipc semaphore,
# to make sure the semaphore is released.
sigprotection=self.threads,
)
self._pool.append(w)
self._process_register_queues(w, (inq, outq, synq))
w.name = w.name.replace('Process', 'PoolWorker')
w.daemon = True
w.index = i
w.start()
self._poolctrl[w.pid] = sentinel
if self.on_process_up:
self.on_process_up(w)
return w
def process_flush_queues(self, worker):
pass
def _join_exited_workers(self, shutdown=False):
"""Cleanup after any worker processes which have exited due to
reaching their specified lifetime. Returns True if any workers were
cleaned up.
"""
now = None
# The worker may have published a result before being terminated,
# but we have no way to accurately tell if it did. So we wait for
# _lost_worker_timeout seconds before we mark the job with
# WorkerLostError.
for job in [job for job in list(self._cache.values())
if not job.ready() and job._worker_lost]:
now = now or monotonic()
lost_time, lost_ret = job._worker_lost
if now - lost_time > job._lost_worker_timeout:
self.mark_as_worker_lost(job, lost_ret)
if shutdown and not len(self._pool):
raise WorkersJoined()
cleaned, exitcodes = {}, {}
for i in reversed(range(len(self._pool))):
worker = self._pool[i]
exitcode = worker.exitcode
popen = worker._popen
if popen is None or exitcode is not None:
# worker exited
debug('Supervisor: cleaning up worker %d', i)
if popen is not None:
worker.join()
debug('Supervisor: worked %d joined', i)
cleaned[worker.pid] = worker
exitcodes[worker.pid] = exitcode
if exitcode not in (EX_OK, EX_RECYCLE) and \
not getattr(worker, '_controlled_termination', False):
error(
'Process %r pid:%r exited with %r',
worker.name, worker.pid, human_status(exitcode),
exc_info=0,
)
self.process_flush_queues(worker)
del self._pool[i]
del self._poolctrl[worker.pid]
if cleaned:
all_pids = [w.pid for w in self._pool]
for job in list(self._cache.values()):
acked_by_gone = next(
(pid for pid in job.worker_pids()
if pid in cleaned or pid not in all_pids),
None
)
# already accepted by process
if acked_by_gone:
self.on_job_process_down(job, acked_by_gone)
if not job.ready():
exitcode = exitcodes.get(acked_by_gone) or 0
proc = cleaned.get(acked_by_gone)
if proc and getattr(proc, '_job_terminated', False):
job._set_terminated(exitcode)
else:
self.on_job_process_lost(
job, acked_by_gone, exitcode,
)
else:
# started writing to
write_to = job._write_to
# was scheduled to write to
sched_for = job._scheduled_for
if write_to and not write_to._is_alive():
self.on_job_process_down(job, write_to.pid)
elif sched_for and not sched_for._is_alive():
self.on_job_process_down(job, sched_for.pid)
for worker in values(cleaned):
if self.on_process_down:
if not shutdown:
self._process_cleanup_queues(worker)
self.on_process_down(worker)
return list(exitcodes.values())
return []
def on_partial_read(self, job, worker):
pass
def _process_cleanup_queues(self, worker):
pass
def on_job_process_down(self, job, pid_gone):
pass
def on_job_process_lost(self, job, pid, exitcode):
job._worker_lost = (monotonic(), exitcode)
def mark_as_worker_lost(self, job, exitcode):
try:
raise WorkerLostError(
'Worker exited prematurely: {0}.'.format(
human_status(exitcode)),
)
except WorkerLostError:
job._set(None, (False, ExceptionInfo()))
else: # pragma: no cover
pass
def __enter__(self):
return self
def __exit__(self, *exc_info):
return self.terminate()
def on_grow(self, n):
pass
def on_shrink(self, n):
pass
def shrink(self, n=1):
for i, worker in enumerate(self._iterinactive()):
self._processes -= 1
if self._putlock:
self._putlock.shrink()
worker.terminate_controlled()
self.on_shrink(1)
if i >= n - 1:
break
else:
raise ValueError("Can't shrink pool. All processes busy!")
def grow(self, n=1):
for i in range(n):
self._processes += 1
if self._putlock:
self._putlock.grow()
self.on_grow(n)
def _iterinactive(self):
for worker in self._pool:
if not self._worker_active(worker):
yield worker
raise StopIteration()
def _worker_active(self, worker):
for job in values(self._cache):
if worker.pid in job.worker_pids():
return True
return False
def _repopulate_pool(self, exitcodes):
"""Bring the number of pool processes up to the specified number,
for use after reaping workers which have exited.
"""
for i in range(self._processes - len(self._pool)):
if self._state != RUN:
return
try:
if exitcodes and exitcodes[i] not in (EX_OK, EX_RECYCLE):
self.restart_state.step()
except IndexError:
self.restart_state.step()
self._create_worker_process(self._avail_index())
debug('added worker')
def _avail_index(self):
assert len(self._pool) < self._processes
indices = set(p.index for p in self._pool)
return next(i for i in range(self._processes) if i not in indices)
def did_start_ok(self):
return not self._join_exited_workers()
def _maintain_pool(self):
""""Clean up any exited workers and start replacements for them.
"""
joined = self._join_exited_workers()
self._repopulate_pool(joined)
for i in range(len(joined)):
if self._putlock is not None:
self._putlock.release()
def maintain_pool(self):
if self._worker_handler._state == RUN and self._state == RUN:
try:
self._maintain_pool()
except RestartFreqExceeded:
self.close()
self.join()
raise
except OSError as exc:
if get_errno(exc) == errno.ENOMEM:
reraise(MemoryError,
MemoryError(str(exc)),
sys.exc_info()[2])
raise
def _setup_queues(self):
from billiard.queues import SimpleQueue
self._inqueue = SimpleQueue()
self._outqueue = SimpleQueue()
self._quick_put = self._inqueue._writer.send
self._quick_get = self._outqueue._reader.recv
def _poll_result(timeout):
if self._outqueue._reader.poll(timeout):
return True, self._quick_get()
return False, None
self._poll_result = _poll_result
def _start_timeout_handler(self):
# ensure more than one thread does not start the timeout handler
# thread at once.
if self.threads:
with self._timeout_handler_mutex:
if not self._timeout_handler_started:
self._timeout_handler_started = True
self._timeout_handler.start()
def apply(self, func, args=(), kwds={}):
'''
Equivalent of `func(*args, **kwargs)`.
'''
if self._state == RUN:
return self.apply_async(func, args, kwds).get()
def starmap(self, func, iterable, chunksize=None):
'''
Like `map()` method but the elements of the `iterable` are expected to
be iterables as well and will be unpacked as arguments. Hence
`func` and (a, b) becomes func(a, b).
'''
if self._state == RUN:
return self._map_async(func, iterable,
starmapstar, chunksize).get()
def starmap_async(self, func, iterable, chunksize=None,
callback=None, error_callback=None):
'''
Asynchronous version of `starmap()` method.
'''
if self._state == RUN:
return self._map_async(func, iterable, starmapstar, chunksize,
callback, error_callback)
def map(self, func, iterable, chunksize=None):
'''
Apply `func` to each element in `iterable`, collecting the results
in a list that is returned.
'''
if self._state == RUN:
return self.map_async(func, iterable, chunksize).get()
def imap(self, func, iterable, chunksize=1, lost_worker_timeout=None):
'''
Equivalent of `map()` -- can be MUCH slower than `Pool.map()`.
'''
if self._state != RUN:
return
lost_worker_timeout = lost_worker_timeout or self.lost_worker_timeout
if chunksize == 1:
result = IMapIterator(self._cache,
lost_worker_timeout=lost_worker_timeout)
self._taskqueue.put((
((TASK, (result._job, i, func, (x,), {}))
for i, x in enumerate(iterable)),
result._set_length,
))
return result
else:
assert chunksize > 1
task_batches = Pool._get_tasks(func, iterable, chunksize)
result = IMapIterator(self._cache,
lost_worker_timeout=lost_worker_timeout)
self._taskqueue.put((
((TASK, (result._job, i, mapstar, (x,), {}))
for i, x in enumerate(task_batches)),
result._set_length,
))
return (item for chunk in result for item in chunk)
def imap_unordered(self, func, iterable, chunksize=1,
lost_worker_timeout=None):
'''
Like `imap()` method but ordering of results is arbitrary.
'''
if self._state != RUN:
return
lost_worker_timeout = lost_worker_timeout or self.lost_worker_timeout
if chunksize == 1:
result = IMapUnorderedIterator(
self._cache, lost_worker_timeout=lost_worker_timeout,
)
self._taskqueue.put((
((TASK, (result._job, i, func, (x,), {}))
for i, x in enumerate(iterable)),
result._set_length,
))
return result
else:
assert chunksize > 1
task_batches = Pool._get_tasks(func, iterable, chunksize)
result = IMapUnorderedIterator(
self._cache, lost_worker_timeout=lost_worker_timeout,
)
self._taskqueue.put((
((TASK, (result._job, i, mapstar, (x,), {}))
for i, x in enumerate(task_batches)),
result._set_length,
))
return (item for chunk in result for item in chunk)
def apply_async(self, func, args=(), kwds={},
callback=None, error_callback=None, accept_callback=None,
timeout_callback=None, waitforslot=None,
soft_timeout=None, timeout=None, lost_worker_timeout=None,
callbacks_propagate=(),
correlation_id=None):
'''
Asynchronous equivalent of `apply()` method.
Callback is called when the functions return value is ready.
The accept callback is called when the job is accepted to be executed.
Simplified the flow is like this:
>>> def apply_async(func, args, kwds, callback, accept_callback):
... if accept_callback:
... accept_callback()
... retval = func(*args, **kwds)
... if callback:
... callback(retval)
'''
if self._state != RUN:
return
soft_timeout = soft_timeout or self.soft_timeout
timeout = timeout or self.timeout
lost_worker_timeout = lost_worker_timeout or self.lost_worker_timeout
if soft_timeout and SIG_SOFT_TIMEOUT is None:
warnings.warn(UserWarning(
"Soft timeouts are not supported: "
"on this platform: It does not have the SIGUSR1 signal.",
))
soft_timeout = None
if self._state == RUN:
waitforslot = self.putlocks if waitforslot is None else waitforslot
if waitforslot and self._putlock is not None:
self._putlock.acquire()
result = ApplyResult(
self._cache, callback, accept_callback, timeout_callback,
error_callback, soft_timeout, timeout, lost_worker_timeout,
on_timeout_set=self.on_timeout_set,
on_timeout_cancel=self.on_timeout_cancel,
callbacks_propagate=callbacks_propagate,
send_ack=self.send_ack if self.synack else None,
correlation_id=correlation_id,
)
if timeout or soft_timeout:
# start the timeout handler thread when required.
self._start_timeout_handler()
if self.threads:
self._taskqueue.put(([(TASK, (result._job, None,
func, args, kwds))], None))
else:
self._quick_put((TASK, (result._job, None, func, args, kwds)))
return result
def send_ack(self, response, job, i, fd):
pass
def terminate_job(self, pid, sig=None):
proc, _ = self._process_by_pid(pid)
if proc is not None:
try:
_kill(pid, sig or signal.SIGTERM)
except OSError as exc:
if get_errno(exc) != errno.ESRCH:
raise
else:
proc._controlled_termination = True
proc._job_terminated = True
def map_async(self, func, iterable, chunksize=None,
callback=None, error_callback=None):
'''
Asynchronous equivalent of `map()` method.
'''
return self._map_async(
func, iterable, mapstar, chunksize, callback, error_callback,
)
def _map_async(self, func, iterable, mapper, chunksize=None,
callback=None, error_callback=None):
'''
Helper function to implement map, starmap and their async counterparts.
'''
if self._state != RUN:
return
if not hasattr(iterable, '__len__'):
iterable = list(iterable)
if chunksize is None:
chunksize, extra = divmod(len(iterable), len(self._pool) * 4)
if extra:
chunksize += 1
if len(iterable) == 0:
chunksize = 0
task_batches = Pool._get_tasks(func, iterable, chunksize)
result = MapResult(self._cache, chunksize, len(iterable), callback,
error_callback=error_callback)
self._taskqueue.put((((TASK, (result._job, i, mapper, (x,), {}))
for i, x in enumerate(task_batches)), None))
return result
@staticmethod
def _get_tasks(func, it, size):
it = iter(it)
while 1:
x = tuple(itertools.islice(it, size))
if not x:
return
yield (func, x)
def __reduce__(self):
raise NotImplementedError(
'pool objects cannot be passed between processes or pickled',
)
def close(self):
debug('closing pool')
if self._state == RUN:
self._state = CLOSE
if self._putlock:
self._putlock.clear()
self._worker_handler.close()
self._taskqueue.put(None)
stop_if_not_current(self._worker_handler)
def terminate(self):
debug('terminating pool')
self._state = TERMINATE
self._worker_handler.terminate()
self._terminate()
@staticmethod
def _stop_task_handler(task_handler):
stop_if_not_current(task_handler)
def join(self):
assert self._state in (CLOSE, TERMINATE)
debug('joining worker handler')
stop_if_not_current(self._worker_handler)
debug('joining task handler')
self._stop_task_handler(self._task_handler)
debug('joining result handler')
stop_if_not_current(self._result_handler)
debug('result handler joined')
for i, p in enumerate(self._pool):
debug('joining worker %s/%s (%r)', i+1, len(self._pool), p)
if p._popen is not None: # process started?
p.join()
debug('pool join complete')
def restart(self):
for e in values(self._poolctrl):
e.set()
@staticmethod
def _help_stuff_finish(inqueue, task_handler, _pool):
# task_handler may be blocked trying to put items on inqueue
debug('removing tasks from inqueue until task handler finished')
inqueue._rlock.acquire()
while task_handler.is_alive() and inqueue._reader.poll():
inqueue._reader.recv()
time.sleep(0)
@classmethod
def _set_result_sentinel(cls, outqueue, pool):
outqueue.put(None)
@classmethod
def _terminate_pool(cls, taskqueue, inqueue, outqueue, pool,
worker_handler, task_handler,
result_handler, cache, timeout_handler,
help_stuff_finish_args):
# this is guaranteed to only be called once
debug('finalizing pool')
worker_handler.terminate()
task_handler.terminate()
taskqueue.put(None) # sentinel
debug('helping task handler/workers to finish')
cls._help_stuff_finish(*help_stuff_finish_args)
result_handler.terminate()
cls._set_result_sentinel(outqueue, pool)
if timeout_handler is not None:
timeout_handler.terminate()
# Terminate workers which haven't already finished
if pool and hasattr(pool[0], 'terminate'):
debug('terminating workers')
for p in pool:
if p._is_alive():
p.terminate()
debug('joining task handler')
cls._stop_task_handler(task_handler)
debug('joining result handler')
result_handler.stop()
if timeout_handler is not None:
debug('joining timeout handler')
timeout_handler.stop(TIMEOUT_MAX)
if pool and hasattr(pool[0], 'terminate'):
debug('joining pool workers')
for p in pool:
if p.is_alive():
# worker has not yet exited
debug('cleaning up worker %d', p.pid)
if p._popen is not None:
p.join()
debug('pool workers joined')
@property
def process_sentinels(self):
return [w._popen.sentinel for w in self._pool]
#
# Class whose instances are returned by `Pool.apply_async()`
#
class ApplyResult(object):
_worker_lost = None
_write_to = None
_scheduled_for = None
def __init__(self, cache, callback, accept_callback=None,
timeout_callback=None, error_callback=None, soft_timeout=None,
timeout=None, lost_worker_timeout=LOST_WORKER_TIMEOUT,
on_timeout_set=None, on_timeout_cancel=None,
callbacks_propagate=(), send_ack=None,
correlation_id=None):
self.correlation_id = correlation_id
self._mutex = Lock()
self._event = threading.Event()
self._job = next(job_counter)
self._cache = cache
self._callback = callback
self._accept_callback = accept_callback
self._error_callback = error_callback
self._timeout_callback = timeout_callback
self._timeout = timeout
self._soft_timeout = soft_timeout
self._lost_worker_timeout = lost_worker_timeout
self._on_timeout_set = on_timeout_set
self._on_timeout_cancel = on_timeout_cancel
self._callbacks_propagate = callbacks_propagate or ()
self._send_ack = send_ack
self._accepted = False
self._cancelled = False
self._worker_pid = None
self._time_accepted = None
self._terminated = None
cache[self._job] = self
def __repr__(self):
return '<Result: {id} ack:{ack} ready:{ready}>'.format(
id=self._job, ack=self._accepted, ready=self.ready(),
)
def ready(self):
return self._event.isSet()
def accepted(self):
return self._accepted
def successful(self):
assert self.ready()
return self._success
def _cancel(self):
"""Only works if synack is used."""
self._cancelled = True
def discard(self):
self._cache.pop(self._job, None)
def terminate(self, signum):
self._terminated = signum
def _set_terminated(self, signum=None):
try:
raise Terminated(-(signum or 0))
except Terminated:
self._set(None, (False, ExceptionInfo()))
def worker_pids(self):
return [self._worker_pid] if self._worker_pid else []
def wait(self, timeout=None):
self._event.wait(timeout)
def get(self, timeout=None):
self.wait(timeout)
if not self.ready():
raise TimeoutError
if self._success:
return self._value
else:
raise self._value.exception
def safe_apply_callback(self, fun, *args, **kwargs):
if fun:
try:
fun(*args, **kwargs)
except self._callbacks_propagate:
raise
except Exception as exc:
error('Pool callback raised exception: %r', exc,
exc_info=1)
def handle_timeout(self, soft=False):
if self._timeout_callback is not None:
self.safe_apply_callback(
self._timeout_callback, soft=soft,
timeout=self._soft_timeout if soft else self._timeout,
)
def _set(self, i, obj):
with self._mutex:
if self._on_timeout_cancel:
self._on_timeout_cancel(self)
self._success, self._value = obj
self._event.set()
if self._accepted:
# if not accepted yet, then the set message
# was received before the ack, which means
# the ack will remove the entry.
self._cache.pop(self._job, None)
# apply callbacks last
if self._callback and self._success:
self.safe_apply_callback(
self._callback, self._value)
if (self._value is not None and
self._error_callback and not self._success):
self.safe_apply_callback(
self._error_callback, self._value)
def _ack(self, i, time_accepted, pid, synqW_fd):
with self._mutex:
if self._cancelled and self._send_ack:
self._accepted = True
if synqW_fd:
return self._send_ack(NACK, pid, self._job, synqW_fd)
return
self._accepted = True
self._time_accepted = time_accepted
self._worker_pid = pid
if self.ready():
# ack received after set()
self._cache.pop(self._job, None)
if self._on_timeout_set:
self._on_timeout_set(self, self._soft_timeout, self._timeout)
response = ACK
if self._accept_callback:
try:
self._accept_callback(pid, time_accepted)
except self._propagate_errors:
response = NACK
raise
except Exception:
response = NACK
# ignore other errors
finally:
if self._send_ack and synqW_fd:
return self._send_ack(
response, pid, self._job, synqW_fd
)
if self._send_ack and synqW_fd:
self._send_ack(response, pid, self._job, synqW_fd)
#
# Class whose instances are returned by `Pool.map_async()`
#
class MapResult(ApplyResult):
def __init__(self, cache, chunksize, length, callback, error_callback):
ApplyResult.__init__(
self, cache, callback, error_callback=error_callback,
)
self._success = True
self._length = length
self._value = [None] * length
self._accepted = [False] * length
self._worker_pid = [None] * length
self._time_accepted = [None] * length
self._chunksize = chunksize
if chunksize <= 0:
self._number_left = 0
self._event.set()
del cache[self._job]
else:
self._number_left = length // chunksize + bool(length % chunksize)
def _set(self, i, success_result):
success, result = success_result
if success:
self._value[i * self._chunksize:(i + 1) * self._chunksize] = result
self._number_left -= 1
if self._number_left == 0:
if self._callback:
self._callback(self._value)
if self._accepted:
self._cache.pop(self._job, None)
self._event.set()
else:
self._success = False
self._value = result
if self._error_callback:
self._error_callback(self._value)
if self._accepted:
self._cache.pop(self._job, None)
self._event.set()
def _ack(self, i, time_accepted, pid, *args):
start = i * self._chunksize
stop = min((i + 1) * self._chunksize, self._length)
for j in range(start, stop):
self._accepted[j] = True
self._worker_pid[j] = pid
self._time_accepted[j] = time_accepted
if self.ready():
self._cache.pop(self._job, None)
def accepted(self):
return all(self._accepted)
def worker_pids(self):
return [pid for pid in self._worker_pid if pid]
#
# Class whose instances are returned by `Pool.imap()`
#
class IMapIterator(object):
_worker_lost = None
def __init__(self, cache, lost_worker_timeout=LOST_WORKER_TIMEOUT):
self._cond = threading.Condition(threading.Lock())
self._job = next(job_counter)
self._cache = cache
self._items = deque()
self._index = 0
self._length = None
self._ready = False
self._unsorted = {}
self._worker_pids = []
self._lost_worker_timeout = lost_worker_timeout
cache[self._job] = self
def __iter__(self):
return self
def next(self, timeout=None):
with self._cond:
try:
item = self._items.popleft()
except IndexError:
if self._index == self._length:
self._ready = True
raise StopIteration
self._cond.wait(timeout)
try:
item = self._items.popleft()
except IndexError:
if self._index == self._length:
self._ready = True
raise StopIteration
raise TimeoutError
success, value = item
if success:
return value
raise Exception(value)
__next__ = next # XXX
def _set(self, i, obj):
with self._cond:
if self._index == i:
self._items.append(obj)
self._index += 1
while self._index in self._unsorted:
obj = self._unsorted.pop(self._index)
self._items.append(obj)
self._index += 1
self._cond.notify()
else:
self._unsorted[i] = obj
if self._index == self._length:
self._ready = True
del self._cache[self._job]
def _set_length(self, length):
with self._cond:
self._length = length
if self._index == self._length:
self._ready = True
self._cond.notify()
del self._cache[self._job]
def _ack(self, i, time_accepted, pid, *args):
self._worker_pids.append(pid)
def ready(self):
return self._ready
def worker_pids(self):
return self._worker_pids
#
# Class whose instances are returned by `Pool.imap_unordered()`
#
class IMapUnorderedIterator(IMapIterator):
def _set(self, i, obj):
with self._cond:
self._items.append(obj)
self._index += 1
self._cond.notify()
if self._index == self._length:
self._ready = True
del self._cache[self._job]
#
#
#
class ThreadPool(Pool):
from billiard.dummy import Process as DummyProcess
Process = DummyProcess
def __init__(self, processes=None, initializer=None, initargs=()):
Pool.__init__(self, processes, initializer, initargs)
def _setup_queues(self):
self._inqueue = Queue()
self._outqueue = Queue()
self._quick_put = self._inqueue.put
self._quick_get = self._outqueue.get
def _poll_result(timeout):
try:
return True, self._quick_get(timeout=timeout)
except Empty:
return False, None
self._poll_result = _poll_result
@staticmethod
def _help_stuff_finish(inqueue, task_handler, pool):
# put sentinels at head of inqueue to make workers finish
with inqueue.not_empty:
inqueue.queue.clear()
inqueue.queue.extend([None] * len(pool))
inqueue.not_empty.notify_all()
|