/usr/lib/python2.7/dist-packages/cherrypy/wsgiserver/__init__.py is in python-cherrypy3 8.9.1-2.
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 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 2031 2032 2033 2034 2035 2036 2037 2038 2039 2040 2041 2042 2043 2044 2045 2046 2047 2048 2049 2050 2051 2052 2053 2054 2055 2056 2057 2058 2059 2060 2061 2062 2063 2064 2065 2066 2067 2068 2069 2070 2071 2072 2073 2074 2075 2076 2077 2078 2079 2080 2081 2082 2083 2084 2085 2086 2087 2088 2089 2090 2091 2092 2093 2094 2095 2096 2097 2098 2099 2100 2101 2102 2103 2104 2105 2106 2107 2108 2109 2110 2111 2112 2113 2114 2115 2116 2117 2118 2119 2120 2121 2122 2123 2124 2125 2126 2127 2128 2129 2130 2131 2132 2133 2134 2135 2136 2137 2138 2139 2140 2141 2142 2143 2144 2145 2146 2147 2148 2149 2150 2151 2152 2153 2154 2155 2156 2157 2158 2159 2160 2161 2162 2163 2164 2165 2166 2167 2168 2169 2170 2171 2172 2173 2174 2175 2176 2177 2178 2179 2180 2181 2182 2183 2184 2185 2186 2187 2188 2189 2190 2191 2192 2193 2194 2195 2196 2197 2198 2199 2200 2201 2202 2203 2204 2205 2206 2207 2208 2209 2210 2211 2212 2213 2214 2215 2216 2217 2218 2219 2220 2221 2222 2223 2224 2225 2226 2227 2228 2229 2230 2231 2232 2233 2234 2235 2236 2237 2238 2239 2240 2241 2242 2243 2244 2245 2246 2247 2248 2249 2250 2251 2252 2253 2254 2255 2256 2257 2258 2259 2260 2261 2262 2263 2264 2265 2266 2267 2268 2269 2270 2271 2272 2273 2274 2275 2276 2277 2278 2279 2280 2281 2282 2283 2284 2285 2286 2287 2288 2289 2290 2291 2292 2293 2294 2295 2296 2297 2298 2299 2300 2301 2302 2303 2304 2305 2306 2307 2308 2309 2310 2311 2312 2313 2314 2315 2316 2317 2318 2319 2320 2321 2322 2323 2324 2325 2326 2327 2328 2329 2330 2331 2332 2333 2334 2335 2336 2337 2338 2339 2340 2341 2342 2343 2344 2345 2346 2347 2348 2349 2350 2351 2352 2353 2354 2355 2356 2357 2358 2359 2360 2361 2362 2363 2364 2365 2366 2367 2368 2369 2370 2371 2372 2373 2374 2375 2376 2377 2378 2379 2380 2381 2382 2383 2384 2385 2386 2387 2388 2389 2390 2391 2392 2393 2394 2395 2396 2397 2398 2399 2400 2401 2402 2403 2404 2405 2406 2407 2408 2409 2410 2411 2412 2413 2414 2415 2416 2417 2418 2419 2420 2421 2422 2423 2424 2425 2426 2427 2428 2429 2430 2431 2432 2433 2434 2435 2436 2437 2438 2439 2440 2441 2442 2443 2444 2445 2446 2447 2448 2449 2450 2451 2452 2453 2454 2455 2456 2457 2458 2459 2460 2461 2462 2463 2464 2465 2466 2467 2468 2469 2470 2471 2472 2473 2474 2475 2476 2477 2478 2479 2480 2481 2482 2483 2484 2485 2486 2487 2488 2489 2490 2491 2492 2493 2494 2495 2496 2497 2498 2499 2500 2501 2502 2503 2504 2505 2506 2507 2508 2509 2510 2511 2512 2513 2514 2515 2516 2517 2518 2519 2520 2521 2522 2523 2524 2525 2526 2527 2528 2529 2530 2531 2532 2533 2534 2535 2536 2537 2538 2539 2540 2541 2542 2543 2544 2545 2546 2547 2548 2549 2550 2551 2552 2553 2554 2555 2556 2557 2558 2559 2560 2561 2562 2563 2564 2565 2566 2567 2568 2569 2570 2571 2572 2573 2574 2575 | """A high-speed, production ready, thread pooled, generic HTTP server.
Simplest example on how to use this module directly
(without using CherryPy's application machinery)::
from cherrypy import wsgiserver
def my_crazy_app(environ, start_response):
status = '200 OK'
response_headers = [('Content-type','text/plain')]
start_response(status, response_headers)
return ['Hello world!']
server = wsgiserver.CherryPyWSGIServer(
('0.0.0.0', 8070), my_crazy_app,
server_name='www.cherrypy.example')
server.start()
The CherryPy WSGI server can serve as many WSGI applications
as you want in one instance by using a WSGIPathInfoDispatcher::
d = WSGIPathInfoDispatcher({'/': my_crazy_app, '/blog': my_blog_app})
server = wsgiserver.CherryPyWSGIServer(('0.0.0.0', 80), d)
Want SSL support? Just set server.ssl_adapter to an SSLAdapter instance.
This won't call the CherryPy engine (application side) at all, only the
HTTP server, which is independent from the rest of CherryPy. Don't
let the name "CherryPyWSGIServer" throw you; the name merely reflects
its origin, not its coupling.
For those of you wanting to understand internals of this module, here's the
basic call flow. The server's listening thread runs a very tight loop,
sticking incoming connections onto a Queue::
server = CherryPyWSGIServer(...)
server.start()
while True:
tick()
# This blocks until a request comes in:
child = socket.accept()
conn = HTTPConnection(child, ...)
server.requests.put(conn)
Worker threads are kept in a pool and poll the Queue, popping off and then
handling each connection in turn. Each connection can consist of an arbitrary
number of requests and their responses, so we run a nested loop::
while True:
conn = server.requests.get()
conn.communicate()
-> while True:
req = HTTPRequest(...)
req.parse_request()
-> # Read the Request-Line, e.g. "GET /page HTTP/1.1"
req.rfile.readline()
read_headers(req.rfile, req.inheaders)
req.respond()
-> response = app(...)
try:
for chunk in response:
if chunk:
req.write(chunk)
finally:
if hasattr(response, "close"):
response.close()
if req.close_connection:
return
"""
import os
try:
import queue
except:
import Queue as queue
import re
import email.utils
import socket
import sys
import threading
import time
import traceback as traceback_
try:
from urllib.parse import unquote_to_bytes, urlparse
except ImportError:
from urlparse import unquote as unquote_to_bytes
from urlparse import urlparse
import errno
import logging
import six
from six.moves import filter
try:
# prefer slower Python-based io module
import _pyio as io
except ImportError:
# Python 2.6
import io
try:
import pkg_resources
except ImportError:
pass
__all__ = ['HTTPRequest', 'HTTPConnection', 'HTTPServer',
'SizeCheckWrapper', 'KnownLengthRFile', 'ChunkedRFile',
'CP_makefile',
'MaxSizeExceeded', 'NoSSLError', 'FatalSSLAlert',
'WorkerThread', 'ThreadPool', 'SSLAdapter',
'CherryPyWSGIServer',
'Gateway', 'WSGIGateway', 'WSGIGateway_10', 'WSGIGateway_u0',
'WSGIPathInfoDispatcher', 'get_ssl_adapter_class',
'socket_errors_to_ignore']
if 'win' in sys.platform and hasattr(socket, 'AF_INET6'):
if not hasattr(socket, 'IPPROTO_IPV6'):
socket.IPPROTO_IPV6 = 41
if not hasattr(socket, 'IPV6_V6ONLY'):
socket.IPV6_V6ONLY = 27
DEFAULT_BUFFER_SIZE = io.DEFAULT_BUFFER_SIZE
try:
cp_version = pkg_resources.require('cherrypy')[0].version
except Exception:
cp_version = 'unknown'
if six.PY3:
def ntob(n, encoding='ISO-8859-1'):
"""Return the given native string as a byte string in the given
encoding.
"""
# In Python 3, the native string type is unicode
return n.encode(encoding)
def bton(b, encoding='ISO-8859-1'):
return b.decode(encoding)
else:
def ntob(n, encoding='ISO-8859-1'):
"""Return the given native string as a byte string in the given
encoding.
"""
# In Python 2, the native string type is bytes. Assume it's already
# in the given encoding, which for ISO-8859-1 is almost always what
# was intended.
return n
def bton(b, encoding='ISO-8859-1'):
return b
LF = b'\n'
CRLF = b'\r\n'
TAB = b'\t'
SPACE = b' '
COLON = b':'
SEMICOLON = b';'
EMPTY = b''
NUMBER_SIGN = b'#'
QUESTION_MARK = b'?'
ASTERISK = b'*'
FORWARD_SLASH = b'/'
quoted_slash = re.compile(b'(?i)%2F')
def plat_specific_errors(*errnames):
"""Return error numbers for all errors in errnames on this platform.
The 'errno' module contains different global constants depending on
the specific platform (OS). This function will return the list of
numeric values for a given list of potential names.
"""
errno_names = dir(errno)
nums = [getattr(errno, k) for k in errnames if k in errno_names]
# de-dupe the list
return list(dict.fromkeys(nums).keys())
socket_error_eintr = plat_specific_errors('EINTR', 'WSAEINTR')
socket_errors_to_ignore = plat_specific_errors(
'EPIPE',
'EBADF', 'WSAEBADF',
'ENOTSOCK', 'WSAENOTSOCK',
'ETIMEDOUT', 'WSAETIMEDOUT',
'ECONNREFUSED', 'WSAECONNREFUSED',
'ECONNRESET', 'WSAECONNRESET',
'ECONNABORTED', 'WSAECONNABORTED',
'ENETRESET', 'WSAENETRESET',
'EHOSTDOWN', 'EHOSTUNREACH',
)
socket_errors_to_ignore.append('timed out')
socket_errors_to_ignore.append('The read operation timed out')
if sys.platform == 'darwin':
socket_errors_to_ignore.append(plat_specific_errors('EPROTOTYPE'))
socket_errors_nonblocking = plat_specific_errors(
'EAGAIN', 'EWOULDBLOCK', 'WSAEWOULDBLOCK')
comma_separated_headers = [
ntob(h) for h in
['Accept', 'Accept-Charset', 'Accept-Encoding',
'Accept-Language', 'Accept-Ranges', 'Allow', 'Cache-Control',
'Connection', 'Content-Encoding', 'Content-Language', 'Expect',
'If-Match', 'If-None-Match', 'Pragma', 'Proxy-Authenticate', 'TE',
'Trailer', 'Transfer-Encoding', 'Upgrade', 'Vary', 'Via', 'Warning',
'WWW-Authenticate']
]
if not hasattr(logging, 'statistics'):
logging.statistics = {}
def read_headers(rfile, hdict=None):
"""Read headers from the given stream into the given header dict.
If hdict is None, a new header dict is created. Returns the populated
header dict.
Headers which are repeated are folded together using a comma if their
specification so dictates.
This function raises ValueError when the read bytes violate the HTTP spec.
You should probably return "400 Bad Request" if this happens.
"""
if hdict is None:
hdict = {}
while True:
line = rfile.readline()
if not line:
# No more data--illegal end of headers
raise ValueError('Illegal end of headers.')
if line == CRLF:
# Normal end of headers
break
if not line.endswith(CRLF):
raise ValueError('HTTP requires CRLF terminators')
if line[0] in (SPACE, TAB):
# It's a continuation line.
v = line.strip()
else:
try:
k, v = line.split(COLON, 1)
except ValueError:
raise ValueError('Illegal header line.')
# TODO: what about TE and WWW-Authenticate?
k = k.strip().title()
v = v.strip()
hname = k
if k in comma_separated_headers:
existing = hdict.get(hname)
if existing:
v = b', '.join((existing, v))
hdict[hname] = v
return hdict
class MaxSizeExceeded(Exception):
pass
class SizeCheckWrapper(object):
"""Wraps a file-like object, raising MaxSizeExceeded if too large."""
def __init__(self, rfile, maxlen):
self.rfile = rfile
self.maxlen = maxlen
self.bytes_read = 0
def _check_length(self):
if self.maxlen and self.bytes_read > self.maxlen:
raise MaxSizeExceeded()
def read(self, size=None):
data = self.rfile.read(size)
self.bytes_read += len(data)
self._check_length()
return data
def readline(self, size=None):
if size is not None:
data = self.rfile.readline(size)
self.bytes_read += len(data)
self._check_length()
return data
# User didn't specify a size ...
# We read the line in chunks to make sure it's not a 100MB line !
res = []
while True:
data = self.rfile.readline(256)
self.bytes_read += len(data)
self._check_length()
res.append(data)
# See https://github.com/cherrypy/cherrypy/issues/421
if len(data) < 256 or data[-1:] == LF:
return EMPTY.join(res)
def readlines(self, sizehint=0):
# Shamelessly stolen from StringIO
total = 0
lines = []
line = self.readline()
while line:
lines.append(line)
total += len(line)
if 0 < sizehint <= total:
break
line = self.readline()
return lines
def close(self):
self.rfile.close()
def __iter__(self):
return self
def __next__(self):
data = next(self.rfile)
self.bytes_read += len(data)
self._check_length()
return data
def next(self):
data = self.rfile.next()
self.bytes_read += len(data)
self._check_length()
return data
class KnownLengthRFile(object):
"""Wraps a file-like object, returning an empty string when exhausted."""
def __init__(self, rfile, content_length):
self.rfile = rfile
self.remaining = content_length
def read(self, size=None):
if self.remaining == 0:
return b''
if size is None:
size = self.remaining
else:
size = min(size, self.remaining)
data = self.rfile.read(size)
self.remaining -= len(data)
return data
def readline(self, size=None):
if self.remaining == 0:
return b''
if size is None:
size = self.remaining
else:
size = min(size, self.remaining)
data = self.rfile.readline(size)
self.remaining -= len(data)
return data
def readlines(self, sizehint=0):
# Shamelessly stolen from StringIO
total = 0
lines = []
line = self.readline(sizehint)
while line:
lines.append(line)
total += len(line)
if 0 < sizehint <= total:
break
line = self.readline(sizehint)
return lines
def close(self):
self.rfile.close()
def __iter__(self):
return self
def __next__(self):
data = next(self.rfile)
self.remaining -= len(data)
return data
class ChunkedRFile(object):
"""Wraps a file-like object, returning an empty string when exhausted.
This class is intended to provide a conforming wsgi.input value for
request entities that have been encoded with the 'chunked' transfer
encoding.
"""
def __init__(self, rfile, maxlen, bufsize=8192):
self.rfile = rfile
self.maxlen = maxlen
self.bytes_read = 0
self.buffer = EMPTY
self.bufsize = bufsize
self.closed = False
def _fetch(self):
if self.closed:
return
line = self.rfile.readline()
self.bytes_read += len(line)
if self.maxlen and self.bytes_read > self.maxlen:
raise MaxSizeExceeded('Request Entity Too Large', self.maxlen)
line = line.strip().split(SEMICOLON, 1)
try:
chunk_size = line.pop(0)
chunk_size = int(chunk_size, 16)
except ValueError:
raise ValueError('Bad chunked transfer size: ' + repr(chunk_size))
if chunk_size <= 0:
self.closed = True
return
## if line: chunk_extension = line[0]
if self.maxlen and self.bytes_read + chunk_size > self.maxlen:
raise IOError('Request Entity Too Large')
chunk = self.rfile.read(chunk_size)
self.bytes_read += len(chunk)
self.buffer += chunk
crlf = self.rfile.read(2)
if crlf != CRLF:
raise ValueError(
"Bad chunked transfer coding (expected '\\r\\n', "
'got ' + repr(crlf) + ')')
def read(self, size=None):
data = EMPTY
while True:
if size and len(data) >= size:
return data
if not self.buffer:
self._fetch()
if not self.buffer:
# EOF
return data
if size:
remaining = size - len(data)
data += self.buffer[:remaining]
self.buffer = self.buffer[remaining:]
else:
data += self.buffer
def readline(self, size=None):
data = EMPTY
while True:
if size and len(data) >= size:
return data
if not self.buffer:
self._fetch()
if not self.buffer:
# EOF
return data
newline_pos = self.buffer.find(LF)
if size:
if newline_pos == -1:
remaining = size - len(data)
data += self.buffer[:remaining]
self.buffer = self.buffer[remaining:]
else:
remaining = min(size - len(data), newline_pos)
data += self.buffer[:remaining]
self.buffer = self.buffer[remaining:]
else:
if newline_pos == -1:
data += self.buffer
else:
data += self.buffer[:newline_pos]
self.buffer = self.buffer[newline_pos:]
def readlines(self, sizehint=0):
# Shamelessly stolen from StringIO
total = 0
lines = []
line = self.readline(sizehint)
while line:
lines.append(line)
total += len(line)
if 0 < sizehint <= total:
break
line = self.readline(sizehint)
return lines
def read_trailer_lines(self):
if not self.closed:
raise ValueError(
'Cannot read trailers until the request body has been read.')
while True:
line = self.rfile.readline()
if not line:
# No more data--illegal end of headers
raise ValueError('Illegal end of headers.')
self.bytes_read += len(line)
if self.maxlen and self.bytes_read > self.maxlen:
raise IOError('Request Entity Too Large')
if line == CRLF:
# Normal end of headers
break
if not line.endswith(CRLF):
raise ValueError('HTTP requires CRLF terminators')
yield line
def close(self):
self.rfile.close()
class HTTPRequest(object):
"""An HTTP Request (and response).
A single HTTP connection may consist of multiple request/response pairs.
"""
server = None
"""The HTTPServer object which is receiving this request."""
conn = None
"""The HTTPConnection object on which this request connected."""
inheaders = {}
"""A dict of request headers."""
outheaders = []
"""A list of header tuples to write in the response."""
ready = False
"""When True, the request has been parsed and is ready to begin generating
the response. When False, signals the calling Connection that the response
should not be generated and the connection should close."""
close_connection = False
"""Signals the calling Connection that the request should close. This does
not imply an error! The client and/or server may each request that the
connection be closed."""
chunked_write = False
"""If True, output will be encoded with the "chunked" transfer-coding.
This value is set automatically inside send_headers."""
def __init__(self, server, conn):
self.server = server
self.conn = conn
self.ready = False
self.started_request = False
self.scheme = b'http'
if self.server.ssl_adapter is not None:
self.scheme = b'https'
# Use the lowest-common protocol in case read_request_line errors.
self.response_protocol = 'HTTP/1.0'
self.inheaders = {}
self.status = ''
self.outheaders = []
self.sent_headers = False
self.close_connection = self.__class__.close_connection
self.chunked_read = False
self.chunked_write = self.__class__.chunked_write
def parse_request(self):
"""Parse the next HTTP request start-line and message-headers."""
self.rfile = SizeCheckWrapper(self.conn.rfile,
self.server.max_request_header_size)
try:
success = self.read_request_line()
except MaxSizeExceeded:
self.simple_response(
'414 Request-URI Too Long',
'The Request-URI sent with the request exceeds the maximum '
'allowed bytes.')
return
else:
if not success:
return
try:
success = self.read_request_headers()
except MaxSizeExceeded:
self.simple_response(
'413 Request Entity Too Large',
'The headers sent with the request exceed the maximum '
'allowed bytes.')
return
else:
if not success:
return
self.ready = True
def read_request_line(self):
# HTTP/1.1 connections are persistent by default. If a client
# requests a page, then idles (leaves the connection open),
# then rfile.readline() will raise socket.error("timed out").
# Note that it does this based on the value given to settimeout(),
# and doesn't need the client to request or acknowledge the close
# (although your TCP stack might suffer for it: cf Apache's history
# with FIN_WAIT_2).
request_line = self.rfile.readline()
# Set started_request to True so communicate() knows to send 408
# from here on out.
self.started_request = True
if not request_line:
return False
if request_line == CRLF:
# RFC 2616 sec 4.1: "...if the server is reading the protocol
# stream at the beginning of a message and receives a CRLF
# first, it should ignore the CRLF."
# But only ignore one leading line! else we enable a DoS.
request_line = self.rfile.readline()
if not request_line:
return False
if not request_line.endswith(CRLF):
self.simple_response(
'400 Bad Request', 'HTTP requires CRLF terminators')
return False
try:
method, uri, req_protocol = request_line.strip().split(SPACE, 2)
req_protocol_str = req_protocol.decode('ascii')
rp = int(req_protocol_str[5]), int(req_protocol_str[7])
except (ValueError, IndexError):
self.simple_response('400 Bad Request', 'Malformed Request-Line')
return False
self.uri = uri
self.method = method
# uri may be an abs_path (including "http://host.domain.tld");
scheme, authority, path = self.parse_request_uri(uri)
if path is None:
self.simple_response('400 Bad Request',
'Invalid path in Request-URI.')
return False
if NUMBER_SIGN in path:
self.simple_response('400 Bad Request',
'Illegal #fragment in Request-URI.')
return False
if scheme:
self.scheme = scheme
qs = EMPTY
if QUESTION_MARK in path:
path, qs = path.split(QUESTION_MARK, 1)
# Unquote the path+params (e.g. "/this%20path" -> "/this path").
# http://www.w3.org/Protocols/rfc2616/rfc2616-sec5.html#sec5.1.2
#
# But note that "...a URI must be separated into its components
# before the escaped characters within those components can be
# safely decoded." http://www.ietf.org/rfc/rfc2396.txt, sec 2.4.2
# Therefore, "/this%2Fpath" becomes "/this%2Fpath", not "/this/path".
try:
atoms = [unquote_to_bytes(x) for x in quoted_slash.split(path)]
except ValueError:
ex = sys.exc_info()[1]
self.simple_response('400 Bad Request', ex.args[0])
return False
path = b'%2F'.join(atoms)
self.path = path
# Note that, like wsgiref and most other HTTP servers,
# we "% HEX HEX"-unquote the path but not the query string.
self.qs = qs
# Compare request and server HTTP protocol versions, in case our
# server does not support the requested protocol. Limit our output
# to min(req, server). We want the following output:
# request server actual written supported response
# protocol protocol response protocol feature set
# a 1.0 1.0 1.0 1.0
# b 1.0 1.1 1.1 1.0
# c 1.1 1.0 1.0 1.0
# d 1.1 1.1 1.1 1.1
# Notice that, in (b), the response will be "HTTP/1.1" even though
# the client only understands 1.0. RFC 2616 10.5.6 says we should
# only return 505 if the _major_ version is different.
sp = int(self.server.protocol[5]), int(self.server.protocol[7])
if sp[0] != rp[0]:
self.simple_response('505 HTTP Version Not Supported')
return False
self.request_protocol = req_protocol
self.response_protocol = 'HTTP/%s.%s' % min(rp, sp)
return True
def read_request_headers(self):
"""Read self.rfile into self.inheaders. Return success."""
# then all the http headers
try:
read_headers(self.rfile, self.inheaders)
except ValueError:
ex = sys.exc_info()[1]
self.simple_response('400 Bad Request', ex.args[0])
return False
mrbs = self.server.max_request_body_size
if mrbs and int(self.inheaders.get(b'Content-Length', 0)) > mrbs:
self.simple_response(
'413 Request Entity Too Large',
'The entity sent with the request exceeds the maximum '
'allowed bytes.')
return False
# Persistent connection support
if self.response_protocol == 'HTTP/1.1':
# Both server and client are HTTP/1.1
if self.inheaders.get(b'Connection', b'') == b'close':
self.close_connection = True
else:
# Either the server or client (or both) are HTTP/1.0
if self.inheaders.get(b'Connection', b'') != b'Keep-Alive':
self.close_connection = True
# Transfer-Encoding support
te = None
if self.response_protocol == 'HTTP/1.1':
te = self.inheaders.get(b'Transfer-Encoding')
if te:
te = [x.strip().lower() for x in te.split(b',') if x.strip()]
self.chunked_read = False
if te:
for enc in te:
if enc == b'chunked':
self.chunked_read = True
else:
# Note that, even if we see "chunked", we must reject
# if there is an extension we don't recognize.
self.simple_response('501 Unimplemented')
self.close_connection = True
return False
# From PEP 333:
# "Servers and gateways that implement HTTP 1.1 must provide
# transparent support for HTTP 1.1's "expect/continue" mechanism.
# This may be done in any of several ways:
# 1. Respond to requests containing an Expect: 100-continue request
# with an immediate "100 Continue" response, and proceed normally.
# 2. Proceed with the request normally, but provide the application
# with a wsgi.input stream that will send the "100 Continue"
# response if/when the application first attempts to read from
# the input stream. The read request must then remain blocked
# until the client responds.
# 3. Wait until the client decides that the server does not support
# expect/continue, and sends the request body on its own.
# (This is suboptimal, and is not recommended.)
#
# We used to do 3, but are now doing 1. Maybe we'll do 2 someday,
# but it seems like it would be a big slowdown for such a rare case.
if self.inheaders.get(b'Expect', b'') == b'100-continue':
# Don't use simple_response here, because it emits headers
# we don't want. See
# https://github.com/cherrypy/cherrypy/issues/951
msg = self.server.protocol.encode('ascii')
msg += b' 100 Continue\r\n\r\n'
try:
self.conn.wfile.write(msg)
except socket.error:
x = sys.exc_info()[1]
if x.args[0] not in socket_errors_to_ignore:
raise
return True
def parse_request_uri(self, uri):
"""Parse a Request-URI into (scheme, authority, path).
Note that Request-URI's must be one of::
Request-URI = "*" | absoluteURI | abs_path | authority
Therefore, a Request-URI which starts with a double forward-slash
cannot be a "net_path"::
net_path = "//" authority [ abs_path ]
Instead, it must be interpreted as an "abs_path" with an empty first
path segment::
abs_path = "/" path_segments
path_segments = segment *( "/" segment )
segment = *pchar *( ";" param )
param = *pchar
"""
if uri == ASTERISK:
return None, None, uri
scheme, authority, path, params, query, fragment = urlparse(uri)
if scheme and QUESTION_MARK not in scheme:
# An absoluteURI.
# If there's a scheme (and it must be http or https), then:
# http_URL = "http:" "//" host [ ":" port ] [ abs_path [ "?" query
# ]]
return scheme, authority, path
if uri.startswith(FORWARD_SLASH):
# An abs_path.
return None, None, uri
else:
# An authority.
return None, uri, None
def respond(self):
"""Call the gateway and write its iterable output."""
mrbs = self.server.max_request_body_size
if self.chunked_read:
self.rfile = ChunkedRFile(self.conn.rfile, mrbs)
else:
cl = int(self.inheaders.get(b'Content-Length', 0))
if mrbs and mrbs < cl:
if not self.sent_headers:
self.simple_response(
'413 Request Entity Too Large',
'The entity sent with the request exceeds the '
'maximum allowed bytes.')
return
self.rfile = KnownLengthRFile(self.conn.rfile, cl)
self.server.gateway(self).respond()
if (self.ready and not self.sent_headers):
self.sent_headers = True
self.send_headers()
if self.chunked_write:
self.conn.wfile.write(b'0\r\n\r\n')
def simple_response(self, status, msg=''):
"""Write a simple response back to the client."""
status = str(status)
proto_status = '%s %s\r\n' % (self.server.protocol, status)
content_length = 'Content-Length: %s\r\n' % len(msg)
content_type = 'Content-Type: text/plain\r\n'
buf = [
proto_status.encode('ISO-8859-1'),
content_length.encode('ISO-8859-1'),
content_type.encode('ISO-8859-1'),
]
if status[:3] in ('413', '414'):
# Request Entity Too Large / Request-URI Too Long
self.close_connection = True
if self.response_protocol == 'HTTP/1.1':
# This will not be true for 414, since read_request_line
# usually raises 414 before reading the whole line, and we
# therefore cannot know the proper response_protocol.
buf.append(b'Connection: close\r\n')
else:
# HTTP/1.0 had no 413/414 status nor Connection header.
# Emit 400 instead and trust the message body is enough.
status = '400 Bad Request'
buf.append(CRLF)
if msg:
if isinstance(msg, six.text_type):
msg = msg.encode('ISO-8859-1')
buf.append(msg)
try:
self.conn.wfile.write(EMPTY.join(buf))
except socket.error:
x = sys.exc_info()[1]
if x.args[0] not in socket_errors_to_ignore:
raise
def write(self, chunk):
"""Write unbuffered data to the client."""
if self.chunked_write and chunk:
chunk_size_hex = hex(len(chunk))[2:].encode('ascii')
buf = [chunk_size_hex, CRLF, chunk, CRLF]
self.conn.wfile.write(EMPTY.join(buf))
else:
self.conn.wfile.write(chunk)
def send_headers(self):
"""Assert, process, and send the HTTP response message-headers.
You must set self.status, and self.outheaders before calling this.
"""
hkeys = [key.lower() for key, value in self.outheaders]
status = int(self.status[:3])
if status == 413:
# Request Entity Too Large. Close conn to avoid garbage.
self.close_connection = True
elif b'content-length' not in hkeys:
# "All 1xx (informational), 204 (no content),
# and 304 (not modified) responses MUST NOT
# include a message-body." So no point chunking.
if status < 200 or status in (204, 205, 304):
pass
else:
if (self.response_protocol == 'HTTP/1.1'
and self.method != b'HEAD'):
# Use the chunked transfer-coding
self.chunked_write = True
self.outheaders.append((b'Transfer-Encoding', b'chunked'))
else:
# Closing the conn is the only way to determine len.
self.close_connection = True
if b'connection' not in hkeys:
if self.response_protocol == 'HTTP/1.1':
# Both server and client are HTTP/1.1 or better
if self.close_connection:
self.outheaders.append((b'Connection', b'close'))
else:
# Server and/or client are HTTP/1.0
if not self.close_connection:
self.outheaders.append((b'Connection', b'Keep-Alive'))
if (not self.close_connection) and (not self.chunked_read):
# Read any remaining request body data on the socket.
# "If an origin server receives a request that does not include an
# Expect request-header field with the "100-continue" expectation,
# the request includes a request body, and the server responds
# with a final status code before reading the entire request body
# from the transport connection, then the server SHOULD NOT close
# the transport connection until it has read the entire request,
# or until the client closes the connection. Otherwise, the client
# might not reliably receive the response message. However, this
# requirement is not be construed as preventing a server from
# defending itself against denial-of-service attacks, or from
# badly broken client implementations."
remaining = getattr(self.rfile, 'remaining', 0)
if remaining > 0:
self.rfile.read(remaining)
if b'date' not in hkeys:
self.outheaders.append((
b'Date',
email.utils.formatdate(usegmt=True).encode('ISO-8859-1'),
))
if b'server' not in hkeys:
self.outheaders.append((
b'Server',
self.server.server_name.encode('ISO-8859-1'),
))
proto = self.server.protocol.encode('ascii')
buf = [proto + SPACE + self.status + CRLF]
for k, v in self.outheaders:
buf.append(k + COLON + SPACE + v + CRLF)
buf.append(CRLF)
self.conn.wfile.write(EMPTY.join(buf))
class NoSSLError(Exception):
"""Exception raised when a client speaks HTTP to an HTTPS socket."""
pass
class FatalSSLAlert(Exception):
"""Exception raised when the SSL implementation signals a fatal alert."""
pass
class CP_BufferedWriter(io.BufferedWriter):
"""Faux file object attached to a socket object."""
def write(self, b):
self._checkClosed()
if isinstance(b, str):
raise TypeError("can't write str to binary stream")
with self._write_lock:
self._write_buf.extend(b)
self._flush_unlocked()
return len(b)
def _flush_unlocked(self):
self._checkClosed('flush of closed file')
while self._write_buf:
try:
# ssl sockets only except 'bytes', not bytearrays
# so perhaps we should conditionally wrap this for perf?
n = self.raw.write(bytes(self._write_buf))
except io.BlockingIOError as e:
n = e.characters_written
del self._write_buf[:n]
def CP_makefile_PY3(sock, mode='r', bufsize=DEFAULT_BUFFER_SIZE):
if 'r' in mode:
return io.BufferedReader(socket.SocketIO(sock, mode), bufsize)
else:
return CP_BufferedWriter(socket.SocketIO(sock, mode), bufsize)
class CP_makefile_PY2(getattr(socket, '_fileobject', object)):
"""Faux file object attached to a socket object."""
def __init__(self, *args, **kwargs):
self.bytes_read = 0
self.bytes_written = 0
socket._fileobject.__init__(self, *args, **kwargs)
def write(self, data):
"""Sendall for non-blocking sockets."""
while data:
try:
bytes_sent = self.send(data)
data = data[bytes_sent:]
except socket.error as e:
if e.args[0] not in socket_errors_nonblocking:
raise
def send(self, data):
bytes_sent = self._sock.send(data)
self.bytes_written += bytes_sent
return bytes_sent
def flush(self):
if self._wbuf:
buffer = ''.join(self._wbuf)
self._wbuf = []
self.write(buffer)
def recv(self, size):
while True:
try:
data = self._sock.recv(size)
self.bytes_read += len(data)
return data
except socket.error as e:
if (e.args[0] not in socket_errors_nonblocking
and e.args[0] not in socket_error_eintr):
raise
class FauxSocket(object):
"""Faux socket with the minimal interface required by pypy"""
def _reuse(self):
pass
_fileobject_uses_str_type = six.PY2 and isinstance(
socket._fileobject(FauxSocket())._rbuf, six.string_types)
# FauxSocket is no longer needed
del FauxSocket
if not _fileobject_uses_str_type:
def read(self, size=-1):
# Use max, disallow tiny reads in a loop as they are very
# inefficient.
# We never leave read() with any leftover data from a new recv()
# call in our internal buffer.
rbufsize = max(self._rbufsize, self.default_bufsize)
# Our use of StringIO rather than lists of string objects returned
# by recv() minimizes memory usage and fragmentation that occurs
# when rbufsize is large compared to the typical return value of
# recv().
buf = self._rbuf
buf.seek(0, 2) # seek end
if size < 0:
# Read until EOF
# reset _rbuf. we consume it via buf.
self._rbuf = io.BytesIO()
while True:
data = self.recv(rbufsize)
if not data:
break
buf.write(data)
return buf.getvalue()
else:
# Read until size bytes or EOF seen, whichever comes first
buf_len = buf.tell()
if buf_len >= size:
# Already have size bytes in our buffer? Extract and
# return.
buf.seek(0)
rv = buf.read(size)
self._rbuf = io.BytesIO()
self._rbuf.write(buf.read())
return rv
# reset _rbuf. we consume it via buf.
self._rbuf = io.BytesIO()
while True:
left = size - buf_len
# recv() will malloc the amount of memory given as its
# parameter even though it often returns much less data
# than that. The returned data string is short lived
# as we copy it into a StringIO and free it. This avoids
# fragmentation issues on many platforms.
data = self.recv(left)
if not data:
break
n = len(data)
if n == size and not buf_len:
# Shortcut. Avoid buffer data copies when:
# - We have no data in our buffer.
# AND
# - Our call to recv returned exactly the
# number of bytes we were asked to read.
return data
if n == left:
buf.write(data)
del data # explicit free
break
assert n <= left, 'recv(%d) returned %d bytes' % (left, n)
buf.write(data)
buf_len += n
del data # explicit free
# assert buf_len == buf.tell()
return buf.getvalue()
def readline(self, size=-1):
buf = self._rbuf
buf.seek(0, 2) # seek end
if buf.tell() > 0:
# check if we already have it in our buffer
buf.seek(0)
bline = buf.readline(size)
if bline.endswith('\n') or len(bline) == size:
self._rbuf = io.BytesIO()
self._rbuf.write(buf.read())
return bline
del bline
if size < 0:
# Read until \n or EOF, whichever comes first
if self._rbufsize <= 1:
# Speed up unbuffered case
buf.seek(0)
buffers = [buf.read()]
# reset _rbuf. we consume it via buf.
self._rbuf = io.BytesIO()
data = None
recv = self.recv
while data != '\n':
data = recv(1)
if not data:
break
buffers.append(data)
return ''.join(buffers)
buf.seek(0, 2) # seek end
# reset _rbuf. we consume it via buf.
self._rbuf = io.BytesIO()
while True:
data = self.recv(self._rbufsize)
if not data:
break
nl = data.find('\n')
if nl >= 0:
nl += 1
buf.write(data[:nl])
self._rbuf.write(data[nl:])
del data
break
buf.write(data)
return buf.getvalue()
else:
# Read until size bytes or \n or EOF seen, whichever comes
# first
buf.seek(0, 2) # seek end
buf_len = buf.tell()
if buf_len >= size:
buf.seek(0)
rv = buf.read(size)
self._rbuf = io.BytesIO()
self._rbuf.write(buf.read())
return rv
# reset _rbuf. we consume it via buf.
self._rbuf = io.BytesIO()
while True:
data = self.recv(self._rbufsize)
if not data:
break
left = size - buf_len
# did we just receive a newline?
nl = data.find('\n', 0, left)
if nl >= 0:
nl += 1
# save the excess data to _rbuf
self._rbuf.write(data[nl:])
if buf_len:
buf.write(data[:nl])
break
else:
# Shortcut. Avoid data copy through buf when
# returning a substring of our first recv().
return data[:nl]
n = len(data)
if n == size and not buf_len:
# Shortcut. Avoid data copy through buf when
# returning exactly all of our first recv().
return data
if n >= left:
buf.write(data[:left])
self._rbuf.write(data[left:])
break
buf.write(data)
buf_len += n
# assert buf_len == buf.tell()
return buf.getvalue()
else:
def read(self, size=-1):
if size < 0:
# Read until EOF
buffers = [self._rbuf]
self._rbuf = ''
if self._rbufsize <= 1:
recv_size = self.default_bufsize
else:
recv_size = self._rbufsize
while True:
data = self.recv(recv_size)
if not data:
break
buffers.append(data)
return ''.join(buffers)
else:
# Read until size bytes or EOF seen, whichever comes first
data = self._rbuf
buf_len = len(data)
if buf_len >= size:
self._rbuf = data[size:]
return data[:size]
buffers = []
if data:
buffers.append(data)
self._rbuf = ''
while True:
left = size - buf_len
recv_size = max(self._rbufsize, left)
data = self.recv(recv_size)
if not data:
break
buffers.append(data)
n = len(data)
if n >= left:
self._rbuf = data[left:]
buffers[-1] = data[:left]
break
buf_len += n
return ''.join(buffers)
def readline(self, size=-1):
data = self._rbuf
if size < 0:
# Read until \n or EOF, whichever comes first
if self._rbufsize <= 1:
# Speed up unbuffered case
assert data == ''
buffers = []
while data != '\n':
data = self.recv(1)
if not data:
break
buffers.append(data)
return ''.join(buffers)
nl = data.find('\n')
if nl >= 0:
nl += 1
self._rbuf = data[nl:]
return data[:nl]
buffers = []
if data:
buffers.append(data)
self._rbuf = ''
while True:
data = self.recv(self._rbufsize)
if not data:
break
buffers.append(data)
nl = data.find('\n')
if nl >= 0:
nl += 1
self._rbuf = data[nl:]
buffers[-1] = data[:nl]
break
return ''.join(buffers)
else:
# Read until size bytes or \n or EOF seen, whichever comes
# first
nl = data.find('\n', 0, size)
if nl >= 0:
nl += 1
self._rbuf = data[nl:]
return data[:nl]
buf_len = len(data)
if buf_len >= size:
self._rbuf = data[size:]
return data[:size]
buffers = []
if data:
buffers.append(data)
self._rbuf = ''
while True:
data = self.recv(self._rbufsize)
if not data:
break
buffers.append(data)
left = size - buf_len
nl = data.find('\n', 0, left)
if nl >= 0:
nl += 1
self._rbuf = data[nl:]
buffers[-1] = data[:nl]
break
n = len(data)
if n >= left:
self._rbuf = data[left:]
buffers[-1] = data[:left]
break
buf_len += n
return ''.join(buffers)
CP_makefile = CP_makefile_PY2 if six.PY2 else CP_makefile_PY3
class HTTPConnection(object):
"""An HTTP connection (active socket).
server: the Server object which received this connection.
socket: the raw socket object (usually TCP) for this connection.
makefile: a fileobject class for reading from the socket.
"""
remote_addr = None
remote_port = None
ssl_env = None
rbufsize = DEFAULT_BUFFER_SIZE
wbufsize = DEFAULT_BUFFER_SIZE
RequestHandlerClass = HTTPRequest
def __init__(self, server, sock, makefile=CP_makefile):
self.server = server
self.socket = sock
self.rfile = makefile(sock, 'rb', self.rbufsize)
self.wfile = makefile(sock, 'wb', self.wbufsize)
self.requests_seen = 0
def communicate(self):
"""Read each request and respond appropriately."""
request_seen = False
try:
while True:
# (re)set req to None so that if something goes wrong in
# the RequestHandlerClass constructor, the error doesn't
# get written to the previous request.
req = None
req = self.RequestHandlerClass(self.server, self)
# This order of operations should guarantee correct pipelining.
req.parse_request()
if self.server.stats['Enabled']:
self.requests_seen += 1
if not req.ready:
# Something went wrong in the parsing (and the server has
# probably already made a simple_response). Return and
# let the conn close.
return
request_seen = True
req.respond()
if req.close_connection:
return
except socket.error:
e = sys.exc_info()[1]
errnum = e.args[0]
# sadly SSL sockets return a different (longer) time out string
if (
errnum == 'timed out' or
errnum == 'The read operation timed out'
):
# Don't error if we're between requests; only error
# if 1) no request has been started at all, or 2) we're
# in the middle of a request.
# See https://github.com/cherrypy/cherrypy/issues/853
if (not request_seen) or (req and req.started_request):
# Don't bother writing the 408 if the response
# has already started being written.
if req and not req.sent_headers:
try:
req.simple_response('408 Request Timeout')
except FatalSSLAlert:
# Close the connection.
return
except NoSSLError:
self._handle_no_ssl()
elif errnum not in socket_errors_to_ignore:
self.server.error_log('socket.error %s' % repr(errnum),
level=logging.WARNING, traceback=True)
if req and not req.sent_headers:
try:
req.simple_response('500 Internal Server Error')
except FatalSSLAlert:
# Close the connection.
return
except NoSSLError:
self._handle_no_ssl()
return
except (KeyboardInterrupt, SystemExit):
raise
except FatalSSLAlert:
# Close the connection.
return
except NoSSLError:
self._handle_no_ssl(req)
except Exception:
e = sys.exc_info()[1]
self.server.error_log(repr(e), level=logging.ERROR, traceback=True)
if req and not req.sent_headers:
try:
req.simple_response('500 Internal Server Error')
except FatalSSLAlert:
# Close the connection.
return
linger = False
def _handle_no_ssl(self, req):
if not req or req.sent_headers:
return
# Unwrap wfile
self.wfile = CP_makefile(self.socket._sock, 'wb', self.wbufsize)
msg = (
'The client sent a plain HTTP request, but '
'this server only speaks HTTPS on this port.'
)
req.simple_response('400 Bad Request', msg)
self.linger = True
def close(self):
"""Close the socket underlying this connection."""
self.rfile.close()
if not self.linger:
self._close_kernel_socket()
self.socket.close()
else:
# On the other hand, sometimes we want to hang around for a bit
# to make sure the client has a chance to read our entire
# response. Skipping the close() calls here delays the FIN
# packet until the socket object is garbage-collected later.
# Someday, perhaps, we'll do the full lingering_close that
# Apache does, but not today.
pass
def _close_kernel_socket(self):
"""
On old Python versions,
Python's socket module does NOT call close on the kernel
socket when you call socket.close(). We do so manually here
because we want this server to send a FIN TCP segment
immediately. Note this must be called *before* calling
socket.close(), because the latter drops its reference to
the kernel socket.
"""
if six.PY2 and hasattr(self.socket, '_sock'):
self.socket._sock.close()
class TrueyZero(object):
"""An object which equals and does math like the integer 0 but evals True.
"""
def __add__(self, other):
return other
def __radd__(self, other):
return other
trueyzero = TrueyZero()
_SHUTDOWNREQUEST = None
class WorkerThread(threading.Thread):
"""Thread which continuously polls a Queue for Connection objects.
Due to the timing issues of polling a Queue, a WorkerThread does not
check its own 'ready' flag after it has started. To stop the thread,
it is necessary to stick a _SHUTDOWNREQUEST object onto the Queue
(one for each running WorkerThread).
"""
conn = None
"""The current connection pulled off the Queue, or None."""
server = None
"""The HTTP Server which spawned this thread, and which owns the
Queue and is placing active connections into it."""
ready = False
"""A simple flag for the calling server to know when this thread
has begun polling the Queue."""
def __init__(self, server):
self.ready = False
self.server = server
self.requests_seen = 0
self.bytes_read = 0
self.bytes_written = 0
self.start_time = None
self.work_time = 0
self.stats = {
'Requests': lambda s: self.requests_seen + (
(self.start_time is None) and
trueyzero or
self.conn.requests_seen
),
'Bytes Read': lambda s: self.bytes_read + (
(self.start_time is None) and
trueyzero or
self.conn.rfile.bytes_read
),
'Bytes Written': lambda s: self.bytes_written + (
(self.start_time is None) and
trueyzero or
self.conn.wfile.bytes_written
),
'Work Time': lambda s: self.work_time + (
(self.start_time is None) and
trueyzero or
time.time() - self.start_time
),
'Read Throughput': lambda s: s['Bytes Read'](s) / (
s['Work Time'](s) or 1e-6),
'Write Throughput': lambda s: s['Bytes Written'](s) / (
s['Work Time'](s) or 1e-6),
}
threading.Thread.__init__(self)
def run(self):
self.server.stats['Worker Threads'][self.getName()] = self.stats
try:
self.ready = True
while True:
conn = self.server.requests.get()
if conn is _SHUTDOWNREQUEST:
return
self.conn = conn
if self.server.stats['Enabled']:
self.start_time = time.time()
try:
conn.communicate()
finally:
conn.close()
if self.server.stats['Enabled']:
self.requests_seen += self.conn.requests_seen
self.bytes_read += self.conn.rfile.bytes_read
self.bytes_written += self.conn.wfile.bytes_written
self.work_time += time.time() - self.start_time
self.start_time = None
self.conn = None
except (KeyboardInterrupt, SystemExit):
exc = sys.exc_info()[1]
self.server.interrupt = exc
class ThreadPool(object):
"""A Request Queue for an HTTPServer which pools threads.
ThreadPool objects must provide min, get(), put(obj), start()
and stop(timeout) attributes.
"""
def __init__(self, server, min=10, max=-1,
accepted_queue_size=-1, accepted_queue_timeout=10):
self.server = server
self.min = min
self.max = max
self._threads = []
self._queue = queue.Queue(maxsize=accepted_queue_size)
self._queue_put_timeout = accepted_queue_timeout
self.get = self._queue.get
def start(self):
"""Start the pool of threads."""
for i in range(self.min):
self._threads.append(WorkerThread(self.server))
for worker in self._threads:
worker.setName('CP Server ' + worker.getName())
worker.start()
for worker in self._threads:
while not worker.ready:
time.sleep(.1)
def _get_idle(self):
"""Number of worker threads which are idle. Read-only."""
return len([t for t in self._threads if t.conn is None])
idle = property(_get_idle, doc=_get_idle.__doc__)
def put(self, obj):
self._queue.put(obj, block=True, timeout=self._queue_put_timeout)
if obj is _SHUTDOWNREQUEST:
return
def grow(self, amount):
"""Spawn new worker threads (not above self.max)."""
if self.max > 0:
budget = max(self.max - len(self._threads), 0)
else:
# self.max <= 0 indicates no maximum
budget = float('inf')
n_new = min(amount, budget)
workers = [self._spawn_worker() for i in range(n_new)]
while not all(worker.ready for worker in workers):
time.sleep(.1)
self._threads.extend(workers)
def _spawn_worker(self):
worker = WorkerThread(self.server)
worker.setName('CP Server ' + worker.getName())
worker.start()
return worker
def shrink(self, amount):
"""Kill off worker threads (not below self.min)."""
# Grow/shrink the pool if necessary.
# Remove any dead threads from our list
for t in self._threads:
if not t.isAlive():
self._threads.remove(t)
amount -= 1
# calculate the number of threads above the minimum
n_extra = max(len(self._threads) - self.min, 0)
# don't remove more than amount
n_to_remove = min(amount, n_extra)
# put shutdown requests on the queue equal to the number of threads
# to remove. As each request is processed by a worker, that worker
# will terminate and be culled from the list.
for n in range(n_to_remove):
self._queue.put(_SHUTDOWNREQUEST)
def stop(self, timeout=5):
# Must shut down threads here so the code that calls
# this method can know when all threads are stopped.
for worker in self._threads:
self._queue.put(_SHUTDOWNREQUEST)
# Don't join currentThread (when stop is called inside a request).
current = threading.currentThread()
if timeout is not None and timeout >= 0:
endtime = time.time() + timeout
while self._threads:
worker = self._threads.pop()
if worker is not current and worker.isAlive():
try:
if timeout is None or timeout < 0:
worker.join()
else:
remaining_time = endtime - time.time()
if remaining_time > 0:
worker.join(remaining_time)
if worker.isAlive():
# We exhausted the timeout.
# Forcibly shut down the socket.
c = worker.conn
if c and not c.rfile.closed:
try:
c.socket.shutdown(socket.SHUT_RD)
except TypeError:
# pyOpenSSL sockets don't take an arg
c.socket.shutdown()
worker.join()
except (AssertionError,
# Ignore repeated Ctrl-C.
# See
# https://github.com/cherrypy/cherrypy/issues/691.
KeyboardInterrupt):
pass
def _get_qsize(self):
return self._queue.qsize()
qsize = property(_get_qsize)
try:
import fcntl
except ImportError:
try:
from ctypes import windll, WinError
import ctypes.wintypes
_SetHandleInformation = windll.kernel32.SetHandleInformation
_SetHandleInformation.argtypes = [
ctypes.wintypes.HANDLE,
ctypes.wintypes.DWORD,
ctypes.wintypes.DWORD,
]
_SetHandleInformation.restype = ctypes.wintypes.BOOL
except ImportError:
def prevent_socket_inheritance(sock):
"""Dummy function, since neither fcntl nor ctypes are available."""
pass
else:
def prevent_socket_inheritance(sock):
"""Mark the given socket fd as non-inheritable (Windows)."""
if not _SetHandleInformation(sock.fileno(), 1, 0):
raise WinError()
else:
def prevent_socket_inheritance(sock):
"""Mark the given socket fd as non-inheritable (POSIX)."""
fd = sock.fileno()
old_flags = fcntl.fcntl(fd, fcntl.F_GETFD)
fcntl.fcntl(fd, fcntl.F_SETFD, old_flags | fcntl.FD_CLOEXEC)
class SSLAdapter(object):
"""Base class for SSL driver library adapters.
Required methods:
* ``wrap(sock) -> (wrapped socket, ssl environ dict)``
* ``makefile(sock, mode='r', bufsize=DEFAULT_BUFFER_SIZE) ->
socket file object``
"""
def __init__(self, certificate, private_key, certificate_chain=None):
self.certificate = certificate
self.private_key = private_key
self.certificate_chain = certificate_chain
def wrap(self, sock):
raise NotImplemented
def makefile(self, sock, mode='r', bufsize=DEFAULT_BUFFER_SIZE):
raise NotImplemented
class HTTPServer(object):
"""An HTTP server."""
_bind_addr = '127.0.0.1'
_interrupt = None
gateway = None
"""A Gateway instance."""
minthreads = None
"""The minimum number of worker threads to create (default 10)."""
maxthreads = None
"""The maximum number of worker threads to create (default -1 = no limit).
"""
server_name = None
"""The name of the server; defaults to socket.gethostname()."""
protocol = 'HTTP/1.1'
"""The version string to write in the Status-Line of all HTTP responses.
For example, "HTTP/1.1" is the default. This also limits the supported
features used in the response."""
request_queue_size = 5
"""The 'backlog' arg to socket.listen(); max queued connections
(default 5).
"""
shutdown_timeout = 5
"""The total time, in seconds, to wait for worker threads to cleanly exit.
"""
timeout = 10
"""The timeout in seconds for accepted connections (default 10)."""
version = 'CherryPy/' + cp_version
"""A version string for the HTTPServer."""
software = None
"""The value to set for the SERVER_SOFTWARE entry in the WSGI environ.
If None, this defaults to ``'%s Server' % self.version``."""
ready = False
"""An internal flag which marks whether the socket is accepting
connections.
"""
max_request_header_size = 0
"""The maximum size, in bytes, for request headers, or 0 for no limit."""
max_request_body_size = 0
"""The maximum size, in bytes, for request bodies, or 0 for no limit."""
nodelay = True
"""If True (the default since 3.1), sets the TCP_NODELAY socket option."""
ConnectionClass = HTTPConnection
"""The class to use for handling HTTP connections."""
ssl_adapter = None
"""An instance of SSLAdapter (or a subclass).
You must have the corresponding SSL driver library installed."""
def __init__(self, bind_addr, gateway, minthreads=10, maxthreads=-1,
server_name=None):
self.bind_addr = bind_addr
self.gateway = gateway
self.requests = ThreadPool(self, min=minthreads or 1, max=maxthreads)
if not server_name:
server_name = socket.gethostname()
self.server_name = server_name
self.clear_stats()
def clear_stats(self):
self._start_time = None
self._run_time = 0
self.stats = {
'Enabled': False,
'Bind Address': lambda s: repr(self.bind_addr),
'Run time': lambda s: (not s['Enabled']) and -1 or self.runtime(),
'Accepts': 0,
'Accepts/sec': lambda s: s['Accepts'] / self.runtime(),
'Queue': lambda s: getattr(self.requests, 'qsize', None),
'Threads': lambda s: len(getattr(self.requests, '_threads', [])),
'Threads Idle': lambda s: getattr(self.requests, 'idle', None),
'Socket Errors': 0,
'Requests': lambda s: (not s['Enabled']) and -1 or sum(
[w['Requests'](w) for w in s['Worker Threads'].values()], 0),
'Bytes Read': lambda s: (not s['Enabled']) and -1 or sum(
[w['Bytes Read'](w) for w in s['Worker Threads'].values()], 0),
'Bytes Written': lambda s: (not s['Enabled']) and -1 or sum(
[w['Bytes Written'](w) for w in s['Worker Threads'].values()],
0),
'Work Time': lambda s: (not s['Enabled']) and -1 or sum(
[w['Work Time'](w) for w in s['Worker Threads'].values()], 0),
'Read Throughput': lambda s: (not s['Enabled']) and -1 or sum(
[w['Bytes Read'](w) / (w['Work Time'](w) or 1e-6)
for w in s['Worker Threads'].values()], 0),
'Write Throughput': lambda s: (not s['Enabled']) and -1 or sum(
[w['Bytes Written'](w) / (w['Work Time'](w) or 1e-6)
for w in s['Worker Threads'].values()], 0),
'Worker Threads': {},
}
logging.statistics['CherryPy HTTPServer %d' % id(self)] = self.stats
def runtime(self):
if self._start_time is None:
return self._run_time
else:
return self._run_time + (time.time() - self._start_time)
def __str__(self):
return '%s.%s(%r)' % (self.__module__, self.__class__.__name__,
self.bind_addr)
def _get_bind_addr(self):
return self._bind_addr
def _set_bind_addr(self, value):
if isinstance(value, tuple) and value[0] in ('', None):
# Despite the socket module docs, using '' does not
# allow AI_PASSIVE to work. Passing None instead
# returns '0.0.0.0' like we want. In other words:
# host AI_PASSIVE result
# '' Y 192.168.x.y
# '' N 192.168.x.y
# None Y 0.0.0.0
# None N 127.0.0.1
# But since you can get the same effect with an explicit
# '0.0.0.0', we deny both the empty string and None as values.
raise ValueError("Host values of '' or None are not allowed. "
"Use '0.0.0.0' (IPv4) or '::' (IPv6) instead "
'to listen on all active interfaces.')
self._bind_addr = value
bind_addr = property(
_get_bind_addr,
_set_bind_addr,
doc="""The interface on which to listen for connections.
For TCP sockets, a (host, port) tuple. Host values may be any IPv4
or IPv6 address, or any valid hostname. The string 'localhost' is a
synonym for '127.0.0.1' (or '::1', if your hosts file prefers IPv6).
The string '0.0.0.0' is a special IPv4 entry meaning "any active
interface" (INADDR_ANY), and '::' is the similar IN6ADDR_ANY for
IPv6. The empty string or None are not allowed.
For UNIX sockets, supply the filename as a string.
Systemd socket activation is automatic and doesn't require tempering
with this variable""")
def start(self):
"""Run the server forever."""
# We don't have to trap KeyboardInterrupt or SystemExit here,
# because cherrpy.server already does so, calling self.stop() for us.
# If you're using this server with another framework, you should
# trap those exceptions in whatever code block calls start().
self._interrupt = None
if self.software is None:
self.software = '%s Server' % self.version
# Select the appropriate socket
self.socket = None
if os.getenv('LISTEN_PID', None):
# systemd socket activation
self.socket = socket.fromfd(3, socket.AF_INET, socket.SOCK_STREAM)
elif isinstance(self.bind_addr, six.string_types):
# AF_UNIX socket
# So we can reuse the socket...
try:
os.unlink(self.bind_addr)
except:
pass
# So everyone can access the socket...
try:
os.chmod(self.bind_addr, 0o777)
except:
pass
info = [
(socket.AF_UNIX, socket.SOCK_STREAM, 0, '', self.bind_addr)]
else:
# AF_INET or AF_INET6 socket
# Get the correct address family for our host (allows IPv6
# addresses)
host, port = self.bind_addr
try:
info = socket.getaddrinfo(
host, port, socket.AF_UNSPEC,
socket.SOCK_STREAM, 0, socket.AI_PASSIVE)
except socket.gaierror:
if ':' in self.bind_addr[0]:
info = [(socket.AF_INET6, socket.SOCK_STREAM,
0, '', self.bind_addr + (0, 0))]
else:
info = [(socket.AF_INET, socket.SOCK_STREAM,
0, '', self.bind_addr)]
if not self.socket:
msg = 'No socket could be created'
for res in info:
af, socktype, proto, canonname, sa = res
try:
self.bind(af, socktype, proto)
break
except socket.error as serr:
msg = '%s -- (%s: %s)' % (msg, sa, serr)
if self.socket:
self.socket.close()
self.socket = None
if not self.socket:
raise socket.error(msg)
# Timeout so KeyboardInterrupt can be caught on Win32
self.socket.settimeout(1)
self.socket.listen(self.request_queue_size)
# Create worker threads
self.requests.start()
self.ready = True
self._start_time = time.time()
while self.ready:
try:
self.tick()
except (KeyboardInterrupt, SystemExit):
raise
except:
self.error_log('Error in HTTPServer.tick', level=logging.ERROR,
traceback=True)
if self.interrupt:
while self.interrupt is True:
# Wait for self.stop() to complete. See _set_interrupt.
time.sleep(0.1)
if self.interrupt:
raise self.interrupt
def error_log(self, msg='', level=20, traceback=False):
# Override this in subclasses as desired
sys.stderr.write(msg + '\n')
sys.stderr.flush()
if traceback:
tblines = traceback_.format_exc()
sys.stderr.write(tblines)
sys.stderr.flush()
def bind(self, family, type, proto=0):
"""Create (or recreate) the actual socket object."""
self.socket = socket.socket(family, type, proto)
prevent_socket_inheritance(self.socket)
self.socket.setsockopt(socket.SOL_SOCKET, socket.SO_REUSEADDR, 1)
if self.nodelay and not isinstance(self.bind_addr, str):
self.socket.setsockopt(socket.IPPROTO_TCP, socket.TCP_NODELAY, 1)
if self.ssl_adapter is not None:
self.socket = self.ssl_adapter.bind(self.socket)
host, port = self.bind_addr[:2]
# If listening on the IPV6 any address ('::' = IN6ADDR_ANY),
# activate dual-stack. See
# https://github.com/cherrypy/cherrypy/issues/871.
if (hasattr(socket, 'AF_INET6') and family == socket.AF_INET6
and host in ('::', '::0', '::0.0.0.0')):
try:
self.socket.setsockopt(
socket.IPPROTO_IPV6, socket.IPV6_V6ONLY, 0)
except (AttributeError, socket.error):
# Apparently, the socket option is not available in
# this machine's TCP stack
pass
self.socket.bind(self.bind_addr)
def tick(self):
"""Accept a new connection and put it on the Queue."""
try:
s, addr = self.socket.accept()
if self.stats['Enabled']:
self.stats['Accepts'] += 1
if not self.ready:
return
prevent_socket_inheritance(s)
if hasattr(s, 'settimeout'):
s.settimeout(self.timeout)
makefile = CP_makefile
ssl_env = {}
# if ssl cert and key are set, we try to be a secure HTTP server
if self.ssl_adapter is not None:
try:
s, ssl_env = self.ssl_adapter.wrap(s)
except NoSSLError:
msg = ('The client sent a plain HTTP request, but '
'this server only speaks HTTPS on this port.')
buf = ['%s 400 Bad Request\r\n' % self.protocol,
'Content-Length: %s\r\n' % len(msg),
'Content-Type: text/plain\r\n\r\n',
msg]
sock_to_make = s if six.PY3 else s._sock
wfile = makefile(sock_to_make, 'wb', DEFAULT_BUFFER_SIZE)
try:
wfile.write(''.join(buf).encode('ISO-8859-1'))
except socket.error:
x = sys.exc_info()[1]
if x.args[0] not in socket_errors_to_ignore:
raise
return
if not s:
return
makefile = self.ssl_adapter.makefile
# Re-apply our timeout since we may have a new socket object
if hasattr(s, 'settimeout'):
s.settimeout(self.timeout)
conn = self.ConnectionClass(self, s, makefile)
if not isinstance(self.bind_addr, six.string_types):
# optional values
# Until we do DNS lookups, omit REMOTE_HOST
if addr is None: # sometimes this can happen
# figure out if AF_INET or AF_INET6.
if len(s.getsockname()) == 2:
# AF_INET
addr = ('0.0.0.0', 0)
else:
# AF_INET6
addr = ('::', 0)
conn.remote_addr = addr[0]
conn.remote_port = addr[1]
conn.ssl_env = ssl_env
try:
self.requests.put(conn)
except queue.Full:
# Just drop the conn. TODO: write 503 back?
conn.close()
return
except socket.timeout:
# The only reason for the timeout in start() is so we can
# notice keyboard interrupts on Win32, which don't interrupt
# accept() by default
return
except socket.error:
x = sys.exc_info()[1]
if self.stats['Enabled']:
self.stats['Socket Errors'] += 1
if x.args[0] in socket_error_eintr:
# I *think* this is right. EINTR should occur when a signal
# is received during the accept() call; all docs say retry
# the call, and I *think* I'm reading it right that Python
# will then go ahead and poll for and handle the signal
# elsewhere. See
# https://github.com/cherrypy/cherrypy/issues/707.
return
if x.args[0] in socket_errors_nonblocking:
# Just try again. See
# https://github.com/cherrypy/cherrypy/issues/479.
return
if x.args[0] in socket_errors_to_ignore:
# Our socket was closed.
# See https://github.com/cherrypy/cherrypy/issues/686.
return
raise
def _get_interrupt(self):
return self._interrupt
def _set_interrupt(self, interrupt):
self._interrupt = True
self.stop()
self._interrupt = interrupt
interrupt = property(_get_interrupt, _set_interrupt,
doc='Set this to an Exception instance to '
'interrupt the server.')
def stop(self):
"""Gracefully shutdown a server that is serving forever."""
self.ready = False
if self._start_time is not None:
self._run_time += (time.time() - self._start_time)
self._start_time = None
sock = getattr(self, 'socket', None)
if sock:
if not isinstance(self.bind_addr, six.string_types):
# Touch our own socket to make accept() return immediately.
try:
host, port = sock.getsockname()[:2]
except socket.error:
x = sys.exc_info()[1]
if x.args[0] not in socket_errors_to_ignore:
# Changed to use error code and not message
# See
# https://github.com/cherrypy/cherrypy/issues/860.
raise
else:
# Note that we're explicitly NOT using AI_PASSIVE,
# here, because we want an actual IP to touch.
# localhost won't work if we've bound to a public IP,
# but it will if we bound to '0.0.0.0' (INADDR_ANY).
for res in socket.getaddrinfo(host, port, socket.AF_UNSPEC,
socket.SOCK_STREAM):
af, socktype, proto, canonname, sa = res
s = None
try:
s = socket.socket(af, socktype, proto)
# See
# http://groups.google.com/group/cherrypy-users/
# browse_frm/thread/bbfe5eb39c904fe0
s.settimeout(1.0)
s.connect((host, port))
s.close()
except socket.error:
if s:
s.close()
if hasattr(sock, 'close'):
sock.close()
self.socket = None
self.requests.stop(self.shutdown_timeout)
class Gateway(object):
"""A base class to interface HTTPServer with other systems, such as WSGI.
"""
def __init__(self, req):
self.req = req
def respond(self):
"""Process the current request. Must be overridden in a subclass."""
raise NotImplemented
# These may either be wsgiserver.SSLAdapter subclasses or the string names
# of such classes (in which case they will be lazily loaded).
ssl_adapters = {
'builtin': 'cherrypy.wsgiserver.ssl_builtin.BuiltinSSLAdapter',
'pyopenssl': 'cherrypy.wsgiserver.ssl_pyopenssl.pyOpenSSLAdapter',
}
def get_ssl_adapter_class(name='builtin'):
"""Return an SSL adapter class for the given name."""
adapter = ssl_adapters[name.lower()]
if isinstance(adapter, six.string_types):
last_dot = adapter.rfind('.')
attr_name = adapter[last_dot + 1:]
mod_path = adapter[:last_dot]
try:
mod = sys.modules[mod_path]
if mod is None:
raise KeyError()
except KeyError:
# The last [''] is important.
mod = __import__(mod_path, globals(), locals(), [''])
# Let an AttributeError propagate outward.
try:
adapter = getattr(mod, attr_name)
except AttributeError:
raise AttributeError("'%s' object has no attribute '%s'"
% (mod_path, attr_name))
return adapter
# ------------------------------- WSGI Stuff -------------------------------- #
class CherryPyWSGIServer(HTTPServer):
"""A subclass of HTTPServer which calls a WSGI application."""
wsgi_version = (1, 0)
"""The version of WSGI to produce."""
def __init__(self, bind_addr, wsgi_app, numthreads=10, server_name=None,
max=-1, request_queue_size=5, timeout=10, shutdown_timeout=5,
accepted_queue_size=-1, accepted_queue_timeout=10):
self.requests = ThreadPool(self, min=numthreads or 1, max=max,
accepted_queue_size=accepted_queue_size,
accepted_queue_timeout=accepted_queue_timeout)
self.wsgi_app = wsgi_app
self.gateway = wsgi_gateways[self.wsgi_version]
self.bind_addr = bind_addr
if not server_name:
server_name = socket.gethostname()
self.server_name = server_name
self.request_queue_size = request_queue_size
self.timeout = timeout
self.shutdown_timeout = shutdown_timeout
self.clear_stats()
def _get_numthreads(self):
return self.requests.min
def _set_numthreads(self, value):
self.requests.min = value
numthreads = property(_get_numthreads, _set_numthreads)
class WSGIGateway(Gateway):
"""A base class to interface HTTPServer with WSGI."""
def __init__(self, req):
self.req = req
self.started_response = False
self.env = self.get_environ()
self.remaining_bytes_out = None
def get_environ(self):
"""Return a new environ dict targeting the given wsgi.version"""
raise NotImplemented
def respond(self):
"""Process the current request."""
"""
From PEP 333:
The start_response callable must not actually transmit
the response headers. Instead, it must store them for the
server or gateway to transmit only after the first
iteration of the application return value that yields
a NON-EMPTY string, or upon the application's first
invocation of the write() callable.
"""
response = self.req.server.wsgi_app(self.env, self.start_response)
try:
for chunk in filter(None, response):
if not isinstance(chunk, six.binary_type):
raise ValueError('WSGI Applications must yield bytes')
self.write(chunk)
finally:
if hasattr(response, 'close'):
response.close()
def start_response(self, status, headers, exc_info=None):
"""
WSGI callable to begin the HTTP response.
"""
# "The application may call start_response more than once,
# if and only if the exc_info argument is provided."
if self.started_response and not exc_info:
raise AssertionError('WSGI start_response called a second '
'time with no exc_info.')
self.started_response = True
# "if exc_info is provided, and the HTTP headers have already been
# sent, start_response must raise an error, and should raise the
# exc_info tuple."
if self.req.sent_headers:
try:
six.reraise(*exc_info)
finally:
exc_info = None
self.req.status = self._encode_status(status)
for k, v in headers:
if not isinstance(k, str):
raise TypeError(
'WSGI response header key %r is not of type str.' % k)
if not isinstance(v, str):
raise TypeError(
'WSGI response header value %r is not of type str.' % v)
if k.lower() == 'content-length':
self.remaining_bytes_out = int(v)
out_header = ntob(k), ntob(v)
self.req.outheaders.append(out_header)
return self.write
@staticmethod
def _encode_status(status):
"""
According to PEP 3333, when using Python 3, the response status
and headers must be bytes masquerading as unicode; that is, they
must be of type "str" but are restricted to code points in the
"latin-1" set.
"""
if six.PY2:
return status
if not isinstance(status, str):
raise TypeError('WSGI response status is not of type str.')
return status.encode('ISO-8859-1')
def write(self, chunk):
"""WSGI callable to write unbuffered data to the client.
This method is also used internally by start_response (to write
data from the iterable returned by the WSGI application).
"""
if not self.started_response:
raise AssertionError('WSGI write called before start_response.')
chunklen = len(chunk)
rbo = self.remaining_bytes_out
if rbo is not None and chunklen > rbo:
if not self.req.sent_headers:
# Whew. We can send a 500 to the client.
self.req.simple_response(
'500 Internal Server Error',
'The requested resource returned more bytes than the '
'declared Content-Length.')
else:
# Dang. We have probably already sent data. Truncate the chunk
# to fit (so the client doesn't hang) and raise an error later.
chunk = chunk[:rbo]
if not self.req.sent_headers:
self.req.sent_headers = True
self.req.send_headers()
self.req.write(chunk)
if rbo is not None:
rbo -= chunklen
if rbo < 0:
raise ValueError(
'Response body exceeds the declared Content-Length.')
class WSGIGateway_10(WSGIGateway):
"""A Gateway class to interface HTTPServer with WSGI 1.0.x."""
def get_environ(self):
"""Return a new environ dict targeting the given wsgi.version"""
req = self.req
env = {
# set a non-standard environ entry so the WSGI app can know what
# the *real* server protocol is (and what features to support).
# See http://www.faqs.org/rfcs/rfc2145.html.
'ACTUAL_SERVER_PROTOCOL': req.server.protocol,
'PATH_INFO': bton(req.path),
'QUERY_STRING': bton(req.qs),
'REMOTE_ADDR': req.conn.remote_addr or '',
'REMOTE_PORT': str(req.conn.remote_port or ''),
'REQUEST_METHOD': bton(req.method),
'REQUEST_URI': bton(req.uri),
'SCRIPT_NAME': '',
'SERVER_NAME': req.server.server_name,
# Bah. "SERVER_PROTOCOL" is actually the REQUEST protocol.
'SERVER_PROTOCOL': bton(req.request_protocol),
'SERVER_SOFTWARE': req.server.software,
'wsgi.errors': sys.stderr,
'wsgi.input': req.rfile,
'wsgi.multiprocess': False,
'wsgi.multithread': True,
'wsgi.run_once': False,
'wsgi.url_scheme': bton(req.scheme),
'wsgi.version': (1, 0),
}
if isinstance(req.server.bind_addr, six.string_types):
# AF_UNIX. This isn't really allowed by WSGI, which doesn't
# address unix domain sockets. But it's better than nothing.
env['SERVER_PORT'] = ''
else:
env['SERVER_PORT'] = str(req.server.bind_addr[1])
# Request headers
env.update(
('HTTP_' + bton(k).upper().replace('-', '_'), bton(v))
for k, v in req.inheaders.items()
)
# CONTENT_TYPE/CONTENT_LENGTH
ct = env.pop('HTTP_CONTENT_TYPE', None)
if ct is not None:
env['CONTENT_TYPE'] = ct
cl = env.pop('HTTP_CONTENT_LENGTH', None)
if cl is not None:
env['CONTENT_LENGTH'] = cl
if req.conn.ssl_env:
env.update(req.conn.ssl_env)
return env
class WSGIGateway_u0(WSGIGateway_10):
"""A Gateway class to interface HTTPServer with WSGI u.0.
WSGI u.0 is an experimental protocol, which uses unicode for keys
and values in both Python 2 and Python 3.
"""
def get_environ(self):
"""Return a new environ dict targeting the given wsgi.version"""
req = self.req
env_10 = WSGIGateway_10.get_environ(self)
env = dict(map(self._decode_key, env_10.items()))
env[six.u('wsgi.version')] = ('u', 0)
# Request-URI
enc = env.setdefault(six.u('wsgi.url_encoding'), six.u('utf-8'))
try:
env['PATH_INFO'] = req.path.decode(enc)
env['QUERY_STRING'] = req.qs.decode(enc)
except UnicodeDecodeError:
# Fall back to latin 1 so apps can transcode if needed.
env['wsgi.url_encoding'] = 'ISO-8859-1'
env['PATH_INFO'] = env_10['PATH_INFO']
env['QUERY_STRING'] = env_10['QUERY_STRING']
env.update(map(self._decode_value, env.items()))
return env
@staticmethod
def _decode_key(item):
k, v = item
if six.PY2:
k = k.decode('ISO-8859-1')
return k, v
@staticmethod
def _decode_value(item):
k, v = item
skip_keys = 'REQUEST_URI', 'wsgi.input'
if six.PY3 or not isinstance(v, bytes) or k in skip_keys:
return k, v
return k, v.decode('ISO-8859-1')
wsgi_gateways = {
(1, 0): WSGIGateway_10,
('u', 0): WSGIGateway_u0,
}
class WSGIPathInfoDispatcher(object):
"""A WSGI dispatcher for dispatch based on the PATH_INFO.
apps: a dict or list of (path_prefix, app) pairs.
"""
def __init__(self, apps):
try:
apps = list(apps.items())
except AttributeError:
pass
# Sort the apps by len(path), descending
by_path_len = lambda app: len(app[0])
apps.sort(key=by_path_len, reverse=True)
# The path_prefix strings must start, but not end, with a slash.
# Use "" instead of "/".
self.apps = [(p.rstrip('/'), a) for p, a in apps]
def __call__(self, environ, start_response):
path = environ['PATH_INFO'] or '/'
for p, app in self.apps:
# The apps list should be sorted by length, descending.
if path.startswith(p + '/') or path == p:
environ = environ.copy()
environ['SCRIPT_NAME'] = environ['SCRIPT_NAME'] + p
environ['PATH_INFO'] = path[len(p):]
return app(environ, start_response)
start_response('404 Not Found', [('Content-Type', 'text/plain'),
('Content-Length', '0')])
return ['']
|