/usr/lib/python3/dist-packages/werkzeug/wsgi.py is in python3-werkzeug 0.14.1+dfsg1-1.
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 | # -*- coding: utf-8 -*-
"""
werkzeug.wsgi
~~~~~~~~~~~~~
This module implements WSGI related helpers.
:copyright: (c) 2014 by the Werkzeug Team, see AUTHORS for more details.
:license: BSD, see LICENSE for more details.
"""
import io
try:
import httplib
except ImportError:
from http import client as httplib
import mimetypes
import os
import posixpath
import re
import socket
from datetime import datetime
from functools import partial, update_wrapper
from itertools import chain
from time import mktime, time
from zlib import adler32
from werkzeug._compat import BytesIO, PY2, implements_iterator, iteritems, \
make_literal_wrapper, string_types, text_type, to_bytes, to_unicode, \
try_coerce_native, wsgi_get_bytes
from werkzeug._internal import _empty_stream, _encode_idna
from werkzeug.filesystem import get_filesystem_encoding
from werkzeug.http import http_date, is_resource_modified, \
is_hop_by_hop_header
from werkzeug.urls import uri_to_iri, url_join, url_parse, url_quote
from werkzeug.datastructures import EnvironHeaders
def responder(f):
"""Marks a function as responder. Decorate a function with it and it
will automatically call the return value as WSGI application.
Example::
@responder
def application(environ, start_response):
return Response('Hello World!')
"""
return update_wrapper(lambda *a: f(*a)(*a[-2:]), f)
def get_current_url(environ, root_only=False, strip_querystring=False,
host_only=False, trusted_hosts=None):
"""A handy helper function that recreates the full URL as IRI for the
current request or parts of it. Here's an example:
>>> from werkzeug.test import create_environ
>>> env = create_environ("/?param=foo", "http://localhost/script")
>>> get_current_url(env)
'http://localhost/script/?param=foo'
>>> get_current_url(env, root_only=True)
'http://localhost/script/'
>>> get_current_url(env, host_only=True)
'http://localhost/'
>>> get_current_url(env, strip_querystring=True)
'http://localhost/script/'
This optionally it verifies that the host is in a list of trusted hosts.
If the host is not in there it will raise a
:exc:`~werkzeug.exceptions.SecurityError`.
Note that the string returned might contain unicode characters as the
representation is an IRI not an URI. If you need an ASCII only
representation you can use the :func:`~werkzeug.urls.iri_to_uri`
function:
>>> from werkzeug.urls import iri_to_uri
>>> iri_to_uri(get_current_url(env))
'http://localhost/script/?param=foo'
:param environ: the WSGI environment to get the current URL from.
:param root_only: set `True` if you only want the root URL.
:param strip_querystring: set to `True` if you don't want the querystring.
:param host_only: set to `True` if the host URL should be returned.
:param trusted_hosts: a list of trusted hosts, see :func:`host_is_trusted`
for more information.
"""
tmp = [environ['wsgi.url_scheme'], '://', get_host(environ, trusted_hosts)]
cat = tmp.append
if host_only:
return uri_to_iri(''.join(tmp) + '/')
cat(url_quote(wsgi_get_bytes(environ.get('SCRIPT_NAME', ''))).rstrip('/'))
cat('/')
if not root_only:
cat(url_quote(wsgi_get_bytes(environ.get('PATH_INFO', '')).lstrip(b'/')))
if not strip_querystring:
qs = get_query_string(environ)
if qs:
cat('?' + qs)
return uri_to_iri(''.join(tmp))
def host_is_trusted(hostname, trusted_list):
"""Checks if a host is trusted against a list. This also takes care
of port normalization.
.. versionadded:: 0.9
:param hostname: the hostname to check
:param trusted_list: a list of hostnames to check against. If a
hostname starts with a dot it will match against
all subdomains as well.
"""
if not hostname:
return False
if isinstance(trusted_list, string_types):
trusted_list = [trusted_list]
def _normalize(hostname):
if ':' in hostname:
hostname = hostname.rsplit(':', 1)[0]
return _encode_idna(hostname)
try:
hostname = _normalize(hostname)
except UnicodeError:
return False
for ref in trusted_list:
if ref.startswith('.'):
ref = ref[1:]
suffix_match = True
else:
suffix_match = False
try:
ref = _normalize(ref)
except UnicodeError:
return False
if ref == hostname:
return True
if suffix_match and hostname.endswith(b'.' + ref):
return True
return False
def get_host(environ, trusted_hosts=None):
"""Return the real host for the given WSGI environment. This first checks
the `X-Forwarded-Host` header, then the normal `Host` header, and finally
the `SERVER_NAME` environment variable (using the first one it finds).
Optionally it verifies that the host is in a list of trusted hosts.
If the host is not in there it will raise a
:exc:`~werkzeug.exceptions.SecurityError`.
:param environ: the WSGI environment to get the host of.
:param trusted_hosts: a list of trusted hosts, see :func:`host_is_trusted`
for more information.
"""
if 'HTTP_X_FORWARDED_HOST' in environ:
rv = environ['HTTP_X_FORWARDED_HOST'].split(',', 1)[0].strip()
elif 'HTTP_HOST' in environ:
rv = environ['HTTP_HOST']
else:
rv = environ['SERVER_NAME']
if (environ['wsgi.url_scheme'], environ['SERVER_PORT']) not \
in (('https', '443'), ('http', '80')):
rv += ':' + environ['SERVER_PORT']
if trusted_hosts is not None:
if not host_is_trusted(rv, trusted_hosts):
from werkzeug.exceptions import SecurityError
raise SecurityError('Host "%s" is not trusted' % rv)
return rv
def get_content_length(environ):
"""Returns the content length from the WSGI environment as
integer. If it's not available or chunked transfer encoding is used,
``None`` is returned.
.. versionadded:: 0.9
:param environ: the WSGI environ to fetch the content length from.
"""
if environ.get('HTTP_TRANSFER_ENCODING', '') == 'chunked':
return None
content_length = environ.get('CONTENT_LENGTH')
if content_length is not None:
try:
return max(0, int(content_length))
except (ValueError, TypeError):
pass
def get_input_stream(environ, safe_fallback=True):
"""Returns the input stream from the WSGI environment and wraps it
in the most sensible way possible. The stream returned is not the
raw WSGI stream in most cases but one that is safe to read from
without taking into account the content length.
If content length is not set, the stream will be empty for safety reasons.
If the WSGI server supports chunked or infinite streams, it should set
the ``wsgi.input_terminated`` value in the WSGI environ to indicate that.
.. versionadded:: 0.9
:param environ: the WSGI environ to fetch the stream from.
:param safe_fallback: use an empty stream as a safe fallback when the
content length is not set. Disabling this allows infinite streams,
which can be a denial-of-service risk.
"""
stream = environ['wsgi.input']
content_length = get_content_length(environ)
# A wsgi extension that tells us if the input is terminated. In
# that case we return the stream unchanged as we know we can safely
# read it until the end.
if environ.get('wsgi.input_terminated'):
return stream
# If the request doesn't specify a content length, returning the stream is
# potentially dangerous because it could be infinite, malicious or not. If
# safe_fallback is true, return an empty stream instead for safety.
if content_length is None:
return safe_fallback and _empty_stream or stream
# Otherwise limit the stream to the content length
return LimitedStream(stream, content_length)
def get_query_string(environ):
"""Returns the `QUERY_STRING` from the WSGI environment. This also takes
care about the WSGI decoding dance on Python 3 environments as a
native string. The string returned will be restricted to ASCII
characters.
.. versionadded:: 0.9
:param environ: the WSGI environment object to get the query string from.
"""
qs = wsgi_get_bytes(environ.get('QUERY_STRING', ''))
# QUERY_STRING really should be ascii safe but some browsers
# will send us some unicode stuff (I am looking at you IE).
# In that case we want to urllib quote it badly.
return try_coerce_native(url_quote(qs, safe=':&%=+$!*\'(),'))
def get_path_info(environ, charset='utf-8', errors='replace'):
"""Returns the `PATH_INFO` from the WSGI environment and properly
decodes it. This also takes care about the WSGI decoding dance
on Python 3 environments. if the `charset` is set to `None` a
bytestring is returned.
.. versionadded:: 0.9
:param environ: the WSGI environment object to get the path from.
:param charset: the charset for the path info, or `None` if no
decoding should be performed.
:param errors: the decoding error handling.
"""
path = wsgi_get_bytes(environ.get('PATH_INFO', ''))
return to_unicode(path, charset, errors, allow_none_charset=True)
def get_script_name(environ, charset='utf-8', errors='replace'):
"""Returns the `SCRIPT_NAME` from the WSGI environment and properly
decodes it. This also takes care about the WSGI decoding dance
on Python 3 environments. if the `charset` is set to `None` a
bytestring is returned.
.. versionadded:: 0.9
:param environ: the WSGI environment object to get the path from.
:param charset: the charset for the path, or `None` if no
decoding should be performed.
:param errors: the decoding error handling.
"""
path = wsgi_get_bytes(environ.get('SCRIPT_NAME', ''))
return to_unicode(path, charset, errors, allow_none_charset=True)
def pop_path_info(environ, charset='utf-8', errors='replace'):
"""Removes and returns the next segment of `PATH_INFO`, pushing it onto
`SCRIPT_NAME`. Returns `None` if there is nothing left on `PATH_INFO`.
If the `charset` is set to `None` a bytestring is returned.
If there are empty segments (``'/foo//bar``) these are ignored but
properly pushed to the `SCRIPT_NAME`:
>>> env = {'SCRIPT_NAME': '/foo', 'PATH_INFO': '/a/b'}
>>> pop_path_info(env)
'a'
>>> env['SCRIPT_NAME']
'/foo/a'
>>> pop_path_info(env)
'b'
>>> env['SCRIPT_NAME']
'/foo/a/b'
.. versionadded:: 0.5
.. versionchanged:: 0.9
The path is now decoded and a charset and encoding
parameter can be provided.
:param environ: the WSGI environment that is modified.
"""
path = environ.get('PATH_INFO')
if not path:
return None
script_name = environ.get('SCRIPT_NAME', '')
# shift multiple leading slashes over
old_path = path
path = path.lstrip('/')
if path != old_path:
script_name += '/' * (len(old_path) - len(path))
if '/' not in path:
environ['PATH_INFO'] = ''
environ['SCRIPT_NAME'] = script_name + path
rv = wsgi_get_bytes(path)
else:
segment, path = path.split('/', 1)
environ['PATH_INFO'] = '/' + path
environ['SCRIPT_NAME'] = script_name + segment
rv = wsgi_get_bytes(segment)
return to_unicode(rv, charset, errors, allow_none_charset=True)
def peek_path_info(environ, charset='utf-8', errors='replace'):
"""Returns the next segment on the `PATH_INFO` or `None` if there
is none. Works like :func:`pop_path_info` without modifying the
environment:
>>> env = {'SCRIPT_NAME': '/foo', 'PATH_INFO': '/a/b'}
>>> peek_path_info(env)
'a'
>>> peek_path_info(env)
'a'
If the `charset` is set to `None` a bytestring is returned.
.. versionadded:: 0.5
.. versionchanged:: 0.9
The path is now decoded and a charset and encoding
parameter can be provided.
:param environ: the WSGI environment that is checked.
"""
segments = environ.get('PATH_INFO', '').lstrip('/').split('/', 1)
if segments:
return to_unicode(wsgi_get_bytes(segments[0]),
charset, errors, allow_none_charset=True)
def extract_path_info(environ_or_baseurl, path_or_url, charset='utf-8',
errors='replace', collapse_http_schemes=True):
"""Extracts the path info from the given URL (or WSGI environment) and
path. The path info returned is a unicode string, not a bytestring
suitable for a WSGI environment. The URLs might also be IRIs.
If the path info could not be determined, `None` is returned.
Some examples:
>>> extract_path_info('http://example.com/app', '/app/hello')
u'/hello'
>>> extract_path_info('http://example.com/app',
... 'https://example.com/app/hello')
u'/hello'
>>> extract_path_info('http://example.com/app',
... 'https://example.com/app/hello',
... collapse_http_schemes=False) is None
True
Instead of providing a base URL you can also pass a WSGI environment.
.. versionadded:: 0.6
:param environ_or_baseurl: a WSGI environment dict, a base URL or
base IRI. This is the root of the
application.
:param path_or_url: an absolute path from the server root, a
relative path (in which case it's the path info)
or a full URL. Also accepts IRIs and unicode
parameters.
:param charset: the charset for byte data in URLs
:param errors: the error handling on decode
:param collapse_http_schemes: if set to `False` the algorithm does
not assume that http and https on the
same server point to the same
resource.
"""
def _normalize_netloc(scheme, netloc):
parts = netloc.split(u'@', 1)[-1].split(u':', 1)
if len(parts) == 2:
netloc, port = parts
if (scheme == u'http' and port == u'80') or \
(scheme == u'https' and port == u'443'):
port = None
else:
netloc = parts[0]
port = None
if port is not None:
netloc += u':' + port
return netloc
# make sure whatever we are working on is a IRI and parse it
path = uri_to_iri(path_or_url, charset, errors)
if isinstance(environ_or_baseurl, dict):
environ_or_baseurl = get_current_url(environ_or_baseurl,
root_only=True)
base_iri = uri_to_iri(environ_or_baseurl, charset, errors)
base_scheme, base_netloc, base_path = url_parse(base_iri)[:3]
cur_scheme, cur_netloc, cur_path, = \
url_parse(url_join(base_iri, path))[:3]
# normalize the network location
base_netloc = _normalize_netloc(base_scheme, base_netloc)
cur_netloc = _normalize_netloc(cur_scheme, cur_netloc)
# is that IRI even on a known HTTP scheme?
if collapse_http_schemes:
for scheme in base_scheme, cur_scheme:
if scheme not in (u'http', u'https'):
return None
else:
if not (base_scheme in (u'http', u'https') and
base_scheme == cur_scheme):
return None
# are the netlocs compatible?
if base_netloc != cur_netloc:
return None
# are we below the application path?
base_path = base_path.rstrip(u'/')
if not cur_path.startswith(base_path):
return None
return u'/' + cur_path[len(base_path):].lstrip(u'/')
class ProxyMiddleware(object):
"""This middleware routes some requests to the provided WSGI app and
proxies some requests to an external server. This is not something that
can generally be done on the WSGI layer and some HTTP requests will not
tunnel through correctly (for instance websocket requests cannot be
proxied through WSGI). As a result this is only really useful for some
basic requests that can be forwarded.
Example configuration::
app = ProxyMiddleware(app, {
'/static/': {
'target': 'http://127.0.0.1:5001/',
}
})
For each host options can be specified. The following options are
supported:
``target``:
the target URL to dispatch to
``remove_prefix``:
if set to `True` the prefix is chopped off the URL before
dispatching it to the server.
``host``:
When set to ``'<auto>'`` which is the default the host header is
automatically rewritten to the URL of the target. If set to `None`
then the host header is unmodified from the client request. Any
other value overwrites the host header with that value.
``headers``:
An optional dictionary of headers that should be sent with the
request to the target host.
``ssl_context``:
In case this is an HTTPS target host then an SSL context can be
provided here (:class:`ssl.SSLContext`). This can be used for instance
to disable SSL verification.
In this case everything below ``'/static/'`` is proxied to the server on
port 5001. The host header is automatically rewritten and so are request
URLs (eg: the leading `/static/` prefix here gets chopped off).
.. versionadded:: 0.14
"""
def __init__(self, app, targets, chunk_size=2 << 13, timeout=10):
def _set_defaults(opts):
opts.setdefault('remove_prefix', False)
opts.setdefault('host', '<auto>')
opts.setdefault('headers', {})
opts.setdefault('ssl_context', None)
return opts
self.app = app
self.targets = dict(('/%s/' % k.strip('/'), _set_defaults(v))
for k, v in iteritems(targets))
self.chunk_size = chunk_size
self.timeout = timeout
def proxy_to(self, opts, path, prefix):
target = url_parse(opts['target'])
def application(environ, start_response):
headers = list(EnvironHeaders(environ).items())
headers[:] = [(k, v) for k, v in headers
if not is_hop_by_hop_header(k) and
k.lower() not in ('content-length', 'host')]
headers.append(('Connection', 'close'))
if opts['host'] == '<auto>':
headers.append(('Host', target.ascii_host))
elif opts['host'] is None:
headers.append(('Host', environ['HTTP_HOST']))
else:
headers.append(('Host', opts['host']))
headers.extend(opts['headers'].items())
remote_path = path
if opts['remove_prefix']:
remote_path = '%s/%s' % (
target.path.rstrip('/'),
remote_path[len(prefix):].lstrip('/')
)
content_length = environ.get('CONTENT_LENGTH')
chunked = False
if content_length not in ('', None):
headers.append(('Content-Length', content_length))
elif content_length is not None:
headers.append(('Transfer-Encoding', 'chunked'))
chunked = True
try:
if target.scheme == 'http':
con = httplib.HTTPConnection(
target.ascii_host, target.port or 80,
timeout=self.timeout)
elif target.scheme == 'https':
con = httplib.HTTPSConnection(
target.ascii_host, target.port or 443,
timeout=self.timeout,
context=opts['ssl_context'])
con.connect()
con.putrequest(environ['REQUEST_METHOD'], url_quote(remote_path),
skip_host=True)
for k, v in headers:
if k.lower() == 'connection':
v = 'close'
con.putheader(k, v)
con.endheaders()
stream = get_input_stream(environ)
while 1:
data = stream.read(self.chunk_size)
if not data:
break
if chunked:
con.send(b'%x\r\n%s\r\n' % (len(data), data))
else:
con.send(data)
resp = con.getresponse()
except socket.error:
from werkzeug.exceptions import BadGateway
return BadGateway()(environ, start_response)
start_response('%d %s' % (resp.status, resp.reason),
[(k.title(), v) for k, v in resp.getheaders()
if not is_hop_by_hop_header(k)])
def read():
while 1:
try:
data = resp.read(self.chunk_size)
except socket.error:
break
if not data:
break
yield data
return read()
return application
def __call__(self, environ, start_response):
path = environ['PATH_INFO']
app = self.app
for prefix, opts in iteritems(self.targets):
if path.startswith(prefix):
app = self.proxy_to(opts, path, prefix)
break
return app(environ, start_response)
class SharedDataMiddleware(object):
"""A WSGI middleware that provides static content for development
environments or simple server setups. Usage is quite simple::
import os
from werkzeug.wsgi import SharedDataMiddleware
app = SharedDataMiddleware(app, {
'/shared': os.path.join(os.path.dirname(__file__), 'shared')
})
The contents of the folder ``./shared`` will now be available on
``http://example.com/shared/``. This is pretty useful during development
because a standalone media server is not required. One can also mount
files on the root folder and still continue to use the application because
the shared data middleware forwards all unhandled requests to the
application, even if the requests are below one of the shared folders.
If `pkg_resources` is available you can also tell the middleware to serve
files from package data::
app = SharedDataMiddleware(app, {
'/shared': ('myapplication', 'shared_files')
})
This will then serve the ``shared_files`` folder in the `myapplication`
Python package.
The optional `disallow` parameter can be a list of :func:`~fnmatch.fnmatch`
rules for files that are not accessible from the web. If `cache` is set to
`False` no caching headers are sent.
Currently the middleware does not support non ASCII filenames. If the
encoding on the file system happens to be the encoding of the URI it may
work but this could also be by accident. We strongly suggest using ASCII
only file names for static files.
The middleware will guess the mimetype using the Python `mimetype`
module. If it's unable to figure out the charset it will fall back
to `fallback_mimetype`.
.. versionchanged:: 0.5
The cache timeout is configurable now.
.. versionadded:: 0.6
The `fallback_mimetype` parameter was added.
:param app: the application to wrap. If you don't want to wrap an
application you can pass it :exc:`NotFound`.
:param exports: a list or dict of exported files and folders.
:param disallow: a list of :func:`~fnmatch.fnmatch` rules.
:param fallback_mimetype: the fallback mimetype for unknown files.
:param cache: enable or disable caching headers.
:param cache_timeout: the cache timeout in seconds for the headers.
"""
def __init__(self, app, exports, disallow=None, cache=True,
cache_timeout=60 * 60 * 12, fallback_mimetype='text/plain'):
self.app = app
self.exports = []
self.cache = cache
self.cache_timeout = cache_timeout
if hasattr(exports, 'items'):
exports = iteritems(exports)
for key, value in exports:
if isinstance(value, tuple):
loader = self.get_package_loader(*value)
elif isinstance(value, string_types):
if os.path.isfile(value):
loader = self.get_file_loader(value)
else:
loader = self.get_directory_loader(value)
else:
raise TypeError('unknown def %r' % value)
self.exports.append((key, loader))
if disallow is not None:
from fnmatch import fnmatch
self.is_allowed = lambda x: not fnmatch(x, disallow)
self.fallback_mimetype = fallback_mimetype
def is_allowed(self, filename):
"""Subclasses can override this method to disallow the access to
certain files. However by providing `disallow` in the constructor
this method is overwritten.
"""
return True
def _opener(self, filename):
return lambda: (
open(filename, 'rb'),
datetime.utcfromtimestamp(os.path.getmtime(filename)),
int(os.path.getsize(filename))
)
def get_file_loader(self, filename):
return lambda x: (os.path.basename(filename), self._opener(filename))
def get_package_loader(self, package, package_path):
from pkg_resources import DefaultProvider, ResourceManager, \
get_provider
loadtime = datetime.utcnow()
provider = get_provider(package)
manager = ResourceManager()
filesystem_bound = isinstance(provider, DefaultProvider)
def loader(path):
if path is None:
return None, None
path = posixpath.join(package_path, path)
if not provider.has_resource(path):
return None, None
basename = posixpath.basename(path)
if filesystem_bound:
return basename, self._opener(
provider.get_resource_filename(manager, path))
s = provider.get_resource_string(manager, path)
return basename, lambda: (
BytesIO(s),
loadtime,
len(s)
)
return loader
def get_directory_loader(self, directory):
def loader(path):
if path is not None:
path = os.path.join(directory, path)
else:
path = directory
if os.path.isfile(path):
return os.path.basename(path), self._opener(path)
return None, None
return loader
def generate_etag(self, mtime, file_size, real_filename):
if not isinstance(real_filename, bytes):
real_filename = real_filename.encode(get_filesystem_encoding())
return 'wzsdm-%d-%s-%s' % (
mktime(mtime.timetuple()),
file_size,
adler32(real_filename) & 0xffffffff
)
def __call__(self, environ, start_response):
cleaned_path = get_path_info(environ)
if PY2:
cleaned_path = cleaned_path.encode(get_filesystem_encoding())
# sanitize the path for non unix systems
cleaned_path = cleaned_path.strip('/')
for sep in os.sep, os.altsep:
if sep and sep != '/':
cleaned_path = cleaned_path.replace(sep, '/')
path = '/' + '/'.join(x for x in cleaned_path.split('/')
if x and x != '..')
file_loader = None
for search_path, loader in self.exports:
if search_path == path:
real_filename, file_loader = loader(None)
if file_loader is not None:
break
if not search_path.endswith('/'):
search_path += '/'
if path.startswith(search_path):
real_filename, file_loader = loader(path[len(search_path):])
if file_loader is not None:
break
if file_loader is None or not self.is_allowed(real_filename):
return self.app(environ, start_response)
guessed_type = mimetypes.guess_type(real_filename)
mime_type = guessed_type[0] or self.fallback_mimetype
f, mtime, file_size = file_loader()
headers = [('Date', http_date())]
if self.cache:
timeout = self.cache_timeout
etag = self.generate_etag(mtime, file_size, real_filename)
headers += [
('Etag', '"%s"' % etag),
('Cache-Control', 'max-age=%d, public' % timeout)
]
if not is_resource_modified(environ, etag, last_modified=mtime):
f.close()
start_response('304 Not Modified', headers)
return []
headers.append(('Expires', http_date(time() + timeout)))
else:
headers.append(('Cache-Control', 'public'))
headers.extend((
('Content-Type', mime_type),
('Content-Length', str(file_size)),
('Last-Modified', http_date(mtime))
))
start_response('200 OK', headers)
return wrap_file(environ, f)
class DispatcherMiddleware(object):
"""Allows one to mount middlewares or applications in a WSGI application.
This is useful if you want to combine multiple WSGI applications::
app = DispatcherMiddleware(app, {
'/app2': app2,
'/app3': app3
})
"""
def __init__(self, app, mounts=None):
self.app = app
self.mounts = mounts or {}
def __call__(self, environ, start_response):
script = environ.get('PATH_INFO', '')
path_info = ''
while '/' in script:
if script in self.mounts:
app = self.mounts[script]
break
script, last_item = script.rsplit('/', 1)
path_info = '/%s%s' % (last_item, path_info)
else:
app = self.mounts.get(script, self.app)
original_script_name = environ.get('SCRIPT_NAME', '')
environ['SCRIPT_NAME'] = original_script_name + script
environ['PATH_INFO'] = path_info
return app(environ, start_response)
@implements_iterator
class ClosingIterator(object):
"""The WSGI specification requires that all middlewares and gateways
respect the `close` callback of an iterator. Because it is useful to add
another close action to a returned iterator and adding a custom iterator
is a boring task this class can be used for that::
return ClosingIterator(app(environ, start_response), [cleanup_session,
cleanup_locals])
If there is just one close function it can be passed instead of the list.
A closing iterator is not needed if the application uses response objects
and finishes the processing if the response is started::
try:
return response(environ, start_response)
finally:
cleanup_session()
cleanup_locals()
"""
def __init__(self, iterable, callbacks=None):
iterator = iter(iterable)
self._next = partial(next, iterator)
if callbacks is None:
callbacks = []
elif callable(callbacks):
callbacks = [callbacks]
else:
callbacks = list(callbacks)
iterable_close = getattr(iterator, 'close', None)
if iterable_close:
callbacks.insert(0, iterable_close)
self._callbacks = callbacks
def __iter__(self):
return self
def __next__(self):
return self._next()
def close(self):
for callback in self._callbacks:
callback()
def wrap_file(environ, file, buffer_size=8192):
"""Wraps a file. This uses the WSGI server's file wrapper if available
or otherwise the generic :class:`FileWrapper`.
.. versionadded:: 0.5
If the file wrapper from the WSGI server is used it's important to not
iterate over it from inside the application but to pass it through
unchanged. If you want to pass out a file wrapper inside a response
object you have to set :attr:`~BaseResponse.direct_passthrough` to `True`.
More information about file wrappers are available in :pep:`333`.
:param file: a :class:`file`-like object with a :meth:`~file.read` method.
:param buffer_size: number of bytes for one iteration.
"""
return environ.get('wsgi.file_wrapper', FileWrapper)(file, buffer_size)
@implements_iterator
class FileWrapper(object):
"""This class can be used to convert a :class:`file`-like object into
an iterable. It yields `buffer_size` blocks until the file is fully
read.
You should not use this class directly but rather use the
:func:`wrap_file` function that uses the WSGI server's file wrapper
support if it's available.
.. versionadded:: 0.5
If you're using this object together with a :class:`BaseResponse` you have
to use the `direct_passthrough` mode.
:param file: a :class:`file`-like object with a :meth:`~file.read` method.
:param buffer_size: number of bytes for one iteration.
"""
def __init__(self, file, buffer_size=8192):
self.file = file
self.buffer_size = buffer_size
def close(self):
if hasattr(self.file, 'close'):
self.file.close()
def seekable(self):
if hasattr(self.file, 'seekable'):
return self.file.seekable()
if hasattr(self.file, 'seek'):
return True
return False
def seek(self, *args):
if hasattr(self.file, 'seek'):
self.file.seek(*args)
def tell(self):
if hasattr(self.file, 'tell'):
return self.file.tell()
return None
def __iter__(self):
return self
def __next__(self):
data = self.file.read(self.buffer_size)
if data:
return data
raise StopIteration()
@implements_iterator
class _RangeWrapper(object):
# private for now, but should we make it public in the future ?
"""This class can be used to convert an iterable object into
an iterable that will only yield a piece of the underlying content.
It yields blocks until the underlying stream range is fully read.
The yielded blocks will have a size that can't exceed the original
iterator defined block size, but that can be smaller.
If you're using this object together with a :class:`BaseResponse` you have
to use the `direct_passthrough` mode.
:param iterable: an iterable object with a :meth:`__next__` method.
:param start_byte: byte from which read will start.
:param byte_range: how many bytes to read.
"""
def __init__(self, iterable, start_byte=0, byte_range=None):
self.iterable = iter(iterable)
self.byte_range = byte_range
self.start_byte = start_byte
self.end_byte = None
if byte_range is not None:
self.end_byte = self.start_byte + self.byte_range
self.read_length = 0
self.seekable = hasattr(iterable, 'seekable') and iterable.seekable()
self.end_reached = False
def __iter__(self):
return self
def _next_chunk(self):
try:
chunk = next(self.iterable)
self.read_length += len(chunk)
return chunk
except StopIteration:
self.end_reached = True
raise
def _first_iteration(self):
chunk = None
if self.seekable:
self.iterable.seek(self.start_byte)
self.read_length = self.iterable.tell()
contextual_read_length = self.read_length
else:
while self.read_length <= self.start_byte:
chunk = self._next_chunk()
if chunk is not None:
chunk = chunk[self.start_byte - self.read_length:]
contextual_read_length = self.start_byte
return chunk, contextual_read_length
def _next(self):
if self.end_reached:
raise StopIteration()
chunk = None
contextual_read_length = self.read_length
if self.read_length == 0:
chunk, contextual_read_length = self._first_iteration()
if chunk is None:
chunk = self._next_chunk()
if self.end_byte is not None and self.read_length >= self.end_byte:
self.end_reached = True
return chunk[:self.end_byte - contextual_read_length]
return chunk
def __next__(self):
chunk = self._next()
if chunk:
return chunk
self.end_reached = True
raise StopIteration()
def close(self):
if hasattr(self.iterable, 'close'):
self.iterable.close()
def _make_chunk_iter(stream, limit, buffer_size):
"""Helper for the line and chunk iter functions."""
if isinstance(stream, (bytes, bytearray, text_type)):
raise TypeError('Passed a string or byte object instead of '
'true iterator or stream.')
if not hasattr(stream, 'read'):
for item in stream:
if item:
yield item
return
if not isinstance(stream, LimitedStream) and limit is not None:
stream = LimitedStream(stream, limit)
_read = stream.read
while 1:
item = _read(buffer_size)
if not item:
break
yield item
def make_line_iter(stream, limit=None, buffer_size=10 * 1024,
cap_at_buffer=False):
"""Safely iterates line-based over an input stream. If the input stream
is not a :class:`LimitedStream` the `limit` parameter is mandatory.
This uses the stream's :meth:`~file.read` method internally as opposite
to the :meth:`~file.readline` method that is unsafe and can only be used
in violation of the WSGI specification. The same problem applies to the
`__iter__` function of the input stream which calls :meth:`~file.readline`
without arguments.
If you need line-by-line processing it's strongly recommended to iterate
over the input stream using this helper function.
.. versionchanged:: 0.8
This function now ensures that the limit was reached.
.. versionadded:: 0.9
added support for iterators as input stream.
.. versionadded:: 0.11.10
added support for the `cap_at_buffer` parameter.
:param stream: the stream or iterate to iterate over.
:param limit: the limit in bytes for the stream. (Usually
content length. Not necessary if the `stream`
is a :class:`LimitedStream`.
:param buffer_size: The optional buffer size.
:param cap_at_buffer: if this is set chunks are split if they are longer
than the buffer size. Internally this is implemented
that the buffer size might be exhausted by a factor
of two however.
"""
_iter = _make_chunk_iter(stream, limit, buffer_size)
first_item = next(_iter, '')
if not first_item:
return
s = make_literal_wrapper(first_item)
empty = s('')
cr = s('\r')
lf = s('\n')
crlf = s('\r\n')
_iter = chain((first_item,), _iter)
def _iter_basic_lines():
_join = empty.join
buffer = []
while 1:
new_data = next(_iter, '')
if not new_data:
break
new_buf = []
buf_size = 0
for item in chain(buffer, new_data.splitlines(True)):
new_buf.append(item)
buf_size += len(item)
if item and item[-1:] in crlf:
yield _join(new_buf)
new_buf = []
elif cap_at_buffer and buf_size >= buffer_size:
rv = _join(new_buf)
while len(rv) >= buffer_size:
yield rv[:buffer_size]
rv = rv[buffer_size:]
new_buf = [rv]
buffer = new_buf
if buffer:
yield _join(buffer)
# This hackery is necessary to merge 'foo\r' and '\n' into one item
# of 'foo\r\n' if we were unlucky and we hit a chunk boundary.
previous = empty
for item in _iter_basic_lines():
if item == lf and previous[-1:] == cr:
previous += item
item = empty
if previous:
yield previous
previous = item
if previous:
yield previous
def make_chunk_iter(stream, separator, limit=None, buffer_size=10 * 1024,
cap_at_buffer=False):
"""Works like :func:`make_line_iter` but accepts a separator
which divides chunks. If you want newline based processing
you should use :func:`make_line_iter` instead as it
supports arbitrary newline markers.
.. versionadded:: 0.8
.. versionadded:: 0.9
added support for iterators as input stream.
.. versionadded:: 0.11.10
added support for the `cap_at_buffer` parameter.
:param stream: the stream or iterate to iterate over.
:param separator: the separator that divides chunks.
:param limit: the limit in bytes for the stream. (Usually
content length. Not necessary if the `stream`
is otherwise already limited).
:param buffer_size: The optional buffer size.
:param cap_at_buffer: if this is set chunks are split if they are longer
than the buffer size. Internally this is implemented
that the buffer size might be exhausted by a factor
of two however.
"""
_iter = _make_chunk_iter(stream, limit, buffer_size)
first_item = next(_iter, '')
if not first_item:
return
_iter = chain((first_item,), _iter)
if isinstance(first_item, text_type):
separator = to_unicode(separator)
_split = re.compile(r'(%s)' % re.escape(separator)).split
_join = u''.join
else:
separator = to_bytes(separator)
_split = re.compile(b'(' + re.escape(separator) + b')').split
_join = b''.join
buffer = []
while 1:
new_data = next(_iter, '')
if not new_data:
break
chunks = _split(new_data)
new_buf = []
buf_size = 0
for item in chain(buffer, chunks):
if item == separator:
yield _join(new_buf)
new_buf = []
buf_size = 0
else:
buf_size += len(item)
new_buf.append(item)
if cap_at_buffer and buf_size >= buffer_size:
rv = _join(new_buf)
while len(rv) >= buffer_size:
yield rv[:buffer_size]
rv = rv[buffer_size:]
new_buf = [rv]
buf_size = len(rv)
buffer = new_buf
if buffer:
yield _join(buffer)
@implements_iterator
class LimitedStream(io.IOBase):
"""Wraps a stream so that it doesn't read more than n bytes. If the
stream is exhausted and the caller tries to get more bytes from it
:func:`on_exhausted` is called which by default returns an empty
string. The return value of that function is forwarded
to the reader function. So if it returns an empty string
:meth:`read` will return an empty string as well.
The limit however must never be higher than what the stream can
output. Otherwise :meth:`readlines` will try to read past the
limit.
.. admonition:: Note on WSGI compliance
calls to :meth:`readline` and :meth:`readlines` are not
WSGI compliant because it passes a size argument to the
readline methods. Unfortunately the WSGI PEP is not safely
implementable without a size argument to :meth:`readline`
because there is no EOF marker in the stream. As a result
of that the use of :meth:`readline` is discouraged.
For the same reason iterating over the :class:`LimitedStream`
is not portable. It internally calls :meth:`readline`.
We strongly suggest using :meth:`read` only or using the
:func:`make_line_iter` which safely iterates line-based
over a WSGI input stream.
:param stream: the stream to wrap.
:param limit: the limit for the stream, must not be longer than
what the string can provide if the stream does not
end with `EOF` (like `wsgi.input`)
"""
def __init__(self, stream, limit):
self._read = stream.read
self._readline = stream.readline
self._pos = 0
self.limit = limit
def __iter__(self):
return self
@property
def is_exhausted(self):
"""If the stream is exhausted this attribute is `True`."""
return self._pos >= self.limit
def on_exhausted(self):
"""This is called when the stream tries to read past the limit.
The return value of this function is returned from the reading
function.
"""
# Read null bytes from the stream so that we get the
# correct end of stream marker.
return self._read(0)
def on_disconnect(self):
"""What should happen if a disconnect is detected? The return
value of this function is returned from read functions in case
the client went away. By default a
:exc:`~werkzeug.exceptions.ClientDisconnected` exception is raised.
"""
from werkzeug.exceptions import ClientDisconnected
raise ClientDisconnected()
def exhaust(self, chunk_size=1024 * 64):
"""Exhaust the stream. This consumes all the data left until the
limit is reached.
:param chunk_size: the size for a chunk. It will read the chunk
until the stream is exhausted and throw away
the results.
"""
to_read = self.limit - self._pos
chunk = chunk_size
while to_read > 0:
chunk = min(to_read, chunk)
self.read(chunk)
to_read -= chunk
def read(self, size=None):
"""Read `size` bytes or if size is not provided everything is read.
:param size: the number of bytes read.
"""
if self._pos >= self.limit:
return self.on_exhausted()
if size is None or size == -1: # -1 is for consistence with file
size = self.limit
to_read = min(self.limit - self._pos, size)
try:
read = self._read(to_read)
except (IOError, ValueError):
return self.on_disconnect()
if to_read and len(read) != to_read:
return self.on_disconnect()
self._pos += len(read)
return read
def readline(self, size=None):
"""Reads one line from the stream."""
if self._pos >= self.limit:
return self.on_exhausted()
if size is None:
size = self.limit - self._pos
else:
size = min(size, self.limit - self._pos)
try:
line = self._readline(size)
except (ValueError, IOError):
return self.on_disconnect()
if size and not line:
return self.on_disconnect()
self._pos += len(line)
return line
def readlines(self, size=None):
"""Reads a file into a list of strings. It calls :meth:`readline`
until the file is read to the end. It does support the optional
`size` argument if the underlaying stream supports it for
`readline`.
"""
last_pos = self._pos
result = []
if size is not None:
end = min(self.limit, last_pos + size)
else:
end = self.limit
while 1:
if size is not None:
size -= last_pos - self._pos
if self._pos >= end:
break
result.append(self.readline(size))
if size is not None:
last_pos = self._pos
return result
def tell(self):
"""Returns the position of the stream.
.. versionadded:: 0.9
"""
return self._pos
def __next__(self):
line = self.readline()
if not line:
raise StopIteration()
return line
def readable(self):
return True
|