aboutsummaryrefslogtreecommitdiffstats
path: root/netlib/http/http1/read.py
blob: 6e3a1b93ab485ed1b5b1cb469a6043c9d98d4108 (plain)
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
from __future__ import absolute_import, print_function, division
import time
import sys
import re

from ... import utils
from ...exceptions import HttpReadDisconnect, HttpSyntaxException, HttpException, TcpDisconnect
from .. import Request, Response, Headers


def read_request(rfile, body_size_limit=None):
    request = read_request_head(rfile)
    expected_body_size = expected_http_body_size(request)
    request.data.content = b"".join(read_body(rfile, expected_body_size, limit=body_size_limit))
    request.timestamp_end = time.time()
    return request


def read_request_head(rfile):
    """
    Parse an HTTP request head (request line + headers) from an input stream

    Args:
        rfile: The input stream

    Returns:
        The HTTP request object (without body)

    Raises:
        HttpReadDisconnect: No bytes can be read from rfile.
        HttpSyntaxException: The input is malformed HTTP.
        HttpException: Any other error occured.
    """
    timestamp_start = time.time()
    if hasattr(rfile, "reset_timestamps"):
        rfile.reset_timestamps()

    form, method, scheme, host, port, path, http_version = _read_request_line(rfile)
    headers = _read_headers(rfile)

    if hasattr(rfile, "first_byte_timestamp"):
        # more accurate timestamp_start
        timestamp_start = rfile.first_byte_timestamp

    return Request(
        form, method, scheme, host, port, path, http_version, headers, None, timestamp_start
    )


def read_response(rfile, request, body_size_limit=None):
    response = read_response_head(rfile)
    expected_body_size = expected_http_body_size(request, response)
    response.data.content = b"".join(read_body(rfile, expected_body_size, body_size_limit))
    response.timestamp_end = time.time()
    return response


def read_response_head(rfile):
    """
    Parse an HTTP response head (response line + headers) from an input stream

    Args:
        rfile: The input stream

    Returns:
        The HTTP request object (without body)

    Raises:
        HttpReadDisconnect: No bytes can be read from rfile.
        HttpSyntaxException: The input is malformed HTTP.
        HttpException: Any other error occured.
    """

    timestamp_start = time.time()
    if hasattr(rfile, "reset_timestamps"):
        rfile.reset_timestamps()

    http_version, status_code, message = _read_response_line(rfile)
    headers = _read_headers(rfile)

    if hasattr(rfile, "first_byte_timestamp"):
        # more accurate timestamp_start
        timestamp_start = rfile.first_byte_timestamp

    return Response(http_version, status_code, message, headers, None, timestamp_start)


def read_body(rfile, expected_size, limit=None, max_chunk_size=4096):
    """
        Read an HTTP message body

        Args:
            rfile: The input stream
            expected_size: The expected body size (see :py:meth:`expected_body_size`)
            limit: Maximum body size
            max_chunk_size: Maximium chunk size that gets yielded

        Returns:
            A generator that yields byte chunks of the content.

        Raises:
            HttpException, if an error occurs

        Caveats:
            max_chunk_size is not considered if the transfer encoding is chunked.
    """
    if not limit or limit < 0:
        limit = sys.maxsize
    if not max_chunk_size:
        max_chunk_size = limit

    if expected_size is None:
        for x in _read_chunked(rfile, limit):
            yield x
    elif expected_size >= 0:
        if limit is not None and expected_size > limit:
            raise HttpException(
                "HTTP Body too large. "
                "Limit is {}, content length was advertised as {}".format(limit, expected_size)
            )
        bytes_left = expected_size
        while bytes_left:
            chunk_size = min(bytes_left, max_chunk_size)
            content = rfile.read(chunk_size)
            if len(content) < chunk_size:
                raise HttpException("Unexpected EOF")
            yield content
            bytes_left -= chunk_size
    else:
        bytes_left = limit
        while bytes_left:
            chunk_size = min(bytes_left, max_chunk_size)
            content = rfile.read(chunk_size)
            if not content:
                return
            yield content
            bytes_left -= chunk_size
        not_done = rfile.read(1)
        if not_done:
            raise HttpException("HTTP body too large. Limit is {}.".format(limit))


def connection_close(http_version, headers):
    """
        Checks the message to see if the client connection should be closed
        according to RFC 2616 Section 8.1.
    """
    # At first, check if we have an explicit Connection header.
    if "connection" in headers:
        tokens = utils.get_header_tokens(headers, "connection")
        if "close" in tokens:
            return True
        elif "keep-alive" in tokens:
            return False

    # If we don't have a Connection header, HTTP 1.1 connections are assumed to
    # be persistent
    return http_version != "HTTP/1.1" and http_version != b"HTTP/1.1"  # FIXME: Remove one case.


def expected_http_body_size(request, response=None):
    """
        Returns:
            The expected body length:
            - a positive integer, if the size is known in advance
            - None, if the size in unknown in advance (chunked encoding)
            - -1, if all data should be read until end of stream.

        Raises:
            HttpSyntaxException, if the content length header is invalid
    """
    # Determine response size according to
    # http://tools.ietf.org/html/rfc7230#section-3.3
    if not response:
        headers = request.headers
        response_code = None
        is_request = True
    else:
        headers = response.headers
        response_code = response.status_code
        is_request = False

    if is_request:
        if headers.get("expect", "").lower() == "100-continue":
            return 0
    else:
        if request.method.upper() == "HEAD":
            return 0
        if 100 <= response_code <= 199:
            return 0
        if response_code == 200 and request.method.upper() == "CONNECT":
            return 0
        if response_code in (204, 304):
            return 0

    if "chunked" in headers.get("transfer-encoding", "").lower():
        return None
    if "content-length" in headers:
        try:
            size = int(headers["content-length"])
            if size < 0:
                raise ValueError()
            return size
        except ValueError:
            raise HttpSyntaxException("Unparseable Content Length")
    if is_request:
        return 0
    return -1


def _get_first_line(rfile):
    try:
        line = rfile.readline()
        if line == b"\r\n" or line == b"\n":
            # Possible leftover from previous message
            line = rfile.readline()
    except TcpDisconnect:
        raise HttpReadDisconnect("Remote disconnected")
    if not line:
        raise HttpReadDisconnect("Remote disconnected")
    return line.strip()


def _read_request_line(rfile):
    try:
        line = _get_first_line(rfile)
    except HttpReadDisconnect:
        # We want to provide a better error message.
        raise HttpReadDisconnect("Client disconnected")

    try:
        method, path, http_version = line.split(b" ")

        if path == b"*" or path.startswith(b"/"):
            form = "relative"
            scheme, host, port = None, None, None
        elif method == b"CONNECT":
            form = "authority"
            host, port = _parse_authority_form(path)
            scheme, path = None, None
        else:
            form = "absolute"
            scheme, host, port, path = utils.parse_url(path)

        _check_http_version(http_version)
    except ValueError:
        raise HttpSyntaxException("Bad HTTP request line: {}".format(line))

    return form, method, scheme, host, port, path, http_version


def _parse_authority_form(hostport):
    """
        Returns (host, port) if hostport is a valid authority-form host specification.
        http://tools.ietf.org/html/draft-luotonen-web-proxy-tunneling-01 section 3.1

        Raises:
            ValueError, if the input is malformed
    """
    try:
        host, port = hostport.split(b":")
        port = int(port)
        if not utils.is_valid_host(host) or not utils.is_valid_port(port):
            raise ValueError()
    except ValueError:
        raise HttpSyntaxException("Invalid host specification: {}".format(hostport))

    return host, port


def _read_response_line(rfile):
    try:
        line = _get_first_line(rfile)
    except HttpReadDisconnect:
        # We want to provide a better error message.
        raise HttpReadDisconnect("Server disconnected")

    try:

        parts = line.split(b" ", 2)
        if len(parts) == 2:  # handle missing message gracefully
            parts.append(b"")

        http_version, status_code, message = parts
        status_code = int(status_code)
        _check_http_version(http_version)

    except ValueError:
        raise HttpSyntaxException("Bad HTTP response line: {}".format(line))

    return http_version, status_code, message


def _check_http_version(http_version):
    if not re.match(br"^HTTP/\d\.\d$", http_version):
        raise HttpSyntaxException("Unknown HTTP version: {}".format(http_version))


def _read_headers(rfile):
    """
        Read a set of headers.
        Stop once a blank line is reached.

        Returns:
            A headers object

        Raises:
            HttpSyntaxException
    """
    ret = []
    while True:
        line = rfile.readline()
        if not line or line == b"\r\n" or line == b"\n":
            break
        if line[0] in b" \t":
            if not ret:
                raise HttpSyntaxException("Invalid headers")
            # continued header
            ret[-1][1] = ret[-1][1] + b'\r\n ' + line.strip()
        else:
            try:
                name, value = line.split(b":", 1)
                value = value.strip()
                if not name:
                    raise ValueError()
                ret.append([name, value])
            except ValueError:
                raise HttpSyntaxException("Invalid headers")
    return Headers(ret)


def _read_chunked(rfile, limit=sys.maxsize):
    """
    Read a HTTP body with chunked transfer encoding.

    Args:
        rfile: the input file
        limit: A positive integer
    """
    total = 0
    while True:
        line = rfile.readline(128)
        if line == b"":
            raise HttpException("Connection closed prematurely")
        if line != b"\r\n" and line != b"\n":
            try:
                length = int(line, 16)
            except ValueError:
                raise HttpSyntaxException("Invalid chunked encoding length: {}".format(line))
            total += length
            if total > limit:
                raise HttpException(
                    "HTTP Body too large. Limit is {}, "
                    "chunked content longer than {}".format(limit, total)
                )
            chunk = rfile.read(length)
            suffix = rfile.readline(5)
            if suffix != b"\r\n":
                raise HttpSyntaxException("Malformed chunked body")
            if length == 0:
                return
            yield chunk