bzr branch
http://gegoxaren.bato24.eu/bzr/brz/remove-bazaar
| 
2018.5.157
by Andrew Bennetts
 Remove unnecessary trivial divergences from bzr.dev.  | 
1  | 
# Copyright (C) 2006, 2007 Canonical Ltd
 | 
| 
2018.5.3
by Andrew Bennetts
 Split up more smart server code, this time into bzrlib/transport/smart/protocol.py  | 
2  | 
#
 | 
3  | 
# This program is free software; you can redistribute it and/or modify
 | 
|
4  | 
# it under the terms of the GNU General Public License as published by
 | 
|
5  | 
# the Free Software Foundation; either version 2 of the License, or
 | 
|
6  | 
# (at your option) any later version.
 | 
|
7  | 
#
 | 
|
8  | 
# This program is distributed in the hope that it will be useful,
 | 
|
9  | 
# but WITHOUT ANY WARRANTY; without even the implied warranty of
 | 
|
10  | 
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
 | 
|
11  | 
# GNU General Public License for more details.
 | 
|
12  | 
#
 | 
|
13  | 
# You should have received a copy of the GNU General Public License
 | 
|
14  | 
# along with this program; if not, write to the Free Software
 | 
|
15  | 
# Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
 | 
|
16  | 
||
| 
2018.5.19
by Andrew Bennetts
 Add docstrings to all the new modules, and a few other places.  | 
17  | 
"""Wire-level encoding and decoding of requests and responses for the smart
 | 
18  | 
client and server.
 | 
|
19  | 
"""
 | 
|
20  | 
||
21  | 
||
| 
2018.5.3
by Andrew Bennetts
 Split up more smart server code, this time into bzrlib/transport/smart/protocol.py  | 
22  | 
from cStringIO import StringIO  | 
23  | 
||
24  | 
from bzrlib import errors  | 
|
| 
2018.5.23
by Andrew Bennetts
 Use a Registry for smart server command handlers.  | 
25  | 
from bzrlib.smart import request  | 
| 
2018.5.3
by Andrew Bennetts
 Split up more smart server code, this time into bzrlib/transport/smart/protocol.py  | 
26  | 
|
27  | 
||
| 
2432.2.7
by Andrew Bennetts
 Use less confusing version strings, and define REQUEST_VERSION_TWO/RESPONSE_VERSION_TWO constants for them.  | 
28  | 
# Protocol version strings.  These are sent as prefixes of bzr requests and
 | 
29  | 
# responses to identify the protocol version being used. (There are no version
 | 
|
30  | 
# one strings because that version doesn't send any).
 | 
|
31  | 
REQUEST_VERSION_TWO = 'bzr request 2\n'  | 
|
32  | 
RESPONSE_VERSION_TWO = 'bzr response 2\n'  | 
|
33  | 
||
34  | 
||
| 
2018.5.3
by Andrew Bennetts
 Split up more smart server code, this time into bzrlib/transport/smart/protocol.py  | 
35  | 
def _recv_tuple(from_file):  | 
36  | 
req_line = from_file.readline()  | 
|
37  | 
return _decode_tuple(req_line)  | 
|
38  | 
||
39  | 
||
40  | 
def _decode_tuple(req_line):  | 
|
41  | 
if req_line == None or req_line == '':  | 
|
42  | 
return None  | 
|
43  | 
if req_line[-1] != '\n':  | 
|
44  | 
raise errors.SmartProtocolError("request %r not terminated" % req_line)  | 
|
45  | 
return tuple(req_line[:-1].split('\x01'))  | 
|
46  | 
||
47  | 
||
48  | 
def _encode_tuple(args):  | 
|
49  | 
"""Encode the tuple args to a bytestream."""  | 
|
50  | 
return '\x01'.join(args) + '\n'  | 
|
51  | 
||
52  | 
||
53  | 
class SmartProtocolBase(object):  | 
|
54  | 
"""Methods common to client and server"""  | 
|
55  | 
||
56  | 
    # TODO: this only actually accomodates a single block; possibly should
 | 
|
57  | 
    # support multiple chunks?
 | 
|
58  | 
def _encode_bulk_data(self, body):  | 
|
59  | 
"""Encode body as a bulk data chunk."""  | 
|
60  | 
return ''.join(('%d\n' % len(body), body, 'done\n'))  | 
|
61  | 
||
62  | 
def _serialise_offsets(self, offsets):  | 
|
63  | 
"""Serialise a readv offset list."""  | 
|
64  | 
txt = []  | 
|
65  | 
for start, length in offsets:  | 
|
66  | 
txt.append('%d,%d' % (start, length))  | 
|
67  | 
return '\n'.join(txt)  | 
|
68  | 
||
69  | 
||
70  | 
class SmartServerRequestProtocolOne(SmartProtocolBase):  | 
|
71  | 
"""Server-side encoding and decoding logic for smart version 1."""  | 
|
72  | 
||
73  | 
def __init__(self, backing_transport, write_func):  | 
|
74  | 
self._backing_transport = backing_transport  | 
|
75  | 
self.excess_buffer = ''  | 
|
76  | 
self._finished = False  | 
|
77  | 
self.in_buffer = ''  | 
|
78  | 
self.has_dispatched = False  | 
|
79  | 
self.request = None  | 
|
80  | 
self._body_decoder = None  | 
|
81  | 
self._write_func = write_func  | 
|
82  | 
||
83  | 
def accept_bytes(self, bytes):  | 
|
84  | 
"""Take bytes, and advance the internal state machine appropriately.  | 
|
85  | 
        
 | 
|
86  | 
        :param bytes: must be a byte string
 | 
|
87  | 
        """
 | 
|
88  | 
assert isinstance(bytes, str)  | 
|
89  | 
self.in_buffer += bytes  | 
|
90  | 
if not self.has_dispatched:  | 
|
91  | 
if '\n' not in self.in_buffer:  | 
|
92  | 
                # no command line yet
 | 
|
93  | 
                return
 | 
|
94  | 
self.has_dispatched = True  | 
|
95  | 
try:  | 
|
96  | 
first_line, self.in_buffer = self.in_buffer.split('\n', 1)  | 
|
97  | 
first_line += '\n'  | 
|
98  | 
req_args = _decode_tuple(first_line)  | 
|
| 
2018.5.14
by Andrew Bennetts
 Move SmartTCPServer to smart/server.py, and SmartServerRequestHandler to smart/request.py.  | 
99  | 
self.request = request.SmartServerRequestHandler(  | 
| 
2018.5.23
by Andrew Bennetts
 Use a Registry for smart server command handlers.  | 
100  | 
self._backing_transport, commands=request.request_handlers)  | 
| 
2018.5.3
by Andrew Bennetts
 Split up more smart server code, this time into bzrlib/transport/smart/protocol.py  | 
101  | 
self.request.dispatch_command(req_args[0], req_args[1:])  | 
102  | 
if self.request.finished_reading:  | 
|
103  | 
                    # trivial request
 | 
|
104  | 
self.excess_buffer = self.in_buffer  | 
|
105  | 
self.in_buffer = ''  | 
|
| 
2432.4.3
by Robert Collins
 Refactor the HPSS Response code to take SmartServerResponse rather than args and body.  | 
106  | 
self._send_response(self.request.response)  | 
| 
2018.5.3
by Andrew Bennetts
 Split up more smart server code, this time into bzrlib/transport/smart/protocol.py  | 
107  | 
except KeyboardInterrupt:  | 
108  | 
                raise
 | 
|
109  | 
except Exception, exception:  | 
|
110  | 
                # everything else: pass to client, flush, and quit
 | 
|
| 
2432.4.3
by Robert Collins
 Refactor the HPSS Response code to take SmartServerResponse rather than args and body.  | 
111  | 
self._send_response(request.FailedSmartServerResponse(  | 
112  | 
('error', str(exception))))  | 
|
| 
2018.5.3
by Andrew Bennetts
 Split up more smart server code, this time into bzrlib/transport/smart/protocol.py  | 
113  | 
                return
 | 
114  | 
||
115  | 
if self.has_dispatched:  | 
|
116  | 
if self._finished:  | 
|
117  | 
                # nothing to do.XXX: this routine should be a single state 
 | 
|
118  | 
                # machine too.
 | 
|
119  | 
self.excess_buffer += self.in_buffer  | 
|
120  | 
self.in_buffer = ''  | 
|
121  | 
                return
 | 
|
122  | 
if self._body_decoder is None:  | 
|
123  | 
self._body_decoder = LengthPrefixedBodyDecoder()  | 
|
124  | 
self._body_decoder.accept_bytes(self.in_buffer)  | 
|
125  | 
self.in_buffer = self._body_decoder.unused_data  | 
|
126  | 
body_data = self._body_decoder.read_pending_data()  | 
|
127  | 
self.request.accept_body(body_data)  | 
|
128  | 
if self._body_decoder.finished_reading:  | 
|
129  | 
self.request.end_of_body()  | 
|
130  | 
assert self.request.finished_reading, \  | 
|
131  | 
                    "no more body, request not finished"
 | 
|
132  | 
if self.request.response is not None:  | 
|
| 
2432.4.3
by Robert Collins
 Refactor the HPSS Response code to take SmartServerResponse rather than args and body.  | 
133  | 
self._send_response(self.request.response)  | 
| 
2018.5.3
by Andrew Bennetts
 Split up more smart server code, this time into bzrlib/transport/smart/protocol.py  | 
134  | 
self.excess_buffer = self.in_buffer  | 
135  | 
self.in_buffer = ''  | 
|
136  | 
else:  | 
|
137  | 
assert not self.request.finished_reading, \  | 
|
138  | 
                    "no response and we have finished reading."
 | 
|
139  | 
||
| 
2432.4.3
by Robert Collins
 Refactor the HPSS Response code to take SmartServerResponse rather than args and body.  | 
140  | 
def _send_response(self, response):  | 
| 
2018.5.3
by Andrew Bennetts
 Split up more smart server code, this time into bzrlib/transport/smart/protocol.py  | 
141  | 
"""Send a smart server response down the output stream."""  | 
142  | 
assert not self._finished, 'response already sent'  | 
|
| 
2432.4.3
by Robert Collins
 Refactor the HPSS Response code to take SmartServerResponse rather than args and body.  | 
143  | 
args = response.args  | 
144  | 
body = response.body  | 
|
| 
2018.5.3
by Andrew Bennetts
 Split up more smart server code, this time into bzrlib/transport/smart/protocol.py  | 
145  | 
self._finished = True  | 
| 
2432.2.1
by Andrew Bennetts
 Add Smart{Client,Server}RequestProtocolTwo, that prefix args tuples with a version marker.  | 
146  | 
self._write_protocol_version()  | 
| 
2432.4.6
by Robert Collins
 Include success/failure feedback in SmartProtocolTwo responses to allow robust handling in the future.  | 
147  | 
self._write_success_or_failure_prefix(response)  | 
| 
2018.5.3
by Andrew Bennetts
 Split up more smart server code, this time into bzrlib/transport/smart/protocol.py  | 
148  | 
self._write_func(_encode_tuple(args))  | 
149  | 
if body is not None:  | 
|
150  | 
assert isinstance(body, str), 'body must be a str'  | 
|
151  | 
bytes = self._encode_bulk_data(body)  | 
|
152  | 
self._write_func(bytes)  | 
|
153  | 
||
| 
2432.2.1
by Andrew Bennetts
 Add Smart{Client,Server}RequestProtocolTwo, that prefix args tuples with a version marker.  | 
154  | 
def _write_protocol_version(self):  | 
155  | 
"""Write any prefixes this protocol requires.  | 
|
156  | 
        
 | 
|
157  | 
        Version one doesn't send protocol versions.
 | 
|
158  | 
        """
 | 
|
159  | 
||
| 
2432.4.6
by Robert Collins
 Include success/failure feedback in SmartProtocolTwo responses to allow robust handling in the future.  | 
160  | 
def _write_success_or_failure_prefix(self, response):  | 
161  | 
"""Write the protocol specific success/failure prefix.  | 
|
162  | 
||
163  | 
        For SmartServerRequestProtocolOne this is omitted but we
 | 
|
164  | 
        call is_successful to ensure that the response is valid.
 | 
|
165  | 
        """
 | 
|
166  | 
response.is_successful()  | 
|
167  | 
||
| 
2018.5.3
by Andrew Bennetts
 Split up more smart server code, this time into bzrlib/transport/smart/protocol.py  | 
168  | 
def next_read_size(self):  | 
169  | 
if self._finished:  | 
|
170  | 
return 0  | 
|
171  | 
if self._body_decoder is None:  | 
|
172  | 
return 1  | 
|
173  | 
else:  | 
|
174  | 
return self._body_decoder.next_read_size()  | 
|
175  | 
||
176  | 
||
| 
2432.2.1
by Andrew Bennetts
 Add Smart{Client,Server}RequestProtocolTwo, that prefix args tuples with a version marker.  | 
177  | 
class SmartServerRequestProtocolTwo(SmartServerRequestProtocolOne):  | 
178  | 
r"""Version two of the server side of the smart protocol.  | 
|
179  | 
   
 | 
|
| 
2432.2.7
by Andrew Bennetts
 Use less confusing version strings, and define REQUEST_VERSION_TWO/RESPONSE_VERSION_TWO constants for them.  | 
180  | 
    This prefixes responses with the value of RESPONSE_VERSION_TWO.
 | 
| 
2432.2.1
by Andrew Bennetts
 Add Smart{Client,Server}RequestProtocolTwo, that prefix args tuples with a version marker.  | 
181  | 
    """
 | 
182  | 
||
| 
2432.4.6
by Robert Collins
 Include success/failure feedback in SmartProtocolTwo responses to allow robust handling in the future.  | 
183  | 
def _write_success_or_failure_prefix(self, response):  | 
184  | 
"""Write the protocol specific success/failure prefix."""  | 
|
185  | 
if response.is_successful():  | 
|
186  | 
self._write_func('success\n')  | 
|
187  | 
else:  | 
|
188  | 
self._write_func('failed\n')  | 
|
189  | 
||
| 
2432.2.1
by Andrew Bennetts
 Add Smart{Client,Server}RequestProtocolTwo, that prefix args tuples with a version marker.  | 
190  | 
def _write_protocol_version(self):  | 
191  | 
r"""Write any prefixes this protocol requires.  | 
|
192  | 
        
 | 
|
| 
2432.2.7
by Andrew Bennetts
 Use less confusing version strings, and define REQUEST_VERSION_TWO/RESPONSE_VERSION_TWO constants for them.  | 
193  | 
        Version two sends the value of RESPONSE_VERSION_TWO.
 | 
| 
2432.2.1
by Andrew Bennetts
 Add Smart{Client,Server}RequestProtocolTwo, that prefix args tuples with a version marker.  | 
194  | 
        """
 | 
| 
2432.2.7
by Andrew Bennetts
 Use less confusing version strings, and define REQUEST_VERSION_TWO/RESPONSE_VERSION_TWO constants for them.  | 
195  | 
self._write_func(RESPONSE_VERSION_TWO)  | 
| 
2432.2.1
by Andrew Bennetts
 Add Smart{Client,Server}RequestProtocolTwo, that prefix args tuples with a version marker.  | 
196  | 
|
197  | 
||
| 
2018.5.3
by Andrew Bennetts
 Split up more smart server code, this time into bzrlib/transport/smart/protocol.py  | 
198  | 
class LengthPrefixedBodyDecoder(object):  | 
199  | 
"""Decodes the length-prefixed bulk data."""  | 
|
200  | 
||
201  | 
def __init__(self):  | 
|
202  | 
self.bytes_left = None  | 
|
203  | 
self.finished_reading = False  | 
|
204  | 
self.unused_data = ''  | 
|
205  | 
self.state_accept = self._state_accept_expecting_length  | 
|
206  | 
self.state_read = self._state_read_no_data  | 
|
207  | 
self._in_buffer = ''  | 
|
208  | 
self._trailer_buffer = ''  | 
|
209  | 
||
210  | 
def accept_bytes(self, bytes):  | 
|
211  | 
"""Decode as much of bytes as possible.  | 
|
212  | 
||
213  | 
        If 'bytes' contains too much data it will be appended to
 | 
|
214  | 
        self.unused_data.
 | 
|
215  | 
||
216  | 
        finished_reading will be set when no more data is required.  Further
 | 
|
217  | 
        data will be appended to self.unused_data.
 | 
|
218  | 
        """
 | 
|
219  | 
        # accept_bytes is allowed to change the state
 | 
|
220  | 
current_state = self.state_accept  | 
|
221  | 
self.state_accept(bytes)  | 
|
222  | 
while current_state != self.state_accept:  | 
|
223  | 
current_state = self.state_accept  | 
|
224  | 
self.state_accept('')  | 
|
225  | 
||
226  | 
def next_read_size(self):  | 
|
227  | 
if self.bytes_left is not None:  | 
|
228  | 
            # Ideally we want to read all the remainder of the body and the
 | 
|
229  | 
            # trailer in one go.
 | 
|
230  | 
return self.bytes_left + 5  | 
|
231  | 
elif self.state_accept == self._state_accept_reading_trailer:  | 
|
232  | 
            # Just the trailer left
 | 
|
233  | 
return 5 - len(self._trailer_buffer)  | 
|
234  | 
elif self.state_accept == self._state_accept_expecting_length:  | 
|
235  | 
            # There's still at least 6 bytes left ('\n' to end the length, plus
 | 
|
236  | 
            # 'done\n').
 | 
|
237  | 
return 6  | 
|
238  | 
else:  | 
|
239  | 
            # Reading excess data.  Either way, 1 byte at a time is fine.
 | 
|
240  | 
return 1  | 
|
241  | 
||
242  | 
def read_pending_data(self):  | 
|
243  | 
"""Return any pending data that has been decoded."""  | 
|
244  | 
return self.state_read()  | 
|
245  | 
||
246  | 
def _state_accept_expecting_length(self, bytes):  | 
|
247  | 
self._in_buffer += bytes  | 
|
248  | 
pos = self._in_buffer.find('\n')  | 
|
249  | 
if pos == -1:  | 
|
250  | 
            return
 | 
|
251  | 
self.bytes_left = int(self._in_buffer[:pos])  | 
|
252  | 
self._in_buffer = self._in_buffer[pos+1:]  | 
|
253  | 
self.bytes_left -= len(self._in_buffer)  | 
|
254  | 
self.state_accept = self._state_accept_reading_body  | 
|
255  | 
self.state_read = self._state_read_in_buffer  | 
|
256  | 
||
257  | 
def _state_accept_reading_body(self, bytes):  | 
|
258  | 
self._in_buffer += bytes  | 
|
259  | 
self.bytes_left -= len(bytes)  | 
|
260  | 
if self.bytes_left <= 0:  | 
|
261  | 
            # Finished with body
 | 
|
262  | 
if self.bytes_left != 0:  | 
|
263  | 
self._trailer_buffer = self._in_buffer[self.bytes_left:]  | 
|
264  | 
self._in_buffer = self._in_buffer[:self.bytes_left]  | 
|
265  | 
self.bytes_left = None  | 
|
266  | 
self.state_accept = self._state_accept_reading_trailer  | 
|
267  | 
||
268  | 
def _state_accept_reading_trailer(self, bytes):  | 
|
269  | 
self._trailer_buffer += bytes  | 
|
270  | 
        # TODO: what if the trailer does not match "done\n"?  Should this raise
 | 
|
271  | 
        # a ProtocolViolation exception?
 | 
|
272  | 
if self._trailer_buffer.startswith('done\n'):  | 
|
273  | 
self.unused_data = self._trailer_buffer[len('done\n'):]  | 
|
274  | 
self.state_accept = self._state_accept_reading_unused  | 
|
275  | 
self.finished_reading = True  | 
|
276  | 
||
277  | 
def _state_accept_reading_unused(self, bytes):  | 
|
278  | 
self.unused_data += bytes  | 
|
279  | 
||
280  | 
def _state_read_no_data(self):  | 
|
281  | 
return ''  | 
|
282  | 
||
283  | 
def _state_read_in_buffer(self):  | 
|
284  | 
result = self._in_buffer  | 
|
285  | 
self._in_buffer = ''  | 
|
286  | 
return result  | 
|
287  | 
||
288  | 
||
289  | 
class SmartClientRequestProtocolOne(SmartProtocolBase):  | 
|
290  | 
"""The client-side protocol for smart version 1."""  | 
|
291  | 
||
292  | 
def __init__(self, request):  | 
|
293  | 
"""Construct a SmartClientRequestProtocolOne.  | 
|
294  | 
||
295  | 
        :param request: A SmartClientMediumRequest to serialise onto and
 | 
|
296  | 
            deserialise from.
 | 
|
297  | 
        """
 | 
|
298  | 
self._request = request  | 
|
299  | 
self._body_buffer = None  | 
|
300  | 
||
301  | 
def call(self, *args):  | 
|
| 
2432.2.1
by Andrew Bennetts
 Add Smart{Client,Server}RequestProtocolTwo, that prefix args tuples with a version marker.  | 
302  | 
self._write_args(args)  | 
| 
2018.5.3
by Andrew Bennetts
 Split up more smart server code, this time into bzrlib/transport/smart/protocol.py  | 
303  | 
self._request.finished_writing()  | 
304  | 
||
305  | 
def call_with_body_bytes(self, args, body):  | 
|
306  | 
"""Make a remote call of args with body bytes 'body'.  | 
|
307  | 
||
308  | 
        After calling this, call read_response_tuple to find the result out.
 | 
|
309  | 
        """
 | 
|
| 
2432.2.1
by Andrew Bennetts
 Add Smart{Client,Server}RequestProtocolTwo, that prefix args tuples with a version marker.  | 
310  | 
self._write_args(args)  | 
| 
2018.5.3
by Andrew Bennetts
 Split up more smart server code, this time into bzrlib/transport/smart/protocol.py  | 
311  | 
bytes = self._encode_bulk_data(body)  | 
312  | 
self._request.accept_bytes(bytes)  | 
|
313  | 
self._request.finished_writing()  | 
|
314  | 
||
315  | 
def call_with_body_readv_array(self, args, body):  | 
|
316  | 
"""Make a remote call with a readv array.  | 
|
317  | 
||
318  | 
        The body is encoded with one line per readv offset pair. The numbers in
 | 
|
319  | 
        each pair are separated by a comma, and no trailing \n is emitted.
 | 
|
320  | 
        """
 | 
|
| 
2432.2.1
by Andrew Bennetts
 Add Smart{Client,Server}RequestProtocolTwo, that prefix args tuples with a version marker.  | 
321  | 
self._write_args(args)  | 
| 
2018.5.3
by Andrew Bennetts
 Split up more smart server code, this time into bzrlib/transport/smart/protocol.py  | 
322  | 
readv_bytes = self._serialise_offsets(body)  | 
323  | 
bytes = self._encode_bulk_data(readv_bytes)  | 
|
324  | 
self._request.accept_bytes(bytes)  | 
|
325  | 
self._request.finished_writing()  | 
|
326  | 
||
327  | 
def cancel_read_body(self):  | 
|
328  | 
"""After expecting a body, a response code may indicate one otherwise.  | 
|
329  | 
||
330  | 
        This method lets the domain client inform the protocol that no body
 | 
|
331  | 
        will be transmitted. This is a terminal method: after calling it the
 | 
|
332  | 
        protocol is not able to be used further.
 | 
|
333  | 
        """
 | 
|
334  | 
self._request.finished_reading()  | 
|
335  | 
||
336  | 
def read_response_tuple(self, expect_body=False):  | 
|
337  | 
"""Read a response tuple from the wire.  | 
|
338  | 
||
339  | 
        This should only be called once.
 | 
|
340  | 
        """
 | 
|
341  | 
result = self._recv_tuple()  | 
|
342  | 
if not expect_body:  | 
|
343  | 
self._request.finished_reading()  | 
|
344  | 
return result  | 
|
345  | 
||
346  | 
def read_body_bytes(self, count=-1):  | 
|
347  | 
"""Read bytes from the body, decoding into a byte stream.  | 
|
348  | 
        
 | 
|
349  | 
        We read all bytes at once to ensure we've checked the trailer for 
 | 
|
350  | 
        errors, and then feed the buffer back as read_body_bytes is called.
 | 
|
351  | 
        """
 | 
|
352  | 
if self._body_buffer is not None:  | 
|
353  | 
return self._body_buffer.read(count)  | 
|
354  | 
_body_decoder = LengthPrefixedBodyDecoder()  | 
|
355  | 
||
356  | 
while not _body_decoder.finished_reading:  | 
|
357  | 
bytes_wanted = _body_decoder.next_read_size()  | 
|
358  | 
bytes = self._request.read_bytes(bytes_wanted)  | 
|
359  | 
_body_decoder.accept_bytes(bytes)  | 
|
360  | 
self._request.finished_reading()  | 
|
361  | 
self._body_buffer = StringIO(_body_decoder.read_pending_data())  | 
|
362  | 
        # XXX: TODO check the trailer result.
 | 
|
363  | 
return self._body_buffer.read(count)  | 
|
364  | 
||
365  | 
def _recv_tuple(self):  | 
|
366  | 
"""Receive a tuple from the medium request."""  | 
|
| 
2432.4.6
by Robert Collins
 Include success/failure feedback in SmartProtocolTwo responses to allow robust handling in the future.  | 
367  | 
return _decode_tuple(self._recv_line())  | 
368  | 
||
369  | 
def _recv_line(self):  | 
|
370  | 
"""Read an entire line from the medium request."""  | 
|
| 
2018.5.3
by Andrew Bennetts
 Split up more smart server code, this time into bzrlib/transport/smart/protocol.py  | 
371  | 
line = ''  | 
372  | 
while not line or line[-1] != '\n':  | 
|
373  | 
            # TODO: this is inefficient - but tuples are short.
 | 
|
374  | 
new_char = self._request.read_bytes(1)  | 
|
375  | 
line += new_char  | 
|
376  | 
assert new_char != '', "end of file reading from server."  | 
|
| 
2432.4.6
by Robert Collins
 Include success/failure feedback in SmartProtocolTwo responses to allow robust handling in the future.  | 
377  | 
return line  | 
| 
2018.5.3
by Andrew Bennetts
 Split up more smart server code, this time into bzrlib/transport/smart/protocol.py  | 
378  | 
|
379  | 
def query_version(self):  | 
|
380  | 
"""Return protocol version number of the server."""  | 
|
381  | 
self.call('hello')  | 
|
382  | 
resp = self.read_response_tuple()  | 
|
383  | 
if resp == ('ok', '1'):  | 
|
384  | 
return 1  | 
|
| 
2432.2.1
by Andrew Bennetts
 Add Smart{Client,Server}RequestProtocolTwo, that prefix args tuples with a version marker.  | 
385  | 
elif resp == ('ok', '2'):  | 
386  | 
return 2  | 
|
| 
2018.5.3
by Andrew Bennetts
 Split up more smart server code, this time into bzrlib/transport/smart/protocol.py  | 
387  | 
else:  | 
388  | 
raise errors.SmartProtocolError("bad response %r" % (resp,))  | 
|
389  | 
||
| 
2432.2.1
by Andrew Bennetts
 Add Smart{Client,Server}RequestProtocolTwo, that prefix args tuples with a version marker.  | 
390  | 
def _write_args(self, args):  | 
391  | 
self._write_protocol_version()  | 
|
392  | 
bytes = _encode_tuple(args)  | 
|
393  | 
self._request.accept_bytes(bytes)  | 
|
394  | 
||
395  | 
def _write_protocol_version(self):  | 
|
396  | 
"""Write any prefixes this protocol requires.  | 
|
397  | 
        
 | 
|
398  | 
        Version one doesn't send protocol versions.
 | 
|
399  | 
        """
 | 
|
400  | 
||
401  | 
||
402  | 
class SmartClientRequestProtocolTwo(SmartClientRequestProtocolOne):  | 
|
| 
2432.2.7
by Andrew Bennetts
 Use less confusing version strings, and define REQUEST_VERSION_TWO/RESPONSE_VERSION_TWO constants for them.  | 
403  | 
"""Version two of the client side of the smart protocol.  | 
| 
2432.2.1
by Andrew Bennetts
 Add Smart{Client,Server}RequestProtocolTwo, that prefix args tuples with a version marker.  | 
404  | 
    
 | 
| 
2432.2.7
by Andrew Bennetts
 Use less confusing version strings, and define REQUEST_VERSION_TWO/RESPONSE_VERSION_TWO constants for them.  | 
405  | 
    This prefixes the request with the value of REQUEST_VERSION_TWO.
 | 
| 
2432.2.1
by Andrew Bennetts
 Add Smart{Client,Server}RequestProtocolTwo, that prefix args tuples with a version marker.  | 
406  | 
    """
 | 
407  | 
||
408  | 
def read_response_tuple(self, expect_body=False):  | 
|
409  | 
"""Read a response tuple from the wire.  | 
|
410  | 
||
411  | 
        This should only be called once.
 | 
|
412  | 
        """
 | 
|
| 
2432.2.7
by Andrew Bennetts
 Use less confusing version strings, and define REQUEST_VERSION_TWO/RESPONSE_VERSION_TWO constants for them.  | 
413  | 
version = self._request.read_line()  | 
414  | 
if version != RESPONSE_VERSION_TWO:  | 
|
| 
2432.2.1
by Andrew Bennetts
 Add Smart{Client,Server}RequestProtocolTwo, that prefix args tuples with a version marker.  | 
415  | 
raise errors.SmartProtocolError('bad protocol marker %r' % version)  | 
| 
2432.4.6
by Robert Collins
 Include success/failure feedback in SmartProtocolTwo responses to allow robust handling in the future.  | 
416  | 
response_status = self._recv_line()  | 
417  | 
if response_status not in ('success\n', 'failed\n'):  | 
|
418  | 
raise errors.SmartProtocolError(  | 
|
419  | 
'bad protocol status %r' % response_status)  | 
|
420  | 
self.response_status = response_status == 'success\n'  | 
|
| 
2432.2.1
by Andrew Bennetts
 Add Smart{Client,Server}RequestProtocolTwo, that prefix args tuples with a version marker.  | 
421  | 
return SmartClientRequestProtocolOne.read_response_tuple(self, expect_body)  | 
422  | 
||
423  | 
def _write_protocol_version(self):  | 
|
424  | 
r"""Write any prefixes this protocol requires.  | 
|
425  | 
        
 | 
|
| 
2432.2.7
by Andrew Bennetts
 Use less confusing version strings, and define REQUEST_VERSION_TWO/RESPONSE_VERSION_TWO constants for them.  | 
426  | 
        Version two sends the value of REQUEST_VERSION_TWO.
 | 
| 
2432.2.1
by Andrew Bennetts
 Add Smart{Client,Server}RequestProtocolTwo, that prefix args tuples with a version marker.  | 
427  | 
        """
 | 
| 
2432.2.7
by Andrew Bennetts
 Use less confusing version strings, and define REQUEST_VERSION_TWO/RESPONSE_VERSION_TWO constants for them.  | 
428  | 
self._request.accept_bytes(REQUEST_VERSION_TWO)  | 
| 
2018.5.3
by Andrew Bennetts
 Split up more smart server code, this time into bzrlib/transport/smart/protocol.py  | 
429  |