1
# Copyright (C) 2007, 2009, 2010 Canonical Ltd
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.
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.
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., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
17
"""Container format for Bazaar data.
19
"Containers" and "records" are described in
20
doc/developers/container-format.txt.
23
from __future__ import absolute_import
25
from cStringIO import StringIO
28
from bzrlib import errors
31
FORMAT_ONE = "Bazaar pack format 1 (introduced in 0.18)"
34
_whitespace_re = re.compile('[\t\n\x0b\x0c\r ]')
37
def _check_name(name):
38
"""Do some basic checking of 'name'.
40
At the moment, this just checks that there are no whitespace characters in a
43
:raises InvalidRecordError: if name is not valid.
44
:seealso: _check_name_encoding
46
if _whitespace_re.search(name) is not None:
47
raise errors.InvalidRecordError("%r is not a valid name." % (name,))
50
def _check_name_encoding(name):
51
"""Check that 'name' is valid UTF-8.
53
This is separate from _check_name because UTF-8 decoding is relatively
54
expensive, and we usually want to avoid it.
56
:raises InvalidRecordError: if name is not valid UTF-8.
60
except UnicodeDecodeError, e:
61
raise errors.InvalidRecordError(str(e))
64
class ContainerSerialiser(object):
65
"""A helper class for serialising containers.
67
It simply returns bytes from method calls to 'begin', 'end' and
68
'bytes_record'. You may find ContainerWriter to be a more convenient
73
"""Return the bytes to begin a container."""
74
return FORMAT_ONE + "\n"
77
"""Return the bytes to finish a container."""
80
def bytes_header(self, length, names):
81
"""Return the header for a Bytes record."""
85
byte_sections.append(str(length) + "\n")
87
for name_tuple in names:
88
# Make sure we're writing valid names. Note that we will leave a
89
# half-written record if a name is bad!
90
for name in name_tuple:
92
byte_sections.append('\x00'.join(name_tuple) + "\n")
94
byte_sections.append("\n")
95
return ''.join(byte_sections)
97
def bytes_record(self, bytes, names):
98
"""Return the bytes for a Bytes record with the given name and
101
If the content may be large, construct the header separately and then
102
stream out the contents.
104
return self.bytes_header(len(bytes), names) + bytes
107
class ContainerWriter(object):
108
"""A class for writing containers to a file.
110
:attribute records_written: The number of user records added to the
111
container. This does not count the prelude or suffix of the container
112
introduced by the begin() and end() methods.
115
# Join up headers with the body if writing fewer than this many bytes:
116
# trades off memory usage and copying to do less IO ops.
117
_JOIN_WRITES_THRESHOLD = 100000
119
def __init__(self, write_func):
122
:param write_func: a callable that will be called when this
123
ContainerWriter needs to write some bytes.
125
self._write_func = write_func
126
self.current_offset = 0
127
self.records_written = 0
128
self._serialiser = ContainerSerialiser()
131
"""Begin writing a container."""
132
self.write_func(self._serialiser.begin())
134
def write_func(self, bytes):
135
self._write_func(bytes)
136
self.current_offset += len(bytes)
139
"""Finish writing a container."""
140
self.write_func(self._serialiser.end())
142
def add_bytes_record(self, bytes, names):
143
"""Add a Bytes record with the given names.
145
:param bytes: The bytes to insert.
146
:param names: The names to give the inserted bytes. Each name is
147
a tuple of bytestrings. The bytestrings may not contain
149
:return: An offset, length tuple. The offset is the offset
150
of the record within the container, and the length is the
151
length of data that will need to be read to reconstitute the
152
record. These offset and length can only be used with the pack
153
interface - they might be offset by headers or other such details
154
and thus are only suitable for use by a ContainerReader.
156
current_offset = self.current_offset
158
if length < self._JOIN_WRITES_THRESHOLD:
159
self.write_func(self._serialiser.bytes_header(length, names)
162
self.write_func(self._serialiser.bytes_header(length, names))
163
self.write_func(bytes)
164
self.records_written += 1
165
# return a memo of where we wrote data to allow random access.
166
return current_offset, self.current_offset - current_offset
169
class ReadVFile(object):
170
"""Adapt a readv result iterator to a file like protocol.
172
The readv result must support the iterator protocol returning (offset,
176
# XXX: This could be a generic transport class, as other code may want to
177
# gradually consume the readv result.
179
def __init__(self, readv_result):
180
"""Construct a new ReadVFile wrapper.
182
:seealso: make_readv_reader
184
:param readv_result: the most recent readv result - list or generator
186
# readv can return a sequence or an iterator, but we require an
187
# iterator to know how much has been consumed.
188
readv_result = iter(readv_result)
189
self.readv_result = readv_result
193
if (self._string is None or
194
self._string.tell() == self._string_length):
195
offset, data = self.readv_result.next()
196
self._string_length = len(data)
197
self._string = StringIO(data)
199
def read(self, length):
201
result = self._string.read(length)
202
if len(result) < length:
203
raise errors.BzrError('wanted %d bytes but next '
204
'hunk only contains %d: %r...' %
205
(length, len(result), result[:20]))
209
"""Note that readline will not cross readv segments."""
211
result = self._string.readline()
212
if self._string.tell() == self._string_length and result[-1] != '\n':
213
raise errors.BzrError('short readline in the readvfile hunk: %r'
218
def make_readv_reader(transport, filename, requested_records):
219
"""Create a ContainerReader that will read selected records only.
221
:param transport: The transport the pack file is located on.
222
:param filename: The filename of the pack file.
223
:param requested_records: The record offset, length tuples as returned
224
by add_bytes_record for the desired records.
226
readv_blocks = [(0, len(FORMAT_ONE)+1)]
227
readv_blocks.extend(requested_records)
228
result = ContainerReader(ReadVFile(
229
transport.readv(filename, readv_blocks)))
233
class BaseReader(object):
235
def __init__(self, source_file):
238
:param source_file: a file-like object with `read` and `readline`
241
self._source = source_file
243
def reader_func(self, length=None):
244
return self._source.read(length)
246
def _read_line(self):
247
line = self._source.readline()
248
if not line.endswith('\n'):
249
raise errors.UnexpectedEndOfContainerError()
250
return line.rstrip('\n')
253
class ContainerReader(BaseReader):
254
"""A class for reading Bazaar's container format."""
256
def iter_records(self):
257
"""Iterate over the container, yielding each record as it is read.
259
Each yielded record will be a 2-tuple of (names, callable), where names
260
is a ``list`` and bytes is a function that takes one argument,
263
You **must not** call the callable after advancing the iterator to the
264
next record. That is, this code is invalid::
266
record_iter = container.iter_records()
267
names1, callable1 = record_iter.next()
268
names2, callable2 = record_iter.next()
269
bytes1 = callable1(None)
271
As it will give incorrect results and invalidate the state of the
274
:raises ContainerError: if any sort of container corruption is
275
detected, e.g. UnknownContainerFormatError is the format of the
276
container is unrecognised.
277
:seealso: ContainerReader.read
280
return self._iter_records()
282
def iter_record_objects(self):
283
"""Iterate over the container, yielding each record as it is read.
285
Each yielded record will be an object with ``read`` and ``validate``
286
methods. Like with iter_records, it is not safe to use a record object
287
after advancing the iterator to yield next record.
289
:raises ContainerError: if any sort of container corruption is
290
detected, e.g. UnknownContainerFormatError is the format of the
291
container is unrecognised.
292
:seealso: iter_records
295
return self._iter_record_objects()
297
def _iter_records(self):
298
for record in self._iter_record_objects():
301
def _iter_record_objects(self):
303
record_kind = self.reader_func(1)
304
if record_kind == 'B':
306
reader = BytesRecordReader(self._source)
308
elif record_kind == 'E':
309
# End marker. There are no more records.
311
elif record_kind == '':
312
# End of stream encountered, but no End Marker record seen, so
313
# this container is incomplete.
314
raise errors.UnexpectedEndOfContainerError()
316
# Unknown record type.
317
raise errors.UnknownRecordTypeError(record_kind)
319
def _read_format(self):
320
format = self._read_line()
321
if format != FORMAT_ONE:
322
raise errors.UnknownContainerFormatError(format)
325
"""Validate this container and its records.
327
Validating consumes the data stream just like iter_records and
328
iter_record_objects, so you cannot call it after
329
iter_records/iter_record_objects.
331
:raises ContainerError: if something is invalid.
334
for record_names, read_bytes in self.iter_records():
336
for name_tuple in record_names:
337
for name in name_tuple:
338
_check_name_encoding(name)
339
# Check that the name is unique. Note that Python will refuse
340
# to decode non-shortest forms of UTF-8 encoding, so there is no
341
# risk that the same unicode string has been encoded two
343
if name_tuple in all_names:
344
raise errors.DuplicateRecordNameError(name_tuple[0])
345
all_names.add(name_tuple)
346
excess_bytes = self.reader_func(1)
347
if excess_bytes != '':
348
raise errors.ContainerHasExcessDataError(excess_bytes)
351
class BytesRecordReader(BaseReader):
356
You can either validate or read a record, you can't do both.
358
:returns: A tuple of (names, callable). The callable can be called
359
repeatedly to obtain the bytes for the record, with a max_length
360
argument. If max_length is None, returns all the bytes. Because
361
records can be arbitrarily large, using None is not recommended
362
unless you have reason to believe the content will fit in memory.
364
# Read the content length.
365
length_line = self._read_line()
367
length = int(length_line)
369
raise errors.InvalidRecordError(
370
"%r is not a valid length." % (length_line,))
372
# Read the list of names.
375
name_line = self._read_line()
378
name_tuple = tuple(name_line.split('\x00'))
379
for name in name_tuple:
381
names.append(name_tuple)
383
self._remaining_length = length
384
return names, self._content_reader
386
def _content_reader(self, max_length):
387
if max_length is None:
388
length_to_read = self._remaining_length
390
length_to_read = min(max_length, self._remaining_length)
391
self._remaining_length -= length_to_read
392
bytes = self.reader_func(length_to_read)
393
if len(bytes) != length_to_read:
394
raise errors.UnexpectedEndOfContainerError()
398
"""Validate this record.
400
You can either validate or read, you can't do both.
402
:raises ContainerError: if this record is invalid.
404
names, read_bytes = self.read()
405
for name_tuple in names:
406
for name in name_tuple:
407
_check_name_encoding(name)
411
class ContainerPushParser(object):
412
"""A "push" parser for container format 1.
414
It accepts bytes via the ``accept_bytes`` method, and parses them into
415
records which can be retrieved via the ``read_pending_records`` method.
420
self._state_handler = self._state_expecting_format_line
421
self._parsed_records = []
422
self._reset_current_record()
423
self.finished = False
425
def _reset_current_record(self):
426
self._current_record_length = None
427
self._current_record_names = []
429
def accept_bytes(self, bytes):
430
self._buffer += bytes
431
# Keep iterating the state machine until it stops consuming bytes from
433
last_buffer_length = None
434
cur_buffer_length = len(self._buffer)
435
last_state_handler = None
436
while (cur_buffer_length != last_buffer_length
437
or last_state_handler != self._state_handler):
438
last_buffer_length = cur_buffer_length
439
last_state_handler = self._state_handler
440
self._state_handler()
441
cur_buffer_length = len(self._buffer)
443
def read_pending_records(self, max=None):
445
records = self._parsed_records[:max]
446
del self._parsed_records[:max]
449
records = self._parsed_records
450
self._parsed_records = []
453
def _consume_line(self):
454
"""Take a line out of the buffer, and return the line.
456
If a newline byte is not found in the buffer, the buffer is
457
unchanged and this returns None instead.
459
newline_pos = self._buffer.find('\n')
460
if newline_pos != -1:
461
line = self._buffer[:newline_pos]
462
self._buffer = self._buffer[newline_pos+1:]
467
def _state_expecting_format_line(self):
468
line = self._consume_line()
470
if line != FORMAT_ONE:
471
raise errors.UnknownContainerFormatError(line)
472
self._state_handler = self._state_expecting_record_type
474
def _state_expecting_record_type(self):
475
if len(self._buffer) >= 1:
476
record_type = self._buffer[0]
477
self._buffer = self._buffer[1:]
478
if record_type == 'B':
479
self._state_handler = self._state_expecting_length
480
elif record_type == 'E':
482
self._state_handler = self._state_expecting_nothing
484
raise errors.UnknownRecordTypeError(record_type)
486
def _state_expecting_length(self):
487
line = self._consume_line()
490
self._current_record_length = int(line)
492
raise errors.InvalidRecordError(
493
"%r is not a valid length." % (line,))
494
self._state_handler = self._state_expecting_name
496
def _state_expecting_name(self):
497
encoded_name_parts = self._consume_line()
498
if encoded_name_parts == '':
499
self._state_handler = self._state_expecting_body
500
elif encoded_name_parts:
501
name_parts = tuple(encoded_name_parts.split('\x00'))
502
for name_part in name_parts:
503
_check_name(name_part)
504
self._current_record_names.append(name_parts)
506
def _state_expecting_body(self):
507
if len(self._buffer) >= self._current_record_length:
508
body_bytes = self._buffer[:self._current_record_length]
509
self._buffer = self._buffer[self._current_record_length:]
510
record = (self._current_record_names, body_bytes)
511
self._parsed_records.append(record)
512
self._reset_current_record()
513
self._state_handler = self._state_expecting_record_type
515
def _state_expecting_nothing(self):
518
def read_size_hint(self):
520
if self._state_handler == self._state_expecting_body:
521
remaining = self._current_record_length - len(self._buffer)
524
return max(hint, remaining)
528
def iter_records_from_file(source_file):
529
parser = ContainerPushParser()
531
bytes = source_file.read(parser.read_size_hint())
532
parser.accept_bytes(bytes)
533
for record in parser.read_pending_records():