1
# Copyright (C) 2007, 2008 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
"""Python implementations of Dirstate Helper functions."""
19
from __future__ import absolute_import
25
# We cannot import the dirstate module, because it loads this module
26
# All we really need is the IN_MEMORY_MODIFIED constant
27
from breezy import errors
28
from .dirstate import DirState
34
def pack_stat(st, _b64=binascii.b2a_base64, _pack=struct.Struct('>6L').pack):
35
"""Convert stat values into a packed representation
37
Not all of the fields from the stat included are strictly needed, and by
38
just encoding the mtime and mode a slight speed increase could be gained.
39
However, using the pyrex version instead is a bigger win.
41
# base64 encoding always adds a final newline, so strip it off
42
return _b64(_pack(st.st_size & 0xFFFFFFFF, int(st.st_mtime) & 0xFFFFFFFF,
43
int(st.st_ctime) & 0xFFFFFFFF, st.st_dev & 0xFFFFFFFF,
44
st.st_ino & 0xFFFFFFFF, st.st_mode))[:-1]
47
def _unpack_stat(packed_stat):
48
"""Turn a packed_stat back into the stat fields.
50
This is meant as a debugging tool, should not be used in real code.
52
(st_size, st_mtime, st_ctime, st_dev, st_ino,
53
st_mode) = struct.unpack('>6L', binascii.a2b_base64(packed_stat))
54
return dict(st_size=st_size, st_mtime=st_mtime, st_ctime=st_ctime,
55
st_dev=st_dev, st_ino=st_ino, st_mode=st_mode)
58
def _bisect_path_left(paths, path):
59
"""Return the index where to insert path into paths.
61
This uses the dirblock sorting. So all children in a directory come before
62
the children of children. For example::
86
:param paths: A list of paths to search through
87
:param path: A single path to insert
88
:return: An offset where 'path' can be inserted.
89
:seealso: bisect.bisect_left
95
# Grab the dirname for the current dirblock
97
if _cmp_path_by_dirblock(cur, path) < 0:
104
def _bisect_path_right(paths, path):
105
"""Return the index where to insert path into paths.
107
This uses a path-wise comparison so we get::
117
:param paths: A list of paths to search through
118
:param path: A single path to insert
119
:return: An offset where 'path' can be inserted.
120
:seealso: bisect.bisect_right
126
# Grab the dirname for the current dirblock
128
if _cmp_path_by_dirblock(path, cur) < 0:
135
def bisect_dirblock(dirblocks, dirname, lo=0, hi=None, cache={}):
136
"""Return the index where to insert dirname into the dirblocks.
138
The return value idx is such that all directories blocks in dirblock[:idx]
139
have names < dirname, and all blocks in dirblock[idx:] have names >=
142
Optional args lo (default 0) and hi (default len(dirblocks)) bound the
143
slice of a to be searched.
148
dirname_split = cache[dirname]
150
dirname_split = dirname.split('/')
151
cache[dirname] = dirname_split
154
# Grab the dirname for the current dirblock
155
cur = dirblocks[mid][0]
157
cur_split = cache[cur]
159
cur_split = cur.split('/')
160
cache[cur] = cur_split
161
if cur_split < dirname_split: lo = mid + 1
166
def cmp_by_dirs(path1, path2):
167
"""Compare two paths directory by directory.
169
This is equivalent to doing::
171
cmp(path1.split('/'), path2.split('/'))
173
The idea is that you should compare path components separately. This
174
differs from plain ``cmp(path1, path2)`` for paths like ``'a-b'`` and
175
``a/b``. "a-b" comes after "a" but would come before "a/b" lexically.
177
:param path1: first path
178
:param path2: second path
179
:return: negative number if ``path1`` comes first,
180
0 if paths are equal,
181
and positive number if ``path2`` sorts first
183
if not isinstance(path1, str):
184
raise TypeError("'path1' must be a plain string, not %s: %r"
185
% (type(path1), path1))
186
if not isinstance(path2, str):
187
raise TypeError("'path2' must be a plain string, not %s: %r"
188
% (type(path2), path2))
189
return cmp(path1.split('/'), path2.split('/'))
192
def _cmp_path_by_dirblock(path1, path2):
193
"""Compare two paths based on what directory they are in.
195
This generates a sort order, such that all children of a directory are
196
sorted together, and grandchildren are in the same order as the
197
children appear. But all grandchildren come after all children.
199
:param path1: first path
200
:param path2: the second path
201
:return: negative number if ``path1`` comes first,
203
and a positive number if ``path2`` sorts first
205
if not isinstance(path1, str):
206
raise TypeError("'path1' must be a plain string, not %s: %r"
207
% (type(path1), path1))
208
if not isinstance(path2, str):
209
raise TypeError("'path2' must be a plain string, not %s: %r"
210
% (type(path2), path2))
211
dirname1, basename1 = os.path.split(path1)
212
key1 = (dirname1.split('/'), basename1)
213
dirname2, basename2 = os.path.split(path2)
214
key2 = (dirname2.split('/'), basename2)
215
return cmp(key1, key2)
218
def _read_dirblocks(state):
219
"""Read in the dirblocks for the given DirState object.
221
This is tightly bound to the DirState internal representation. It should be
222
thought of as a member function, which is only separated out so that we can
223
re-write it in pyrex.
225
:param state: A DirState object.
228
state._state_file.seek(state._end_of_header)
229
text = state._state_file.read()
230
# TODO: check the crc checksums. crc_measured = zlib.crc32(text)
232
fields = text.split('\0')
233
# Remove the last blank entry
234
trailing = fields.pop()
236
raise errors.DirstateCorrupt(state,
237
'trailing garbage: %r' % (trailing,))
238
# consider turning fields into a tuple.
240
# skip the first field which is the trailing null from the header.
242
# Each line now has an extra '\n' field which is not used
243
# so we just skip over it
245
# 3 fields for the key
246
# + number of fields per tree_data (5) * tree count
248
num_present_parents = state._num_present_parents()
249
tree_count = 1 + num_present_parents
250
entry_size = state._fields_per_entry()
251
expected_field_count = entry_size * state._num_entries
252
field_count = len(fields)
253
# this checks our adjustment, and also catches file too short.
254
if field_count - cur != expected_field_count:
255
raise errors.DirstateCorrupt(state,
256
'field count incorrect %s != %s, entry_size=%s, '\
257
'num_entries=%s fields=%r' % (
258
field_count - cur, expected_field_count, entry_size,
259
state._num_entries, fields))
261
if num_present_parents == 1:
262
# Bind external functions to local names
264
# We access all fields in order, so we can just iterate over
265
# them. Grab an straight iterator over the fields. (We use an
266
# iterator because we don't want to do a lot of additions, nor
267
# do we want to do a lot of slicing)
269
# Get a local reference to the compatible next method
270
next = getattr(_iter, '__next__', None)
273
# Move the iterator to the current position
276
# The two blocks here are deliberate: the root block and the
277
# contents-of-root block.
278
state._dirblocks = [('', []), ('', [])]
279
current_block = state._dirblocks[0][1]
281
append_entry = current_block.append
282
for count in range(state._num_entries):
286
if dirname != current_dirname:
287
# new block - different dirname
289
current_dirname = dirname
290
state._dirblocks.append((current_dirname, current_block))
291
append_entry = current_block.append
292
# we know current_dirname == dirname, so re-use it to avoid
293
# creating new strings
294
entry = ((current_dirname, name, file_id),
297
next(), # fingerprint
299
next() == 'y', # executable
300
next(), # packed_stat or revision_id
304
next(), # fingerprint
306
next() == 'y', # executable
307
next(), # packed_stat or revision_id
312
raise ValueError("trailing garbage in dirstate: %r" % trailing)
313
# append the entry to the current block
315
state._split_root_dirblock_into_contents()
317
fields_to_entry = state._get_fields_to_entry()
318
entries = [fields_to_entry(fields[pos:pos+entry_size])
319
for pos in range(cur, field_count, entry_size)]
320
state._entries_to_current_state(entries)
321
# To convert from format 2 => format 3
322
# state._dirblocks = sorted(state._dirblocks,
323
# key=lambda blk:blk[0].split('/'))
324
# To convert from format 3 => format 2
325
# state._dirblocks = sorted(state._dirblocks)
326
state._dirblock_state = DirState.IN_MEMORY_UNMODIFIED