1
# Copyright (C) 2006, 2007 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., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
17
"""DirState objects record the state of a directory and its bzr metadata.
19
Pseudo EBNF grammar for the state file. Fields are separated by NULLs, and
20
lines by NL. The field delimiters are ommitted in the grammar, line delimiters
21
are not - this is done for clarity of reading. All string data is in utf8.
23
MINIKIND = "f" | "d" | "l" | "a" | "r";
26
WHOLE_NUMBER = {digit}, digit;
28
REVISION_ID = a non-empty utf8 string;
30
dirstate format = header line, full checksum, row count, parent details,
31
ghost_details, entries;
32
header line = "#bazaar dirstate flat format 2", NL;
33
full checksum = "adler32: ", ["-"], WHOLE_NUMBER, NL;
34
row count = "num_entries: ", digit, NL;
35
parent_details = WHOLE NUMBER, {REVISION_ID}* NL;
36
ghost_details = WHOLE NUMBER, {REVISION_ID}*, NL;
38
entry = entry_key, current_entry_details, {parent_entry_details};
39
entry_key = dirname, basename, fileid;
40
current_entry_details = common_entry_details, working_entry_details;
41
parent_entry_details = common_entry_details, history_entry_details;
42
common_entry_details = MINIKIND, fingerprint, size, executable
43
working_entry_details = packed_stat
44
history_entry_details = REVISION_ID;
47
fingerprint = a nonempty utf8 sequence with meaning defined by minikind.
49
Given this definition, the following is useful to know:
50
entry (aka row) - all the data for a given key.
51
entry[0]: The key (dirname, basename, fileid)
55
entry[1]: The tree(s) data for this path and id combination.
56
entry[1][0]: The current tree
57
entry[1][1]: The second tree
59
For an entry for a tree, we have (using tree 0 - current tree) to demonstrate:
61
entry[1][0][1]: fingerprint
63
entry[1][0][3]: executable
64
entry[1][0][4]: packed_stat
66
entry[1][1][4]: revision_id
68
There may be multiple rows at the root, one per id present in the root, so the
69
in memory root row is now:
70
self._dirblocks[0] -> ('', [entry ...]),
71
and the entries in there are
74
entries[0][2]: file_id
75
entries[1][0]: The tree data for the current tree for this fileid at /
79
'r' is a relocated entry: This path is not present in this tree with this id,
80
but the id can be found at another location. The fingerprint is used to
81
point to the target location.
82
'a' is an absent entry: In that tree the id is not present at this path.
83
'd' is a directory entry: This path in this tree is a directory with the
84
current file id. There is no fingerprint for directories.
85
'f' is a file entry: As for directory, but its a file. The fingerprint is a
87
'l' is a symlink entry: As for directory, but a symlink. The fingerprint is the
91
--- Format 1 had the following different definition: ---
92
rows = dirname, NULL, basename, NULL, MINIKIND, NULL, fileid_utf8, NULL,
93
WHOLE NUMBER (* size *), NULL, packed stat, NULL, sha1|symlink target,
95
PARENT ROW = NULL, revision_utf8, NULL, MINIKIND, NULL, dirname, NULL,
96
basename, NULL, WHOLE NUMBER (* size *), NULL, "y" | "n", NULL,
99
PARENT ROW's are emitted for every parent that is not in the ghosts details
100
line. That is, if the parents are foo, bar, baz, and the ghosts are bar, then
101
each row will have a PARENT ROW for foo and baz, but not for bar.
104
In any tree, a kind of 'moved' indicates that the fingerprint field
105
(which we treat as opaque data specific to the 'kind' anyway) has the
106
details for the id of this row in that tree.
108
I'm strongly tempted to add a id->path index as well, but I think that
109
where we need id->path mapping; we also usually read the whole file, so
110
I'm going to skip that for the moment, as we have the ability to locate
111
via bisect any path in any tree, and if we lookup things by path, we can
112
accumulate a id->path mapping as we go, which will tend to match what we
115
I plan to implement this asap, so please speak up now to alter/tweak the
116
design - and once we stabilise on this, I'll update the wiki page for
119
The rationale for all this is that we want fast operations for the
120
common case (diff/status/commit/merge on all files) and extremely fast
121
operations for the less common but still occurs a lot status/diff/commit
122
on specific files). Operations on specific files involve a scan for all
123
the children of a path, *in every involved tree*, which the current
124
format did not accommodate.
128
1) Fast end to end use for bzr's top 5 uses cases. (commmit/diff/status/merge/???)
129
2) fall back current object model as needed.
130
3) scale usably to the largest trees known today - say 50K entries. (mozilla
131
is an example of this)
135
Eventually reuse dirstate objects across locks IFF the dirstate file has not
136
been modified, but will require that we flush/ignore cached stat-hit data
137
because we wont want to restat all files on disk just because a lock was
138
acquired, yet we cannot trust the data after the previous lock was released.
140
Memory representation:
141
vector of all directories, and vector of the childen ?
143
root_entrie = (direntry for root, [parent_direntries_for_root]),
145
('', ['data for achild', 'data for bchild', 'data for cchild'])
146
('dir', ['achild', 'cchild', 'echild'])
148
- single bisect to find N subtrees from a path spec
149
- in-order for serialisation - this is 'dirblock' grouping.
150
- insertion of a file '/a' affects only the '/' child-vector, that is, to
151
insert 10K elements from scratch does not generates O(N^2) memoves of a
152
single vector, rather each individual, which tends to be limited to a
153
manageable number. Will scale badly on trees with 10K entries in a
154
single directory. compare with Inventory.InventoryDirectory which has
155
a dictionary for the children. No bisect capability, can only probe for
156
exact matches, or grab all elements and sorta.
157
- Whats the risk of error here? Once we have the base format being processed
158
we should have a net win regardless of optimality. So we are going to
159
go with what seems reasonably.
162
maybe we should do a test profile of these core structure - 10K simulated searches/lookups/etc?
164
Objects for each row?
165
The lifetime of Dirstate objects is current per lock, but see above for
166
possible extensions. The lifetime of a row from a dirstate is expected to be
167
very short in the optimistic case: which we are optimising for. For instance,
168
subtree status will determine from analysis of the disk data what rows need to
169
be examined at all, and will be able to determine from a single row whether
170
that file has altered or not, so we are aiming to process tens of thousands of
171
entries each second within the dirstate context, before exposing anything to
172
the larger codebase. This suggests we want the time for a single file
173
comparison to be < 0.1 milliseconds. That would give us 10000 paths per second
174
processed, and to scale to 100 thousand we'll another order of magnitude to do
175
that. Now, as the lifetime for all unchanged entries is the time to parse, stat
176
the file on disk, and then immediately discard, the overhead of object creation
177
becomes a significant cost.
179
Figures: Creating a tuple from from 3 elements was profiled at 0.0625
180
microseconds, whereas creating a object which is subclassed from tuple was
181
0.500 microseconds, and creating an object with 3 elements and slots was 3
182
microseconds long. 0.1 milliseconds is 100 microseconds, and ideally we'll get
183
down to 10 microseconds for the total processing - having 33% of that be object
184
creation is a huge overhead. There is a potential cost in using tuples within
185
each row which is that the conditional code to do comparisons may be slower
186
than method invocation, but method invocation is known to be slow due to stack
187
frame creation, so avoiding methods in these tight inner loops in unfortunately
188
desirable. We can consider a pyrex version of this with objects in future if
206
import bzrlib.inventory
207
from bzrlib import osutils
208
from bzrlib.osutils import (
216
class DirState(object):
217
"""Record directory and metadata state for fast access.
219
A dirstate is a specialised data structure for managing local working
220
tree state information. Its not yet well defined whether it is platform
221
specific, and if it is how we detect/parameterise that.
224
_kind_to_minikind = {'absent':'a', 'file':'f', 'directory':'d', 'relocated':'r', 'symlink':'l'}
225
_minikind_to_kind = {'a':'absent', 'f':'file', 'd':'directory', 'l':'symlink', 'r':'relocated'}
226
_to_yesno = {True:'y', False: 'n'} # TODO profile the performance gain
227
# of using int conversion rather than a dict here. AND BLAME ANDREW IF
230
# TODO: jam 20070221 Make sure we handle when there are duplicated records
231
# (like when we remove + add the same path, or we have a rename)
232
# TODO: jam 20070221 Figure out what to do if we have a record that exceeds
233
# the BISECT_PAGE_SIZE.
234
BISECT_PAGE_SIZE = 4096
237
IN_MEMORY_UNMODIFIED = 1
238
IN_MEMORY_MODIFIED = 2
240
# A pack_stat (the x's) that is just noise and will never match the output
243
NULL_PARENT_DETAILS = ('absent', '', 0, False, '')
246
"""Create a DirState object.
250
:attr _root_entrie: The root row of the directory/file information,
251
- contains the path to / - '', ''
252
- kind of 'directory',
253
- the file id of the root in utf8
256
- and no sha information.
258
# _header_state and _dirblock_state represent the current state
259
# of the dirstate metadata and the per-row data respectiely.
260
# NOT_IN_MEMORY indicates that no data is in memory
261
# IN_MEMORY_UNMODIFIED indicates that what we have in memory
262
# is the same as is on disk
263
# IN_MEMORY_MODIFIED indicates that we have a modified version
264
# of what is on disk.
265
# In future we will add more granularity, for instance _dirblock_state
266
# will probably support partially-in-memory as a separate variable,
267
# allowing for partially-in-memory unmodified and partially-in-memory
269
self._header_state = DirState.NOT_IN_MEMORY
270
self._dirblock_state = DirState.NOT_IN_MEMORY
274
self._state_file=None
276
def add(self, path, file_id, kind, stat, link_or_sha1):
277
"""Add a path to be tracked.
279
:param path: The path within the dirstate - '' is the root, 'foo' is the
280
path foo within the root, 'foo/bar' is the path bar within foo
282
:param file_id: The file id of the path being added.
283
:param kind: The kind of the path.
284
:param stat: The output of os.lstate for the path.
285
:param link_or_sha1: The sha value of the file, or the target of a
286
symlink. '' for directories.
289
# find the block its in.
290
# find the location in the block.
291
# check its not there
293
#------- copied from bzrlib.inventory.make_entry
294
# --- normalized_filename wants a unicode basename only, so get one.
295
dirname, basename = os.path.split(path)
296
# we dont import normalized_filename directly because we want to be
297
# able to change the implementation at runtime for tests.
298
norm_name, can_access = osutils.normalized_filename(basename)
299
if norm_name != basename:
303
raise errors.InvalidNormalization(path)
304
# now that we've normalised, we need the correct utf8 path and
305
# dirname and basename elements. This single encode and split should be
306
# faster than three separate encodes.
307
utf8path = (dirname + '/' + basename).strip('/').encode('utf8')
308
dirname, basename = os.path.split(utf8path)
309
entry_key = (dirname, basename, file_id.encode('utf8'))
310
self._read_dirblocks_if_needed()
311
block_index, present = self._find_block_index_from_key(entry_key)
313
# TODO: This test is not complete - an empty directory, or a
314
# directory for a parent tree will fool it.
315
# some parent path has not been added - its an error to add this
317
raise errors.NotVersionedError(path, str(self))
318
block = self._dirblocks[block_index][1]
321
packed_stat = DirState.NULLSTAT
324
packed_stat = pack_stat(stat)
325
parent_info = self._empty_parent_info()
327
entry_data = entry_key, [
328
(kind, link_or_sha1, size, False, packed_stat),
330
elif kind == 'directory':
331
entry_data = entry_key, [
332
(kind, '', 0, False, packed_stat),
334
elif kind == 'symlink':
335
entry_data = entry_key, [
336
(kind, link_or_sha1, size, False, packed_stat),
339
raise errors.BzrError('unknown kind %r' % kind)
340
entry_index, present = self._find_entry_index(entry_key, block)
341
assert not present, "basename %r already added" % basename
342
block.insert(entry_index, entry_data)
344
if kind == 'directory':
345
# insert a new dirblock
346
self._ensure_block(block_index, entry_index, utf8path)
347
self._dirblock_state = DirState.IN_MEMORY_MODIFIED
349
def _empty_parent_info(self):
350
return [DirState.NULL_PARENT_DETAILS] * (len(self._parents) -
353
def _ensure_block(self, parent_block_index, parent_row_index, dirname):
354
"""Enssure a block for dirname exists.
356
This function exists to let callers which know that there is a
357
directory dirname ensure that the block for it exists. This block can
358
fail to exist because of demand loading, or because a directory had no
359
children. In either case it is not an error. It is however an error to
360
call this if there is no parent entry for the directory, and thus the
361
function requires the coordinates of such an entry to be provided.
363
The root row is special cased and can be indicated with a parent block
366
:param parent_block_index: The index of the block in which dirname's row
368
:param parent_row_index: The index in the parent block where the row
370
:param dirname: The utf8 dirname to ensure there is a block for.
371
:return: The index for the block.
374
# the basename of the directory must be the end of its full name.
375
if not (parent_block_index == -1 and
376
parent_block_index == -1 and dirname == ''):
377
assert dirname.endswith(
378
self._dirblocks[parent_block_index][1][parent_row_index][0][1])
379
block_index, present = self._find_block_index_from_key((dirname, '', ''))
381
## In future, when doing partial parsing, this should load and
382
# populate the entire block.
383
self._dirblocks.insert(block_index, (dirname, []))
386
def _entries_to_current_state(self, new_entries):
387
"""Load new_entries into self.dirblocks.
389
Process new_entries into the current state object, making them the active
392
:param new_entries: A sorted list of entries. This function does not sort
393
to prevent unneeded overhead when callers have a sorted list already.
396
assert new_entries[0][0][0:2] == ('', ''), \
397
"Missing root row %r" % new_entries[0][0]
398
# The two blocks here are deliberate: the root block and the
399
# contents-of-root block.
400
self._dirblocks = [('', []), ('', [])]
401
current_block = self._dirblocks[0][1]
404
for entry in new_entries:
405
if entry[0][0] != current_dirname:
406
# new block - different dirname
408
current_dirname = entry[0][0]
409
self._dirblocks.append((current_dirname, current_block))
411
# this is not a root entry for a tree (it has a basename)
412
current_block = self._dirblocks[-1][1]
413
# append the entry to the current block
414
current_block.append(entry)
416
def _entry_to_line(self, entry):
417
"""Serialize entry to a NULL delimited line ready for _get_output_lines.
419
:param entry: An entry_tuple as defined in the module docstring.
421
entire_entry = list(entry[0])
422
for tree_number, tree_data in enumerate(entry[1]):
423
# (kind, fingerprint, size, executable, tree_specific_string)
424
entire_entry.extend(tree_data)
425
# 3 for the key, 5 for the fields per tree.
426
tree_offset = 3 + tree_number * 5
428
entire_entry[tree_offset + 0] = DirState._kind_to_minikind[tree_data[0]]
430
entire_entry[tree_offset + 2] = str(tree_data[2])
432
entire_entry[tree_offset + 3] = DirState._to_yesno[tree_data[3]]
433
return '\0'.join(entire_entry)
435
def _fields_per_row(self):
436
"""How many null separated fields should be in each entry row.
438
Each line now has an extra '\n' field which is not used
439
so we just skip over it
442
+ number of fields per tree_data (5) * tree count
445
tree_count = 1 + self._num_present_parents()
446
return 3 + 5 * tree_count + 1
448
def _find_block(self, key, add_if_missing=False):
449
"""Return the block that key should be present in.
451
:param key: A dirstate entry key.
452
:return: The block tuple.
454
block_index, present = self._find_block_index_from_key(key)
456
if not add_if_missing:
457
# check to see if key is versioned itself - we might want to
458
# add it anyway, because dirs with no entries dont get a
459
# dirblock at parse time.
460
# This is an uncommon branch to take: most dirs have children,
461
# and most code works with versioned paths.
462
parent_base, parent_name = os.path.split(key[0])
463
if not self._get_block_entry_index(parent_base, parent_name, 0)[3]:
464
# some parent path has not been added - its an error to add
466
raise errors.NotVersionedError(key[0:2], str(self))
467
self._dirblocks.insert(block_index, (key[0], []))
468
return self._dirblocks[block_index]
470
def _find_block_index_from_key(self, key):
471
"""Find the dirblock index for a key.
473
:return: The block index, True if the block for the key is present.
475
if key[0:2] == ('', ''):
477
block_index = bisect.bisect_left(self._dirblocks, (key[0], []), 1)
478
# _right returns one-past-where-key is so we have to subtract
479
# one to use it. we use _right here because there are two
480
# '' blocks - the root, and the contents of root
481
# we always have a minimum of 2 in self._dirblocks: root and
482
# root-contents, and for '', we get 2 back, so this is
483
# simple and correct:
484
present = (block_index < len(self._dirblocks) and
485
self._dirblocks[block_index][0] == key[0])
486
return block_index, present
488
def _find_entry_index(self, key, block):
489
"""Find the entry index for a key in a block.
491
:return: The entry index, True if the entry for the key is present.
493
entry_index = bisect.bisect_left(block, (key, []))
494
present = (entry_index < len(block) and
495
block[entry_index][0] == key)
496
return entry_index, present
499
def from_tree(tree, dir_state_filename):
500
"""Create a dirstate from a bzr Tree.
502
:param tree: The tree which should provide parent information and
505
result = DirState.initialize(dir_state_filename)
507
parent_ids = tree.get_parent_ids()
508
num_parents = len(parent_ids)
510
for parent_id in parent_ids:
511
parent_trees.append((parent_id, tree.branch.repository.revision_tree(parent_id)))
512
parent_trees[-1][1].lock_read()
513
result.set_parent_trees(parent_trees, [])
514
result.set_state_from_inventory(tree.inventory)
516
for revid, parent in parent_trees:
521
def get_ghosts(self):
522
"""Return a list of the parent tree revision ids that are ghosts."""
523
self._read_header_if_needed()
527
"""Serialise the entire dirstate to a sequence of lines."""
528
if (self._header_state == DirState.IN_MEMORY_UNMODIFIED and
529
self._dirblock_state == DirState.IN_MEMORY_UNMODIFIED):
530
# read whats on disk.
531
self._state_file.seek(0)
532
return self._state_file.readlines()
534
lines.append(self._get_parents_line(self.get_parent_ids()))
535
lines.append(self._get_ghosts_line(self._ghosts))
536
# append the root line which is special cased
537
lines.extend(map(self._entry_to_line, self._iter_entries()))
538
return self._get_output_lines(lines)
540
def _get_ghosts_line(self, ghost_ids):
541
"""Create a line for the state file for ghost information."""
542
return '\0'.join([str(len(ghost_ids))] + ghost_ids)
544
def _get_parents_line(self, parent_ids):
545
"""Create a line for the state file for parents information."""
546
return '\0'.join([str(len(parent_ids))] + parent_ids)
548
def _get_fields_to_entry(self):
549
"""Get a function which converts entry fields into a entry record.
551
This handles kind, size, and executable, as well as parent records.
553
:return: A function which takes a list of fields, and returns an
554
appropriate record for storing in memory.
556
# This is intentionally unrolled for performance
557
num_present_parents = self._num_present_parents()
558
if num_present_parents == 0:
559
def fields_to_entry_0_parents(fields, _int=int, _tuple=tuple,
560
_mini_to_kind=self._minikind_to_kind):
561
path_name_file_id_key = _tuple(fields[:3])
562
return (path_name_file_id_key, [
564
_mini_to_kind[fields[3]], # kind
565
fields[4], # fingerprint
566
_int(fields[5]), # size
567
fields[6] == 'y', # executable
568
fields[7], # packed_stat or revision_id
570
return fields_to_entry_0_parents
571
elif num_present_parents == 1:
572
def fields_to_entry_1_parent(fields, _int=int, _tuple=tuple,
573
_mini_to_kind=self._minikind_to_kind):
574
path_name_file_id_key = _tuple(fields[:3])
575
return (path_name_file_id_key, [
577
_mini_to_kind[fields[3]], # kind
578
fields[4], # fingerprint
579
_int(fields[5]), # size
580
fields[6] == 'y', # executable
581
fields[7], # packed_stat or revision_id
584
_mini_to_kind[fields[8]], # kind
585
fields[9], # fingerprint
586
_int(fields[10]), # size
587
fields[11] == 'y', # executable
588
fields[12], # packed_stat or revision_id
591
return fields_to_entry_1_parent
592
elif num_present_parents == 2:
593
def fields_to_entry_2_parents(fields, _int=int, _tuple=tuple,
594
_mini_to_kind=self._minikind_to_kind):
595
path_name_file_id_key = _tuple(fields[:3])
596
return (path_name_file_id_key, [
598
_mini_to_kind[fields[3]], # kind
599
fields[4], # fingerprint
600
_int(fields[5]), # size
601
fields[6] == 'y', # executable
602
fields[7], # packed_stat or revision_id
605
_mini_to_kind[fields[8]], # kind
606
fields[9], # fingerprint
607
_int(fields[10]), # size
608
fields[11] == 'y', # executable
609
fields[12], # packed_stat or revision_id
612
_mini_to_kind[fields[13]],# kind
613
fields[14], # fingerprint
614
_int(fields[15]), # size
615
fields[16] == 'y', # executable
616
fields[17], # packed_stat or revision_id
619
return fields_to_entry_2_parents
621
def fields_to_entry_n_parents(fields, _int=int, _tuple=tuple,
622
_mini_to_kind=self._minikind_to_kind):
623
path_name_file_id_key = _tuple(fields[:3])
624
trees = [(_mini_to_kind[fields[cur]], # kind
625
fields[cur+1], # fingerprint
626
_int(fields[cur+2]), # size
627
fields[cur+3] == 'y', # executable
628
fields[cur+4], # stat or revision_id
629
) for cur in xrange(3, len(fields)-1, 5)]
630
return path_name_file_id_key, trees
631
return fields_to_entry_n_parents
633
def get_parent_ids(self):
634
"""Return a list of the parent tree ids for the directory state."""
635
self._read_header_if_needed()
636
return list(self._parents)
638
def _get_block_entry_index(self, dirname, basename, tree_index):
639
"""Get the coordinates for a path in the state structure.
641
:param dirname: The utf8 dirname to lookup.
642
:param basename: The utf8 basename to lookup.
643
:param tree_index: The index of the tree for which this lookup should
645
:return: A tuple describing where the path is located, or should be
646
inserted. The tuple contains four fields: the block index, the row
647
index, anda two booleans are True when the directory is present, and
648
when the entire path is present. There is no guarantee that either
649
coordinate is currently reachable unless the found field for it is
650
True. For instance, a directory not present in the searched tree
651
may be returned with a value one greater than the current highest
652
block offset. The directory present field will always be True when
653
the path present field is True. The directory present field does
654
NOT indicate that the directory is present in the searched tree,
655
rather it indicates that there are at least some files in some
658
self._read_dirblocks_if_needed()
659
key = dirname, basename, ''
660
block_index, present = self._find_block_index_from_key(key)
662
# no such directory - return the dir index and 0 for the row.
663
return block_index, 0, False, False
664
block = self._dirblocks[block_index][1] # access the entries only
665
entry_index, present = self._find_entry_index(key, block)
666
# linear search through present entries at this path to find the one
668
while entry_index < len(block) and block[entry_index][0][1] == basename:
669
if block[entry_index][1][tree_index][0] not in \
670
('absent', 'relocated'):
671
return block_index, entry_index, True, True
673
return block_index, entry_index, True, False
675
def _get_entry(self, tree_index, fileid_utf8=None, path_utf8=None):
676
"""Get the dirstate entry for path in tree tree_index
678
If either file_id or path is supplied, it is used as the key to lookup.
679
If both are supplied, the fastest lookup is used, and an error is
680
raised if they do not both point at the same row.
682
:param tree_index: The index of the tree we wish to locate this path
683
in. If the path is present in that tree, the entry containing its
684
details is returned, otherwise (None, None) is returned
685
:param fileid_utf8: A utf8 file_id to look up.
686
:param path_utf8: An utf8 path to be looked up.
687
:return: The dirstate entry tuple for path, or (None, None)
689
self._read_dirblocks_if_needed()
690
if path_utf8 is not None:
691
assert path_utf8.__class__ == str, 'path_utf8 is not a str: %s %s' % (type(path_utf8), path_utf8)
692
# path lookups are faster
693
dirname, basename = os.path.split(path_utf8)
694
block_index, entry_index, dir_present, file_present = \
695
self._get_block_entry_index(dirname, basename, tree_index)
698
entry = self._dirblocks[block_index][1][entry_index]
699
assert entry[0][2] and entry[1][tree_index][0] not in ('absent', 'relocated'), 'unversioned entry?!?!'
701
if entry[0][2] != fileid_utf8:
702
raise BzrError('integrity error ? : mismatching tree_index, file_id and path')
705
for entry in self._iter_entries():
706
if entry[0][2] == fileid_utf8:
707
if entry[1][tree_index][0] == 'relocated':
708
# look up the real location directly by path
709
return self._get_entry(tree_index,
710
fileid_utf8=fileid_utf8,
711
path_utf8=entry[1][tree_index][1])
712
if entry[1][tree_index][0] == 'absent':
713
# not in the tree at all.
719
def initialize(path):
720
"""Create a new dirstate on path.
722
The new dirstate will be an empty tree - that is it has no parents,
723
and only a root node - which has id ROOT_ID.
725
:param path: The name of the file for the dirstate.
726
:return: A DirState object.
728
# This constructs a new DirState object on a path, sets the _state_file
729
# to a new empty file for that path. It then calls _set_data() with our
730
# stock empty dirstate information - a root with ROOT_ID, no children,
731
# and no parents. Finally it calls save() to ensure that this data will
734
result._state_file = open(path, 'wb+')
735
# root dir and root dir contents with no children.
736
empty_tree_dirblocks = [('', []), ('', [])]
737
# a new root directory, with a NULLSTAT.
738
empty_tree_dirblocks[0][1].append(
739
(('', '', bzrlib.inventory.ROOT_ID), [
740
('directory', '', 0, False, DirState.NULLSTAT),
742
result._set_data([], empty_tree_dirblocks)
746
result._state_file.close()
750
def _inv_entry_to_details(self, inv_entry):
751
"""Convert an inventory entry (from a revision tree) to state details.
753
:param inv_entry: An inventory entry whose sha1 and link targets can be
754
relied upon, and which has a revision set.
755
:return: A details tuple - the details for a single tree at a path +
758
kind = inv_entry.kind
759
tree_data = inv_entry.revision.encode('utf8')
760
assert len(tree_data) > 0, 'empty revision for the inv_entry.'
761
if kind == 'directory':
765
elif kind == 'symlink':
766
fingerprint = inv_entry.symlink_target or ''
770
fingerprint = inv_entry.text_sha1 or ''
771
size = inv_entry.text_size or 0
772
executable = inv_entry.executable
775
return (kind, fingerprint, size, executable, tree_data)
777
def _iter_entries(self):
778
"""Iterate over all the entries in the dirstate.
780
Each yelt item is an entry in the standard format described in the
781
docstring of bzrlib.dirstate.
783
self._read_dirblocks_if_needed()
784
for directory in self._dirblocks:
785
for entry in directory[1]:
788
def _get_id_index(self):
789
"""Get an id index of self._dirblocks."""
791
for key, tree_details in self._iter_entries():
792
id_index.setdefault(key[2], set()).add(key)
795
def _get_output_lines(self, lines):
796
"""format lines for final output.
798
:param lines: A sequece of lines containing the parents list and the
801
output_lines = ['#bazaar dirstate flat format 2\n']
802
lines.append('') # a final newline
803
inventory_text = '\0\n\0'.join(lines)
804
output_lines.append('adler32: %s\n' % (zlib.adler32(inventory_text),))
805
# -3, 1 for num parents, 1 for ghosts, 1 for final newline
806
num_entries = len(lines)-3
807
output_lines.append('num_entries: %s\n' % (num_entries,))
808
output_lines.append(inventory_text)
811
def _make_deleted_row(self, fileid_utf8, parents):
812
"""Return a deleted for for fileid_utf8."""
813
return ('/', 'RECYCLED.BIN', 'file', fileid_utf8, 0, DirState.NULLSTAT,
816
def _num_present_parents(self):
817
"""The number of parent entries in each record row."""
818
return len(self._parents) - len(self._ghosts)
822
"""Construct a DirState on the file at path path."""
824
result._state_file = open(path, 'rb+')
827
def _read_dirblocks_if_needed(self):
828
"""Read in all the dirblocks from the file if they are not in memory.
830
This populates self._dirblocks, and sets self._dirblock_state to
831
IN_MEMORY_UNMODIFIED. It is not currently ready for incremental block
834
self._read_header_if_needed()
835
if self._dirblock_state == DirState.NOT_IN_MEMORY:
836
# the _state_file pointer will be positioned at the start of the
838
text = self._state_file.read()
839
# TODO: check the adler checksums. adler_measured = zlib.adler32(text)
841
fields = text.split('\0')
842
# Remove the last blank entry
843
trailing = fields.pop()
844
assert trailing == ''
845
# consider turning fields into a tuple.
847
# skip the first field which is the trailing null from the header.
849
# Each line now has an extra '\n' field which is not used
850
# so we just skip over it
852
# 3 fields for the key
853
# + number of fields per tree_data (5) * tree count
855
num_present_parents = self._num_present_parents()
856
tree_count = 1 + num_present_parents
857
entry_size = self._fields_per_row()
858
expected_field_count = entry_size * self._num_entries
859
if len(fields) - cur > expected_field_count:
860
fields = fields[:expected_field_count + cur]
861
trace.mutter('Unexpectedly long dirstate field count!')
862
print "XXX: incorrectly truncated dirstate file bug triggered."
863
field_count = len(fields)
864
# this checks our adjustment, and also catches file too short.
865
assert field_count - cur == expected_field_count, \
866
'field count incorrect %s != %s, entry_size=%s, '\
867
'num_entries=%s fields=%r' % (
868
field_count - cur, expected_field_count, entry_size,
869
self._num_entries, fields)
871
fields_to_entry = self._get_fields_to_entry()
872
entries = [fields_to_entry(fields[pos:pos+entry_size])
873
for pos in xrange(cur, field_count, entry_size)]
874
self._entries_to_current_state(entries)
875
self._dirblock_state = DirState.IN_MEMORY_UNMODIFIED
877
def _read_header(self):
878
"""This reads in the metadata header, and the parent ids.
880
After reading in, the file should be positioned at the null
881
just before the start of the first record in the file.
883
:return: (expected adler checksum, number of entries, parent list)
886
parent_line = self._state_file.readline()
887
info = parent_line.split('\0')
888
num_parents = int(info[0])
889
assert num_parents == len(info)-2, 'incorrect parent info line'
890
self._parents = info[1:-1]
892
ghost_line = self._state_file.readline()
893
info = ghost_line.split('\0')
894
num_ghosts = int(info[1])
895
assert num_ghosts == len(info)-3, 'incorrect ghost info line'
896
self._ghosts = info[2:-1]
897
self._header_state = DirState.IN_MEMORY_UNMODIFIED
899
def _read_header_if_needed(self):
900
"""Read the header of the dirstate file if needed."""
901
if self._header_state == DirState.NOT_IN_MEMORY:
904
def _read_prelude(self):
905
"""Read in the prelude header of the dirstate file
907
This only reads in the stuff that is not connected to the adler
908
checksum. The position will be correct to read in the rest of
909
the file and check the checksum after this point.
910
The next entry in the file should be the number of parents,
911
and their ids. Followed by a newline.
913
header = self._state_file.readline()
914
assert header == '#bazaar dirstate flat format 2\n', \
915
'invalid header line: %r' % (header,)
916
adler_line = self._state_file.readline()
917
assert adler_line.startswith('adler32: '), 'missing adler32 checksum'
918
self.adler_expected = int(adler_line[len('adler32: '):-1])
919
num_entries_line = self._state_file.readline()
920
assert num_entries_line.startswith('num_entries: '), 'missing num_entries line'
921
self._num_entries = int(num_entries_line[len('num_entries: '):-1])
924
"""Save any pending changes created during this session.
926
We reuse the existing file, because that prevents race conditions with
927
file creation, and we expect to be using oslocks on it in the near
928
future to prevent concurrent modification and reads - because dirstates
929
incremental data aggretation is not compatible with reading a modified
930
file, and replacing a file in use by another process is impossible on
933
A dirstate in read only mode should be smart enough though to validate
934
that the file has not changed, and otherwise discard its cache and
935
start over, to allow for fine grained read lock duration, so 'status'
936
wont block 'commit' - for example.
938
if (self._header_state == DirState.IN_MEMORY_MODIFIED or
939
self._dirblock_state == DirState.IN_MEMORY_MODIFIED):
940
self._state_file.seek(0)
941
self._state_file.writelines(self.get_lines())
942
self._state_file.truncate()
943
self._state_file.flush()
944
self._header_state = DirState.IN_MEMORY_UNMODIFIED
945
self._dirblock_state = DirState.IN_MEMORY_UNMODIFIED
947
def _set_data(self, parent_ids, dirblocks):
948
"""Set the full dirstate data in memory.
950
This is an internal function used to completely replace the objects
951
in memory state. It puts the dirstate into state 'full-dirty'.
953
:param parent_ids: A list of parent tree revision ids.
954
:param dirblocks: A list containing one tuple for each directory in the
955
tree. Each tuple contains the directory path and a list of entries
956
found in that directory.
958
# our memory copy is now authoritative.
959
self._dirblocks = dirblocks
960
self._header_state = DirState.IN_MEMORY_MODIFIED
961
self._dirblock_state = DirState.IN_MEMORY_MODIFIED
962
self._parents = list(parent_ids)
964
def set_path_id(self, path, new_id):
965
"""Change the id of path to new_id in the current working tree.
967
:param path: The path inside the tree to set - '' is the root, 'foo'
968
is the path foo in the root.
969
:param new_id: The new id to assign to the path. If unicode, it will
970
be encoded to utf8. In future this will be deprecated: avoid using
971
unicode ids if possible.
973
# TODO: start warning here.
974
if new_id.__class__ == unicode:
975
new_id = new_id.encode('utf8')
976
self._read_dirblocks_if_needed()
978
import pdb;pdb.set_trace()
980
raise NotImplementedError(self.set_path_id)
981
# TODO: check new id is unique
982
entry = self._get_entry(0, path_utf8=path)
983
# mark the old path absent, and insert a new root path
984
present_parents = len(entry[1]) - 1
985
self._make_absent(entry)
986
id_index = self._get_id_index()
987
self.update_minimal(('', '', new_id), 'directory', present_parents,
988
path_utf8='', id_index=id_index, packed_stat=entry[1][0][4])
989
self._dirblock_state = DirState.IN_MEMORY_MODIFIED
991
def set_parent_trees(self, trees, ghosts):
992
"""Set the parent trees for the dirstate.
994
:param trees: A list of revision_id, tree tuples. tree must be provided
995
even if the revision_id refers to a ghost: supply an empty tree in
997
:param ghosts: A list of the revision_ids that are ghosts at the time
1000
# TODO: generate a list of parent indexes to preserve to save
1001
# processing specific parent trees. In the common case one tree will
1002
# be preserved - the left most parent.
1003
# TODO: if the parent tree is a dirstate, we might want to walk them
1004
# all by path in parallel for 'optimal' common-case performance.
1005
# generate new root row.
1006
self._read_dirblocks_if_needed()
1007
# TODO future sketch: Examine the existing parents to generate a change
1008
# map and then walk the new parent trees only, mapping them into the
1009
# dirstate. Walk the dirstate at the same time to remove unreferenced
1012
# sketch: loop over all entries in the dirstate, cherry picking
1013
# entries from the parent trees, if they are not ghost trees.
1014
# after we finish walking the dirstate, all entries not in the dirstate
1015
# are deletes, so we want to append them to the end as per the design
1016
# discussions. So do a set difference on ids with the parents to
1017
# get deletes, and add them to the end.
1018
# During the update process we need to answer the following questions:
1019
# - find other keys containing a fileid in order to create cross-path
1020
# links. We dont't trivially use the inventory from other trees
1021
# because this leads to either double touching, or to accessing
1023
# - find other keys containing a path
1024
# We accumulate each entry via this dictionary, including the root
1027
# we could do parallel iterators, but because file id data may be
1028
# scattered throughout, we dont save on index overhead: we have to look
1029
# at everything anyway. We can probably save cycles by reusing parent
1030
# data and doing an incremental update when adding an additional
1031
# parent, but for now the common cases are adding a new parent (merge),
1032
# and replacing completely (commit), and commit is more common: so
1033
# optimise merge later.
1035
# ---- start generation of full tree mapping data
1036
# what trees should we use?
1037
parent_trees = [tree for rev_id, tree in trees if rev_id not in ghosts]
1038
# how many trees do we end up with
1039
parent_count = len(parent_trees)
1041
# one: the current tree
1042
for entry in self._iter_entries():
1043
# skip entries not in the current tree
1044
if entry[1][0][0] in ('absent', 'relocated'):
1046
by_path[entry[0]] = [entry[1][0]] + \
1047
[DirState.NULL_PARENT_DETAILS] * parent_count
1048
id_index[entry[0][2]] = set([entry[0]])
1050
# now the parent trees:
1051
for tree_index, tree in enumerate(parent_trees):
1052
# the index is off by one, adjust it.
1053
tree_index = tree_index + 1
1054
# when we add new locations for a fileid we need these ranges for
1055
# any fileid in this tree as we set the by_path[id] to:
1056
# already_processed_tree_details + new_details + new_location_suffix
1057
# the suffix is from tree_index+1:parent_count+1.
1058
new_location_suffix = [DirState.NULL_PARENT_DETAILS] * (parent_count - tree_index)
1059
# now stitch in all the entries from this tree
1060
for path, entry in tree.inventory.iter_entries_by_dir():
1061
# here we process each trees details for each item in the tree.
1062
# we first update any existing entries for the id at other paths,
1063
# then we either create or update the entry for the id at the
1064
# right path, and finally we add (if needed) a mapping from
1065
# file_id to this path. We do it in this order to allow us to
1066
# avoid checking all known paths for the id when generating a
1067
# new entry at this path: by adding the id->path mapping last,
1068
# all the mappings are valid and have correct relocation
1069
# records where needed.
1070
file_id = entry.file_id.encode('utf8')
1071
path_utf8 = path.encode('utf8')
1072
dirname, basename = os.path.split(path_utf8)
1073
new_entry_key = (dirname, basename, file_id)
1074
# tree index consistency: All other paths for this id in this tree
1075
# index must point to the correct path.
1076
for entry_key in id_index.setdefault(file_id, set()):
1077
# TODO:PROFILING: It might be faster to just update
1078
# rather than checking if we need to, and then overwrite
1079
# the one we are located at.
1080
if entry_key != new_entry_key:
1081
# this file id is at a different path in one of the
1082
# other trees, so put absent pointers there
1083
# This is the vertical axis in the matrix, all pointing
1085
by_path[entry_key][tree_index] = ('relocated', path_utf8, 0, False, '')
1086
# by path consistency: Insert into an existing path record (trivial), or
1087
# add a new one with relocation pointers for the other tree indexes.
1088
if new_entry_key in id_index[file_id]:
1089
# there is already an entry where this data belongs, just insert it.
1090
by_path[new_entry_key][tree_index] = \
1091
self._inv_entry_to_details(entry)
1093
# add relocated entries to the horizontal axis - this row
1094
# mapping from path,id. We need to look up the correct path
1095
# for the indexes from 0 to tree_index -1
1097
for lookup_index in xrange(tree_index):
1098
# boundary case: this is the first occurence of file_id
1099
# so there are no id_indexs, possibly take this out of
1101
if not len(id_index[file_id]):
1102
new_details.append(DirState.NULL_PARENT_DETAILS)
1104
# grab any one entry, use it to find the right path.
1105
# TODO: optimise this to reduce memory use in highly
1106
# fragmented situations by reusing the relocation
1108
a_key = iter(id_index[file_id]).next()
1109
if by_path[a_key][lookup_index][0] in ('relocated', 'absent'):
1110
# its a pointer or missing statement, use it as is.
1111
new_details.append(by_path[a_key][lookup_index])
1113
# we have the right key, make a pointer to it.
1114
real_path = ('/'.join(a_key[0:2])).strip('/')
1115
new_details.append(('relocated', real_path, 0, False, ''))
1116
new_details.append(self._inv_entry_to_details(entry))
1117
new_details.extend(new_location_suffix)
1118
by_path[new_entry_key] = new_details
1119
id_index[file_id].add(new_entry_key)
1120
# --- end generation of full tree mappings
1122
# sort and output all the entries
1123
new_entries = sorted(by_path.items())
1124
self._entries_to_current_state(new_entries)
1125
self._parents = [rev_id for rev_id, tree in trees]
1126
self._ghosts = list(ghosts)
1127
self._header_state = DirState.IN_MEMORY_MODIFIED
1128
self._dirblock_state = DirState.IN_MEMORY_MODIFIED
1130
def set_state_from_inventory(self, new_inv):
1131
"""Set new_inv as the current state.
1133
This API is called by tree transform, and will usually occur with
1134
existing parent trees.
1136
:param new_inv: The inventory object to set current state from.
1138
self._read_dirblocks_if_needed()
1140
# generate a byid index of the dirstate
1141
id_index = self._get_id_index()
1143
num_present_parents = len(self._parents) - len(self._ghosts)
1144
# incremental algorithm:
1145
# two iterators: current data and new data, both in dirblock order.
1146
new_iterator = new_inv.iter_entries_by_dir()
1147
# we will be modifying the dirstate, so we need a stable iterator. In
1148
# future we might write one, for now we just clone the state into a
1149
# list - which is a shallow copy, so each
1150
old_iterator = iter(list(self._iter_entries()))
1151
# both must have roots so this is safe:
1152
current_new = new_iterator.next()
1153
current_old = old_iterator.next()
1154
def advance(iterator):
1156
return iterator.next()
1157
except StopIteration:
1159
while current_new or current_old:
1160
# skip entries in old that are not really there
1161
if current_old and current_old[1][0][0] in ('relocated', 'absent'):
1162
current_old = advance(old_iterator)
1165
# convert new into dirblock style
1166
new_path_utf8 = current_new[0].encode('utf8')
1167
new_dirname, new_basename = os.path.split(new_path_utf8)
1168
new_id = current_new[1].file_id.encode('utf8')
1169
new_entry_key = (new_dirname, new_basename, new_id)
1171
# for safety disable variables
1172
new_path_utf8 = new_dirname = new_basename = new_id = new_entry_key = None
1173
# 5 cases, we dont have a value that is strictly greater than everything, so
1174
# we make both end conditions explicit
1176
# old is finished: insert current_new into the state.
1177
self.update_minimal(new_entry_key, current_new[1].kind,
1178
num_present_parents, executable=current_new[1].executable,
1179
id_index=id_index, path_utf8=new_path_utf8)
1180
current_new = advance(new_iterator)
1181
elif not current_new:
1183
self._make_absent(current_old, id_index)
1184
current_old = advance(old_iterator)
1185
elif new_entry_key == current_old[0]:
1186
# same - common case
1187
# TODO: update the record if anything significant has changed.
1188
# the minimal required trigger is if the execute bit or cached
1190
if (current_old[1][0][3] != current_new[1].executable or
1191
current_old[1][0][0] != current_new[1].kind):
1192
self.update_minimal(current_old[0], current_new[1].kind,
1193
num_present_parents,
1194
executable=current_new[1].executable,
1195
id_index=id_index, path_utf8=new_path_utf8)
1196
# both sides are dealt with, move on
1197
current_old = advance(old_iterator)
1198
current_new = advance(new_iterator)
1199
elif new_entry_key < current_old[0]:
1201
# add a entry for this and advance new
1202
self.update_minimal(new_entry_key, current_new[1].kind,
1203
num_present_parents, executable=current_new[1].executable,
1204
id_index=id_index, path_utf8=new_path_utf8)
1205
current_new = advance(new_iterator)
1208
self._make_absent(current_old, id_index)
1209
current_old = advance(old_iterator)
1210
self._dirblock_state = DirState.IN_MEMORY_MODIFIED
1212
def _make_absent(self, current_old, id_index=None):
1213
"""Mark current_old - an entry - as absent for tree 0.
1215
:param id_index: An index from fileid_utf8 to sets of keys, used by
1216
some functions. If provided it will be updated if needed.
1217
:return: True if this was the last details entry for they entry key:
1218
that is, if the underlying block has had the entry removed, thus
1219
shrinking in legnth.
1221
# build up paths that this id will be left at after the change is made,
1222
# so we can update their cross references in tree 0
1223
all_remaining_keys = set()
1224
# Dont check the working tree, because its going.
1225
for details in current_old[1][1:]:
1226
if details[0] not in ('absent', 'relocated'):
1227
all_remaining_keys.add(current_old[0])
1228
elif details[0] == 'relocated':
1229
# record the key for the real path.
1230
all_remaining_keys.add(tuple(os.path.split(details[1])) + (current_old[0][2],))
1231
# absent rows are not present at any path.
1232
last_reference = current_old[0] not in all_remaining_keys
1234
# the current row consists entire of the current item (being marked
1235
# absent), and relocated or absent entries for the other trees:
1236
# Remove it, its meaningless.
1237
block = self._find_block(current_old[0])
1238
entry_index, present = self._find_entry_index(current_old[0], block[1])
1240
block[1].pop(entry_index)
1241
# if we have an id_index in use, remove this key from it for this id.
1242
if id_index is not None:
1243
id_index[current_old[0][2]].remove(current_old[0])
1244
# update all remaining keys for this id to record it as absent. The
1245
# existing details may either be the record we are making as deleted
1246
# (if there were other trees with the id present at this path), or may
1248
for update_key in all_remaining_keys:
1249
update_block_index, present = \
1250
self._find_block_index_from_key(update_key)
1252
update_entry_index, present = \
1253
self._find_entry_index(update_key, self._dirblocks[update_block_index][1])
1255
update_tree_details = self._dirblocks[update_block_index][1][update_entry_index][1]
1256
# it must not be absent at the moment
1257
assert update_tree_details[0][0] != 'absent'
1258
update_tree_details[0] = DirState.NULL_PARENT_DETAILS
1259
self._dirblock_state = DirState.IN_MEMORY_MODIFIED
1260
return last_reference
1262
def update_minimal(self, key, kind, num_present_parents, executable=False,
1263
fingerprint='', packed_stat=None, size=0, id_index=None,
1265
"""Update an entry to the state in tree 0."""
1266
block = self._find_block(key)[1]
1267
if packed_stat is None:
1268
packed_stat = DirState.NULLSTAT
1269
entry_index, present = self._find_entry_index(key, block)
1270
new_details = (kind, fingerprint, size, executable, packed_stat)
1271
assert id_index is not None, 'need an id index to do updates for now !'
1273
# new entry, synthesis cross reference here,
1274
existing_keys = id_index.setdefault(key[2], set())
1275
if not existing_keys:
1276
# not currently in the state, simplest case
1277
new_entry = key, [new_details] + self._empty_parent_info()
1279
# present at one or more existing other paths.
1280
# grab one of them and use it to generate parent
1281
# relocation/absent entries.
1282
new_entry = key, [new_details]
1283
for other_key in existing_keys:
1284
# change the record at other to be a pointer to this new
1285
# record. The loop looks similar to the change to
1286
# relocations when updating an existing record but its not:
1287
# the test for existing kinds is different: this can be
1288
# factored out to a helper though.
1289
other_block_index, present = self._find_block_index_from_key(other_key)
1291
other_entry_index, present = self._find_entry_index(other_key, self._dirblocks[other_block_index][1])
1293
assert path_utf8 is not None
1294
self._dirblocks[other_block_index][1][other_entry_index][1][0] = \
1295
('relocated', path_utf8, 0, False, '')
1297
for lookup_index in xrange(1, num_present_parents + 1):
1298
# grab any one entry, use it to find the right path.
1299
# TODO: optimise this to reduce memory use in highly
1300
# fragmented situations by reusing the relocation
1302
update_block_index, present = \
1303
self._find_block_index_from_key(other_key)
1305
update_entry_index, present = \
1306
self._find_entry_index(other_key, self._dirblocks[update_block_index][1])
1308
update_details = self._dirblocks[update_block_index][1][update_entry_index][1][lookup_index]
1309
if update_details[0] in ('relocated', 'absent'):
1310
# its a pointer or absent in lookup_index's tree, use
1312
new_entry[1].append(update_details)
1314
# we have the right key, make a pointer to it.
1315
pointer_path = os.path.join(*other_key[0:2])
1316
new_entry[1].append(('relocated', pointer_path, 0, False, ''))
1317
block.insert(entry_index, new_entry)
1318
existing_keys.add(key)
1320
# Does the new state matter?
1321
block[entry_index][1][0] = new_details
1322
# parents cannot be affected by what we do.
1323
# other occurences of this id can be found
1324
# from the id index.
1326
# tree index consistency: All other paths for this id in this tree
1327
# index must point to the correct path. We have to loop here because
1328
# we may have passed entries in the state with this file id already
1329
# that were absent - where parent entries are - and they need to be
1330
# converted to relocated.
1331
assert path_utf8 is not None
1332
for entry_key in id_index.setdefault(key[2], set()):
1333
# TODO:PROFILING: It might be faster to just update
1334
# rather than checking if we need to, and then overwrite
1335
# the one we are located at.
1336
if entry_key != key:
1337
# this file id is at a different path in one of the
1338
# other trees, so put absent pointers there
1339
# This is the vertical axis in the matrix, all pointing
1341
block_index, present = self._find_block_index_from_key(entry_key)
1343
entry_index, present = self._find_entry_index(entry_key, self._dirblocks[block_index][1])
1345
self._dirblocks[block_index][1][entry_index][1][0] = \
1346
('relocated', path_utf8, 0, False, '')
1347
# add a containing dirblock if needed.
1348
if new_details[0] == 'directory':
1349
subdir_key = (os.path.join(*key[0:2]), '', '')
1350
block_index, present = self._find_block_index_from_key(subdir_key)
1352
self._dirblocks.insert(block_index, (subdir_key[0], []))
1354
self._dirblock_state = DirState.IN_MEMORY_MODIFIED
1358
def pack_stat(st, _encode=base64.encodestring, _pack=struct.pack):
1359
"""Convert stat values into a packed representation."""
1360
# jam 20060614 it isn't really worth removing more entries if we
1361
# are going to leave it in packed form.
1362
# With only st_mtime and st_mode filesize is 5.5M and read time is 275ms
1363
# With all entries filesize is 5.9M and read time is mabye 280ms
1364
# well within the noise margin
1366
# base64.encode always adds a final newline, so strip it off
1367
return _encode(_pack('>llllll'
1368
, st.st_size, st.st_mtime, st.st_ctime
1369
, st.st_dev, st.st_ino, st.st_mode))[:-1]
added
removed
bzrlib/dirstate.py
