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" | "t";
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 = "crc32: ", ["-"], 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:
60
entry[1][0][0]: minikind
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
89
't' is a reference to a nested subtree; the fingerprint is the referenced
94
The entries on disk and in memory are ordered according to the following keys:
96
directory, as a list of components
100
--- Format 1 had the following different definition: ---
101
rows = dirname, NULL, basename, NULL, MINIKIND, NULL, fileid_utf8, NULL,
102
WHOLE NUMBER (* size *), NULL, packed stat, NULL, sha1|symlink target,
104
PARENT ROW = NULL, revision_utf8, NULL, MINIKIND, NULL, dirname, NULL,
105
basename, NULL, WHOLE NUMBER (* size *), NULL, "y" | "n", NULL,
108
PARENT ROW's are emitted for every parent that is not in the ghosts details
109
line. That is, if the parents are foo, bar, baz, and the ghosts are bar, then
110
each row will have a PARENT ROW for foo and baz, but not for bar.
113
In any tree, a kind of 'moved' indicates that the fingerprint field
114
(which we treat as opaque data specific to the 'kind' anyway) has the
115
details for the id of this row in that tree.
117
I'm strongly tempted to add a id->path index as well, but I think that
118
where we need id->path mapping; we also usually read the whole file, so
119
I'm going to skip that for the moment, as we have the ability to locate
120
via bisect any path in any tree, and if we lookup things by path, we can
121
accumulate a id->path mapping as we go, which will tend to match what we
124
I plan to implement this asap, so please speak up now to alter/tweak the
125
design - and once we stabilise on this, I'll update the wiki page for
128
The rationale for all this is that we want fast operations for the
129
common case (diff/status/commit/merge on all files) and extremely fast
130
operations for the less common but still occurs a lot status/diff/commit
131
on specific files). Operations on specific files involve a scan for all
132
the children of a path, *in every involved tree*, which the current
133
format did not accommodate.
137
1) Fast end to end use for bzr's top 5 uses cases. (commmit/diff/status/merge/???)
138
2) fall back current object model as needed.
139
3) scale usably to the largest trees known today - say 50K entries. (mozilla
140
is an example of this)
144
Eventually reuse dirstate objects across locks IFF the dirstate file has not
145
been modified, but will require that we flush/ignore cached stat-hit data
146
because we wont want to restat all files on disk just because a lock was
147
acquired, yet we cannot trust the data after the previous lock was released.
149
Memory representation:
150
vector of all directories, and vector of the childen ?
152
root_entrie = (direntry for root, [parent_direntries_for_root]),
154
('', ['data for achild', 'data for bchild', 'data for cchild'])
155
('dir', ['achild', 'cchild', 'echild'])
157
- single bisect to find N subtrees from a path spec
158
- in-order for serialisation - this is 'dirblock' grouping.
159
- insertion of a file '/a' affects only the '/' child-vector, that is, to
160
insert 10K elements from scratch does not generates O(N^2) memoves of a
161
single vector, rather each individual, which tends to be limited to a
162
manageable number. Will scale badly on trees with 10K entries in a
163
single directory. compare with Inventory.InventoryDirectory which has
164
a dictionary for the children. No bisect capability, can only probe for
165
exact matches, or grab all elements and sorta.
166
- Whats the risk of error here? Once we have the base format being processed
167
we should have a net win regardless of optimality. So we are going to
168
go with what seems reasonably.
171
maybe we should do a test profile of these core structure - 10K simulated searches/lookups/etc?
173
Objects for each row?
174
The lifetime of Dirstate objects is current per lock, but see above for
175
possible extensions. The lifetime of a row from a dirstate is expected to be
176
very short in the optimistic case: which we are optimising for. For instance,
177
subtree status will determine from analysis of the disk data what rows need to
178
be examined at all, and will be able to determine from a single row whether
179
that file has altered or not, so we are aiming to process tens of thousands of
180
entries each second within the dirstate context, before exposing anything to
181
the larger codebase. This suggests we want the time for a single file
182
comparison to be < 0.1 milliseconds. That would give us 10000 paths per second
183
processed, and to scale to 100 thousand we'll another order of magnitude to do
184
that. Now, as the lifetime for all unchanged entries is the time to parse, stat
185
the file on disk, and then immediately discard, the overhead of object creation
186
becomes a significant cost.
188
Figures: Creating a tuple from from 3 elements was profiled at 0.0625
189
microseconds, whereas creating a object which is subclassed from tuple was
190
0.500 microseconds, and creating an object with 3 elements and slots was 3
191
microseconds long. 0.1 milliseconds is 100 microseconds, and ideally we'll get
192
down to 10 microseconds for the total processing - having 33% of that be object
193
creation is a huge overhead. There is a potential cost in using tuples within
194
each row which is that the conditional code to do comparisons may be slower
195
than method invocation, but method invocation is known to be slow due to stack
196
frame creation, so avoiding methods in these tight inner loops in unfortunately
197
desirable. We can consider a pyrex version of this with objects in future if
206
from stat import S_IEXEC
222
def pack_stat(st, _encode=binascii.b2a_base64, _pack=struct.pack):
223
"""Convert stat values into a packed representation."""
224
# jam 20060614 it isn't really worth removing more entries if we
225
# are going to leave it in packed form.
226
# With only st_mtime and st_mode filesize is 5.5M and read time is 275ms
227
# With all entries filesize is 5.9M and read time is mabye 280ms
228
# well within the noise margin
230
# base64 encoding always adds a final newline, so strip it off
231
# The current version
232
return _encode(_pack('>LLLLLL'
233
, st.st_size, int(st.st_mtime), int(st.st_ctime)
234
, st.st_dev, st.st_ino & 0xFFFFFFFF, st.st_mode))[:-1]
235
# This is 0.060s / 1.520s faster by not encoding as much information
236
# return _encode(_pack('>LL', int(st.st_mtime), st.st_mode))[:-1]
237
# This is not strictly faster than _encode(_pack())[:-1]
238
# return '%X.%X.%X.%X.%X.%X' % (
239
# st.st_size, int(st.st_mtime), int(st.st_ctime),
240
# st.st_dev, st.st_ino, st.st_mode)
241
# Similar to the _encode(_pack('>LL'))
242
# return '%X.%X' % (int(st.st_mtime), st.st_mode)
245
class DirState(object):
246
"""Record directory and metadata state for fast access.
248
A dirstate is a specialised data structure for managing local working
249
tree state information. Its not yet well defined whether it is platform
250
specific, and if it is how we detect/parameterise that.
252
Dirstates use the usual lock_write, lock_read and unlock mechanisms.
253
Unlike most bzr disk formats, DirStates must be locked for reading, using
254
lock_read. (This is an os file lock internally.) This is necessary
255
because the file can be rewritten in place.
257
DirStates must be explicitly written with save() to commit changes; just
258
unlocking them does not write the changes to disk.
261
_kind_to_minikind = {
267
'tree-reference': 't',
269
_minikind_to_kind = {
275
't': 'tree-reference',
277
_stat_to_minikind = {
282
_to_yesno = {True:'y', False: 'n'} # TODO profile the performance gain
283
# of using int conversion rather than a dict here. AND BLAME ANDREW IF
286
# TODO: jam 20070221 Figure out what to do if we have a record that exceeds
287
# the BISECT_PAGE_SIZE. For now, we just have to make it large enough
288
# that we are sure a single record will always fit.
289
BISECT_PAGE_SIZE = 4096
292
IN_MEMORY_UNMODIFIED = 1
293
IN_MEMORY_MODIFIED = 2
295
# A pack_stat (the x's) that is just noise and will never match the output
298
NULL_PARENT_DETAILS = ('a', '', 0, False, '')
300
HEADER_FORMAT_2 = '#bazaar dirstate flat format 2\n'
301
HEADER_FORMAT_3 = '#bazaar dirstate flat format 3\n'
303
def __init__(self, path):
304
"""Create a DirState object.
308
:attr _root_entrie: The root row of the directory/file information,
309
- contains the path to / - '', ''
310
- kind of 'directory',
311
- the file id of the root in utf8
314
- and no sha information.
315
:param path: The path at which the dirstate file on disk should live.
317
# _header_state and _dirblock_state represent the current state
318
# of the dirstate metadata and the per-row data respectiely.
319
# NOT_IN_MEMORY indicates that no data is in memory
320
# IN_MEMORY_UNMODIFIED indicates that what we have in memory
321
# is the same as is on disk
322
# IN_MEMORY_MODIFIED indicates that we have a modified version
323
# of what is on disk.
324
# In future we will add more granularity, for instance _dirblock_state
325
# will probably support partially-in-memory as a separate variable,
326
# allowing for partially-in-memory unmodified and partially-in-memory
328
self._header_state = DirState.NOT_IN_MEMORY
329
self._dirblock_state = DirState.NOT_IN_MEMORY
333
self._state_file = None
334
self._filename = path
335
self._lock_token = None
336
self._lock_state = None
337
self._id_index = None
338
self._end_of_header = None
339
self._cutoff_time = None
340
self._split_path_cache = {}
341
self._bisect_page_size = DirState.BISECT_PAGE_SIZE
345
(self.__class__.__name__, self._filename)
347
def add(self, path, file_id, kind, stat, fingerprint):
348
"""Add a path to be tracked.
350
:param path: The path within the dirstate - '' is the root, 'foo' is the
351
path foo within the root, 'foo/bar' is the path bar within foo
353
:param file_id: The file id of the path being added.
354
:param kind: The kind of the path, as a string like 'file',
356
:param stat: The output of os.lstat for the path.
357
:param fingerprint: The sha value of the file,
358
or the target of a symlink,
359
or the referenced revision id for tree-references,
360
or '' for directories.
363
# find the block its in.
364
# find the location in the block.
365
# check its not there
367
#------- copied from inventory.make_entry
368
# --- normalized_filename wants a unicode basename only, so get one.
369
if path.__class__ == unicode:
370
utf8path = path.encode('utf8')
373
dirname, basename = osutils.split(utf8path)
374
# you should never have files called . or ..; just add the directory
375
# in the parent, or according to the special treatment for the root
376
if basename == '.' or basename == '..':
377
raise errors.InvalidEntryName(path)
378
# now that we've normalised, we need the correct utf8 path and
379
# dirname and basename elements. This single encode and split should be
380
# faster than three separate encodes.
381
assert file_id.__class__ == str, \
382
"must be a utf8 file_id not %s" % (type(file_id))
383
# Make sure the file_id does not exist in this tree
384
file_id_entry = self._get_entry(0, fileid_utf8=file_id)
385
if file_id_entry != (None, None):
386
path = osutils.pathjoin(file_id_entry[0][0], file_id_entry[0][1])
387
kind = DirState._minikind_to_kind[file_id_entry[1][0][0]]
388
info = '%s:%s' % (kind, path)
389
raise errors.DuplicateFileId(file_id, info)
390
first_key = (dirname, basename, '')
391
block_index, present = self._find_block_index_from_key(first_key)
393
# check the path is not in the tree
394
block = self._dirblocks[block_index][1]
395
entry_index, _ = self._find_entry_index(first_key, block)
396
while (entry_index < len(block) and
397
block[entry_index][0][0:2] == first_key[0:2]):
398
if block[entry_index][1][0][0] not in 'ar':
399
# this path is in the dirstate in the current tree.
400
raise Exception, "adding already added path!"
403
# The block where we want to put the file is not present. But it
404
# might be because the directory was empty, or not loaded yet. Look
405
# for a parent entry, if not found, raise NotVersionedError
406
parent_dir, parent_base = osutils.split(dirname)
407
parent_block_idx, parent_entry_idx, _, parent_present = \
408
self._get_block_entry_index(parent_dir, parent_base, 0)
409
if not parent_present:
410
raise errors.NotVersionedError(path, str(self))
411
self._ensure_block(parent_block_idx, parent_entry_idx, dirname)
412
block = self._dirblocks[block_index][1]
413
entry_key = (dirname, basename, file_id)
416
packed_stat = DirState.NULLSTAT
419
packed_stat = pack_stat(stat)
420
parent_info = self._empty_parent_info()
421
minikind = DirState._kind_to_minikind[kind]
423
entry_data = entry_key, [
424
(minikind, fingerprint, size, False, packed_stat),
426
elif kind == 'directory':
427
entry_data = entry_key, [
428
(minikind, '', 0, False, packed_stat),
430
elif kind == 'symlink':
431
entry_data = entry_key, [
432
(minikind, fingerprint, size, False, packed_stat),
434
elif kind == 'tree-reference':
435
entry_data = entry_key, [
436
(minikind, fingerprint, 0, False, packed_stat),
439
raise errors.BzrError('unknown kind %r' % kind)
440
entry_index, present = self._find_entry_index(entry_key, block)
442
block.insert(entry_index, entry_data)
444
assert block[entry_index][1][0][0] == 'a', " %r(%r) already added" % (basename, file_id)
445
block[entry_index][1][0] = entry_data[1][0]
447
if kind == 'directory':
448
# insert a new dirblock
449
self._ensure_block(block_index, entry_index, utf8path)
450
self._dirblock_state = DirState.IN_MEMORY_MODIFIED
452
self._id_index.setdefault(entry_key[2], set()).add(entry_key)
454
def _bisect(self, paths):
455
"""Bisect through the disk structure for specific rows.
457
:param paths: A list of paths to find
458
:return: A dict mapping path => entries for found entries. Missing
459
entries will not be in the map.
460
The list is not sorted, and entries will be populated
461
based on when they were read.
463
self._requires_lock()
464
# We need the file pointer to be right after the initial header block
465
self._read_header_if_needed()
466
# If _dirblock_state was in memory, we should just return info from
467
# there, this function is only meant to handle when we want to read
469
assert self._dirblock_state == DirState.NOT_IN_MEMORY
471
# The disk representation is generally info + '\0\n\0' at the end. But
472
# for bisecting, it is easier to treat this as '\0' + info + '\0\n'
473
# Because it means we can sync on the '\n'
474
state_file = self._state_file
475
file_size = os.fstat(state_file.fileno()).st_size
476
# We end up with 2 extra fields, we should have a trailing '\n' to
477
# ensure that we read the whole record, and we should have a precursur
478
# '' which ensures that we start after the previous '\n'
479
entry_field_count = self._fields_per_entry() + 1
481
low = self._end_of_header
482
high = file_size - 1 # Ignore the final '\0'
483
# Map from (dir, name) => entry
486
# Avoid infinite seeking
487
max_count = 30*len(paths)
489
# pending is a list of places to look.
490
# each entry is a tuple of low, high, dir_names
491
# low -> the first byte offset to read (inclusive)
492
# high -> the last byte offset (inclusive)
493
# dir_names -> The list of (dir, name) pairs that should be found in
494
# the [low, high] range
495
pending = [(low, high, paths)]
497
page_size = self._bisect_page_size
499
fields_to_entry = self._get_fields_to_entry()
502
low, high, cur_files = pending.pop()
504
if not cur_files or low >= high:
509
if count > max_count:
510
raise errors.BzrError('Too many seeks, most likely a bug.')
512
mid = max(low, (low+high-page_size)/2)
515
# limit the read size, so we don't end up reading data that we have
517
read_size = min(page_size, (high-mid)+1)
518
block = state_file.read(read_size)
521
entries = block.split('\n')
524
# We didn't find a '\n', so we cannot have found any records.
525
# So put this range back and try again. But we know we have to
526
# increase the page size, because a single read did not contain
527
# a record break (so records must be larger than page_size)
529
pending.append((low, high, cur_files))
532
# Check the first and last entries, in case they are partial, or if
533
# we don't care about the rest of this page
535
first_fields = entries[0].split('\0')
536
if len(first_fields) < entry_field_count:
537
# We didn't get the complete first entry
538
# so move start, and grab the next, which
539
# should be a full entry
540
start += len(entries[0])+1
541
first_fields = entries[1].split('\0')
544
if len(first_fields) <= 2:
545
# We didn't even get a filename here... what do we do?
546
# Try a large page size and repeat this query
548
pending.append((low, high, cur_files))
551
# Find what entries we are looking for, which occur before and
552
# after this first record.
555
first_path = first_fields[1] + '/' + first_fields[2]
557
first_path = first_fields[2]
558
first_loc = _bisect_path_left(cur_files, first_path)
560
# These exist before the current location
561
pre = cur_files[:first_loc]
562
# These occur after the current location, which may be in the
563
# data we read, or might be after the last entry
564
post = cur_files[first_loc:]
566
if post and len(first_fields) >= entry_field_count:
567
# We have files after the first entry
569
# Parse the last entry
570
last_entry_num = len(entries)-1
571
last_fields = entries[last_entry_num].split('\0')
572
if len(last_fields) < entry_field_count:
573
# The very last hunk was not complete,
574
# read the previous hunk
575
after = mid + len(block) - len(entries[-1])
577
last_fields = entries[last_entry_num].split('\0')
579
after = mid + len(block)
582
last_path = last_fields[1] + '/' + last_fields[2]
584
last_path = last_fields[2]
585
last_loc = _bisect_path_right(post, last_path)
587
middle_files = post[:last_loc]
588
post = post[last_loc:]
591
# We have files that should occur in this block
592
# (>= first, <= last)
593
# Either we will find them here, or we can mark them as
596
if middle_files[0] == first_path:
597
# We might need to go before this location
598
pre.append(first_path)
599
if middle_files[-1] == last_path:
600
post.insert(0, last_path)
602
# Find out what paths we have
603
paths = {first_path:[first_fields]}
604
# last_path might == first_path so we need to be
605
# careful if we should append rather than overwrite
606
if last_entry_num != first_entry_num:
607
paths.setdefault(last_path, []).append(last_fields)
608
for num in xrange(first_entry_num+1, last_entry_num):
609
# TODO: jam 20070223 We are already splitting here, so
610
# shouldn't we just split the whole thing rather
611
# than doing the split again in add_one_record?
612
fields = entries[num].split('\0')
614
path = fields[1] + '/' + fields[2]
617
paths.setdefault(path, []).append(fields)
619
for path in middle_files:
620
for fields in paths.get(path, []):
621
# offset by 1 because of the opening '\0'
622
# consider changing fields_to_entry to avoid the
624
entry = fields_to_entry(fields[1:])
625
found.setdefault(path, []).append(entry)
627
# Now we have split up everything into pre, middle, and post, and
628
# we have handled everything that fell in 'middle'.
629
# We add 'post' first, so that we prefer to seek towards the
630
# beginning, so that we will tend to go as early as we need, and
631
# then only seek forward after that.
633
pending.append((after, high, post))
635
pending.append((low, start-1, pre))
637
# Consider that we may want to return the directory entries in sorted
638
# order. For now, we just return them in whatever order we found them,
639
# and leave it up to the caller if they care if it is ordered or not.
642
def _bisect_dirblocks(self, dir_list):
643
"""Bisect through the disk structure to find entries in given dirs.
645
_bisect_dirblocks is meant to find the contents of directories, which
646
differs from _bisect, which only finds individual entries.
648
:param dir_list: An sorted list of directory names ['', 'dir', 'foo'].
649
:return: A map from dir => entries_for_dir
651
# TODO: jam 20070223 A lot of the bisecting logic could be shared
652
# between this and _bisect. It would require parameterizing the
653
# inner loop with a function, though. We should evaluate the
654
# performance difference.
655
self._requires_lock()
656
# We need the file pointer to be right after the initial header block
657
self._read_header_if_needed()
658
# If _dirblock_state was in memory, we should just return info from
659
# there, this function is only meant to handle when we want to read
661
assert self._dirblock_state == DirState.NOT_IN_MEMORY
663
# The disk representation is generally info + '\0\n\0' at the end. But
664
# for bisecting, it is easier to treat this as '\0' + info + '\0\n'
665
# Because it means we can sync on the '\n'
666
state_file = self._state_file
667
file_size = os.fstat(state_file.fileno()).st_size
668
# We end up with 2 extra fields, we should have a trailing '\n' to
669
# ensure that we read the whole record, and we should have a precursur
670
# '' which ensures that we start after the previous '\n'
671
entry_field_count = self._fields_per_entry() + 1
673
low = self._end_of_header
674
high = file_size - 1 # Ignore the final '\0'
675
# Map from dir => entry
678
# Avoid infinite seeking
679
max_count = 30*len(dir_list)
681
# pending is a list of places to look.
682
# each entry is a tuple of low, high, dir_names
683
# low -> the first byte offset to read (inclusive)
684
# high -> the last byte offset (inclusive)
685
# dirs -> The list of directories that should be found in
686
# the [low, high] range
687
pending = [(low, high, dir_list)]
689
page_size = self._bisect_page_size
691
fields_to_entry = self._get_fields_to_entry()
694
low, high, cur_dirs = pending.pop()
696
if not cur_dirs or low >= high:
701
if count > max_count:
702
raise errors.BzrError('Too many seeks, most likely a bug.')
704
mid = max(low, (low+high-page_size)/2)
707
# limit the read size, so we don't end up reading data that we have
709
read_size = min(page_size, (high-mid)+1)
710
block = state_file.read(read_size)
713
entries = block.split('\n')
716
# We didn't find a '\n', so we cannot have found any records.
717
# So put this range back and try again. But we know we have to
718
# increase the page size, because a single read did not contain
719
# a record break (so records must be larger than page_size)
721
pending.append((low, high, cur_dirs))
724
# Check the first and last entries, in case they are partial, or if
725
# we don't care about the rest of this page
727
first_fields = entries[0].split('\0')
728
if len(first_fields) < entry_field_count:
729
# We didn't get the complete first entry
730
# so move start, and grab the next, which
731
# should be a full entry
732
start += len(entries[0])+1
733
first_fields = entries[1].split('\0')
736
if len(first_fields) <= 1:
737
# We didn't even get a dirname here... what do we do?
738
# Try a large page size and repeat this query
740
pending.append((low, high, cur_dirs))
743
# Find what entries we are looking for, which occur before and
744
# after this first record.
746
first_dir = first_fields[1]
747
first_loc = bisect.bisect_left(cur_dirs, first_dir)
749
# These exist before the current location
750
pre = cur_dirs[:first_loc]
751
# These occur after the current location, which may be in the
752
# data we read, or might be after the last entry
753
post = cur_dirs[first_loc:]
755
if post and len(first_fields) >= entry_field_count:
756
# We have records to look at after the first entry
758
# Parse the last entry
759
last_entry_num = len(entries)-1
760
last_fields = entries[last_entry_num].split('\0')
761
if len(last_fields) < entry_field_count:
762
# The very last hunk was not complete,
763
# read the previous hunk
764
after = mid + len(block) - len(entries[-1])
766
last_fields = entries[last_entry_num].split('\0')
768
after = mid + len(block)
770
last_dir = last_fields[1]
771
last_loc = bisect.bisect_right(post, last_dir)
773
middle_files = post[:last_loc]
774
post = post[last_loc:]
777
# We have files that should occur in this block
778
# (>= first, <= last)
779
# Either we will find them here, or we can mark them as
782
if middle_files[0] == first_dir:
783
# We might need to go before this location
784
pre.append(first_dir)
785
if middle_files[-1] == last_dir:
786
post.insert(0, last_dir)
788
# Find out what paths we have
789
paths = {first_dir:[first_fields]}
790
# last_dir might == first_dir so we need to be
791
# careful if we should append rather than overwrite
792
if last_entry_num != first_entry_num:
793
paths.setdefault(last_dir, []).append(last_fields)
794
for num in xrange(first_entry_num+1, last_entry_num):
795
# TODO: jam 20070223 We are already splitting here, so
796
# shouldn't we just split the whole thing rather
797
# than doing the split again in add_one_record?
798
fields = entries[num].split('\0')
799
paths.setdefault(fields[1], []).append(fields)
801
for cur_dir in middle_files:
802
for fields in paths.get(cur_dir, []):
803
# offset by 1 because of the opening '\0'
804
# consider changing fields_to_entry to avoid the
806
entry = fields_to_entry(fields[1:])
807
found.setdefault(cur_dir, []).append(entry)
809
# Now we have split up everything into pre, middle, and post, and
810
# we have handled everything that fell in 'middle'.
811
# We add 'post' first, so that we prefer to seek towards the
812
# beginning, so that we will tend to go as early as we need, and
813
# then only seek forward after that.
815
pending.append((after, high, post))
817
pending.append((low, start-1, pre))
821
def _bisect_recursive(self, paths):
822
"""Bisect for entries for all paths and their children.
824
This will use bisect to find all records for the supplied paths. It
825
will then continue to bisect for any records which are marked as
826
directories. (and renames?)
828
:param paths: A sorted list of (dir, name) pairs
829
eg: [('', 'a'), ('', 'f'), ('a/b', 'c')]
830
:return: A dictionary mapping (dir, name, file_id) => [tree_info]
832
# Map from (dir, name, file_id) => [tree_info]
835
found_dir_names = set()
837
# Directories that have been read
838
processed_dirs = set()
839
# Get the ball rolling with the first bisect for all entries.
840
newly_found = self._bisect(paths)
843
# Directories that need to be read
845
paths_to_search = set()
846
for entry_list in newly_found.itervalues():
847
for dir_name_id, trees_info in entry_list:
848
found[dir_name_id] = trees_info
849
found_dir_names.add(dir_name_id[:2])
851
for tree_info in trees_info:
852
minikind = tree_info[0]
855
# We already processed this one as a directory,
856
# we don't need to do the extra work again.
858
subdir, name, file_id = dir_name_id
859
path = osutils.pathjoin(subdir, name)
861
if path not in processed_dirs:
862
pending_dirs.add(path)
863
elif minikind == 'r':
864
# Rename, we need to directly search the target
865
# which is contained in the fingerprint column
866
dir_name = osutils.split(tree_info[1])
867
if dir_name[0] in pending_dirs:
868
# This entry will be found in the dir search
870
if dir_name not in found_dir_names:
871
paths_to_search.add(tree_info[1])
872
# Now we have a list of paths to look for directly, and
873
# directory blocks that need to be read.
874
# newly_found is mixing the keys between (dir, name) and path
875
# entries, but that is okay, because we only really care about the
877
newly_found = self._bisect(sorted(paths_to_search))
878
newly_found.update(self._bisect_dirblocks(sorted(pending_dirs)))
879
processed_dirs.update(pending_dirs)
882
def _empty_parent_info(self):
883
return [DirState.NULL_PARENT_DETAILS] * (len(self._parents) -
886
def _ensure_block(self, parent_block_index, parent_row_index, dirname):
887
"""Ensure a block for dirname exists.
889
This function exists to let callers which know that there is a
890
directory dirname ensure that the block for it exists. This block can
891
fail to exist because of demand loading, or because a directory had no
892
children. In either case it is not an error. It is however an error to
893
call this if there is no parent entry for the directory, and thus the
894
function requires the coordinates of such an entry to be provided.
896
The root row is special cased and can be indicated with a parent block
899
:param parent_block_index: The index of the block in which dirname's row
901
:param parent_row_index: The index in the parent block where the row
903
:param dirname: The utf8 dirname to ensure there is a block for.
904
:return: The index for the block.
906
if dirname == '' and parent_row_index == 0 and parent_block_index == 0:
907
# This is the signature of the root row, and the
908
# contents-of-root row is always index 1
910
# the basename of the directory must be the end of its full name.
911
if not (parent_block_index == -1 and
912
parent_block_index == -1 and dirname == ''):
913
assert dirname.endswith(
914
self._dirblocks[parent_block_index][1][parent_row_index][0][1])
915
block_index, present = self._find_block_index_from_key((dirname, '', ''))
917
## In future, when doing partial parsing, this should load and
918
# populate the entire block.
919
self._dirblocks.insert(block_index, (dirname, []))
922
def _entries_to_current_state(self, new_entries):
923
"""Load new_entries into self.dirblocks.
925
Process new_entries into the current state object, making them the active
926
state. The entries are grouped together by directory to form dirblocks.
928
:param new_entries: A sorted list of entries. This function does not sort
929
to prevent unneeded overhead when callers have a sorted list already.
932
assert new_entries[0][0][0:2] == ('', ''), \
933
"Missing root row %r" % (new_entries[0][0],)
934
# The two blocks here are deliberate: the root block and the
935
# contents-of-root block.
936
self._dirblocks = [('', []), ('', [])]
937
current_block = self._dirblocks[0][1]
940
append_entry = current_block.append
941
for entry in new_entries:
942
if entry[0][0] != current_dirname:
943
# new block - different dirname
945
current_dirname = entry[0][0]
946
self._dirblocks.append((current_dirname, current_block))
947
append_entry = current_block.append
948
# append the entry to the current block
950
self._split_root_dirblock_into_contents()
952
def _split_root_dirblock_into_contents(self):
953
"""Split the root dirblocks into root and contents-of-root.
955
After parsing by path, we end up with root entries and contents-of-root
956
entries in the same block. This loop splits them out again.
958
# The above loop leaves the "root block" entries mixed with the
959
# "contents-of-root block". But we don't want an if check on
960
# all entries, so instead we just fix it up here.
961
assert self._dirblocks[1] == ('', [])
963
contents_of_root_block = []
964
for entry in self._dirblocks[0][1]:
965
if not entry[0][1]: # This is a root entry
966
root_block.append(entry)
968
contents_of_root_block.append(entry)
969
self._dirblocks[0] = ('', root_block)
970
self._dirblocks[1] = ('', contents_of_root_block)
972
def _entry_to_line(self, entry):
973
"""Serialize entry to a NULL delimited line ready for _get_output_lines.
975
:param entry: An entry_tuple as defined in the module docstring.
977
entire_entry = list(entry[0])
978
for tree_number, tree_data in enumerate(entry[1]):
979
# (minikind, fingerprint, size, executable, tree_specific_string)
980
entire_entry.extend(tree_data)
981
# 3 for the key, 5 for the fields per tree.
982
tree_offset = 3 + tree_number * 5
984
entire_entry[tree_offset + 0] = tree_data[0]
986
entire_entry[tree_offset + 2] = str(tree_data[2])
988
entire_entry[tree_offset + 3] = DirState._to_yesno[tree_data[3]]
989
return '\0'.join(entire_entry)
991
def _fields_per_entry(self):
992
"""How many null separated fields should be in each entry row.
994
Each line now has an extra '\n' field which is not used
995
so we just skip over it
998
+ number of fields per tree_data (5) * tree count
1001
tree_count = 1 + self._num_present_parents()
1002
return 3 + 5 * tree_count + 1
1004
def _find_block(self, key, add_if_missing=False):
1005
"""Return the block that key should be present in.
1007
:param key: A dirstate entry key.
1008
:return: The block tuple.
1010
block_index, present = self._find_block_index_from_key(key)
1012
if not add_if_missing:
1013
# check to see if key is versioned itself - we might want to
1014
# add it anyway, because dirs with no entries dont get a
1015
# dirblock at parse time.
1016
# This is an uncommon branch to take: most dirs have children,
1017
# and most code works with versioned paths.
1018
parent_base, parent_name = osutils.split(key[0])
1019
if not self._get_block_entry_index(parent_base, parent_name, 0)[3]:
1020
# some parent path has not been added - its an error to add
1022
raise errors.NotVersionedError(key[0:2], str(self))
1023
self._dirblocks.insert(block_index, (key[0], []))
1024
return self._dirblocks[block_index]
1026
def _find_block_index_from_key(self, key):
1027
"""Find the dirblock index for a key.
1029
:return: The block index, True if the block for the key is present.
1031
if key[0:2] == ('', ''):
1033
block_index = bisect_dirblock(self._dirblocks, key[0], 1,
1034
cache=self._split_path_cache)
1035
# _right returns one-past-where-key is so we have to subtract
1036
# one to use it. we use _right here because there are two
1037
# '' blocks - the root, and the contents of root
1038
# we always have a minimum of 2 in self._dirblocks: root and
1039
# root-contents, and for '', we get 2 back, so this is
1040
# simple and correct:
1041
present = (block_index < len(self._dirblocks) and
1042
self._dirblocks[block_index][0] == key[0])
1043
return block_index, present
1045
def _find_entry_index(self, key, block):
1046
"""Find the entry index for a key in a block.
1048
:return: The entry index, True if the entry for the key is present.
1050
entry_index = bisect.bisect_left(block, (key, []))
1051
present = (entry_index < len(block) and
1052
block[entry_index][0] == key)
1053
return entry_index, present
1056
def from_tree(tree, dir_state_filename):
1057
"""Create a dirstate from a bzr Tree.
1059
:param tree: The tree which should provide parent information and
1061
:return: a DirState object which is currently locked for writing.
1062
(it was locked by DirState.initialize)
1064
result = DirState.initialize(dir_state_filename)
1068
parent_ids = tree.get_parent_ids()
1069
num_parents = len(parent_ids)
1071
for parent_id in parent_ids:
1072
parent_tree = tree.branch.repository.revision_tree(parent_id)
1073
parent_trees.append((parent_id, parent_tree))
1074
parent_tree.lock_read()
1075
result.set_parent_trees(parent_trees, [])
1076
result.set_state_from_inventory(tree.inventory)
1078
for revid, parent_tree in parent_trees:
1079
parent_tree.unlock()
1082
# The caller won't have a chance to unlock this, so make sure we
1088
def update_entry(self, entry, abspath, stat_value,
1089
_stat_to_minikind=_stat_to_minikind,
1090
_pack_stat=pack_stat):
1091
"""Update the entry based on what is actually on disk.
1093
:param entry: This is the dirblock entry for the file in question.
1094
:param abspath: The path on disk for this file.
1095
:param stat_value: (optional) if we already have done a stat on the
1097
:return: The sha1 hexdigest of the file (40 bytes) or link target of a
1101
minikind = _stat_to_minikind[stat_value.st_mode & 0170000]
1105
packed_stat = _pack_stat(stat_value)
1106
(saved_minikind, saved_link_or_sha1, saved_file_size,
1107
saved_executable, saved_packed_stat) = entry[1][0]
1109
if (minikind == saved_minikind
1110
and packed_stat == saved_packed_stat):
1111
# The stat hasn't changed since we saved, so we can re-use the
1116
# size should also be in packed_stat
1117
if saved_file_size == stat_value.st_size:
1118
return saved_link_or_sha1
1120
# If we have gotten this far, that means that we need to actually
1121
# process this entry.
1124
link_or_sha1 = self._sha1_file(abspath, entry)
1125
executable = self._is_executable(stat_value.st_mode,
1127
if self._cutoff_time is None:
1128
self._sha_cutoff_time()
1129
if (stat_value.st_mtime < self._cutoff_time
1130
and stat_value.st_ctime < self._cutoff_time):
1131
entry[1][0] = ('f', link_or_sha1, stat_value.st_size,
1132
executable, packed_stat)
1134
entry[1][0] = ('f', '', stat_value.st_size,
1135
executable, DirState.NULLSTAT)
1136
elif minikind == 'd':
1138
entry[1][0] = ('d', '', 0, False, packed_stat)
1139
if saved_minikind != 'd':
1140
# This changed from something into a directory. Make sure we
1141
# have a directory block for it. This doesn't happen very
1142
# often, so this doesn't have to be super fast.
1143
block_index, entry_index, dir_present, file_present = \
1144
self._get_block_entry_index(entry[0][0], entry[0][1], 0)
1145
self._ensure_block(block_index, entry_index,
1146
osutils.pathjoin(entry[0][0], entry[0][1]))
1147
elif minikind == 'l':
1148
link_or_sha1 = self._read_link(abspath, saved_link_or_sha1)
1149
if self._cutoff_time is None:
1150
self._sha_cutoff_time()
1151
if (stat_value.st_mtime < self._cutoff_time
1152
and stat_value.st_ctime < self._cutoff_time):
1153
entry[1][0] = ('l', link_or_sha1, stat_value.st_size,
1156
entry[1][0] = ('l', '', stat_value.st_size,
1157
False, DirState.NULLSTAT)
1158
self._dirblock_state = DirState.IN_MEMORY_MODIFIED
1161
def _sha_cutoff_time(self):
1162
"""Return cutoff time.
1164
Files modified more recently than this time are at risk of being
1165
undetectably modified and so can't be cached.
1167
# Cache the cutoff time as long as we hold a lock.
1168
# time.time() isn't super expensive (approx 3.38us), but
1169
# when you call it 50,000 times it adds up.
1170
# For comparison, os.lstat() costs 7.2us if it is hot.
1171
self._cutoff_time = int(time.time()) - 3
1172
return self._cutoff_time
1174
def _lstat(self, abspath, entry):
1175
"""Return the os.lstat value for this path."""
1176
return os.lstat(abspath)
1178
def _sha1_file(self, abspath, entry):
1179
"""Calculate the SHA1 of a file by reading the full text"""
1180
f = file(abspath, 'rb', buffering=65000)
1182
return osutils.sha_file(f)
1186
def _is_executable(self, mode, old_executable):
1187
"""Is this file executable?"""
1188
return bool(S_IEXEC & mode)
1190
def _is_executable_win32(self, mode, old_executable):
1191
"""On win32 the executable bit is stored in the dirstate."""
1192
return old_executable
1194
if sys.platform == 'win32':
1195
_is_executable = _is_executable_win32
1197
def _read_link(self, abspath, old_link):
1198
"""Read the target of a symlink"""
1199
# TODO: jam 200700301 On Win32, this could just return the value
1200
# already in memory. However, this really needs to be done at a
1201
# higher level, because there either won't be anything on disk,
1202
# or the thing on disk will be a file.
1203
return os.readlink(abspath)
1205
def get_ghosts(self):
1206
"""Return a list of the parent tree revision ids that are ghosts."""
1207
self._read_header_if_needed()
1210
def get_lines(self):
1211
"""Serialise the entire dirstate to a sequence of lines."""
1212
if (self._header_state == DirState.IN_MEMORY_UNMODIFIED and
1213
self._dirblock_state == DirState.IN_MEMORY_UNMODIFIED):
1214
# read whats on disk.
1215
self._state_file.seek(0)
1216
return self._state_file.readlines()
1218
lines.append(self._get_parents_line(self.get_parent_ids()))
1219
lines.append(self._get_ghosts_line(self._ghosts))
1220
# append the root line which is special cased
1221
lines.extend(map(self._entry_to_line, self._iter_entries()))
1222
return self._get_output_lines(lines)
1224
def _get_ghosts_line(self, ghost_ids):
1225
"""Create a line for the state file for ghost information."""
1226
return '\0'.join([str(len(ghost_ids))] + ghost_ids)
1228
def _get_parents_line(self, parent_ids):
1229
"""Create a line for the state file for parents information."""
1230
return '\0'.join([str(len(parent_ids))] + parent_ids)
1232
def _get_fields_to_entry(self):
1233
"""Get a function which converts entry fields into a entry record.
1235
This handles size and executable, as well as parent records.
1237
:return: A function which takes a list of fields, and returns an
1238
appropriate record for storing in memory.
1240
# This is intentionally unrolled for performance
1241
num_present_parents = self._num_present_parents()
1242
if num_present_parents == 0:
1243
def fields_to_entry_0_parents(fields, _int=int):
1244
path_name_file_id_key = (fields[0], fields[1], fields[2])
1245
return (path_name_file_id_key, [
1247
fields[3], # minikind
1248
fields[4], # fingerprint
1249
_int(fields[5]), # size
1250
fields[6] == 'y', # executable
1251
fields[7], # packed_stat or revision_id
1253
return fields_to_entry_0_parents
1254
elif num_present_parents == 1:
1255
def fields_to_entry_1_parent(fields, _int=int):
1256
path_name_file_id_key = (fields[0], fields[1], fields[2])
1257
return (path_name_file_id_key, [
1259
fields[3], # minikind
1260
fields[4], # fingerprint
1261
_int(fields[5]), # size
1262
fields[6] == 'y', # executable
1263
fields[7], # packed_stat or revision_id
1266
fields[8], # minikind
1267
fields[9], # fingerprint
1268
_int(fields[10]), # size
1269
fields[11] == 'y', # executable
1270
fields[12], # packed_stat or revision_id
1273
return fields_to_entry_1_parent
1274
elif num_present_parents == 2:
1275
def fields_to_entry_2_parents(fields, _int=int):
1276
path_name_file_id_key = (fields[0], fields[1], fields[2])
1277
return (path_name_file_id_key, [
1279
fields[3], # minikind
1280
fields[4], # fingerprint
1281
_int(fields[5]), # size
1282
fields[6] == 'y', # executable
1283
fields[7], # packed_stat or revision_id
1286
fields[8], # minikind
1287
fields[9], # fingerprint
1288
_int(fields[10]), # size
1289
fields[11] == 'y', # executable
1290
fields[12], # packed_stat or revision_id
1293
fields[13], # minikind
1294
fields[14], # fingerprint
1295
_int(fields[15]), # size
1296
fields[16] == 'y', # executable
1297
fields[17], # packed_stat or revision_id
1300
return fields_to_entry_2_parents
1302
def fields_to_entry_n_parents(fields, _int=int):
1303
path_name_file_id_key = (fields[0], fields[1], fields[2])
1304
trees = [(fields[cur], # minikind
1305
fields[cur+1], # fingerprint
1306
_int(fields[cur+2]), # size
1307
fields[cur+3] == 'y', # executable
1308
fields[cur+4], # stat or revision_id
1309
) for cur in xrange(3, len(fields)-1, 5)]
1310
return path_name_file_id_key, trees
1311
return fields_to_entry_n_parents
1313
def get_parent_ids(self):
1314
"""Return a list of the parent tree ids for the directory state."""
1315
self._read_header_if_needed()
1316
return list(self._parents)
1318
def _get_block_entry_index(self, dirname, basename, tree_index):
1319
"""Get the coordinates for a path in the state structure.
1321
:param dirname: The utf8 dirname to lookup.
1322
:param basename: The utf8 basename to lookup.
1323
:param tree_index: The index of the tree for which this lookup should
1325
:return: A tuple describing where the path is located, or should be
1326
inserted. The tuple contains four fields: the block index, the row
1327
index, anda two booleans are True when the directory is present, and
1328
when the entire path is present. There is no guarantee that either
1329
coordinate is currently reachable unless the found field for it is
1330
True. For instance, a directory not present in the searched tree
1331
may be returned with a value one greater than the current highest
1332
block offset. The directory present field will always be True when
1333
the path present field is True. The directory present field does
1334
NOT indicate that the directory is present in the searched tree,
1335
rather it indicates that there are at least some files in some
1338
self._read_dirblocks_if_needed()
1339
key = dirname, basename, ''
1340
block_index, present = self._find_block_index_from_key(key)
1342
# no such directory - return the dir index and 0 for the row.
1343
return block_index, 0, False, False
1344
block = self._dirblocks[block_index][1] # access the entries only
1345
entry_index, present = self._find_entry_index(key, block)
1346
# linear search through present entries at this path to find the one
1348
while entry_index < len(block) and block[entry_index][0][1] == basename:
1349
if block[entry_index][1][tree_index][0] not in \
1350
('a', 'r'): # absent, relocated
1351
return block_index, entry_index, True, True
1353
return block_index, entry_index, True, False
1355
def _get_entry(self, tree_index, fileid_utf8=None, path_utf8=None):
1356
"""Get the dirstate entry for path in tree tree_index
1358
If either file_id or path is supplied, it is used as the key to lookup.
1359
If both are supplied, the fastest lookup is used, and an error is
1360
raised if they do not both point at the same row.
1362
:param tree_index: The index of the tree we wish to locate this path
1363
in. If the path is present in that tree, the entry containing its
1364
details is returned, otherwise (None, None) is returned
1365
0 is the working tree, higher indexes are successive parent
1367
:param fileid_utf8: A utf8 file_id to look up.
1368
:param path_utf8: An utf8 path to be looked up.
1369
:return: The dirstate entry tuple for path, or (None, None)
1371
self._read_dirblocks_if_needed()
1372
if path_utf8 is not None:
1373
assert path_utf8.__class__ == str, 'path_utf8 is not a str: %s %s' % (type(path_utf8), path_utf8)
1374
# path lookups are faster
1375
dirname, basename = osutils.split(path_utf8)
1376
block_index, entry_index, dir_present, file_present = \
1377
self._get_block_entry_index(dirname, basename, tree_index)
1378
if not file_present:
1380
entry = self._dirblocks[block_index][1][entry_index]
1381
assert entry[0][2] and entry[1][tree_index][0] not in ('a', 'r'), 'unversioned entry?!?!'
1383
if entry[0][2] != fileid_utf8:
1384
raise errors.BzrError('integrity error ? : mismatching'
1385
' tree_index, file_id and path')
1388
assert fileid_utf8 is not None
1389
possible_keys = self._get_id_index().get(fileid_utf8, None)
1390
if not possible_keys:
1392
for key in possible_keys:
1393
block_index, present = \
1394
self._find_block_index_from_key(key)
1395
# strange, probably indicates an out of date
1396
# id index - for now, allow this.
1399
# WARNING: DO not change this code to use _get_block_entry_index
1400
# as that function is not suitable: it does not use the key
1401
# to lookup, and thus the wront coordinates are returned.
1402
block = self._dirblocks[block_index][1]
1403
entry_index, present = self._find_entry_index(key, block)
1405
entry = self._dirblocks[block_index][1][entry_index]
1406
if entry[1][tree_index][0] in 'fdlt':
1407
# this is the result we are looking for: the
1408
# real home of this file_id in this tree.
1410
if entry[1][tree_index][0] == 'a':
1411
# there is no home for this entry in this tree
1413
assert entry[1][tree_index][0] == 'r', \
1414
"entry %r has invalid minikind %r for tree %r" \
1416
entry[1][tree_index][0],
1418
real_path = entry[1][tree_index][1]
1419
return self._get_entry(tree_index, fileid_utf8=fileid_utf8,
1420
path_utf8=real_path)
1424
def initialize(cls, path):
1425
"""Create a new dirstate on path.
1427
The new dirstate will be an empty tree - that is it has no parents,
1428
and only a root node - which has id ROOT_ID.
1430
:param path: The name of the file for the dirstate.
1431
:return: A write-locked DirState object.
1433
# This constructs a new DirState object on a path, sets the _state_file
1434
# to a new empty file for that path. It then calls _set_data() with our
1435
# stock empty dirstate information - a root with ROOT_ID, no children,
1436
# and no parents. Finally it calls save() to ensure that this data will
1439
# root dir and root dir contents with no children.
1440
empty_tree_dirblocks = [('', []), ('', [])]
1441
# a new root directory, with a NULLSTAT.
1442
empty_tree_dirblocks[0][1].append(
1443
(('', '', inventory.ROOT_ID), [
1444
('d', '', 0, False, DirState.NULLSTAT),
1448
result._set_data([], empty_tree_dirblocks)
1455
def _inv_entry_to_details(self, inv_entry):
1456
"""Convert an inventory entry (from a revision tree) to state details.
1458
:param inv_entry: An inventory entry whose sha1 and link targets can be
1459
relied upon, and which has a revision set.
1460
:return: A details tuple - the details for a single tree at a path +
1463
kind = inv_entry.kind
1464
minikind = DirState._kind_to_minikind[kind]
1465
tree_data = inv_entry.revision
1466
assert len(tree_data) > 0, 'empty revision for the inv_entry.'
1467
if kind == 'directory':
1471
elif kind == 'symlink':
1472
fingerprint = inv_entry.symlink_target or ''
1475
elif kind == 'file':
1476
fingerprint = inv_entry.text_sha1 or ''
1477
size = inv_entry.text_size or 0
1478
executable = inv_entry.executable
1479
elif kind == 'tree-reference':
1480
fingerprint = inv_entry.reference_revision or ''
1484
raise Exception("can't pack %s" % inv_entry)
1485
return (minikind, fingerprint, size, executable, tree_data)
1487
def _iter_entries(self):
1488
"""Iterate over all the entries in the dirstate.
1490
Each yelt item is an entry in the standard format described in the
1491
docstring of bzrlib.dirstate.
1493
self._read_dirblocks_if_needed()
1494
for directory in self._dirblocks:
1495
for entry in directory[1]:
1498
def _get_id_index(self):
1499
"""Get an id index of self._dirblocks."""
1500
if self._id_index is None:
1502
for key, tree_details in self._iter_entries():
1503
id_index.setdefault(key[2], set()).add(key)
1504
self._id_index = id_index
1505
return self._id_index
1507
def _get_output_lines(self, lines):
1508
"""format lines for final output.
1510
:param lines: A sequece of lines containing the parents list and the
1513
output_lines = [DirState.HEADER_FORMAT_3]
1514
lines.append('') # a final newline
1515
inventory_text = '\0\n\0'.join(lines)
1516
output_lines.append('crc32: %s\n' % (zlib.crc32(inventory_text),))
1517
# -3, 1 for num parents, 1 for ghosts, 1 for final newline
1518
num_entries = len(lines)-3
1519
output_lines.append('num_entries: %s\n' % (num_entries,))
1520
output_lines.append(inventory_text)
1523
def _make_deleted_row(self, fileid_utf8, parents):
1524
"""Return a deleted for for fileid_utf8."""
1525
return ('/', 'RECYCLED.BIN', 'file', fileid_utf8, 0, DirState.NULLSTAT,
1528
def _num_present_parents(self):
1529
"""The number of parent entries in each record row."""
1530
return len(self._parents) - len(self._ghosts)
1534
"""Construct a DirState on the file at path path.
1536
:return: An unlocked DirState object, associated with the given path.
1538
result = DirState(path)
1541
def _read_dirblocks_if_needed(self):
1542
"""Read in all the dirblocks from the file if they are not in memory.
1544
This populates self._dirblocks, and sets self._dirblock_state to
1545
IN_MEMORY_UNMODIFIED. It is not currently ready for incremental block
1548
self._read_header_if_needed()
1549
if self._dirblock_state == DirState.NOT_IN_MEMORY:
1550
_read_dirblocks(self)
1552
def _read_header(self):
1553
"""This reads in the metadata header, and the parent ids.
1555
After reading in, the file should be positioned at the null
1556
just before the start of the first record in the file.
1558
:return: (expected crc checksum, number of entries, parent list)
1560
self._read_prelude()
1561
parent_line = self._state_file.readline()
1562
info = parent_line.split('\0')
1563
num_parents = int(info[0])
1564
assert num_parents == len(info)-2, 'incorrect parent info line'
1565
self._parents = info[1:-1]
1567
ghost_line = self._state_file.readline()
1568
info = ghost_line.split('\0')
1569
num_ghosts = int(info[1])
1570
assert num_ghosts == len(info)-3, 'incorrect ghost info line'
1571
self._ghosts = info[2:-1]
1572
self._header_state = DirState.IN_MEMORY_UNMODIFIED
1573
self._end_of_header = self._state_file.tell()
1575
def _read_header_if_needed(self):
1576
"""Read the header of the dirstate file if needed."""
1577
# inline this as it will be called a lot
1578
if not self._lock_token:
1579
raise errors.ObjectNotLocked(self)
1580
if self._header_state == DirState.NOT_IN_MEMORY:
1583
def _read_prelude(self):
1584
"""Read in the prelude header of the dirstate file
1586
This only reads in the stuff that is not connected to the crc
1587
checksum. The position will be correct to read in the rest of
1588
the file and check the checksum after this point.
1589
The next entry in the file should be the number of parents,
1590
and their ids. Followed by a newline.
1592
header = self._state_file.readline()
1593
assert header == DirState.HEADER_FORMAT_3, \
1594
'invalid header line: %r' % (header,)
1595
crc_line = self._state_file.readline()
1596
assert crc_line.startswith('crc32: '), 'missing crc32 checksum'
1597
self.crc_expected = int(crc_line[len('crc32: '):-1])
1598
num_entries_line = self._state_file.readline()
1599
assert num_entries_line.startswith('num_entries: '), 'missing num_entries line'
1600
self._num_entries = int(num_entries_line[len('num_entries: '):-1])
1603
"""Save any pending changes created during this session.
1605
We reuse the existing file, because that prevents race conditions with
1606
file creation, and use oslocks on it to prevent concurrent modification
1607
and reads - because dirstates incremental data aggretation is not
1608
compatible with reading a modified file, and replacing a file in use by
1609
another process is impossible on windows.
1611
A dirstate in read only mode should be smart enough though to validate
1612
that the file has not changed, and otherwise discard its cache and
1613
start over, to allow for fine grained read lock duration, so 'status'
1614
wont block 'commit' - for example.
1616
if (self._header_state == DirState.IN_MEMORY_MODIFIED or
1617
self._dirblock_state == DirState.IN_MEMORY_MODIFIED):
1619
grabbed_write_lock = False
1620
if self._lock_state != 'w':
1621
grabbed_write_lock, new_lock = self._lock_token.temporary_write_lock()
1622
# Switch over to the new lock, as the old one may be closed.
1623
# TODO: jam 20070315 We should validate the disk file has
1624
# not changed contents. Since temporary_write_lock may
1625
# not be an atomic operation.
1626
self._lock_token = new_lock
1627
self._state_file = new_lock.f
1628
if not grabbed_write_lock:
1629
# We couldn't grab a write lock, so we switch back to a read one
1632
self._state_file.seek(0)
1633
self._state_file.writelines(self.get_lines())
1634
self._state_file.truncate()
1635
self._state_file.flush()
1636
self._header_state = DirState.IN_MEMORY_UNMODIFIED
1637
self._dirblock_state = DirState.IN_MEMORY_UNMODIFIED
1639
if grabbed_write_lock:
1640
self._lock_token = self._lock_token.restore_read_lock()
1641
self._state_file = self._lock_token.f
1642
# TODO: jam 20070315 We should validate the disk file has
1643
# not changed contents. Since restore_read_lock may
1644
# not be an atomic operation.
1646
def _set_data(self, parent_ids, dirblocks):
1647
"""Set the full dirstate data in memory.
1649
This is an internal function used to completely replace the objects
1650
in memory state. It puts the dirstate into state 'full-dirty'.
1652
:param parent_ids: A list of parent tree revision ids.
1653
:param dirblocks: A list containing one tuple for each directory in the
1654
tree. Each tuple contains the directory path and a list of entries
1655
found in that directory.
1657
# our memory copy is now authoritative.
1658
self._dirblocks = dirblocks
1659
self._header_state = DirState.IN_MEMORY_MODIFIED
1660
self._dirblock_state = DirState.IN_MEMORY_MODIFIED
1661
self._parents = list(parent_ids)
1662
self._id_index = None
1664
def set_path_id(self, path, new_id):
1665
"""Change the id of path to new_id in the current working tree.
1667
:param path: The path inside the tree to set - '' is the root, 'foo'
1668
is the path foo in the root.
1669
:param new_id: The new id to assign to the path. This must be a utf8
1670
file id (not unicode, and not None).
1672
assert new_id.__class__ == str, \
1673
"path_id %r is not a plain string" % (new_id,)
1674
self._read_dirblocks_if_needed()
1677
raise NotImplementedError(self.set_path_id)
1678
# TODO: check new id is unique
1679
entry = self._get_entry(0, path_utf8=path)
1680
if entry[0][2] == new_id:
1681
# Nothing to change.
1683
# mark the old path absent, and insert a new root path
1684
self._make_absent(entry)
1685
self.update_minimal(('', '', new_id), 'd',
1686
path_utf8='', packed_stat=entry[1][0][4])
1687
self._dirblock_state = DirState.IN_MEMORY_MODIFIED
1688
if self._id_index is not None:
1689
self._id_index.setdefault(new_id, set()).add(entry[0])
1691
def set_parent_trees(self, trees, ghosts):
1692
"""Set the parent trees for the dirstate.
1694
:param trees: A list of revision_id, tree tuples. tree must be provided
1695
even if the revision_id refers to a ghost: supply an empty tree in
1697
:param ghosts: A list of the revision_ids that are ghosts at the time
1700
# TODO: generate a list of parent indexes to preserve to save
1701
# processing specific parent trees. In the common case one tree will
1702
# be preserved - the left most parent.
1703
# TODO: if the parent tree is a dirstate, we might want to walk them
1704
# all by path in parallel for 'optimal' common-case performance.
1705
# generate new root row.
1706
self._read_dirblocks_if_needed()
1707
# TODO future sketch: Examine the existing parents to generate a change
1708
# map and then walk the new parent trees only, mapping them into the
1709
# dirstate. Walk the dirstate at the same time to remove unreferenced
1712
# sketch: loop over all entries in the dirstate, cherry picking
1713
# entries from the parent trees, if they are not ghost trees.
1714
# after we finish walking the dirstate, all entries not in the dirstate
1715
# are deletes, so we want to append them to the end as per the design
1716
# discussions. So do a set difference on ids with the parents to
1717
# get deletes, and add them to the end.
1718
# During the update process we need to answer the following questions:
1719
# - find other keys containing a fileid in order to create cross-path
1720
# links. We dont't trivially use the inventory from other trees
1721
# because this leads to either double touching, or to accessing
1723
# - find other keys containing a path
1724
# We accumulate each entry via this dictionary, including the root
1727
# we could do parallel iterators, but because file id data may be
1728
# scattered throughout, we dont save on index overhead: we have to look
1729
# at everything anyway. We can probably save cycles by reusing parent
1730
# data and doing an incremental update when adding an additional
1731
# parent, but for now the common cases are adding a new parent (merge),
1732
# and replacing completely (commit), and commit is more common: so
1733
# optimise merge later.
1735
# ---- start generation of full tree mapping data
1736
# what trees should we use?
1737
parent_trees = [tree for rev_id, tree in trees if rev_id not in ghosts]
1738
# how many trees do we end up with
1739
parent_count = len(parent_trees)
1741
# one: the current tree
1742
for entry in self._iter_entries():
1743
# skip entries not in the current tree
1744
if entry[1][0][0] in ('a', 'r'): # absent, relocated
1746
by_path[entry[0]] = [entry[1][0]] + \
1747
[DirState.NULL_PARENT_DETAILS] * parent_count
1748
id_index[entry[0][2]] = set([entry[0]])
1750
# now the parent trees:
1751
for tree_index, tree in enumerate(parent_trees):
1752
# the index is off by one, adjust it.
1753
tree_index = tree_index + 1
1754
# when we add new locations for a fileid we need these ranges for
1755
# any fileid in this tree as we set the by_path[id] to:
1756
# already_processed_tree_details + new_details + new_location_suffix
1757
# the suffix is from tree_index+1:parent_count+1.
1758
new_location_suffix = [DirState.NULL_PARENT_DETAILS] * (parent_count - tree_index)
1759
# now stitch in all the entries from this tree
1760
for path, entry in tree.inventory.iter_entries_by_dir():
1761
# here we process each trees details for each item in the tree.
1762
# we first update any existing entries for the id at other paths,
1763
# then we either create or update the entry for the id at the
1764
# right path, and finally we add (if needed) a mapping from
1765
# file_id to this path. We do it in this order to allow us to
1766
# avoid checking all known paths for the id when generating a
1767
# new entry at this path: by adding the id->path mapping last,
1768
# all the mappings are valid and have correct relocation
1769
# records where needed.
1770
file_id = entry.file_id
1771
path_utf8 = path.encode('utf8')
1772
dirname, basename = osutils.split(path_utf8)
1773
new_entry_key = (dirname, basename, file_id)
1774
# tree index consistency: All other paths for this id in this tree
1775
# index must point to the correct path.
1776
for entry_key in id_index.setdefault(file_id, set()):
1777
# TODO:PROFILING: It might be faster to just update
1778
# rather than checking if we need to, and then overwrite
1779
# the one we are located at.
1780
if entry_key != new_entry_key:
1781
# this file id is at a different path in one of the
1782
# other trees, so put absent pointers there
1783
# This is the vertical axis in the matrix, all pointing
1785
by_path[entry_key][tree_index] = ('r', path_utf8, 0, False, '')
1786
# by path consistency: Insert into an existing path record (trivial), or
1787
# add a new one with relocation pointers for the other tree indexes.
1788
if new_entry_key in id_index[file_id]:
1789
# there is already an entry where this data belongs, just insert it.
1790
by_path[new_entry_key][tree_index] = \
1791
self._inv_entry_to_details(entry)
1793
# add relocated entries to the horizontal axis - this row
1794
# mapping from path,id. We need to look up the correct path
1795
# for the indexes from 0 to tree_index -1
1797
for lookup_index in xrange(tree_index):
1798
# boundary case: this is the first occurence of file_id
1799
# so there are no id_indexs, possibly take this out of
1801
if not len(id_index[file_id]):
1802
new_details.append(DirState.NULL_PARENT_DETAILS)
1804
# grab any one entry, use it to find the right path.
1805
# TODO: optimise this to reduce memory use in highly
1806
# fragmented situations by reusing the relocation
1808
a_key = iter(id_index[file_id]).next()
1809
if by_path[a_key][lookup_index][0] in ('r', 'a'):
1810
# its a pointer or missing statement, use it as is.
1811
new_details.append(by_path[a_key][lookup_index])
1813
# we have the right key, make a pointer to it.
1814
real_path = ('/'.join(a_key[0:2])).strip('/')
1815
new_details.append(('r', real_path, 0, False, ''))
1816
new_details.append(self._inv_entry_to_details(entry))
1817
new_details.extend(new_location_suffix)
1818
by_path[new_entry_key] = new_details
1819
id_index[file_id].add(new_entry_key)
1820
# --- end generation of full tree mappings
1822
# sort and output all the entries
1823
new_entries = self._sort_entries(by_path.items())
1824
self._entries_to_current_state(new_entries)
1825
self._parents = [rev_id for rev_id, tree in trees]
1826
self._ghosts = list(ghosts)
1827
self._header_state = DirState.IN_MEMORY_MODIFIED
1828
self._dirblock_state = DirState.IN_MEMORY_MODIFIED
1829
self._id_index = id_index
1831
def _sort_entries(self, entry_list):
1832
"""Given a list of entries, sort them into the right order.
1834
This is done when constructing a new dirstate from trees - normally we
1835
try to keep everything in sorted blocks all the time, but sometimes
1836
it's easier to sort after the fact.
1838
# TODO: Might be faster to do a schwartzian transform?
1840
# sort by: directory parts, file name, file id
1841
return entry[0][0].split('/'), entry[0][1], entry[0][2]
1842
return sorted(entry_list, key=_key)
1844
def set_state_from_inventory(self, new_inv):
1845
"""Set new_inv as the current state.
1847
This API is called by tree transform, and will usually occur with
1848
existing parent trees.
1850
:param new_inv: The inventory object to set current state from.
1852
self._read_dirblocks_if_needed()
1854
# incremental algorithm:
1855
# two iterators: current data and new data, both in dirblock order.
1856
new_iterator = new_inv.iter_entries_by_dir()
1857
# we will be modifying the dirstate, so we need a stable iterator. In
1858
# future we might write one, for now we just clone the state into a
1859
# list - which is a shallow copy, so each
1860
old_iterator = iter(list(self._iter_entries()))
1861
# both must have roots so this is safe:
1862
current_new = new_iterator.next()
1863
current_old = old_iterator.next()
1864
def advance(iterator):
1866
return iterator.next()
1867
except StopIteration:
1869
while current_new or current_old:
1870
# skip entries in old that are not really there
1871
if current_old and current_old[1][0][0] in ('r', 'a'):
1872
# relocated or absent
1873
current_old = advance(old_iterator)
1876
# convert new into dirblock style
1877
new_path_utf8 = current_new[0].encode('utf8')
1878
new_dirname, new_basename = osutils.split(new_path_utf8)
1879
new_id = current_new[1].file_id
1880
new_entry_key = (new_dirname, new_basename, new_id)
1881
current_new_minikind = \
1882
DirState._kind_to_minikind[current_new[1].kind]
1883
if current_new_minikind == 't':
1884
fingerprint = current_new[1].reference_revision
1888
# for safety disable variables
1889
new_path_utf8 = new_dirname = new_basename = new_id = new_entry_key = None
1890
# 5 cases, we dont have a value that is strictly greater than everything, so
1891
# we make both end conditions explicit
1893
# old is finished: insert current_new into the state.
1894
self.update_minimal(new_entry_key, current_new_minikind,
1895
executable=current_new[1].executable,
1896
path_utf8=new_path_utf8, fingerprint=fingerprint)
1897
current_new = advance(new_iterator)
1898
elif not current_new:
1900
self._make_absent(current_old)
1901
current_old = advance(old_iterator)
1902
elif new_entry_key == current_old[0]:
1903
# same - common case
1904
# TODO: update the record if anything significant has changed.
1905
# the minimal required trigger is if the execute bit or cached
1907
if (current_old[1][0][3] != current_new[1].executable or
1908
current_old[1][0][0] != current_new_minikind):
1909
self.update_minimal(current_old[0], current_new_minikind,
1910
executable=current_new[1].executable,
1911
path_utf8=new_path_utf8, fingerprint=fingerprint)
1912
# both sides are dealt with, move on
1913
current_old = advance(old_iterator)
1914
current_new = advance(new_iterator)
1915
elif (new_entry_key[0].split('/') < current_old[0][0].split('/')
1916
and new_entry_key[1:] < current_old[0][1:]):
1918
# add a entry for this and advance new
1919
self.update_minimal(new_entry_key, current_new_minikind,
1920
executable=current_new[1].executable,
1921
path_utf8=new_path_utf8, fingerprint=fingerprint)
1922
current_new = advance(new_iterator)
1925
self._make_absent(current_old)
1926
current_old = advance(old_iterator)
1927
self._dirblock_state = DirState.IN_MEMORY_MODIFIED
1928
self._id_index = None
1930
def _make_absent(self, current_old):
1931
"""Mark current_old - an entry - as absent for tree 0.
1933
:return: True if this was the last details entry for they entry key:
1934
that is, if the underlying block has had the entry removed, thus
1935
shrinking in length.
1937
# build up paths that this id will be left at after the change is made,
1938
# so we can update their cross references in tree 0
1939
all_remaining_keys = set()
1940
# Dont check the working tree, because its going.
1941
for details in current_old[1][1:]:
1942
if details[0] not in ('a', 'r'): # absent, relocated
1943
all_remaining_keys.add(current_old[0])
1944
elif details[0] == 'r': # relocated
1945
# record the key for the real path.
1946
all_remaining_keys.add(tuple(osutils.split(details[1])) + (current_old[0][2],))
1947
# absent rows are not present at any path.
1948
last_reference = current_old[0] not in all_remaining_keys
1950
# the current row consists entire of the current item (being marked
1951
# absent), and relocated or absent entries for the other trees:
1952
# Remove it, its meaningless.
1953
block = self._find_block(current_old[0])
1954
entry_index, present = self._find_entry_index(current_old[0], block[1])
1955
assert present, 'could not find entry for %s' % (current_old,)
1956
block[1].pop(entry_index)
1957
# if we have an id_index in use, remove this key from it for this id.
1958
if self._id_index is not None:
1959
self._id_index[current_old[0][2]].remove(current_old[0])
1960
# update all remaining keys for this id to record it as absent. The
1961
# existing details may either be the record we are making as deleted
1962
# (if there were other trees with the id present at this path), or may
1964
for update_key in all_remaining_keys:
1965
update_block_index, present = \
1966
self._find_block_index_from_key(update_key)
1967
assert present, 'could not find block for %s' % (update_key,)
1968
update_entry_index, present = \
1969
self._find_entry_index(update_key, self._dirblocks[update_block_index][1])
1970
assert present, 'could not find entry for %s' % (update_key,)
1971
update_tree_details = self._dirblocks[update_block_index][1][update_entry_index][1]
1972
# it must not be absent at the moment
1973
assert update_tree_details[0][0] != 'a' # absent
1974
update_tree_details[0] = DirState.NULL_PARENT_DETAILS
1975
self._dirblock_state = DirState.IN_MEMORY_MODIFIED
1976
return last_reference
1978
def update_minimal(self, key, minikind, executable=False, fingerprint='',
1979
packed_stat=None, size=0, path_utf8=None):
1980
"""Update an entry to the state in tree 0.
1982
This will either create a new entry at 'key' or update an existing one.
1983
It also makes sure that any other records which might mention this are
1986
:param key: (dir, name, file_id) for the new entry
1987
:param minikind: The type for the entry ('f' == 'file', 'd' ==
1989
:param executable: Should the executable bit be set?
1990
:param fingerprint: Simple fingerprint for new entry.
1991
:param packed_stat: packed stat value for new entry.
1992
:param size: Size information for new entry
1993
:param path_utf8: key[0] + '/' + key[1], just passed in to avoid doing
1996
block = self._find_block(key)[1]
1997
if packed_stat is None:
1998
packed_stat = DirState.NULLSTAT
1999
entry_index, present = self._find_entry_index(key, block)
2000
new_details = (minikind, fingerprint, size, executable, packed_stat)
2001
id_index = self._get_id_index()
2003
# new entry, synthesis cross reference here,
2004
existing_keys = id_index.setdefault(key[2], set())
2005
if not existing_keys:
2006
# not currently in the state, simplest case
2007
new_entry = key, [new_details] + self._empty_parent_info()
2009
# present at one or more existing other paths.
2010
# grab one of them and use it to generate parent
2011
# relocation/absent entries.
2012
new_entry = key, [new_details]
2013
for other_key in existing_keys:
2014
# change the record at other to be a pointer to this new
2015
# record. The loop looks similar to the change to
2016
# relocations when updating an existing record but its not:
2017
# the test for existing kinds is different: this can be
2018
# factored out to a helper though.
2019
other_block_index, present = self._find_block_index_from_key(other_key)
2020
assert present, 'could not find block for %s' % (other_key,)
2021
other_entry_index, present = self._find_entry_index(other_key,
2022
self._dirblocks[other_block_index][1])
2023
assert present, 'could not find entry for %s' % (other_key,)
2024
assert path_utf8 is not None
2025
self._dirblocks[other_block_index][1][other_entry_index][1][0] = \
2026
('r', path_utf8, 0, False, '')
2028
num_present_parents = self._num_present_parents()
2029
for lookup_index in xrange(1, num_present_parents + 1):
2030
# grab any one entry, use it to find the right path.
2031
# TODO: optimise this to reduce memory use in highly
2032
# fragmented situations by reusing the relocation
2034
update_block_index, present = \
2035
self._find_block_index_from_key(other_key)
2036
assert present, 'could not find block for %s' % (other_key,)
2037
update_entry_index, present = \
2038
self._find_entry_index(other_key, self._dirblocks[update_block_index][1])
2039
assert present, 'could not find entry for %s' % (other_key,)
2040
update_details = self._dirblocks[update_block_index][1][update_entry_index][1][lookup_index]
2041
if update_details[0] in ('r', 'a'): # relocated, absent
2042
# its a pointer or absent in lookup_index's tree, use
2044
new_entry[1].append(update_details)
2046
# we have the right key, make a pointer to it.
2047
pointer_path = osutils.pathjoin(*other_key[0:2])
2048
new_entry[1].append(('r', pointer_path, 0, False, ''))
2049
block.insert(entry_index, new_entry)
2050
existing_keys.add(key)
2052
# Does the new state matter?
2053
block[entry_index][1][0] = new_details
2054
# parents cannot be affected by what we do.
2055
# other occurences of this id can be found
2056
# from the id index.
2058
# tree index consistency: All other paths for this id in this tree
2059
# index must point to the correct path. We have to loop here because
2060
# we may have passed entries in the state with this file id already
2061
# that were absent - where parent entries are - and they need to be
2062
# converted to relocated.
2063
assert path_utf8 is not None
2064
for entry_key in id_index.setdefault(key[2], set()):
2065
# TODO:PROFILING: It might be faster to just update
2066
# rather than checking if we need to, and then overwrite
2067
# the one we are located at.
2068
if entry_key != key:
2069
# this file id is at a different path in one of the
2070
# other trees, so put absent pointers there
2071
# This is the vertical axis in the matrix, all pointing
2073
block_index, present = self._find_block_index_from_key(entry_key)
2075
entry_index, present = self._find_entry_index(entry_key, self._dirblocks[block_index][1])
2077
self._dirblocks[block_index][1][entry_index][1][0] = \
2078
('r', path_utf8, 0, False, '')
2079
# add a containing dirblock if needed.
2080
if new_details[0] == 'd':
2081
subdir_key = (osutils.pathjoin(*key[0:2]), '', '')
2082
block_index, present = self._find_block_index_from_key(subdir_key)
2084
self._dirblocks.insert(block_index, (subdir_key[0], []))
2086
self._dirblock_state = DirState.IN_MEMORY_MODIFIED
2088
def _validate(self):
2089
"""Check that invariants on the dirblock are correct.
2091
This can be useful in debugging; it shouldn't be necessary in
2094
This must be called with a lock held.
2096
# NOTE: This must always raise AssertionError not just assert,
2097
# otherwise it may not behave properly under python -O
2099
# TODO: All entries must have some content that's not 'a' or 'r',
2100
# otherwise it could just be removed.
2102
# TODO: All relocations must point directly to a real entry.
2104
# TODO: No repeated keys.
2107
from pprint import pformat
2108
self._read_dirblocks_if_needed()
2109
if len(self._dirblocks) > 0:
2110
if not self._dirblocks[0][0] == '':
2111
raise AssertionError(
2112
"dirblocks don't start with root block:\n" + \
2114
if len(self._dirblocks) > 1:
2115
if not self._dirblocks[1][0] == '':
2116
raise AssertionError(
2117
"dirblocks missing root directory:\n" + \
2119
# the dirblocks are sorted by their path components, name, and dir id
2120
dir_names = [d[0].split('/')
2121
for d in self._dirblocks[1:]]
2122
if dir_names != sorted(dir_names):
2123
raise AssertionError(
2124
"dir names are not in sorted order:\n" + \
2125
pformat(self._dirblocks) + \
2128
for dirblock in self._dirblocks:
2129
# within each dirblock, the entries are sorted by filename and
2131
for entry in dirblock[1]:
2132
if dirblock[0] != entry[0][0]:
2133
raise AssertionError(
2135
"doesn't match directory name in\n%r" %
2136
(entry, pformat(dirblock)))
2137
if dirblock[1] != sorted(dirblock[1]):
2138
raise AssertionError(
2139
"dirblock for %r is not sorted:\n%s" % \
2140
(dirblock[0], pformat(dirblock)))
2143
def check_valid_parent():
2144
"""Check that the current entry has a valid parent.
2146
This makes sure that the parent has a record,
2147
and that the parent isn't marked as "absent" in the
2148
current tree. (It is invalid to have a non-absent file in an absent
2151
if entry[0][0:2] == ('', ''):
2152
# There should be no parent for the root row
2154
parent_entry = self._get_entry(tree_index, path_utf8=entry[0][0])
2155
if parent_entry == (None, None):
2156
raise AssertionError(
2157
"no parent entry for: %s in tree %s"
2158
% (this_path, tree_index))
2159
if parent_entry[1][tree_index][0] != 'd':
2160
raise AssertionError(
2161
"Parent entry for %s is not marked as a valid"
2162
" directory. %s" % (this_path, parent_entry,))
2164
# For each file id, for each tree: either
2165
# the file id is not present at all; all rows with that id in the
2166
# key have it marked as 'absent'
2167
# OR the file id is present under exactly one name; any other entries
2168
# that mention that id point to the correct name.
2170
# We check this with a dict per tree pointing either to the present
2171
# name, or None if absent.
2172
tree_count = self._num_present_parents() + 1
2173
id_path_maps = [dict() for i in range(tree_count)]
2174
# Make sure that all renamed entries point to the correct location.
2175
for entry in self._iter_entries():
2176
file_id = entry[0][2]
2177
this_path = osutils.pathjoin(entry[0][0], entry[0][1])
2178
if len(entry[1]) != tree_count:
2179
raise AssertionError(
2180
"wrong number of entry details for row\n%s" \
2181
",\nexpected %d" % \
2182
(pformat(entry), tree_count))
2183
for tree_index, tree_state in enumerate(entry[1]):
2184
this_tree_map = id_path_maps[tree_index]
2185
minikind = tree_state[0]
2186
# have we seen this id before in this column?
2187
if file_id in this_tree_map:
2188
previous_path = this_tree_map[file_id]
2189
# any later mention of this file must be consistent with
2190
# what was said before
2192
if previous_path is not None:
2193
raise AssertionError(
2194
"file %s is absent in row %r but also present " \
2196
(file_id, entry, previous_path))
2197
elif minikind == 'r':
2198
target_location = tree_state[1]
2199
if previous_path != target_location:
2200
raise AssertionError(
2201
"file %s relocation in row %r but also at %r" \
2202
% (file_id, entry, previous_path))
2204
# a file, directory, etc - may have been previously
2205
# pointed to by a relocation, which must point here
2206
if previous_path != this_path:
2207
raise AssertionError(
2208
"entry %r inconsistent with previous path %r" % \
2209
(entry, previous_path))
2210
check_valid_parent()
2213
# absent; should not occur anywhere else
2214
this_tree_map[file_id] = None
2215
elif minikind == 'r':
2216
# relocation, must occur at expected location
2217
this_tree_map[file_id] = tree_state[1]
2219
this_tree_map[file_id] = this_path
2220
check_valid_parent()
2222
def _wipe_state(self):
2223
"""Forget all state information about the dirstate."""
2224
self._header_state = DirState.NOT_IN_MEMORY
2225
self._dirblock_state = DirState.NOT_IN_MEMORY
2228
self._dirblocks = []
2229
self._id_index = None
2230
self._end_of_header = None
2231
self._cutoff_time = None
2232
self._split_path_cache = {}
2234
def lock_read(self):
2235
"""Acquire a read lock on the dirstate"""
2236
if self._lock_token is not None:
2237
raise errors.LockContention(self._lock_token)
2238
# TODO: jam 20070301 Rather than wiping completely, if the blocks are
2239
# already in memory, we could read just the header and check for
2240
# any modification. If not modified, we can just leave things
2242
self._lock_token = lock.ReadLock(self._filename)
2243
self._lock_state = 'r'
2244
self._state_file = self._lock_token.f
2247
def lock_write(self):
2248
"""Acquire a write lock on the dirstate"""
2249
if self._lock_token is not None:
2250
raise errors.LockContention(self._lock_token)
2251
# TODO: jam 20070301 Rather than wiping completely, if the blocks are
2252
# already in memory, we could read just the header and check for
2253
# any modification. If not modified, we can just leave things
2255
self._lock_token = lock.WriteLock(self._filename)
2256
self._lock_state = 'w'
2257
self._state_file = self._lock_token.f
2261
"""Drop any locks held on the dirstate"""
2262
if self._lock_token is None:
2263
raise errors.LockNotHeld(self)
2264
# TODO: jam 20070301 Rather than wiping completely, if the blocks are
2265
# already in memory, we could read just the header and check for
2266
# any modification. If not modified, we can just leave things
2268
self._state_file = None
2269
self._lock_state = None
2270
self._lock_token.unlock()
2271
self._lock_token = None
2272
self._split_path_cache = {}
2274
def _requires_lock(self):
2275
"""Checks that a lock is currently held by someone on the dirstate"""
2276
if not self._lock_token:
2277
raise errors.ObjectNotLocked(self)
2280
# Try to load the compiled form if possible
2282
from bzrlib._dirstate_helpers_c import (
2283
_read_dirblocks_c as _read_dirblocks,
2284
bisect_dirblock_c as bisect_dirblock,
2285
_bisect_path_left_c as _bisect_path_left,
2286
_bisect_path_right_c as _bisect_path_right,
2287
cmp_by_dirs_c as cmp_by_dirs,
2290
from bzrlib._dirstate_helpers_py import (
2291
_read_dirblocks_py as _read_dirblocks,
2292
bisect_dirblock_py as bisect_dirblock,
2293
_bisect_path_left_py as _bisect_path_left,
2294
_bisect_path_right_py as _bisect_path_right,
2295
cmp_by_dirs_py as cmp_by_dirs,
added
removed
bzrlib/dirstate.py
