22
22
"""Weave - storage of related text file versions"""
24
# before intset (r923) 2000 versions in 41.5s
25
# with intset (r926) 2000 versions in 93s !!!
26
# better to just use plain sets.
28
# making _extract build and return a list, rather than being a generator
31
# with python -O, r923 does 2000 versions in 36.87s
33
# with optimizations to avoid mutating lists - 35.75! I guess copying
34
# all the elements every time costs more than the small manipulations.
35
# a surprisingly small change.
37
# r931, which avoids using a generator for extract, does 36.98s
39
# with memoized inclusions, takes 41.49s; not very good
41
# with slots, takes 37.35s; without takes 39.16, a bit surprising
43
# with the delta calculation mixed in with the add method, rather than
44
# separated, takes 36.78s
46
# with delta folded in and mutation of the list, 36.13s
48
# with all this and simplification of add code, 33s
54
# TODO: Perhaps have copy method for Weave instances?
56
25
# XXX: If we do weaves this way, will a merge still behave the same
57
26
# way if it's done in a different order? That's a pretty desirable
75
44
# TODO: Parallel-extract that passes back each line along with a
76
45
# description of which revisions include it. Nice for checking all
46
# shas or calculating stats in parallel.
79
48
# TODO: Using a single _extract routine and then processing the output
80
49
# is probably inefficient. It's simple enough that we can afford to
81
50
# have slight specializations for different ways its used: annotate,
82
51
# basis for add, get, etc.
84
# TODO: Perhaps the API should work only in names to hide the integer
85
# indexes from the user?
53
# TODO: Probably the API should work only in names to hide the integer
54
# indexes from the user.
87
56
# TODO: Is there any potential performance win by having an add()
88
57
# variant that is passed a pre-cooked version of the single basis
60
# TODO: Reweave can possibly be made faster by remembering diffs
61
# where the basis and destination are unchanged.
63
# FIXME: Sometimes we will be given a parents list for a revision
64
# that includes some redundant parents (i.e. already a parent of
65
# something in the list.) We should eliminate them. This can
66
# be done fairly efficiently because the sequence numbers constrain
67
# the possible relationships.
94
71
from difflib import SequenceMatcher
99
class WeaveError(Exception):
100
"""Exception in processing weave"""
103
class WeaveFormatError(WeaveError):
104
"""Weave invariant violated"""
73
from bzrlib.trace import mutter
74
from bzrlib.errors import WeaveError, WeaveFormatError, WeaveParentMismatch, \
75
WeaveRevisionNotPresent, WeaveRevisionAlreadyPresent
76
from bzrlib.tsort import topo_sort
107
79
class Weave(object):
108
80
"""weave - versioned text file storage.
231
221
return self._name_map[name]
233
raise WeaveError("name %r not present in weave %r" %
234
(name, self._weave_name))
223
raise WeaveRevisionNotPresent(name, self)
226
return self._names[:]
228
def iter_names(self):
229
"""Yield a list of all names in this weave."""
230
return iter(self._names)
237
232
def idx_to_name(self, version):
238
233
return self._names[version]
241
235
def _check_repeated_add(self, name, parents, text, sha1):
242
236
"""Check that a duplicated add is OK.
244
238
If it is, return the (old) index; otherwise raise an exception.
246
240
idx = self.lookup(name)
247
if sorted(self._parents[idx]) != sorted(parents):
248
raise WeaveError("name \"%s\" already present in weave "
249
"with different parents" % name)
250
if sha1 != self._sha1s[idx]:
251
raise WeaveError("name \"%s\" already present in weave "
252
"with different text" % name)
241
if sorted(self._parents[idx]) != sorted(parents) \
242
or sha1 != self._sha1s[idx]:
243
raise WeaveRevisionAlreadyPresent(name, self)
257
246
def add(self, name, parents, text, sha1=None):
258
247
"""Add a single text on top of the weave.
760
def join(self, other):
761
"""Integrate versions from other into this weave.
763
The resulting weave contains all the history of both weaves;
764
any version you could retrieve from either self or other can be
765
retrieved from self after this call.
767
It is illegal for the two weaves to contain different values
768
or different parents for any version. See also reweave().
770
if other.numversions() == 0:
771
return # nothing to update, easy
772
# two loops so that we do not change ourselves before verifying it
774
# work through in index order to make sure we get all dependencies
775
for other_idx, name in enumerate(other._names):
776
if self._check_version_consistent(other, other_idx, name):
778
for other_idx, name in enumerate(other._names):
779
# TODO: If all the parents of the other version are already
780
# present then we can avoid some work by just taking the delta
781
# and adjusting the offsets.
782
new_parents = self._imported_parents(other, other_idx)
783
lines = other.get_lines(other_idx)
784
sha1 = other._sha1s[other_idx]
785
self.add(name, new_parents, lines, sha1)
788
def _imported_parents(self, other, other_idx):
789
"""Return list of parents in self corresponding to indexes in other."""
791
for parent_idx in other._parents[other_idx]:
792
parent_name = other._names[parent_idx]
793
if parent_name not in self._names:
794
# should not be possible
795
raise WeaveError("missing parent {%s} of {%s} in %r"
796
% (parent_name, other._name_map[other_idx], self))
797
new_parents.append(self._name_map[parent_name])
800
def _check_version_consistent(self, other, other_idx, name):
801
"""Check if a version in consistent in this and other.
803
To be consistent it must have:
806
* the same direct parents (by name, not index, and disregarding
809
If present & correct return True;
810
if not present in self return False;
811
if inconsistent raise error."""
812
this_idx = self._name_map.get(name, -1)
814
if self._sha1s[this_idx] != other._sha1s[other_idx]:
815
raise WeaveError("inconsistent texts for version {%s} "
816
"when joining weaves"
818
self_parents = self._parents[this_idx]
819
other_parents = other._parents[other_idx]
820
n1 = [self._names[i] for i in self_parents]
821
n2 = [other._names[i] for i in other_parents]
825
raise WeaveParentMismatch("inconsistent parents "
826
"for version {%s}: %s vs %s" % (name, n1, n2))
832
def reweave(self, other):
833
"""Reweave self with other."""
834
new_weave = reweave(self, other)
835
for attr in self.__slots__:
836
setattr(self, attr, getattr(new_weave, attr))
840
"""Combine two weaves and return the result.
842
This works even if a revision R has different parents in
843
wa and wb. In the resulting weave all the parents are given.
845
This is done by just building up a new weave, maintaining ordering
846
of the versions in the two inputs. More efficient approaches
847
might be possible but it should only be necessary to do
848
this operation rarely, when a new previously ghost version is
853
queue_a = range(wa.numversions())
854
queue_b = range(wb.numversions())
855
# first determine combined parents of all versions
856
# map from version name -> all parent names
857
combined_parents = _reweave_parent_graphs(wa, wb)
858
mutter("combined parents: %r", combined_parents)
859
order = topo_sort(combined_parents.iteritems())
860
mutter("order to reweave: %r", order)
862
if name in wa._name_map:
863
lines = wa.get_lines(name)
864
if name in wb._name_map:
865
assert lines == wb.get_lines(name)
867
lines = wb.get_lines(name)
868
wr.add(name, combined_parents[name], lines)
872
def _reweave_parent_graphs(wa, wb):
873
"""Return combined parent ancestry for two weaves.
875
Returned as a list of (version_name, set(parent_names))"""
877
for weave in [wa, wb]:
878
for idx, name in enumerate(weave._names):
879
p = combined.setdefault(name, set())
880
p.update(map(weave.idx_to_name, weave._parents[idx]))
777
884
def weave_toc(w):
added
removed
bzrlib/weave.py
