3
# Copyright (C) 2005 Canonical Ltd
5
# This program is free software; you can redistribute it and/or modify
6
# it under the terms of the GNU General Public License as published by
7
# the Free Software Foundation; either version 2 of the License, or
8
# (at your option) any later version.
10
# This program is distributed in the hope that it will be useful,
11
# but WITHOUT ANY WARRANTY; without even the implied warranty of
12
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
13
# GNU General Public License for more details.
15
# You should have received a copy of the GNU General Public License
16
# along with this program; if not, write to the Free Software
17
# Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
19
"""Experiment in converting existing bzr branches to weaves."""
21
# To make this properly useful
23
# 1. assign text version ids, and put those text versions into
24
# the inventory as they're converted.
26
# 2. keep track of the previous version of each file, rather than
27
# just using the last one imported
29
# 3. assign entry versions when files are added, renamed or moved.
31
# 4. when merged-in versions are observed, walk down through them
32
# to discover everything, then commit bottom-up
34
# 5. track ancestry as things are merged in, and commit that in each
37
# Perhaps it's best to first walk the whole graph and make a plan for
38
# what should be imported in what order? Need a kind of topological
39
# sort of all revisions. (Or do we, can we just before doing a revision
40
# see that all its parents have either been converted or abandoned?)
51
import hotshot, hotshot.stats
56
from bzrlib.branch import Branch, find_branch
57
from bzrlib.revfile import Revfile
58
from bzrlib.weave import Weave
59
from bzrlib.weavefile import read_weave, write_weave
60
from bzrlib.progress import ProgressBar
61
from bzrlib.atomicfile import AtomicFile
62
from bzrlib.xml4 import serializer_v4
63
from bzrlib.xml5 import serializer_v5
68
class Convert(object):
71
self.converted_revs = 0
78
bzrlib.trace.enable_default_logging()
79
self.pb = ProgressBar()
80
self.inv_weave = Weave('__inventory')
81
self.anc_weave = Weave('__ancestry')
85
# holds in-memory weaves for all files
88
b = self.branch = Branch('.', relax_version_check=True)
91
rev_history = b.revision_history()
95
# todo is a stack holding the revisions we still need to process;
96
# appending to it adds new highest-priority revisions
99
self.total_revs = len(todo)
102
self._convert_one_rev(todo.pop())
105
print 'upgraded to weaves:'
106
print ' %6d revisions and inventories' % self.converted_revs
107
print ' %6d texts' % self.text_count
109
self._write_all_weaves()
112
def _write_all_weaves(self):
114
return ############################################
115
# TODO: commit them all atomically at the end, not one by one
116
write_atomic_weave(self.inv_weave, 'weaves/inventory.weave')
117
write_atomic_weave(self.anc_weave, 'weaves/ancestry.weave')
118
for file_id, file_weave in text_weaves.items():
119
self.pb.update('writing weave', i, len(text_weaves))
120
write_atomic_weave(file_weave, 'weaves/%s.weave' % file_id)
126
def _convert_one_rev(self, rev_id):
127
self._bump_progress()
129
rev_xml = b.revision_store[rev_id].read()
130
inv_xml = b.inventory_store[rev_id].read()
132
rev = serializer_v4.read_revision_from_string(rev_xml)
133
inv = serializer_v4.read_inventory_from_string(inv_xml)
135
return ##########################################
137
new_idx = self.inv_weave.add(rev_id, inv_parents, inv_xml)
138
inv_parents = [new_idx]
140
tree = b.revision_tree(rev_id)
143
# for each file in the inventory, put it into its own revfile
146
if ie.kind != 'file':
148
if last_text_sha.get(file_id) == ie.text_sha1:
151
last_text_sha[file_id] = ie.text_sha1
153
# new text (though possibly already stored); need to store it
154
text_lines = tree.get_file(file_id).readlines()
156
# if the file's created for the first time in this
157
# revision then make a new weave; else find the old one
158
if file_id not in text_weaves:
159
text_weaves[file_id] = Weave()
161
w = text_weaves[file_id]
163
# base the new text version off whatever was last
164
# (actually it'd be better to track this, to allow for
165
# files that are deleted and then reappear)
172
w.add(rev_id, parents, text_lines)
177
def _bump_progress(self):
178
self.converted_revs += 1
179
self.pb.update('converting revisions',
184
def write_atomic_weave(weave, filename):
185
inv_wf = AtomicFile(filename)
187
write_weave(weave, inv_wf)
195
def profile_convert():
196
prof_f = tempfile.NamedTemporaryFile()
198
prof = hotshot.Profile(prof_f.name)
200
prof.runcall(Convert)
203
stats = hotshot.stats.load(prof_f.name)
205
stats.sort_stats('time')
206
# XXX: Might like to write to stderr or the trace file instead but
207
# print_stats seems hardcoded to stdout
208
stats.print_stats(20)
211
if '-p' in sys.argv[1:]:
added
removed
tools/history2weaves.py
