File: graph.py

package info (click to toggle)
bzrtools 2.6.0-4
  • links: PTS, VCS
  • area: main
  • in suites: buster, sid
  • size: 740 kB
  • sloc: python: 4,299; sh: 13; makefile: 10
file content (423 lines) | stat: -rw-r--r-- 15,215 bytes parent folder | download | duplicates (4)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
# Copyright (C) 2005, 2008 Aaron Bentley
# <aaron@aaronbentley.com>
#
#    This program is free software; you can redistribute it and/or modify
#    it under the terms of the GNU General Public License as published by
#    the Free Software Foundation; either version 2 of the License, or
#    (at your option) any later version.
#
#    This program is distributed in the hope that it will be useful,
#    but WITHOUT ANY WARRANTY; without even the implied warranty of
#    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
#    GNU General Public License for more details.
#
#    You should have received a copy of the GNU General Public License
#    along with this program; if not, write to the Free Software
#    Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA


import time

from bzrlib.branch import Branch
from bzrlib.errors import BzrCommandError, NoSuchRevision
from bzrlib.revision import NULL_REVISION

from bzrtools import short_committer
from dotgraph import (
    dot_output,
    DOT_OUTPUT_TYPES,
    Edge,
    invoke_dot,
    invoke_dot_aa,
    invoke_dot_html,
    Node,
    NoDot,
    NoRsvg,
    RSVG_OUTPUT_TYPES,
    )


def max_distance(node, ancestors, distances, root_descendants):
    """Calculate the max distance to an ancestor.
    Return None if not all possible ancestors have known distances"""
    best = None
    if node in distances:
        best = distances[node]
    for ancestor in ancestors[node]:
        # skip ancestors we will never traverse:
        if root_descendants is not None and ancestor not in root_descendants:
            continue
        # An ancestor which is not listed in ancestors will never be in
        # distances, so we pretend it never existed.
        if ancestor not in ancestors:
            continue
        if ancestor not in distances:
            return None
        if best is None or distances[ancestor]+1 > best:
            best = distances[ancestor] + 1
    return best


def node_distances(graph, ancestors, start, root_descendants=None):
    """Produce a list of nodes, sorted by distance from a start node.
    This is an algorithm devised by Aaron Bentley, because applying Dijkstra
    backwards seemed too complicated.

    For each node, we walk its descendants.  If all the descendant's ancestors
    have a max-distance-to-start, (excluding ones that can never reach start),
    we calculate their max-distance-to-start, and schedule their descendants.

    So when a node's last parent acquires a distance, it will acquire a
    distance on the next iteration.

    Once we know the max distances for all nodes, we can return a list sorted
    by distance, farthest first.
    """
    distances = {start: 0}
    lines = set([start])
    while len(lines) > 0:
        new_lines = set()
        for line in lines:
            line_descendants = graph[line]
            for descendant in line_descendants:
                distance = max_distance(descendant, ancestors, distances,
                                        root_descendants)
                if distance is None:
                    continue
                distances[descendant] = distance
                new_lines.add(descendant)
        lines = new_lines
    return distances


def nodes_by_distance(distances):
    """Return a list of nodes sorted by distance"""
    def by_distance(n):
        return distances[n],n

    node_list = distances.keys()
    node_list.sort(key=by_distance, reverse=True)
    return node_list


def select_farthest(distances, common):
    """Return the farthest common node, or None if no node qualifies."""
    node_list = nodes_by_distance(distances)
    for node in node_list:
        if node in common:
            return node


mail_map = {'aaron.bentley@utoronto.ca'     : 'Aaron Bentley',
            'abentley@panoramicfeedback.com': 'Aaron Bentley',
            'abentley@lappy'                : 'Aaron Bentley',
            'john@arbash-meinel.com'        : 'John Arbash Meinel',
            'mbp@sourcefrog.net'            : 'Martin Pool',
            'robertc@robertcollins.net'     : 'Robert Collins',
            }

committer_alias = {'abentley': 'Aaron Bentley'}
def can_skip(rev_id, descendants, ancestors):
    if rev_id not in descendants:
        return False
    elif rev_id not in ancestors:
        return False
    elif len(ancestors[rev_id]) != 1:
        return False
    elif len(descendants[list(ancestors[rev_id])[0]]) != 1:
        return False
    elif len(descendants[rev_id]) != 1:
        return False
    else:
        return True

def compact_ancestors(descendants, ancestors, exceptions=()):
    new_ancestors={}
    skip = set()
    for me, my_parents in ancestors.iteritems():
        if me in skip:
            continue
        new_ancestors[me] = {}
        for parent in my_parents:
            new_parent = parent
            distance = 0
            while can_skip(new_parent, descendants, ancestors):
                if new_parent in exceptions:
                    break
                skip.add(new_parent)
                if new_parent in new_ancestors:
                    del new_ancestors[new_parent]
                new_parent = list(ancestors[new_parent])[0]
                distance += 1
            new_ancestors[me][new_parent] = distance
    return new_ancestors

def get_rev_info(rev_id, source):
    """Return the committer, message, nick and date of a revision."""
    committer = None
    message = None
    date = None
    nick = None
    if rev_id == 'null:':
        return None, 'Null Revision', None, None
    try:
        rev = source.get_revision(rev_id)
    except NoSuchRevision:
        try:
            committer = '-'.join(rev_id.split('-')[:-2]).strip(' ')
            if committer == '':
                return None, None, None, None
        except ValueError:
            return None, None, None, None
    else:
        committer = short_committer(rev.committer)
        if rev.message is not None:
            message = rev.message.split('\n')[0]
        gmtime = time.gmtime(rev.timestamp + (rev.timezone or 0))
        date = time.strftime('%Y/%m/%d', gmtime)
        nick = rev.properties.get('branch-nick')
    if '@' in committer:
        try:
            committer = mail_map[committer]
        except KeyError:
            pass
    try:
        committer = committer_alias[committer]
    except KeyError:
        pass
    return committer, message, nick, date

class Grapher(object):

    def __init__(self, branch, other_branch=None):
        object.__init__(self)
        self.branch = branch
        self.other_branch = other_branch
        if other_branch is not None:
            other_repo = other_branch.repository
            revision_b = self.other_branch.last_revision()
        else:
            other_repo = None
            revision_b = None
        self.graph = self.branch.repository.get_graph(other_repo)
        revision_a = self.branch.last_revision()
        self.scan_graph(revision_a, revision_b)
        self.n_history = list(self.graph.iter_lefthand_ancestry(revision_a))
        self.n_history.reverse()
        self.n_revnos = branch.get_revision_id_to_revno_map()
        self.distances = node_distances(self.descendants, self.ancestors,
                                        self.root)
        if other_branch is not None:
            self.base = select_farthest(self.distances, self.common)
            self.m_history = self.graph.iter_lefthand_ancestry(revision_b)
            self.m_history = list(self.m_history)
            self.m_history.reverse()
            self.m_revnos = other_branch.get_revision_id_to_revno_map()
            self.new_base = self.graph.find_unique_lca(revision_a,
                                                       revision_b)
            self.lcas = self.graph.find_lca(revision_a, revision_b)
        else:
            self.base = None
            self.new_base = None
            self.lcas = set()
            self.m_history = []
            self.m_revnos = {}

    def scan_graph(self, revision_a, revision_b):
        a_ancestors = dict(self.graph.iter_ancestry([revision_a]))
        self.ancestors = a_ancestors
        self.root = NULL_REVISION
        if revision_b is not None:
            b_ancestors = dict(self.graph.iter_ancestry([revision_b]))
            self.common = set(a_ancestors.keys())
            self.common.intersection_update(b_ancestors)
            self.ancestors.update(b_ancestors)
        else:
            self.common = []
            revision_b = None
        self.descendants = {}
        ghosts = set()
        for revision, parents in self.ancestors.iteritems():
            self.descendants.setdefault(revision, [])
            if parents is None:
                ghosts.add(revision)
                parents = [NULL_REVISION]
            for parent in parents:
                self.descendants.setdefault(parent, []).append(revision)
        for ghost in ghosts:
            self.ancestors[ghost] = [NULL_REVISION]

    @staticmethod
    def _get_revno_str(prefix, revno_map, revision_id):
        try:
            revno = revno_map[revision_id]
        except KeyError:
            return None
        return '%s%s' % (prefix, '.'.join(str(n) for n in revno))

    def dot_node(self, node, num):
        try:
            n_rev = self.n_history.index(node) + 1
        except ValueError:
            n_rev = None
        try:
            m_rev = self.m_history.index(node) + 1
        except ValueError:
            m_rev = None
        if (n_rev, m_rev) == (None, None):
            name = self._get_revno_str('r', self.n_revnos, node)
            if name is None:
                name = self._get_revno_str('R', self.m_revnos, node)
            if name is None:
                name = node[-5:]
            cluster = None
        elif n_rev == m_rev:
            name = "rR%d" % n_rev
        else:
            namelist = []
            for prefix, revno in (('r', n_rev), ('R', m_rev)):
                if revno is not None:
                    namelist.append("%s%d" % (prefix, revno))
            name = ' '.join(namelist)
        if None not in (n_rev, m_rev):
            cluster = "common_history"
            color = "#ff9900"
        elif (None, None) == (n_rev, m_rev):
            cluster = None
            if node in self.common:
                color = "#6699ff"
            else:
                color = "white"
        elif n_rev is not None:
            cluster = "my_history"
            color = "#ffff00"
        else:
            assert m_rev is not None
            cluster = "other_history"
            color = "#ff0000"
        if node in self.lcas:
            color = "#9933cc"
        if node == self.base:
            color = "#669933"
            if node == self.new_base:
                color = "#33ff33"
        if node == self.new_base:
            color = '#33cc99'

        label = [name]
        committer, message, nick, date = get_rev_info(node,
                                                      self.branch.repository)
        if committer is not None:
            label.append(committer)

        if nick is not None:
            label.append(nick)

        if date is not None:
            label.append(date)

        if node in self.distances:
            rank = self.distances[node]
            label.append('d%d' % self.distances[node])
        else:
            rank = None

        d_node = Node("n%d" % num, color=color, label="\\n".join(label),
                    rev_id=node, cluster=cluster, message=message,
                    date=date)
        d_node.rank = rank

        if node not in self.ancestors:
            d_node.node_style.append('dotted')

        return d_node

    def get_relations(self, collapse=False, max_distance=None):
        dot_nodes = {}
        node_relations = []
        num = 0
        if collapse:
            exceptions = self.lcas.union([self.base, self.new_base])
            visible_ancestors = compact_ancestors(self.descendants,
                                                  self.ancestors,
                                                  exceptions)
        else:
            visible_ancestors = {}
            for revision, parents in self.ancestors.iteritems():
                visible_ancestors[revision] = dict((p, 0) for p in parents)
        if max_distance is not None:
            min_distance = max(self.distances.values()) - max_distance
            visible_ancestors = dict((n, p) for n, p in
                                     visible_ancestors.iteritems() if
                                     self.distances[n] >= min_distance)
        for node, parents in visible_ancestors.iteritems():
            if node not in dot_nodes:
                dot_nodes[node] = self.dot_node(node, num)
                num += 1
            for parent, skipped in parents.iteritems():
                if parent not in dot_nodes:
                    dot_nodes[parent] = self.dot_node(parent, num)
                    num += 1
                edge = Edge(dot_nodes[parent], dot_nodes[node])
                if skipped != 0:
                    edge.label = "%d" % skipped
                node_relations.append(edge)
        return node_relations


def write_ancestry_file(branch, filename, collapse=True, antialias=True,
                        merge_branch=None, ranking="forced", max_distance=None):
    b = Branch.open_containing(branch)[0]
    if merge_branch is not None:
        m = Branch.open_containing(merge_branch)[0]
    else:
        m = None
    b.lock_write()
    try:
        if m is not None:
            m.lock_read()
        try:
            grapher = Grapher(b, m)
            relations = grapher.get_relations(collapse, max_distance)
        finally:
            if m is not None:
                m.unlock()
    finally:
        b.unlock()

    ext = filename.split('.')[-1]
    output = dot_output(relations, ranking)
    done = False
    if ext not in RSVG_OUTPUT_TYPES:
        antialias = False
    if antialias:
        output = list(output)
        try:
            invoke_dot_aa(output, filename, ext)
            done = True
        except NoDot, e:
            raise BzrCommandError("Can't find 'dot'.  Please ensure Graphviz"\
                " is installed correctly.")
        except NoRsvg, e:
            print "Not antialiasing because rsvg (from librsvg-bin) is not"\
                " installed."
            antialias = False
    if ext in DOT_OUTPUT_TYPES and not antialias and not done:
        try:
            invoke_dot(output, filename, ext)
            done = True
        except NoDot, e:
            raise BzrCommandError("Can't find 'dot'.  Please ensure Graphviz"\
                " is installed correctly.")
    elif ext == 'dot' and not done:
        my_file = file(filename, 'wb')
        for fragment in output:
            my_file.write(fragment.encode('utf-8'))
    elif ext == 'html':
        try:
            invoke_dot_html(output, filename)
        except NoDot, e:
            raise BzrCommandError("Can't find 'dot'.  Please ensure Graphviz"\
                " is installed correctly.")
    elif not done:
        print "Unknown file extension: %s" % ext