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
|
# Copyright (C) 2012 Codethink Limited
#
# 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; version 2 of the License.
#
# 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.,
# 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
import collections
import os
import morphlib
class BuildDependencyGraph(object): # pragma: no cover
'''This class constructs a build dependency graph from an input morphology
and provides ways to traverse this graph. It also provides a method to
transform the dependency graph into groups of items that are independent
and can be built in parallel.'''
def __init__(self, loader, morph):
self.loader = loader
self.morph = morph
self.blobs = set()
def create_blob(self, morph, parent=None):
'''Creates a blob from a morphology. The optional parent is used to
associate chunks with their containing stratum.'''
if morph.kind == 'stratum':
return morphlib.blobs.Stratum(parent, morph)
elif morph.kind == 'chunk':
return morphlib.blobs.Chunk(parent, morph)
else:
return morphlib.blobs.System(parent, morph)
def resolve(self):
'''Constructs the dependency graph by resolving dependencies
recursively.'''
self.cached_blobs = {}
self._resolve_strata()
self._resolve_chunks()
def build_order(self):
'''Computes a topological sorting of the dependency graph and
generates a deque of groups, each of which contains a set
of items that are independent and can be built in parallel.'''
sorting = self._compute_topological_sorting()
groups = collections.deque()
# create the first group
group = set()
groups.append(group)
# traverse the graph in topological order
for blob in sorting:
# add the current item to the current group, or a new group
# if one of its dependencies is in the current one
create_group = False
for dependency in blob.dependencies:
if dependency in group:
create_group = True
if create_group:
group = set()
groups.append(group)
group.add(blob)
# return the set of blobs and the build groups
return set(self.blobs), groups
def _get_blob(self, info, parent=None):
'''Takes a (repo, ref, filename) tuple and looks up the blob for it.
It loads the corresponding morphology and blob on-demand if it is
not cached yet.'''
blob = self.cached_blobs.get(info, None)
if not blob:
morphology = self.loader.load(info[0], info[1], info[2])
blob = self.create_blob(morphology, parent)
self.cached_blobs[info] = blob
return blob
def _resolve_strata(self):
'''This method recursively generates a dependency graph of strata
for the input morphology using breadth-first search. It loads
morphologies and blobs on demand.'''
if self.morph.kind == 'stratum':
# turn the morphology into a stratum object
stratum = self.create_blob(self.morph)
# start the BFS at the input stratum
queue = collections.deque()
queue.append(stratum)
self.blobs.add(stratum)
while len(queue) > 0:
stratum = queue.popleft()
# the DFS recursion ends whenever we have a stratum
# that depends on nothing else
if not stratum.morph.build_depends:
continue
# verify that the build-depends format is valid
if isinstance(stratum.morph.build_depends, list):
for depname in stratum.morph.build_depends:
# prepare a tuple for the dependency stratum
repo = stratum.morph.repo
ref = stratum.morph.ref
filename = '%s.morph' % depname
info = (repo, ref, filename)
# load the dependency stratum on demand
depstratum = self._get_blob(info)
# add the dependency stratum to the graph
stratum.add_dependency(depstratum)
queue.append(depstratum)
self.blobs.add(depstratum)
else:
raise Exception('%s uses an invalid "build-depends" format'
% stratum)
def _resolve_chunks(self):
'''Starting with a dependency graph of strata, this method fills the
graph with all contained chunks and creates dependencies where
appropriate. Chunk morphologies and blobs are loaded on demand.'''
if self.morph.kind == 'chunk':
blob = self.create_blob(self.morph)
self.blobs.add(blob)
blobs = list(self.blobs)
for blob in blobs:
if isinstance(blob, morphlib.blobs.Stratum):
self._resolve_stratum_chunks(blob)
def _resolve_stratum_chunks(self, stratum):
# the set of chunks contained in the stratum
stratum_chunks = set()
# dictionary that maps chunk names to chunks
name_to_chunk = {}
# create objects for all chunks in the stratum
for i in xrange(0, len(stratum.morph.sources)):
source = stratum.morph.sources[i]
# construct a tuple for loading the chunk
repo = source['repo']
ref = source['ref']
filename = '%s.morph' % (source['morph']
if 'morph' in source
else source['name'])
info = (repo, ref, filename)
# load the chunk on demand
chunk = self._get_blob(info, stratum)
# store (name -> chunk) association to avoid loading the chunk twice
name_to_chunk[source['name']] = chunk
# read the build-depends information
build_depends = (source['build-depends']
if 'build-depends' in source
else None)
# turn build-depends into proper dependencies in the graph
if build_depends is None:
# chunks with no build-depends implicitly depend on all
# chunks listed earlier in the same stratum
for dependency in stratum_chunks:
chunk.add_dependency(dependency)
elif isinstance(build_depends, list):
for depname in build_depends:
if depname in name_to_chunk:
dependency = name_to_chunk[depname]
chunk.add_dependency(dependency)
else:
filename = os.path.basename(stratum.morph.filename)
raise Exception('%s: source %s references %s before it '
'is defined' % (filename,
source['name'],
depname))
else:
filename = os.path.basename(stratum.morph.filename)
raise Exception('%s: source %s uses an invalid build-depends '
'format' % (filename, source['name']))
# add the chunk to stratum and graph
stratum_chunks.add(chunk)
self.blobs.add(chunk)
# make the chunks in this stratum depend on all
# strata that need to be built first
for chunk in stratum_chunks:
for dependency in stratum.dependencies:
chunk.add_dependency(dependency)
# clear the dependencies of the stratum
stratum.dependencies = set()
# make the stratum depend on all its chunks
for chunk in stratum_chunks:
stratum.add_dependency(chunk)
def _compute_topological_sorting(self):
'''Computes a topological sorting of the dependency graph. A topological
sorting basically is the result of a series of breadth-first searches
starting at each leaf node (blobs with no dependencies). Blobs are
added to the sorting as soon as all their dependencies have been
added (which means that by then, all dependencies are satisfied).
http://en.wikipedia.org/wiki/Topological_sorting.'''
# map blobs to sets of satisfied dependencies. this is to detect when
# we can actually add blobs to the BFS queue. rather than dropping
# links between nodes, like most topological sorting algorithms do,
# we simply remember all satisfied dependencies and check if all
# of them are met repeatedly
satisfied_dependencies = {}
# create an empty sorting
sorting = collections.deque()
# create a set of leafs to start the DFS from
leafs = collections.deque()
for blob in self.blobs:
satisfied_dependencies[blob] = set()
if len(blob.dependencies) == 0:
leafs.append(blob)
while len(leafs) > 0:
# fetch a leaf blob from the DFS queue
blob = leafs.popleft()
# add it to the sorting
sorting.append(blob)
# mark this dependency as resolved
for dependent in blob.dependents:
satisfied_dependencies[dependent].add(blob)
# add the dependent blob as a leaf if all
# its dependencies have been resolved
has = len(satisfied_dependencies[dependent])
needs = len(dependent.dependencies)
if has == needs:
leafs.append(dependent)
# if not all dependencies were resolved on the way, we
# have found at least one cyclic dependency
if len(sorting) < len(self.blobs):
raise Exception('Cyclic dependencies found in the dependency '
'graph of "%s"' % self.morph)
return sorting
|