Hide keyboard shortcuts

Hot-keys on this page

r m x p   toggle line displays

j k   next/prev highlighted chunk

0   (zero) top of page

1   (one) first highlighted chunk

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

# This code is part of Ansible, but is an independent component. 

# This particular file snippet, and this file snippet only, is BSD licensed. 

# Modules you write using this snippet, which is embedded dynamically by Ansible 

# still belong to the author of the module, and may assign their own license 

# to the complete work. 

# 

# (c) 2017 Red Hat Inc. 

# 

# Redistribution and use in source and binary forms, with or without modification, 

# are permitted provided that the following conditions are met: 

# 

# * Redistributions of source code must retain the above copyright 

# notice, this list of conditions and the following disclaimer. 

# * Redistributions in binary form must reproduce the above copyright notice, 

# this list of conditions and the following disclaimer in the documentation 

# and/or other materials provided with the distribution. 

# 

# THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND 

# ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED 

# WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. 

# IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, 

# INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, 

# PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS 

# INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT 

# LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE 

# USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. 

# 

 

from __future__ import (absolute_import, division, print_function) 

__metaclass__ = type 

 

from collections import defaultdict 

 

import platform 

 

from ansible.module_utils.facts import timeout 

 

 

class CycleFoundInFactDeps(Exception): 

'''Indicates there is a cycle in fact collector deps 

 

If collector-B requires collector-A, and collector-A requires 

collector-B, that is a cycle. In that case, there is no ordering 

that will satisfy B before A and A and before B. That will cause this 

error to be raised. 

''' 

pass 

 

 

class UnresolvedFactDep(ValueError): 

pass 

 

 

class CollectorNotFoundError(KeyError): 

pass 

 

 

class BaseFactCollector: 

_fact_ids = set() 

 

_platform = 'Generic' 

name = None 

required_facts = set() 

 

def __init__(self, collectors=None, namespace=None): 

'''Base class for things that collect facts. 

 

'collectors' is an optional list of other FactCollectors for composing.''' 

self.collectors = collectors or [] 

 

# self.namespace is a object with a 'transform' method that transforms 

# the name to indicate the namespace (ie, adds a prefix or suffix). 

self.namespace = namespace 

 

self.fact_ids = set([self.name]) 

self.fact_ids.update(self._fact_ids) 

 

@classmethod 

def platform_match(cls, platform_info): 

if platform_info.get('system', None) == cls._platform: 

return cls 

return None 

 

def _transform_name(self, key_name): 

if self.namespace: 

return self.namespace.transform(key_name) 

return key_name 

 

def _transform_dict_keys(self, fact_dict): 

'''update a dicts keys to use new names as transformed by self._transform_name''' 

 

for old_key in list(fact_dict.keys()): 

new_key = self._transform_name(old_key) 

# pop the item by old_key and replace it using new_key 

fact_dict[new_key] = fact_dict.pop(old_key) 

return fact_dict 

 

# TODO/MAYBE: rename to 'collect' and add 'collect_without_namespace' 

def collect_with_namespace(self, module=None, collected_facts=None): 

# collect, then transform the key names if needed 

facts_dict = self.collect(module=module, collected_facts=collected_facts) 

if self.namespace: 

facts_dict = self._transform_dict_keys(facts_dict) 

return facts_dict 

 

def collect(self, module=None, collected_facts=None): 

'''do the fact collection 

 

'collected_facts' is a object (a dict, likely) that holds all previously 

facts. This is intended to be used if a FactCollector needs to reference 

another fact (for ex, the system arch) and should not be modified (usually). 

 

Returns a dict of facts. 

 

''' 

facts_dict = {} 

return facts_dict 

 

 

def get_collector_names(valid_subsets=None, 

minimal_gather_subset=None, 

gather_subset=None, 

aliases_map=None, 

platform_info=None): 

'''return a set of FactCollector names based on gather_subset spec. 

 

gather_subset is a spec describing which facts to gather. 

valid_subsets is a frozenset of potential matches for gather_subset ('all', 'network') etc 

minimal_gather_subsets is a frozenset of matches to always use, even for gather_subset='!all' 

''' 

 

# Retrieve module parameters 

gather_subset = gather_subset or ['all'] 

 

# the list of everything that 'all' expands to 

valid_subsets = valid_subsets or frozenset() 

 

# if provided, minimal_gather_subset is always added, even after all negations 

minimal_gather_subset = minimal_gather_subset or frozenset() 

 

aliases_map = aliases_map or defaultdict(set) 

 

# Retrieve all facts elements 

additional_subsets = set() 

exclude_subsets = set() 

 

# total always starts with the min set, then 

# adds of the additions in gather_subset, then 

# excludes all of the excludes, then add any explicitly 

# requested subsets. 

gather_subset_with_min = ['min'] 

gather_subset_with_min.extend(gather_subset) 

 

# subsets we mention in gather_subset explicitly, except for 'all'/'min' 

explicitly_added = set() 

 

for subset in gather_subset_with_min: 

subset_id = subset 

if subset_id == 'min': 

additional_subsets.update(minimal_gather_subset) 

continue 

if subset_id == 'all': 

additional_subsets.update(valid_subsets) 

continue 

if subset_id.startswith('!'): 

subset = subset[1:] 

if subset == 'min': 

exclude_subsets.update(minimal_gather_subset) 

continue 

if subset == 'all': 

exclude_subsets.update(valid_subsets - minimal_gather_subset) 

continue 

exclude = True 

else: 

exclude = False 

 

if exclude: 

# include 'devices', 'dmi' etc for '!hardware' 

exclude_subsets.update(aliases_map.get(subset, set())) 

exclude_subsets.add(subset) 

else: 

# NOTE: this only considers adding an unknown gather subsetup an error. Asking to 

# exclude an unknown gather subset is ignored. 

if subset_id not in valid_subsets: 

raise TypeError("Bad subset '%s' given to Ansible. gather_subset options allowed: all, %s" % 

(subset, ", ".join(sorted(valid_subsets)))) 

 

explicitly_added.add(subset) 

additional_subsets.add(subset) 

 

if not additional_subsets: 

additional_subsets.update(valid_subsets) 

 

additional_subsets.difference_update(exclude_subsets - explicitly_added) 

 

return additional_subsets 

 

 

def find_collectors_for_platform(all_collector_classes, compat_platforms): 

found_collectors = set() 

found_collectors_names = set() 

 

# start from specific platform, then try generic 

for compat_platform in compat_platforms: 

platform_match = None 

for all_collector_class in all_collector_classes: 

 

# ask the class if it is compatible with the platform info 

platform_match = all_collector_class.platform_match(compat_platform) 

 

if not platform_match: 

continue 

 

primary_name = all_collector_class.name 

 

if primary_name not in found_collectors_names: 

found_collectors.add(all_collector_class) 

found_collectors_names.add(all_collector_class.name) 

 

return found_collectors 

 

 

def build_fact_id_to_collector_map(collectors_for_platform): 

fact_id_to_collector_map = defaultdict(list) 

aliases_map = defaultdict(set) 

 

for collector_class in collectors_for_platform: 

primary_name = collector_class.name 

 

fact_id_to_collector_map[primary_name].append(collector_class) 

 

for fact_id in collector_class._fact_ids: 

fact_id_to_collector_map[fact_id].append(collector_class) 

aliases_map[primary_name].add(fact_id) 

 

return fact_id_to_collector_map, aliases_map 

 

 

def select_collector_classes(collector_names, all_fact_subsets): 

seen_collector_classes = set() 

 

selected_collector_classes = [] 

 

for collector_name in collector_names: 

collector_classes = all_fact_subsets.get(collector_name, []) 

for collector_class in collector_classes: 

if collector_class not in seen_collector_classes: 

selected_collector_classes.append(collector_class) 

seen_collector_classes.add(collector_class) 

 

return selected_collector_classes 

 

 

def _get_requires_by_collector_name(collector_name, all_fact_subsets): 

required_facts = set() 

 

try: 

collector_classes = all_fact_subsets[collector_name] 

except KeyError: 

raise CollectorNotFoundError('Fact collector "%s" not found' % collector_name) 

for collector_class in collector_classes: 

required_facts.update(collector_class.required_facts) 

return required_facts 

 

 

def find_unresolved_requires(collector_names, all_fact_subsets): 

'''Find any collector names that have unresolved requires 

 

Returns a list of collector names that correspond to collector 

classes whose .requires_facts() are not in collector_names. 

''' 

unresolved = set() 

 

for collector_name in collector_names: 

required_facts = _get_requires_by_collector_name(collector_name, all_fact_subsets) 

for required_fact in required_facts: 

if required_fact not in collector_names: 

unresolved.add(required_fact) 

 

return unresolved 

 

 

def resolve_requires(unresolved_requires, all_fact_subsets): 

new_names = set() 

failed = [] 

for unresolved in unresolved_requires: 

if unresolved in all_fact_subsets: 

new_names.add(unresolved) 

else: 

failed.append(unresolved) 

 

if failed: 

raise UnresolvedFactDep('unresolved fact dep %s' % ','.join(failed)) 

return new_names 

 

 

def build_dep_data(collector_names, all_fact_subsets): 

dep_map = defaultdict(set) 

for collector_name in collector_names: 

collector_deps = set() 

for collector in all_fact_subsets[collector_name]: 

for dep in collector.required_facts: 

collector_deps.add(dep) 

dep_map[collector_name] = collector_deps 

return dep_map 

 

 

def tsort(dep_map): 

sorted_list = [] 

 

unsorted_map = dep_map.copy() 

 

while unsorted_map: 

acyclic = False 

for node, edges in list(unsorted_map.items()): 

for edge in edges: 

if edge in unsorted_map: 

break 

else: 

acyclic = True 

del unsorted_map[node] 

sorted_list.append((node, edges)) 

 

if not acyclic: 

raise CycleFoundInFactDeps('Unable to tsort deps, there was a cycle in the graph. sorted=%s' % sorted_list) 

 

return sorted_list 

 

 

def _solve_deps(collector_names, all_fact_subsets): 

unresolved = collector_names.copy() 

solutions = collector_names.copy() 

 

while True: 

unresolved = find_unresolved_requires(solutions, all_fact_subsets) 

if unresolved == set(): 

break 

 

new_names = resolve_requires(unresolved, all_fact_subsets) 

solutions.update(new_names) 

 

return solutions 

 

 

def collector_classes_from_gather_subset(all_collector_classes=None, 

valid_subsets=None, 

minimal_gather_subset=None, 

gather_subset=None, 

gather_timeout=None, 

platform_info=None): 

'''return a list of collector classes that match the args''' 

 

# use gather_name etc to get the list of collectors 

 

all_collector_classes = all_collector_classes or [] 

 

minimal_gather_subset = minimal_gather_subset or frozenset() 

 

platform_info = platform_info or {'system': platform.system()} 

 

gather_timeout = gather_timeout or timeout.DEFAULT_GATHER_TIMEOUT 

 

# tweak the modules GATHER_TIMEOUT 

timeout.GATHER_TIMEOUT = gather_timeout 

 

valid_subsets = valid_subsets or frozenset() 

 

# maps alias names like 'hardware' to the list of names that are part of hardware 

# like 'devices' and 'dmi' 

aliases_map = defaultdict(set) 

 

compat_platforms = [platform_info, {'system': 'Generic'}] 

 

collectors_for_platform = find_collectors_for_platform(all_collector_classes, compat_platforms) 

 

# all_facts_subsets maps the subset name ('hardware') to the class that provides it. 

 

# TODO: name collisions here? are there facts with the same name as a gather_subset (all, network, hardware, virtual, ohai, facter) 

all_fact_subsets, aliases_map = build_fact_id_to_collector_map(collectors_for_platform) 

 

all_valid_subsets = frozenset(all_fact_subsets.keys()) 

 

# expand any fact_id/collectorname/gather_subset term ('all', 'env', etc) to the list of names that represents 

collector_names = get_collector_names(valid_subsets=all_valid_subsets, 

minimal_gather_subset=minimal_gather_subset, 

gather_subset=gather_subset, 

aliases_map=aliases_map, 

platform_info=platform_info) 

 

complete_collector_names = _solve_deps(collector_names, all_fact_subsets) 

 

dep_map = build_dep_data(complete_collector_names, all_fact_subsets) 

 

ordered_deps = tsort(dep_map) 

ordered_collector_names = [x[0] for x in ordered_deps] 

 

selected_collector_classes = select_collector_classes(ordered_collector_names, 

all_fact_subsets) 

 

return selected_collector_classes