scancode-toolkit/src/licensedcode/cache.py at develop · aboutcode-org/scancode-toolkit · GitHub

History

566 lines (467 loc) · 19.6 KB

Raw

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

424

425

426

427

428

429

430

431

432

433

434

435

436

437

438

439

440

441

442

443

444

445

446

447

448

449

450

451

452

453

454

455

456

457

458

459

460

461

462

463

464

465

466

467

468

469

470

471

472

473

474

475

476

477

478

479

480

481

482

483

484

485

486

487

488

489

490

491

492

493

494

495

496

497

498

499

500

501

502

503

504

505

506

507

508

509

510

511

512

513

514

515

516

517

518

519

520

521

522

523

524

525

526

527

528

529

530

531

532

533

534

535

536

537

538

539

540

541

542

543

544

545

546

547

548

549

550

551

552

553

554

555

556

557

558

559

560

561

562

563

564

565

566

# ScanCode is a trademark of nexB Inc.

# SPDX-License-Identifier: Apache-2.0

# See http://www.apache.org/licenses/LICENSE-2.0 for the license text.

# See https://github.com/nexB/scancode-toolkit for support or download.

# See https://aboutcode.org for more information about nexB OSS projects.

import os

import pickle

from shutil import rmtree

from commoncode.fileutils import create_dir

from scancode_config import licensedcode_cache_dir

from scancode_config import scancode_cache_dir

"""

An on-disk persistent cache of LicenseIndex and related data structures such as

the licenses database. The data are pickled and must be regenerated if there

are any changes in the code or licenses text or rules. Loading and dumping the

cached pickle is safe to use across multiple processes using lock files.

"""

# This is the Pickle protocol we use, which was added in Python 3.4.

PICKLE_PROTOCOL = 4

# global in-memory cache of the LicenseCache

_LICENSE_CACHE = None

LICENSE_INDEX_LOCK_TIMEOUT = 60 * 6

LICENSE_INDEX_DIR = 'license_index'

LICENSE_INDEX_FILENAME = 'index_cache'

LICENSE_LOCKFILE_NAME = 'scancode_license_index_lockfile'

class LicenseCache:

"""

Represent cachable/pickable LicenseIndex and index-related objects.

"""

def __init__(

self,

db=None,

index=None,

licensing=None,

spdx_symbols=None,

unknown_spdx_symbol=None,

additional_license_directory=None,

additional_license_plugins=None,

# mapping of License objects by key

self.db = db

# LicenseIndex object

self.index = index

# Licensing object

self.licensing = licensing

# mapping of LicenseSymbol objects by SPDX key

self.spdx_symbols = spdx_symbols

# LicenseSymbol object

self.unknown_spdx_symbol = unknown_spdx_symbol

# Additional licenses from directory and plugins

self.additional_license_directory = additional_license_directory

self.additional_license_plugins = additional_license_plugins

@staticmethod

def load_or_build(

only_builtin=False,

licensedcode_cache_dir=licensedcode_cache_dir,

scancode_cache_dir=scancode_cache_dir,

force=False,

index_all_languages=False,

# used for testing only

timeout=LICENSE_INDEX_LOCK_TIMEOUT,

licenses_data_dir=None,

rules_data_dir=None,

additional_directory=None,

"""

Load or build and save and return a LicenseCache object.

We either load a cached LicenseIndex or build and cache the index.

On the side, we load cached or build license db, SPDX symbols and other

license-related data structures.

- If the cache exists, it is returned unless corrupted.

- If ``force`` is True, or if the cache does not exist a new index is built

and cached.

- If ``index_all_languages`` is True, include texts in all languages when

building the license index. Otherwise, only include the English license

texts and rules (the default)

- ``additional_directory`` is an optional additional directory

that contain additional licenses and rules in a /licenses and a /rules

directories using the same format that we use for licenses and rules.

"""

idx_cache_dir = os.path.join(licensedcode_cache_dir, LICENSE_INDEX_DIR)

if only_builtin:

rmtree(idx_cache_dir)

create_dir(idx_cache_dir)

cache_file = os.path.join(idx_cache_dir, LICENSE_INDEX_FILENAME)

has_cache = os.path.exists(cache_file) and os.path.getsize(cache_file)

# bypass build if cache exists

if has_cache and not force:

try:

# save the list of additional directories included in the cache, or None if the cache does not

# include any additional directories

return load_cache_file(cache_file)

except Exception as e:

# work around some rare Windows quirks

import traceback

print('Inconsistent License cache: rebuilding index.')

print(str(e))

print(traceback.format_exc())

from licensedcode.models import licenses_data_dir as ldd

from licensedcode.models import rules_data_dir as rdd

from licensedcode.models import load_licenses_from_multiple_dirs

from licensedcode.models import get_license_dirs

from licensedcode.models import validate_additional_license_data

from licensedcode.models import get_paths_to_installed_licenses_and_rules

from scancode import lockfile

licenses_data_dir = licenses_data_dir or ldd

rules_data_dir = rules_data_dir or rdd

lock_file = os.path.join(scancode_cache_dir, LICENSE_LOCKFILE_NAME)

# here, we have no cache: lock, check and rebuild

try:

# acquire lock and wait until timeout to get a lock or die

with lockfile.FileLock(lock_file).locked(timeout=timeout):

additional_directories = []

if only_builtin:

additional_directory = None

plugin_directories = []

else:

plugin_directories = get_paths_to_installed_licenses_and_rules()

if plugin_directories:

additional_directories.extend(plugin_directories)

# include installed licenses

if additional_directory:

# additional_directories is originally a tuple

additional_directories.append(additional_directory)

additional_license_dirs = get_license_dirs(additional_dirs=additional_directories)

validate_additional_license_data(

additional_directories=additional_license_dirs,

scancode_license_dir=licenses_data_dir

)

# Here, the cache is either stale or non-existing: we need to

# rebuild all cached data (e.g. mostly the index) and cache it

licenses_db = load_licenses_from_multiple_dirs(

additional_license_data_dirs=additional_license_dirs,

builtin_license_data_dir=licenses_data_dir,

)

# create a single merged index containing license data from licenses_data_dir

# and data from additional directories

index = build_index(

licenses_db=licenses_db,

licenses_data_dir=licenses_data_dir,

rules_data_dir=rules_data_dir,

index_all_languages=index_all_languages,

additional_directories=plugin_directories,

)

spdx_symbols = build_spdx_symbols(licenses_db=licenses_db)

unknown_spdx_symbol = build_unknown_spdx_symbol(licenses_db=licenses_db)

licensing = build_licensing(licenses_db=licenses_db)

license_cache = LicenseCache(

db=licenses_db,

index=index,

licensing=licensing,

spdx_symbols=spdx_symbols,

unknown_spdx_symbol=unknown_spdx_symbol,

additional_license_directory=additional_directory,

additional_license_plugins=plugin_directories,

)

license_cache.dump(cache_file)

return license_cache

except lockfile.LockTimeout:

# TODO: handle unable to lock in a nicer way

raise

@property

def has_additional_licenses(self):

cache = get_cache()

if cache.additional_license_directory or cache.additional_license_plugins:

return True

def dump(self, cache_file):

"""

Dump this license cache on disk at ``cache_file``.

"""

with open(cache_file, 'wb') as fn:

pickle.dump(self, fn, protocol=PICKLE_PROTOCOL)

def build_index(

licenses_db=None,

licenses_data_dir=None,

rules_data_dir=None,

index_all_languages=False,

additional_directories=None,

"""

Return an index built from rules and licenses directories

If ``index_all_languages`` is True, include texts and rules in all languages.

Otherwise, only include the English license texts and rules (the default)

If ``additional_directories`` is not None, we will include licenses and rules

from these additional directories in the returned index.

"""

from licensedcode.index import LicenseIndex

from licensedcode.models import get_license_dirs

from licensedcode.models import get_rule_dirs

from licensedcode.models import get_rules_from_multiple_dirs

from licensedcode.models import get_all_spdx_key_tokens

from licensedcode.models import get_license_tokens

from licensedcode.models import licenses_data_dir as ldd

from licensedcode.models import rules_data_dir as rdd

from licensedcode.models import load_licenses_from_multiple_dirs

from licensedcode.models import validate_ignorable_clues

from licensedcode.legalese import common_license_words

licenses_data_dir = licenses_data_dir or ldd

rules_data_dir = rules_data_dir or rdd

if not licenses_db:

# combine the licenses in these additional directories with the licenses in the original DB

additional_license_dirs = get_license_dirs(additional_dirs=additional_directories)

# generate a single combined license db with all licenses

licenses_db = load_licenses_from_multiple_dirs(

builtin_license_data_dir=licenses_data_dir,

additional_license_data_dirs=additional_license_dirs,

with_deprecated=False,

)

# if we have additional directories, extract the rules from them

additional_rule_dirs = get_rule_dirs(additional_dirs=additional_directories)

validate_ignorable_clues(rule_directories=additional_rule_dirs, is_builtin=False)

# then combine the rules in these additional directories with the rules in the original rules directory

rules = get_rules_from_multiple_dirs(

licenses_db=licenses_db,

additional_rules_data_dirs=additional_rule_dirs,

builtin_rule_data_dir=rules_data_dir,

)

legalese = common_license_words

spdx_tokens = set(get_all_spdx_key_tokens(licenses_db))

license_tokens = set(get_license_tokens())

# only skip licenses to be indexed

if not index_all_languages:

rules = (r for r in rules if r.language == 'en')

return LicenseIndex(

rules,

_legalese=legalese,

_spdx_tokens=spdx_tokens,

_license_tokens=license_tokens,

_all_languages=index_all_languages,

)

def build_licensing(licenses_db=None):

"""

Return a `license_expression.Licensing` objet built from a `licenses_db`

mapping of {key: License} or the standard license db.

"""

from license_expression import LicenseSymbolLike

from license_expression import Licensing

from licensedcode.models import load_licenses

licenses_db = licenses_db or load_licenses()

return Licensing(symbols=(LicenseSymbolLike(lic) for lic in licenses_db.values()))

def build_spdx_symbols(licenses_db=None):

"""

Return a mapping of {lowercased SPDX license key: LicenseSymbolLike} where

LicenseSymbolLike wraps a License object loaded from a `licenses_db` mapping

of {key: License} or the standard license db.

"""

from license_expression import LicenseSymbolLike

from licensedcode.models import load_licenses

licenses_db = licenses_db or load_licenses()

licenses_by_spdx_key = get_licenses_by_spdx_key(

licenses=licenses_db.values(),

include_deprecated=False,

lowercase_keys=True,

include_other_spdx_license_keys=True,

)

return {

spdx: LicenseSymbolLike(lic)

for spdx, lic in licenses_by_spdx_key.items()

}

def get_licenses_by_spdx_key(

licenses=None,

include_deprecated=False,

lowercase_keys=True,

include_other_spdx_license_keys=False,

"""

Return a mapping of {SPDX license id: License} where license is a License

object loaded from a `licenses` list of License or the standard

license db if not provided.

Optionally include deprecated if ``include_deprecated`` is True.

Optionally make the keys lowercase if ``lowercase_keys`` is True.

Optionally include the license "other_spdx_license_keys" if present and

``include_other_spdx_license_keys`` is True.

"""

from licensedcode.models import load_licenses

if not licenses:

licenses = load_licenses().values()

licenses_by_spdx_key = {}

for lic in licenses:

if not (lic.spdx_license_key or lic.other_spdx_license_keys):

continue

if lic.spdx_license_key:

slk = lic.spdx_license_key

if lowercase_keys:

slk = slk.lower()

existing = licenses_by_spdx_key.get(slk)

if existing and not lic.is_deprecated:

# temp hack for wharty ICU key!!

if slk not in ('icu', 'ICU',):

raise ValueError(

f'Duplicated SPDX license key: {slk!r} defined in '

f'{lic.key!r} and {existing!r}'

)

if (

not lic.is_deprecated

or (lic.is_deprecated and include_deprecated)

licenses_by_spdx_key[slk] = lic

if include_other_spdx_license_keys:

for other_spdx in lic.other_spdx_license_keys:

if not other_spdx or not other_spdx.strip():

continue

slk = other_spdx

if lowercase_keys:

slk = slk.lower()

existing = licenses_by_spdx_key.get(slk)

if existing:

raise ValueError(

f'Duplicated "other" SPDX license key: {slk!r} defined '

f'in {lic.key!r} and {existing!r}'

)

licenses_by_spdx_key[slk] = lic

return licenses_by_spdx_key

def build_unknown_spdx_symbol(licenses_db=None):

"""

Return the unknown SPDX license symbol given a `licenses_db` mapping of

{key: License} or the standard license db.

"""

from license_expression import LicenseSymbolLike

from licensedcode.models import load_licenses

licenses_db = licenses_db or load_licenses()

return LicenseSymbolLike(licenses_db['unknown-spdx'])

def get_cache(

only_builtin=False,

force=False,

index_all_languages=False,

additional_directory=None

"""

Return a LicenseCache either rebuilt, cached or loaded from disk.

If ``index_all_languages`` is True, include texts in all languages when

building the license index. Otherwise, only include the English license

texts and rules (the default)

"""

return populate_cache(

only_builtin=only_builtin,

force=force,

index_all_languages=index_all_languages,

additional_directory=additional_directory,

)

def populate_cache(

only_builtin=False,

force=False,

index_all_languages=False,

additional_directory=None

"""

Return, load or build and cache a LicenseCache.

"""

global _LICENSE_CACHE

if force or not _LICENSE_CACHE:

_LICENSE_CACHE = LicenseCache.load_or_build(

only_builtin=only_builtin,

licensedcode_cache_dir=licensedcode_cache_dir,

scancode_cache_dir=scancode_cache_dir,

force=force,

index_all_languages=index_all_languages,

# used for testing only

timeout=LICENSE_INDEX_LOCK_TIMEOUT,

additional_directory=additional_directory,

)

return _LICENSE_CACHE

def load_cache_file(cache_file):

"""

Return a LicenseCache loaded from ``cache_file``.

"""

with open(cache_file, 'rb') as lfc:

# Note: weird but read() + loads() is much (twice++???) faster than load()

try:

return pickle.load(lfc)

except Exception as e:

msg = (

'ERROR: Failed to load license cache (the file may be corrupted ?).\n'

f'Please delete "{cache_file}" and retry.\n'

'If the problem persists, copy this error message '

'and submit a bug report at https://github.com/nexB/scancode-toolkit/issues/'

)

raise Exception(msg) from e

def get_index(

only_builtin=False,

force=False,

index_all_languages=False,

additional_directory=None

"""

Return and eventually build and cache a LicenseIndex.

"""

return get_cache(

only_builtin=only_builtin,

force=force,

index_all_languages=index_all_languages,

additional_directory=additional_directory

).index

get_cached_index = get_index

def get_licenses_db():

"""

Return a mapping of license key -> license object.

"""

return get_cache().db

def get_licensing():

"""

Return a license_expression.Licensing objet built from the all the licenses.

"""

return get_cache().licensing

def get_unknown_spdx_symbol():

"""

Return the unknown SPDX license symbol.

"""

return get_cache().unknown_spdx_symbol

def get_spdx_symbols(licenses_db=None):

"""

Return a mapping of {lowercased SPDX license key: LicenseSymbolLike} where

LicenseSymbolLike wraps a License object

"""

if licenses_db:

return build_spdx_symbols(licenses_db)

return get_cache().spdx_symbols

def build_spdx_license_expression(license_expression, licensing=None):

"""

Return an SPDX license expression from a ScanCode ``license_expression``

string.

For example::

>>> exp = "mit OR gpl-2.0 with generic-exception"

>>> spdx = "MIT OR GPL-2.0-only WITH LicenseRef-scancode-generic-exception"

>>> assert build_spdx_license_expression(exp) == spdx

"""

if not license_expression:

return

if not licensing:

licensing = get_licensing()

validate_spdx_license_keys(license_expression=license_expression, licensing=licensing)

parsed = licensing.parse(license_expression)

return parsed.render(template='{symbol.wrapped.spdx_license_key}')

def validate_spdx_license_keys(license_expression, licensing):

"""

Raise InvalidLicenseKeyError for the cases where the there is no corresponding :

"""

from licensedcode.models import load_licenses

license_keys = licensing.license_keys(license_expression)

license_db = get_licenses_db()

messages = []

for key in license_keys:

if not type(key) == str:

msg = f"Invalid license key: {key} of type {type(key)}, license key should be a string"

messages.append(msg)

lic = license_db.get(key, None)

if not lic:

licenses = load_licenses(with_deprecated=True)

if licenses.get(key, None):

msg = f"License key: {key} is deprecated license key in LicenseDB"

else:

msg = f"License key: {key} is not a valid license key from LicenseDB"

messages.append(msg)

parsed = licensing.parse(key)

try:

parsed.render(template='{symbol.wrapped.spdx_license_key}')

except AttributeError:

msg = f"Error rendering SPDX license key for: {key}"

messages.append(msg)

pass

if messages:

raise InvalidLicenseKeyError(f"ERROR in parsing license_expression: {license_expression}: type: {type(license_expression)} :{messages}")

class InvalidLicenseKeyError(Exception):

pass

scancode-toolkit/src/licensedcode/cache.py at develop · aboutcode-org/scancode-toolkit · GitHub

Uh oh!

Uh oh!

scancode-toolkit/src/licensedcode/cache.py at develop · aboutcode-org/scancode-toolkit · GitHub

Uh oh!

Uh oh!

Expand file tree

Breadcrumbs

cache.py

Latest commit

History

Breadcrumbs

cache.py

File metadata and controls