You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.

test_runner.py 20KB

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132133134135136137138139140141142143144145146147148149150151152153154155156157158159160161162163164165166167168169170171172173174175176177178179180181182183184185186187188189190191192193194195196197198199200201202203204205206207208209210211212213214215216217218219220221222223224225226227228229230231232233234235236237238239240241242243244245246247248249250251252253254255256257258259260261262263264265266267268269270271272273274275276277278279280281282283284285286287288289290291292293294295296297298299300301302303304305306307308309310311312313314315316317318319320321322323324325326327328329330331332333334335336337338339340341342343344345346347348349350351352353354355356357358359360361362363364365366367368369370371372373374375376377378379380381382383384385386387388389390391392393394395396397398399400401402403404405406407408409410411412413414415416417418419420421422423424425426427428429430431432433434435436437438439440441442443444445446447448449450451452453454455456457458459460461462463464465466467468469470471472473474475476477478479480481482483484485486487488489490491492493494495496497498499500501502503504505506507508509510511512513514515516517518519520521522523524525526527528
  1. #!/usr/bin/env python3
  2. # Copyright (c) 2014-2016 The Bitcoin Core developers
  3. # Distributed under the MIT software license, see the accompanying
  4. # file COPYING or http://www.opensource.org/licenses/mit-license.php.
  5. """Run regression test suite.
  6. This module calls down into individual test cases via subprocess. It will
  7. forward all unrecognized arguments onto the individual test scripts.
  8. Functional tests are disabled on Windows by default. Use --force to run them anyway.
  9. For a description of arguments recognized by test scripts, see
  10. `test/functional/test_framework/test_framework.py:BitcoinTestFramework.main`.
  11. """
  12. import argparse
  13. import configparser
  14. import datetime
  15. import os
  16. import time
  17. import shutil
  18. import signal
  19. import sys
  20. import subprocess
  21. import tempfile
  22. import re
  23. import logging
  24. # Formatting. Default colors to empty strings.
  25. BOLD, BLUE, RED, GREY = ("", ""), ("", ""), ("", ""), ("", "")
  26. try:
  27. # Make sure python thinks it can write unicode to its stdout
  28. "\u2713".encode("utf_8").decode(sys.stdout.encoding)
  29. TICK = "✓ "
  30. CROSS = "✖ "
  31. CIRCLE = "○ "
  32. except UnicodeDecodeError:
  33. TICK = "P "
  34. CROSS = "x "
  35. CIRCLE = "o "
  36. if os.name == 'posix':
  37. # primitive formatting on supported
  38. # terminal via ANSI escape sequences:
  39. BOLD = ('\033[0m', '\033[1m')
  40. BLUE = ('\033[0m', '\033[0;34m')
  41. RED = ('\033[0m', '\033[0;31m')
  42. GREY = ('\033[0m', '\033[1;30m')
  43. TEST_EXIT_PASSED = 0
  44. TEST_EXIT_SKIPPED = 77
  45. BASE_SCRIPTS= [
  46. # Scripts that are run by the travis build process.
  47. # Longest test should go first, to favor running tests in parallel
  48. 'wallet-hd.py',
  49. 'walletbackup.py',
  50. # vv Tests less than 5m vv
  51. 'p2p-fullblocktest.py',
  52. 'fundrawtransaction.py',
  53. 'p2p-compactblocks.py',
  54. 'segwit.py',
  55. # vv Tests less than 2m vv
  56. 'wallet.py',
  57. 'wallet-accounts.py',
  58. 'p2p-segwit.py',
  59. 'wallet-dump.py',
  60. 'listtransactions.py',
  61. # vv Tests less than 60s vv
  62. 'sendheaders.py',
  63. 'zapwallettxes.py',
  64. 'importmulti.py',
  65. 'mempool_limit.py',
  66. 'merkle_blocks.py',
  67. 'receivedby.py',
  68. 'abandonconflict.py',
  69. 'bip68-112-113-p2p.py',
  70. 'rawtransactions.py',
  71. 'reindex.py',
  72. # vv Tests less than 30s vv
  73. 'keypool-topup.py',
  74. 'zmq_test.py',
  75. 'bitcoin_cli.py',
  76. 'mempool_resurrect_test.py',
  77. 'txn_doublespend.py --mineblock',
  78. 'txn_clone.py',
  79. 'getchaintips.py',
  80. 'rest.py',
  81. 'mempool_spendcoinbase.py',
  82. 'mempool_reorg.py',
  83. 'mempool_persist.py',
  84. 'multiwallet.py',
  85. 'httpbasics.py',
  86. 'multi_rpc.py',
  87. 'proxy_test.py',
  88. 'signrawtransactions.py',
  89. 'disconnect_ban.py',
  90. 'decodescript.py',
  91. 'blockchain.py',
  92. 'disablewallet.py',
  93. 'net.py',
  94. 'keypool.py',
  95. 'p2p-mempool.py',
  96. 'prioritise_transaction.py',
  97. 'invalidblockrequest.py',
  98. 'invalidtxrequest.py',
  99. 'p2p-versionbits-warning.py',
  100. 'preciousblock.py',
  101. 'importprunedfunds.py',
  102. 'signmessages.py',
  103. 'nulldummy.py',
  104. 'import-rescan.py',
  105. 'mining.py',
  106. 'bumpfee.py',
  107. 'rpcnamedargs.py',
  108. 'listsinceblock.py',
  109. 'p2p-leaktests.py',
  110. 'wallet-encryption.py',
  111. 'bipdersig-p2p.py',
  112. 'bip65-cltv-p2p.py',
  113. 'uptime.py',
  114. 'resendwallettransactions.py',
  115. 'minchainwork.py',
  116. 'p2p-acceptblock.py',
  117. ]
  118. EXTENDED_SCRIPTS = [
  119. # These tests are not run by the travis build process.
  120. # Longest test should go first, to favor running tests in parallel
  121. 'pruning.py',
  122. # vv Tests less than 20m vv
  123. 'smartfees.py',
  124. # vv Tests less than 5m vv
  125. 'maxuploadtarget.py',
  126. 'mempool_packages.py',
  127. 'dbcrash.py',
  128. # vv Tests less than 2m vv
  129. 'bip68-sequence.py',
  130. 'getblocktemplate_longpoll.py',
  131. 'p2p-timeouts.py',
  132. # vv Tests less than 60s vv
  133. 'bip9-softforks.py',
  134. 'p2p-feefilter.py',
  135. 'rpcbind_test.py',
  136. # vv Tests less than 30s vv
  137. 'assumevalid.py',
  138. 'example_test.py',
  139. 'txn_doublespend.py',
  140. 'txn_clone.py --mineblock',
  141. 'forknotify.py',
  142. 'invalidateblock.py',
  143. 'replace-by-fee.py',
  144. ]
  145. # Place EXTENDED_SCRIPTS first since it has the 3 longest running tests
  146. ALL_SCRIPTS = EXTENDED_SCRIPTS + BASE_SCRIPTS
  147. NON_SCRIPTS = [
  148. # These are python files that live in the functional tests directory, but are not test scripts.
  149. "combine_logs.py",
  150. "create_cache.py",
  151. "test_runner.py",
  152. ]
  153. def main():
  154. # Parse arguments and pass through unrecognised args
  155. parser = argparse.ArgumentParser(add_help=False,
  156. usage='%(prog)s [test_runner.py options] [script options] [scripts]',
  157. description=__doc__,
  158. epilog='''
  159. Help text and arguments for individual test script:''',
  160. formatter_class=argparse.RawTextHelpFormatter)
  161. parser.add_argument('--coverage', action='store_true', help='generate a basic coverage report for the RPC interface')
  162. parser.add_argument('--exclude', '-x', help='specify a comma-seperated-list of scripts to exclude.')
  163. parser.add_argument('--extended', action='store_true', help='run the extended test suite in addition to the basic tests')
  164. parser.add_argument('--force', '-f', action='store_true', help='run tests even on platforms where they are disabled by default (e.g. windows).')
  165. parser.add_argument('--help', '-h', '-?', action='store_true', help='print help text and exit')
  166. parser.add_argument('--jobs', '-j', type=int, default=4, help='how many test scripts to run in parallel. Default=4.')
  167. parser.add_argument('--keepcache', '-k', action='store_true', help='the default behavior is to flush the cache directory on startup. --keepcache retains the cache from the previous testrun.')
  168. parser.add_argument('--quiet', '-q', action='store_true', help='only print results summary and failure logs')
  169. parser.add_argument('--tmpdirprefix', '-t', default=tempfile.gettempdir(), help="Root directory for datadirs")
  170. args, unknown_args = parser.parse_known_args()
  171. # args to be passed on always start with two dashes; tests are the remaining unknown args
  172. tests = [arg for arg in unknown_args if arg[:2] != "--"]
  173. passon_args = [arg for arg in unknown_args if arg[:2] == "--"]
  174. # Read config generated by configure.
  175. config = configparser.ConfigParser()
  176. configfile = os.path.abspath(os.path.dirname(__file__)) + "/../config.ini"
  177. config.read_file(open(configfile))
  178. passon_args.append("--configfile=%s" % configfile)
  179. # Set up logging
  180. logging_level = logging.INFO if args.quiet else logging.DEBUG
  181. logging.basicConfig(format='%(message)s', level=logging_level)
  182. # Create base test directory
  183. tmpdir = "%s/bitcoin_test_runner_%s" % (args.tmpdirprefix, datetime.datetime.now().strftime("%Y%m%d_%H%M%S"))
  184. os.makedirs(tmpdir)
  185. logging.debug("Temporary test directory at %s" % tmpdir)
  186. enable_wallet = config["components"].getboolean("ENABLE_WALLET")
  187. enable_utils = config["components"].getboolean("ENABLE_UTILS")
  188. enable_bitcoind = config["components"].getboolean("ENABLE_BITCOIND")
  189. if config["environment"]["EXEEXT"] == ".exe" and not args.force:
  190. # https://github.com/bitcoin/bitcoin/commit/d52802551752140cf41f0d9a225a43e84404d3e9
  191. # https://github.com/bitcoin/bitcoin/pull/5677#issuecomment-136646964
  192. print("Tests currently disabled on Windows by default. Use --force option to enable")
  193. sys.exit(0)
  194. if not (enable_wallet and enable_utils and enable_bitcoind):
  195. print("No functional tests to run. Wallet, utils, and bitcoind must all be enabled")
  196. print("Rerun `configure` with -enable-wallet, -with-utils and -with-daemon and rerun make")
  197. sys.exit(0)
  198. # Build list of tests
  199. if tests:
  200. # Individual tests have been specified. Run specified tests that exist
  201. # in the ALL_SCRIPTS list. Accept the name with or without .py extension.
  202. tests = [re.sub("\.py$", "", t) + ".py" for t in tests]
  203. test_list = []
  204. for t in tests:
  205. if t in ALL_SCRIPTS:
  206. test_list.append(t)
  207. else:
  208. print("{}WARNING!{} Test '{}' not found in full test list.".format(BOLD[1], BOLD[0], t))
  209. else:
  210. # No individual tests have been specified.
  211. # Run all base tests, and optionally run extended tests.
  212. test_list = BASE_SCRIPTS
  213. if args.extended:
  214. # place the EXTENDED_SCRIPTS first since the three longest ones
  215. # are there and the list is shorter
  216. test_list = EXTENDED_SCRIPTS + test_list
  217. # Remove the test cases that the user has explicitly asked to exclude.
  218. if args.exclude:
  219. tests_excl = [re.sub("\.py$", "", t) + ".py" for t in args.exclude.split(',')]
  220. for exclude_test in tests_excl:
  221. if exclude_test in test_list:
  222. test_list.remove(exclude_test)
  223. else:
  224. print("{}WARNING!{} Test '{}' not found in current test list.".format(BOLD[1], BOLD[0], exclude_test))
  225. if not test_list:
  226. print("No valid test scripts specified. Check that your test is in one "
  227. "of the test lists in test_runner.py, or run test_runner.py with no arguments to run all tests")
  228. sys.exit(0)
  229. if args.help:
  230. # Print help for test_runner.py, then print help of the first script (with args removed) and exit.
  231. parser.print_help()
  232. subprocess.check_call([(config["environment"]["SRCDIR"] + '/test/functional/' + test_list[0].split()[0])] + ['-h'])
  233. sys.exit(0)
  234. check_script_list(config["environment"]["SRCDIR"])
  235. if not args.keepcache:
  236. shutil.rmtree("%s/test/cache" % config["environment"]["BUILDDIR"], ignore_errors=True)
  237. run_tests(test_list, config["environment"]["SRCDIR"], config["environment"]["BUILDDIR"], config["environment"]["EXEEXT"], tmpdir, args.jobs, args.coverage, passon_args)
  238. def run_tests(test_list, src_dir, build_dir, exeext, tmpdir, jobs=1, enable_coverage=False, args=[]):
  239. # Warn if bitcoind is already running (unix only)
  240. try:
  241. if subprocess.check_output(["pidof", "bitcoind"]) is not None:
  242. print("%sWARNING!%s There is already a bitcoind process running on this system. Tests may fail unexpectedly due to resource contention!" % (BOLD[1], BOLD[0]))
  243. except (OSError, subprocess.SubprocessError):
  244. pass
  245. # Warn if there is a cache directory
  246. cache_dir = "%s/test/cache" % build_dir
  247. if os.path.isdir(cache_dir):
  248. print("%sWARNING!%s There is a cache directory here: %s. If tests fail unexpectedly, try deleting the cache directory." % (BOLD[1], BOLD[0], cache_dir))
  249. #Set env vars
  250. if "BITCOIND" not in os.environ:
  251. os.environ["BITCOIND"] = build_dir + '/src/bitcoind' + exeext
  252. os.environ["BITCOINCLI"] = build_dir + '/src/bitcoin-cli' + exeext
  253. tests_dir = src_dir + '/test/functional/'
  254. flags = ["--srcdir={}/src".format(build_dir)] + args
  255. flags.append("--cachedir=%s" % cache_dir)
  256. if enable_coverage:
  257. coverage = RPCCoverage()
  258. flags.append(coverage.flag)
  259. logging.debug("Initializing coverage directory at %s" % coverage.dir)
  260. else:
  261. coverage = None
  262. if len(test_list) > 1 and jobs > 1:
  263. # Populate cache
  264. subprocess.check_output([tests_dir + 'create_cache.py'] + flags + ["--tmpdir=%s/cache" % tmpdir])
  265. #Run Tests
  266. job_queue = TestHandler(jobs, tests_dir, tmpdir, test_list, flags)
  267. time0 = time.time()
  268. test_results = []
  269. max_len_name = len(max(test_list, key=len))
  270. for _ in range(len(test_list)):
  271. test_result, stdout, stderr = job_queue.get_next()
  272. test_results.append(test_result)
  273. if test_result.status == "Passed":
  274. logging.debug("\n%s%s%s passed, Duration: %s s" % (BOLD[1], test_result.name, BOLD[0], test_result.time))
  275. elif test_result.status == "Skipped":
  276. logging.debug("\n%s%s%s skipped" % (BOLD[1], test_result.name, BOLD[0]))
  277. else:
  278. print("\n%s%s%s failed, Duration: %s s\n" % (BOLD[1], test_result.name, BOLD[0], test_result.time))
  279. print(BOLD[1] + 'stdout:\n' + BOLD[0] + stdout + '\n')
  280. print(BOLD[1] + 'stderr:\n' + BOLD[0] + stderr + '\n')
  281. print_results(test_results, max_len_name, (int(time.time() - time0)))
  282. if coverage:
  283. coverage.report_rpc_coverage()
  284. logging.debug("Cleaning up coverage data")
  285. coverage.cleanup()
  286. # Clear up the temp directory if all subdirectories are gone
  287. if not os.listdir(tmpdir):
  288. os.rmdir(tmpdir)
  289. all_passed = all(map(lambda test_result: test_result.was_successful, test_results))
  290. sys.exit(not all_passed)
  291. def print_results(test_results, max_len_name, runtime):
  292. results = "\n" + BOLD[1] + "%s | %s | %s\n\n" % ("TEST".ljust(max_len_name), "STATUS ", "DURATION") + BOLD[0]
  293. test_results.sort(key=lambda result: result.name.lower())
  294. all_passed = True
  295. time_sum = 0
  296. for test_result in test_results:
  297. all_passed = all_passed and test_result.was_successful
  298. time_sum += test_result.time
  299. test_result.padding = max_len_name
  300. results += str(test_result)
  301. status = TICK + "Passed" if all_passed else CROSS + "Failed"
  302. results += BOLD[1] + "\n%s | %s | %s s (accumulated) \n" % ("ALL".ljust(max_len_name), status.ljust(9), time_sum) + BOLD[0]
  303. results += "Runtime: %s s\n" % (runtime)
  304. print(results)
  305. class TestHandler:
  306. """
  307. Trigger the testscrips passed in via the list.
  308. """
  309. def __init__(self, num_tests_parallel, tests_dir, tmpdir, test_list=None, flags=None):
  310. assert(num_tests_parallel >= 1)
  311. self.num_jobs = num_tests_parallel
  312. self.tests_dir = tests_dir
  313. self.tmpdir = tmpdir
  314. self.test_list = test_list
  315. self.flags = flags
  316. self.num_running = 0
  317. # In case there is a graveyard of zombie bitcoinds, we can apply a
  318. # pseudorandom offset to hopefully jump over them.
  319. # (625 is PORT_RANGE/MAX_NODES)
  320. self.portseed_offset = int(time.time() * 1000) % 625
  321. self.jobs = []
  322. def get_next(self):
  323. while self.num_running < self.num_jobs and self.test_list:
  324. # Add tests
  325. self.num_running += 1
  326. t = self.test_list.pop(0)
  327. portseed = len(self.test_list) + self.portseed_offset
  328. portseed_arg = ["--portseed={}".format(portseed)]
  329. log_stdout = tempfile.SpooledTemporaryFile(max_size=2**16)
  330. log_stderr = tempfile.SpooledTemporaryFile(max_size=2**16)
  331. test_argv = t.split()
  332. tmpdir = ["--tmpdir=%s/%s_%s" % (self.tmpdir, re.sub(".py$", "", test_argv[0]), portseed)]
  333. self.jobs.append((t,
  334. time.time(),
  335. subprocess.Popen([self.tests_dir + test_argv[0]] + test_argv[1:] + self.flags + portseed_arg + tmpdir,
  336. universal_newlines=True,
  337. stdout=log_stdout,
  338. stderr=log_stderr),
  339. log_stdout,
  340. log_stderr))
  341. if not self.jobs:
  342. raise IndexError('pop from empty list')
  343. while True:
  344. # Return first proc that finishes
  345. time.sleep(.5)
  346. for j in self.jobs:
  347. (name, time0, proc, log_out, log_err) = j
  348. if os.getenv('TRAVIS') == 'true' and int(time.time() - time0) > 20 * 60:
  349. # In travis, timeout individual tests after 20 minutes (to stop tests hanging and not
  350. # providing useful output.
  351. proc.send_signal(signal.SIGINT)
  352. if proc.poll() is not None:
  353. log_out.seek(0), log_err.seek(0)
  354. [stdout, stderr] = [l.read().decode('utf-8') for l in (log_out, log_err)]
  355. log_out.close(), log_err.close()
  356. if proc.returncode == TEST_EXIT_PASSED and stderr == "":
  357. status = "Passed"
  358. elif proc.returncode == TEST_EXIT_SKIPPED:
  359. status = "Skipped"
  360. else:
  361. status = "Failed"
  362. self.num_running -= 1
  363. self.jobs.remove(j)
  364. return TestResult(name, status, int(time.time() - time0)), stdout, stderr
  365. print('.', end='', flush=True)
  366. class TestResult():
  367. def __init__(self, name, status, time):
  368. self.name = name
  369. self.status = status
  370. self.time = time
  371. self.padding = 0
  372. def __repr__(self):
  373. if self.status == "Passed":
  374. color = BLUE
  375. glyph = TICK
  376. elif self.status == "Failed":
  377. color = RED
  378. glyph = CROSS
  379. elif self.status == "Skipped":
  380. color = GREY
  381. glyph = CIRCLE
  382. return color[1] + "%s | %s%s | %s s\n" % (self.name.ljust(self.padding), glyph, self.status.ljust(7), self.time) + color[0]
  383. @property
  384. def was_successful(self):
  385. return self.status != "Failed"
  386. def check_script_list(src_dir):
  387. """Check scripts directory.
  388. Check that there are no scripts in the functional tests directory which are
  389. not being run by pull-tester.py."""
  390. script_dir = src_dir + '/test/functional/'
  391. python_files = set([t for t in os.listdir(script_dir) if t[-3:] == ".py"])
  392. missed_tests = list(python_files - set(map(lambda x: x.split()[0], ALL_SCRIPTS + NON_SCRIPTS)))
  393. if len(missed_tests) != 0:
  394. print("%sWARNING!%s The following scripts are not being run: %s. Check the test lists in test_runner.py." % (BOLD[1], BOLD[0], str(missed_tests)))
  395. if os.getenv('TRAVIS') == 'true':
  396. # On travis this warning is an error to prevent merging incomplete commits into master
  397. sys.exit(1)
  398. class RPCCoverage(object):
  399. """
  400. Coverage reporting utilities for test_runner.
  401. Coverage calculation works by having each test script subprocess write
  402. coverage files into a particular directory. These files contain the RPC
  403. commands invoked during testing, as well as a complete listing of RPC
  404. commands per `bitcoin-cli help` (`rpc_interface.txt`).
  405. After all tests complete, the commands run are combined and diff'd against
  406. the complete list to calculate uncovered RPC commands.
  407. See also: test/functional/test_framework/coverage.py
  408. """
  409. def __init__(self):
  410. self.dir = tempfile.mkdtemp(prefix="coverage")
  411. self.flag = '--coveragedir=%s' % self.dir
  412. def report_rpc_coverage(self):
  413. """
  414. Print out RPC commands that were unexercised by tests.
  415. """
  416. uncovered = self._get_uncovered_rpc_commands()
  417. if uncovered:
  418. print("Uncovered RPC commands:")
  419. print("".join((" - %s\n" % i) for i in sorted(uncovered)))
  420. else:
  421. print("All RPC commands covered.")
  422. def cleanup(self):
  423. return shutil.rmtree(self.dir)
  424. def _get_uncovered_rpc_commands(self):
  425. """
  426. Return a set of currently untested RPC commands.
  427. """
  428. # This is shared from `test/functional/test-framework/coverage.py`
  429. reference_filename = 'rpc_interface.txt'
  430. coverage_file_prefix = 'coverage.'
  431. coverage_ref_filename = os.path.join(self.dir, reference_filename)
  432. coverage_filenames = set()
  433. all_cmds = set()
  434. covered_cmds = set()
  435. if not os.path.isfile(coverage_ref_filename):
  436. raise RuntimeError("No coverage reference found")
  437. with open(coverage_ref_filename, 'r') as f:
  438. all_cmds.update([i.strip() for i in f.readlines()])
  439. for root, dirs, files in os.walk(self.dir):
  440. for filename in files:
  441. if filename.startswith(coverage_file_prefix):
  442. coverage_filenames.add(os.path.join(root, filename))
  443. for filename in coverage_filenames:
  444. with open(filename, 'r') as f:
  445. covered_cmds.update([i.strip() for i in f.readlines()])
  446. return all_cmds - covered_cmds
  447. if __name__ == '__main__':
  448. main()