class DiskJobResultTests(TestCase):
def setUp(self):
self.scratch_dir = TemporaryDirectory()
def tearDown(self):
self.scratch_dir.cleanup()
def test_smoke(self):
result = DiskJobResult({})
self.assertEqual(str(result), "None")
self.assertEqual(repr(result), "<DiskJobResult outcome:None>")
self.assertIsNone(result.outcome)
self.assertIsNone(result.comments)
self.assertEqual(result.io_log, ())
self.assertIsNone(result.return_code)
def test_everything(self):
result = DiskJobResult({
'outcome': IJobResult.OUTCOME_PASS,
'comments': "it said blah",
'io_log_filename': make_io_log([
(0, 'stdout', b'blah\n')
], self.scratch_dir.name),
'return_code': 0
})
self.assertEqual(str(result), "pass")
self.assertEqual(repr(result), "<DiskJobResult outcome:'pass'>")
self.assertEqual(result.outcome, IJobResult.OUTCOME_PASS)
self.assertEqual(result.comments, "it said blah")
self.assertEqual(result.io_log, ((0, 'stdout', b'blah\n'),))
self.assertEqual(result.return_code, 0)
class TestStatelog(unittest.TestCase):
def setUp(self):
self.tmpdir = TemporaryDirectory()
self.logpath = os.path.join(self.tmpdir.name, 'statelog')
self.nonexist = os.path.join(self.tmpdir.name, 'nonexist')
with open(self.logpath, 'wb') as fw:
fw.write(b'001\n')
fw.write(b'002\n')
def tearDown(self):
self.tmpdir.cleanup()
def test_load(self):
state = pop3.statelog_load(self.logpath)
self.assertEqual(state, {b'001', b'002'})
def test_load_fallback(self):
state = pop3.statelog_load(self.nonexist)
self.assertEqual(state, set())
def test_create(self):
pop3.statelog_save(self.logpath, {b'001', b'002'})
with open(self.logpath, 'rb') as fp:
self.assertEqual(fp.readline(), b'001\n')
self.assertEqual(fp.readline(), b'002\n')
class NamedFileInTemporaryDirectory(object):
def __init__(self, filename, mode='w+b', bufsize=-1, **kwds):
"""
Open a file named `filename` in a temporary directory.
This context manager is preferred over `NamedTemporaryFile` in
stdlib `tempfile` when one needs to reopen the file.
Arguments `mode` and `bufsize` are passed to `open`.
Rest of the arguments are passed to `TemporaryDirectory`.
"""
self._tmpdir = TemporaryDirectory(**kwds)
path = _os.path.join(self._tmpdir.name, filename)
self.file = open(path, mode, bufsize)
def cleanup(self):
self.file.close()
self._tmpdir.cleanup()
__del__ = cleanup
def __enter__(self):
return self.file
def __exit__(self, type, value, traceback):
self.cleanup()
class NamedFileInTemporaryDirectory(object):
"""Open a file named `filename` in a temporary directory.
This context manager is preferred over :class:`tempfile.NamedTemporaryFile`
when one needs to reopen the file, because on Windows only one handle on a
file can be open at a time. You can close the returned handle explicitly
inside the context without deleting the file, and the context manager will
delete the whole directory when it exits.
Arguments `mode` and `bufsize` are passed to `open`.
Rest of the arguments are passed to `TemporaryDirectory`.
Usage example::
with NamedFileInTemporaryDirectory('myfile', 'wb') as f:
f.write('stuff')
f.close()
# You can now pass f.name to things that will re-open the file
"""
def __init__(self, filename, mode='w+b', bufsize=-1, **kwds):
self._tmpdir = TemporaryDirectory(**kwds)
path = _os.path.join(self._tmpdir.name, filename)
self.file = open(path, mode, bufsize)
def cleanup(self):
self.file.close()
self._tmpdir.cleanup()
__del__ = cleanup
def __enter__(self):
return self.file
def __exit__(self, type, value, traceback):
self.cleanup()
def test_dcm2niix_run():
dicomDir = os.path.join(TEST_DATA_DIR, "sourcedata", "sub-01")
tmpBase = os.path.join(TEST_DATA_DIR, "tmp")
#tmpDir = TemporaryDirectory(dir=tmpBase)
tmpDir = TemporaryDirectory()
app = Dcm2niix([dicomDir], tmpDir.name)
app.run()
helperDir = os.path.join(
tmpDir.name, DEFAULT.tmpDirName, DEFAULT.helperDir, "*")
ls = sorted(glob(helperDir))
firstMtime = [os.stat(_).st_mtime for _ in ls]
assert 'localizer_20100603125600' in ls[0]
#files should not be change after a rerun
app.run()
secondMtime = [os.stat(_).st_mtime for _ in ls]
assert firstMtime == secondMtime
#files should be change after a forced rerun
app.run(force=True)
thirdMtime = [os.stat(_).st_mtime for _ in ls]
assert firstMtime != thirdMtime
tmpDir.cleanup()
def test_load_from_file_with_relative_paths(self):
"""
When explicitly setting a config file, paths should be relative to the
config file, not the working directory.
"""
config_dir = TemporaryDirectory()
config_fname = os.path.join(config_dir.name, 'mkdocs.yml')
docs_dir = os.path.join(config_dir.name, 'src')
os.mkdir(docs_dir)
config_file = open(config_fname, 'w')
try:
config_file.write("docs_dir: src\nsite_name: MkDocs Test\n")
config_file.flush()
config_file.close()
cfg = base.load_config(config_file=config_file)
self.assertTrue(isinstance(cfg, base.Config))
self.assertEqual(cfg['site_name'], 'MkDocs Test')
self.assertEqual(cfg['docs_dir'], docs_dir)
self.assertEqual(cfg.config_file_path, config_fname)
self.assertIsInstance(cfg.config_file_path, utils.text_type)
finally:
config_dir.cleanup()
class GitRepositoryTest(TestCase):
def setUp(self):
self.tmpdir = TemporaryDirectory()
self.repo1 = GitRepository(
self.tmpdir.name,
url="https://github.com/st-tu-dresden/inloop.git",
branch="master"
)
self.repo2 = GitRepository(
self.tmpdir.name,
url="https://github.com/st-tu-dresden/inloop-java-repository-example.git",
branch="master"
)
def tearDown(self):
self.tmpdir.cleanup()
def test_git_operations(self):
self.repo1.synchronize()
self.assertTrue(self.get_path(".git").exists())
self.assertTrue(self.get_path("manage.py").exists())
self.assertEqual(b"", self.run_command("git status -s"))
self.repo2.synchronize()
self.assertFalse(self.get_path("manage.py").exists())
self.assertTrue(self.get_path("build.xml").exists())
self.assertEqual(b"", self.run_command("git status -s"))
def get_path(self, name):
return Path(self.tmpdir.name).joinpath(name)
def run_command(self, command):
return check_output(command.split(), cwd=self.tmpdir.name)
class TemporaryRepository(object):
"""A Git repository initialized in a temporary directory as a context manager.
usage:
with TemporaryRepository() as tempRepo:
print("workdir:", tempRepo.workdir)
print("path:", tempRepo.path)
index = repo.index
index.read()
index.add("...")
index.write()
tree = index.write_tree()
repo.create_commit('HEAD', author, comitter, message, tree, [])
"""
def __init__(self, is_bare=False, clone_from_repo=None):
self.temp_dir = TemporaryDirectory()
if clone_from_repo:
self.repo = clone_repository(clone_from_repo.path, self.temp_dir.name)
else:
self.repo = init_repository(self.temp_dir.name, is_bare)
def __enter__(self):
return self.repo
def __exit__(self, type, value, traceback):
self.temp_dir.cleanup()
def testInitNotExistingsRepo(self):
dir = TemporaryDirectory()
repo = quit.git.Repository(dir.name, create=True)
self.assertFalse(repo.is_bare)
self.assertEqual(len(repo.revisions()), 0)
dir.cleanup()
def save_document(self):
tempdirectory = TemporaryDirectory()
document = self.generate_document(tempdirectory.name)
if document:
with open(document, 'rb') as f:
self.data_file.save(path.basename(document), File(f))
self.last_update_of_data_file = datetime.datetime.now()
tempdirectory.cleanup()
def test_dcm2bids():
tmpBase = os.path.join(TEST_DATA_DIR, "tmp")
#bidsDir = TemporaryDirectory(dir=tmpBase)
bidsDir = TemporaryDirectory()
tmpSubDir = os.path.join(bidsDir.name, DEFAULT.tmpDirName, "sub-01")
shutil.copytree(
os.path.join(TEST_DATA_DIR, "sidecars"),
tmpSubDir)
app = Dcm2bids(
[TEST_DATA_DIR], "01",
os.path.join(TEST_DATA_DIR, "config_test.json"),
bidsDir.name
)
app.run()
layout = BIDSLayout(bidsDir.name, validate=False)
assert layout.get_subjects() == ["01"]
assert layout.get_sessions() == []
assert layout.get_tasks() == ["rest"]
assert layout.get_runs() == [1,2,3]
app = Dcm2bids(
[TEST_DATA_DIR], "01",
os.path.join(TEST_DATA_DIR, "config_test.json"),
bidsDir.name
)
app.run()
fmapFile = os.path.join(
bidsDir.name, "sub-01", "fmap", "sub-01_echo-492_fmap.json")
data = load_json(fmapFile)
fmapMtime = os.stat(fmapFile).st_mtime
assert data["IntendedFor"] == "dwi/sub-01_dwi.nii.gz"
data = load_json(os.path.join(
bidsDir.name, "sub-01", "localizer", "sub-01_run-01_localizer.json"))
assert data["ProcedureStepDescription"] == "Modify by dcm2bids"
#rerun
shutil.rmtree(tmpSubDir)
shutil.copytree(
os.path.join(TEST_DATA_DIR, "sidecars"),
tmpSubDir)
app = Dcm2bids(
[TEST_DATA_DIR], "01",
os.path.join(TEST_DATA_DIR, "config_test.json"),
bidsDir.name
)
app.run()
fmapMtimeRerun = os.stat(fmapFile).st_mtime
assert fmapMtime == fmapMtimeRerun
bidsDir.cleanup()
def testCloneNotExistingRepo(self):
environ["QUIT_SSH_KEY_HOME"] = "./tests/assets/sshkey/"
REMOTE_URL = '[email protected]:AKSW/ThereIsNoQuitStoreRepo.git'
dir = TemporaryDirectory()
with self.assertRaises(Exception) as context:
quit.git.Repository(dir.name, create=True, origin=REMOTE_URL)
dir.cleanup()
class MyTest(TestCase):
def setUp(self):
self.test_dir = TemporaryDirectory()
def tearDown(self):
self.test_dir.cleanup()
# Test methods follow
# 2016.07.08 add
def test_sample(self):
print(self.test_dir)
def testCloneRepo(self):
REMOTE_NAME = 'origin'
REMOTE_URL = 'git://github.com/AKSW/QuitStore.example.git'
dir = TemporaryDirectory()
repo = quit.git.Repository(dir.name, create=True, origin=REMOTE_URL)
self.assertTrue(path.exists(path.join(dir.name, 'example.nq')))
self.assertFalse(repo.is_bare)
dir.cleanup()
def testCloneRepoViaSSH(self):
environ["QUIT_SSH_KEY_HOME"] = "./tests/assets/sshkey/"
REMOTE_URL = '[email protected]:AKSW/QuitStore.example.git'
dir = TemporaryDirectory()
repo = quit.git.Repository(dir.name, create=True, origin=REMOTE_URL)
self.assertTrue(path.exists(path.join(dir.name, 'example.nt')))
self.assertFalse(repo.is_bare)
dir.cleanup()
class LoadScriptTest(unittest.TestCase):
def setUp(self):
self.tempdir = TemporaryDirectory()
for filename, contents in self.script_files:
with open(os.path.join(self.tempdir.name, filename), "xt") as f:
f.write(contents)
def tearDown(self):
self.tempdir.cleanup()
def testCloneRepoViaSSHNoKeyFiles(self):
environ["QUIT_SSH_KEY_HOME"] = "./tests/assets/nosshkey/"
if "SSH_AUTH_SOCK" in environ:
del environ["SSH_AUTH_SOCK"]
REMOTE_URL = '[email protected]:AKSW/QuitStore.example.git'
dir = TemporaryDirectory()
with self.assertRaises(Exception) as context:
quit.git.Repository(dir.name, create=True, origin=REMOTE_URL)
dir.cleanup()
class BaseRecorderTest(unittest.TestCase):
def setUp(self):
unittest.TestCase.setUp(self)
self.original_dir = os.getcwd()
self.temp_dir = TemporaryDirectory()
os.chdir(self.temp_dir.name)
def tearDown(self):
self.temp_dir.cleanup()
os.chdir(self.original_dir)
unittest.TestCase.tearDown(self)
class CLITestCase(unittest.TestCase):
def setUp(self):
self.temp_dir = TemporaryDirectory()
self.output_file_index = 0
def tearDown(self):
self.temp_dir.cleanup()
def exec(self, string):
args = ['-b', self.temp_dir.name]
args.extend(shlex.split(string))
process_args(args)
def list_output(self, string=''):
temp_text_file = os.path.join(self.temp_dir.name,
'temp{}.txt'.format(self.output_file_index))
self.output_file_index += 1
self.exec('list --pipe-to "cat > {}" {}'.format(temp_text_file, string))
with open(temp_text_file) as f:
text = f.read()
return text
def populate(self, num_entries=20):
for i in range(num_entries):
self.exec('new -m "{}"'.format(shlex.quote(random_text())))
def test_list(self):
self.populate(20)
text = self.list_output()
self.assertTrue(len(text.split('\n'))>20*3)
def test_edit(self):
self.exec('new --message "Hello world"')
original_text = self.list_output()
self.exec('edit -m "New text"')
modified_text = self.list_output()
self.assertTrue(re.search('Hello world', original_text))
self.assertTrue(re.search('New text', modified_text))
self.assertTrue(re.sub('Hello world', 'New text', original_text))
def test_search(self):
self.populate(2)
search_string = 'stringthatwontgetgeneratedbyaccident'
self.exec('new -m "hello world\n test text {}inthisentry"'.format(search_string))
self.populate(2)
all_entries = self.list_output()
search_matches = self.list_output(search_string)
self.assertNotEqual(all_entries, search_matches)
self.assertTrue(re.search(search_string, all_entries))
self.assertTrue(re.search(search_string, search_matches))
def setUp(self):
self.dir = TemporaryDirectory()
self.remotedir = TemporaryDirectory()
self.file = NamedTemporaryFile(dir=self.dir.name, delete=False)
self.filename = path.basename(self.file.name)
self.author = Signature('QuitStoreTest', '*****@*****.**')
self.comitter = Signature('QuitStoreTest', '*****@*****.**')
# Initialize repository
init_repository(self.dir.name, False)
class TestCaseSnapshotPath(TestCaseCfg):
def setUp(self):
super(TestCaseSnapshotPath, self).setUp()
#use a new TemporaryDirectory for snapshotPath to avoid
#side effects on leftovers
self.tmpDir = TemporaryDirectory()
self.cfg.dict['profile1.snapshots.path'] = self.tmpDir.name
self.snapshotPath = self.cfg.snapshotsFullPath()
def tearDown(self):
super(TestCaseSnapshotPath, self).tearDown()
self.tmpDir.cleanup()
class FileSystemBackendTest(unittest.TestCase):
def setUp(self) -> None:
self.tmpdir = TemporaryDirectory()
self.cwd = os.getcwd()
os.chdir(self.tmpdir.name)
dirname = 'fsbackendtest'
os.mkdir(dirname) # replace by temporary directory
self.backend = FilesystemBackend(dirname)
self.testdata = 'dshiuasduzchjbfdnbewhsdcuzd'
self.alternative_testdata = "8u993zhhbn\nb3tadgadg"
self.identifier = 'some name'
def tearDown(self) -> None:
os.chdir(self.cwd)
self.tmpdir.cleanup()
def test_put_and_get_normal(self) -> None:
# first put the data
self.backend.put(self.identifier, self.testdata)
# then retrieve it again
data = self.backend.get(self.identifier)
self.assertEqual(data, self.testdata)
def test_put_file_exists_no_overwrite(self) -> None:
name = 'test_put_file_exists_no_overwrite'
self.backend.put(name, self.testdata)
with self.assertRaises(FileExistsError):
self.backend.put(name, self.alternative_testdata)
self.assertEqual(self.testdata, self.backend.get(name))
def test_put_file_exists_overwrite(self) -> None:
name = 'test_put_file_exists_overwrite'
self.backend.put(name, self.testdata)
self.backend.put(name, self.alternative_testdata, overwrite=True)
self.assertEqual(self.alternative_testdata, self.backend.get(name))
def test_instantiation_fail(self) -> None:
with self.assertRaises(NotADirectoryError):
FilesystemBackend("C\\#~~")
def test_exists(self) -> None:
name = 'test_exists'
self.backend.put(name, self.testdata)
self.assertTrue(self.backend.exists(name))
self.assertFalse(self.backend.exists('exists_not'))
def test_get_not_existing(self) -> None:
name = 'test_get_not_existing'
with self.assertRaises(FileNotFoundError):
self.backend.get(name)
def create_archive(archive, excluded):
"""Creates a new zip archive file by excluding files at positions found in
excluded.
"""
new_archive_file = NamedTemporaryFile()
temporary_directory = TemporaryDirectory()
new_archive = ZipFile(new_archive_file, 'w')
for index, filename in enumerate(archive.namelist()):
if index not in excluded:
archive.extract(filename, path=temporary_directory.name)
new_archive.write(os.path.join(temporary_directory.name, filename))
temporary_directory.cleanup()
return new_archive_file
class TestCase(unittest.TestCase):
def __init__(self, methodName):
os.environ['LANGUAGE'] = 'en_US.UTF-8'
self.cfgFile = os.path.abspath(os.path.join(__file__, os.pardir, 'config'))
logger.APP_NAME = 'BIT_unittest'
logger.openlog()
super(TestCase, self).__init__(methodName)
def setUp(self):
logger.DEBUG = '-v' in sys.argv
self.run = False
self.sharePathObj = TemporaryDirectory()
self.sharePath = self.sharePathObj.name
def tearDown(self):
self.sharePathObj.cleanup()
def callback(self, func, *args):
func(*args)
self.run = True
def assertExists(self, *path):
full_path = os.path.join(*path)
if not os.path.exists(full_path):
self.fail('File does not exist: {}'.format(full_path))
def assertNotExists(self, *path):
full_path = os.path.join(*path)
if os.path.exists(full_path):
self.fail('File does unexpected exist: {}'.format(full_path))
def assertIsFile(self, *path):
full_path = os.path.join(*path)
if not os.path.isfile(full_path):
self.fail('Not a File: {}'.format(full_path))
def assertIsNoFile(self, *path):
full_path = os.path.join(*path)
if os.path.isfile(full_path):
self.fail('Unexpected File: {}'.format(full_path))
def assertIsDir(self, *path):
full_path = os.path.join(*path)
if not os.path.isdir(full_path):
self.fail('Not a directory: {}'.format(full_path))
def assertIsLink(self, *path):
full_path = os.path.join(*path)
if not os.path.islink(full_path):
self.fail('Not a symlink: {}'.format(full_path))
def _run_interface(self, runtime):
# Get all inputs from the ApplyTransforms object
ifargs = self.inputs.get()
# Extract number of input images and transforms
in_files = ifargs.pop('input_image')
num_files = len(in_files)
transforms = ifargs.pop('transforms')
# Get number of parallel jobs
num_threads = ifargs.pop('num_threads')
save_cmd = ifargs.pop('save_cmd')
# Remove certain keys
for key in ['environ', 'ignore_exception',
'terminal_output', 'output_image']:
ifargs.pop(key, None)
# Get a temp folder ready
tmp_folder = TemporaryDirectory(prefix='tmp-', dir=runtime.cwd)
xfms_list = _arrange_xfms(transforms, num_files, tmp_folder)
assert len(xfms_list) == num_files
# Inputs are ready to run in parallel
if num_threads < 1:
num_threads = None
if num_threads == 1:
out_files = [_applytfms((
in_file, in_xfm, ifargs, i, runtime.cwd))
for i, (in_file, in_xfm) in enumerate(zip(in_files, xfms_list))
]
else:
from concurrent.futures import ThreadPoolExecutor
with ThreadPoolExecutor(max_workers=num_threads) as pool:
out_files = list(pool.map(_applytfms, [
(in_file, in_xfm, ifargs, i, runtime.cwd)
for i, (in_file, in_xfm) in enumerate(zip(in_files, xfms_list))]
))
tmp_folder.cleanup()
# Collect output file names, after sorting by index
self._results['out_files'] = [el[0] for el in out_files]
if save_cmd:
self._results['log_cmdline'] = os.path.join(runtime.cwd, 'command.txt')
with open(self._results['log_cmdline'], 'w') as cmdfile:
print('\n-------\n'.join([el[1] for el in out_files]),
file=cmdfile)
return runtime
class TestDataSet(unittest.TestCase):
def setUp(self):
# Creating a temporary directory
self.output_dir = TemporaryDirectory(prefix="post_hit_dataset_test_")
def test_dataset(self):
self.assertTrue(isinstance(dataset.DataSet(dataset_path = "roadmap_epigenomic.json"), dataset.DataSet))
self.assertTrue(isinstance(dataset.DataSet(dataset_path = "".join([POST_HIT_PATH,"data/datasets/phyloP100way.json"])), dataset.DataSet))
self.assertTrue(isinstance(dataset.DataSet(dataset_path ="ensembl"), dataset.DataSet))
self.assertTrue(isinstance(dataset.DataSet(dataset_path = "gtex.json"), dataset.DataSet))
self.assertRaises(ValueError, dataset.DataSet,"gtex.txt")
self.assertRaises(ValueError, dataset.DataSet,"dummy")
self.assertRaises(ValueError, dataset.DataSet,"dummy.json")
self.dummy_dataset = dataset.DataSet(dataset_path = POST_HIT_PATH + "tmp/dummy_test.json",
project= "TEST",
description= "Dummy Test JSON FILE",
project_link= "dummy.test",
version= 0,
data_path= "tmp/",
protocole= "ftp",
file_type= "gtf",
ids=["DummyTest"],
download_links=["ftp://ftp.ensembl.org/pub/release-75//gtf/homo_sapiens"],
filenames=["DummyTest.gtf.gz"],
metadata= [],
data_representation={})
self.assertTrue(isinstance(self.dummy_dataset, dataset.DataSet))
self.assertTrue(os.path.exists(POST_HIT_PATH + "tmp/DummyTest.gtf.gz"), dataset.DataSet)
folder = POST_HIT_PATH + 'tmp/'
for the_file in os.listdir(folder):
file_path = os.path.join(folder, the_file)
try:
if os.path.isfile(file_path):
os.unlink(file_path)
elif os.path.isdir(file_path):
shutil.rmtree(file_path)
except Exception as e:
print(e)
def tearDown(self):
"""Finishes the test."""
# Deleting the output directory
self.output_dir.cleanup()
class TestRestoreSSH(generic.SSHSnapshotsWithSidTestCase, TestRestoreLocal):
def setUp(self):
super(TestRestoreSSH, self).setUp()
self.include = TemporaryDirectory()
generic.create_test_files(os.path.join(self.remoteSIDBackupPath, self.include.name[1:]))
#mount
self.cfg.setCurrentHashId(mount.Mount(cfg = self.cfg).mount())
def tearDown(self):
#unmount
mount.Mount(cfg = self.cfg).umount(self.cfg.current_hash_id)
super(TestRestoreSSH, self).tearDown()
self.include.cleanup()
class ArchiveCopy(ar.Archive):
def __init__(self, path):
self.__dir = TemporaryDirectory()
dest = '%s/%s' % (self.__dir.name, os.path.basename(path))
shutil.copyfile(path, dest)
super().__init__(dest)
def __enter__(self):
self.__dir.__enter__()
return super().__enter__()
def __exit__(self, type, value, traceback):
super().__exit__(type, value, traceback)
self.__dir.__exit__(type, value, traceback)
def setUp(self):
super(TestTakeSnapshotSSH, self).setUp()
self.include = TemporaryDirectory()
generic.create_test_files(self.include.name)
#mount
self.cfg.setCurrentHashId(mount.Mount(cfg = self.cfg).mount())
def setUp(self):
super(TestRestoreSSH, self).setUp()
self.include = TemporaryDirectory()
generic.create_test_files(os.path.join(self.remoteSIDBackupPath, self.include.name[1:]))
#mount
self.cfg.setCurrentHashId(mount.Mount(cfg = self.cfg).mount())
def setUp(self):
self.test_dir = TemporaryDirectory()
self.test_path = Path(self.test_dir.name)
def run_cli(
self,
hql: Union[str, str],
schema: Optional[str] = None,
verbose: bool = True,
hive_conf: Optional[Dict[Any, Any]] = None,
) -> Any:
"""
Run an hql statement using the hive cli. If hive_conf is specified
it should be a dict and the entries will be set as key/value pairs
in HiveConf
:param hive_conf: if specified these key value pairs will be passed
to hive as ``-hiveconf "key"="value"``. Note that they will be
passed after the ``hive_cli_params`` and thus will override
whatever values are specified in the database.
:type hive_conf: dict
>>> hh = HiveCliHook()
>>> result = hh.run_cli("USE airflow;")
>>> ("OK" in result)
True
"""
conn = self.conn
schema = schema or conn.schema
if schema:
hql = f"USE {schema};\n{hql}"
with TemporaryDirectory(prefix='airflow_hiveop_') as tmp_dir:
with NamedTemporaryFile(dir=tmp_dir) as f:
hql += '\n'
f.write(hql.encode('UTF-8'))
f.flush()
hive_cmd = self._prepare_cli_cmd()
env_context = get_context_from_env_var()
# Only extend the hive_conf if it is defined.
if hive_conf:
env_context.update(hive_conf)
hive_conf_params = self._prepare_hiveconf(env_context)
if self.mapred_queue:
hive_conf_params.extend([
'-hiveconf',
f'mapreduce.job.queuename={self.mapred_queue}',
'-hiveconf',
f'mapred.job.queue.name={self.mapred_queue}',
'-hiveconf',
f'tez.queue.name={self.mapred_queue}',
])
if self.mapred_queue_priority:
hive_conf_params.extend([
'-hiveconf',
f'mapreduce.job.priority={self.mapred_queue_priority}'
])
if self.mapred_job_name:
hive_conf_params.extend([
'-hiveconf', f'mapred.job.name={self.mapred_job_name}'
])
hive_cmd.extend(hive_conf_params)
hive_cmd.extend(['-f', f.name])
if verbose:
self.log.info("%s", " ".join(hive_cmd))
sub_process: Any = subprocess.Popen(hive_cmd,
stdout=subprocess.PIPE,
stderr=subprocess.STDOUT,
cwd=tmp_dir,
close_fds=True)
self.sub_process = sub_process
stdout = ''
while True:
line = sub_process.stdout.readline()
if not line:
break
stdout += line.decode('UTF-8')
if verbose:
self.log.info(line.decode('UTF-8').strip())
sub_process.wait()
if sub_process.returncode:
raise AirflowException(stdout)
return stdout
def hasherAndTree(hasher):
tdir = TemporaryDirectory().name
store = FileHashStore(tdir)
m = CompactMerkleTree(hasher=hasher, hashStore=store)
return hasher, m
def update(packages, no_detect, env_name, eager, all_packages, infra, global_install,
admin, force, dev, as_module, self, quiet):
"""If the option --env is supplied, the update will be applied using
that named virtual env. Unless the option --global is selected, the
update will only affect the current user. Of course, this will have
no effect if a virtual env is in use. The desired name of the admin
user can be set with the `_DEFAULT_ADMIN_` environment variable.
When performing a global update, your system may use an older version
of pip that is incompatible with some features such as --eager. To
force the use of these features, use --force.
With no packages nor options selected, this will update packages by
looking for a `requirements.txt` or a dev version of that in the current
directory.
If no --env is chosen, this will attempt to detect a project and use its
virtual env before resorting to the default pip. No project detection
will occur if a virtual env is active.
To update this tool, use the --self flag. All other methods of updating will
ignore `hatch`. See: https://github.com/pypa/pip/issues/1299
"""
command = ['install', '--upgrade'] + (['-q'] if quiet else [])
if not global_install or force: # no cov
command.extend(['--upgrade-strategy', 'eager' if eager else 'only-if-needed'])
infra_packages = ['pip', 'setuptools', 'wheel']
temp_dir = None
# Windows' `runas` allows only a single argument for the
# command so we catch this case and turn our command into
# a string later.
windows_admin_command = None
if self: # no cov
as_module = True
if not self and env_name:
venv_dir = os.path.join(get_venv_dir(), env_name)
if not os.path.exists(venv_dir):
echo_failure('Virtual env named `{}` does not exist.'.format(env_name))
sys.exit(1)
with venv(venv_dir):
executable = (
[get_proper_python(), '-m', 'pip']
if as_module or (infra and ON_WINDOWS)
else [get_proper_pip()]
)
command = executable + command
if all_packages:
installed_packages = infra_packages if infra else get_installed_packages()
else:
installed_packages = None
elif not self and not venv_active() and not no_detect and is_project():
venv_dir = os.path.join(os.getcwd(), get_venv_folder())
if not is_venv(venv_dir):
echo_info('A project has been detected!')
echo_waiting('Creating a dedicated virtual env... ', nl=False)
create_venv(venv_dir)
echo_success('complete!')
with venv(venv_dir):
echo_waiting('Installing this project in the virtual env... ', nl=False)
install_packages(['-q', '-e', '.'])
echo_success('complete!')
with venv(venv_dir):
executable = (
[get_proper_python(), '-m', 'pip']
if as_module or (infra and ON_WINDOWS)
else [get_proper_pip()]
)
command = executable + command
if all_packages:
installed_packages = infra_packages if infra else get_installed_packages()
else:
installed_packages = None
else:
venv_dir = None
executable = (
[sys.executable if self else get_proper_python(), '-m', 'pip']
if as_module or (infra and ON_WINDOWS)
else [get_proper_pip()]
)
command = executable + command
if all_packages:
installed_packages = infra_packages if infra else get_installed_packages()
else:
installed_packages = None
if not venv_active(): # no cov
if global_install:
if not admin:
if ON_WINDOWS:
windows_admin_command = get_admin_command()
else:
command = get_admin_command() + command
else:
command.append('--user')
if self: # no cov
command.append('hatch')
if ON_WINDOWS:
echo_warning('After the update you may want to press Enter to flush stdout.')
subprocess.Popen(command, shell=NEED_SUBPROCESS_SHELL)
sys.exit()
else:
result = subprocess.run(command, shell=NEED_SUBPROCESS_SHELL)
sys.exit(result.returncode)
elif infra:
command.extend(infra_packages)
elif all_packages:
installed_packages = [
package for package in installed_packages
if package not in infra_packages and package != 'hatch'
]
if not installed_packages:
echo_failure('No packages installed.')
sys.exit(1)
command.extend(installed_packages)
elif packages:
packages = [package for package in packages if package != 'hatch']
if not packages:
echo_failure('No packages to install.')
sys.exit(1)
command.extend(packages)
# When https://github.com/pypa/pipfile is finalized, we'll use it.
else:
reqs = get_requirements_file(os.getcwd(), dev=dev)
if not reqs:
echo_failure('Unable to locate a requirements file.')
sys.exit(1)
with open(reqs, 'r') as f:
lines = f.readlines()
matches = []
for line in lines:
match = re.match(r'^[^=<>]+', line.lstrip())
if match and match.group(0) == 'hatch':
matches.append(line)
if matches:
for line in matches:
lines.remove(line)
temp_dir = TemporaryDirectory()
reqs = os.path.join(temp_dir.name, basepath(reqs))
with open(reqs, 'w') as f:
f.writelines(lines)
command.extend(['-r', reqs])
if windows_admin_command: # no cov
command = windows_admin_command + [' '.join(command)]
if venv_dir:
with venv(venv_dir):
if env_name:
echo_waiting('Updating virtual env `{}`...'.format(env_name))
else:
echo_waiting('Updating for this project...')
result = subprocess.run(command, shell=NEED_SUBPROCESS_SHELL)
else:
echo_waiting('Updating...')
result = subprocess.run(command, shell=NEED_SUBPROCESS_SHELL)
if temp_dir is not None:
temp_dir.cleanup()
sys.exit(result.returncode)
def report_main(args_parser):
"""
report mainline
The mainline for the 'report' action.
Args:
args_parser: the argument parser to use for argument processing
Returns:
the exit code
"""
args_parser.add_argument('--full-config', '-C', action='store_true')
args_parser.add_argument('--no-sanitize', action='store_true')
args_parser.add_argument('--offline', action='store_true')
known_args = sys.argv[1:]
args, unknown_args = args_parser.parse_known_args(known_args)
if unknown_args:
logger.warn('unknown arguments: {}'.format(' '.join(unknown_args)))
rv = 0
work_dir = args.work_dir if args.work_dir else os.getcwd()
# setup sphinx engine to extract configuration
config = {}
configuration_load_issue = None
confluence_instance_info = None
publisher = ConfluencePublisher()
try:
with TemporaryDirectory() as tmp_dir:
with docutils_namespace():
print('fetching configuration information...')
app = Sphinx(
work_dir, # document sources
work_dir, # directory with configuration
tmp_dir, # output for built documents
tmp_dir, # output for doctree files
ConfluenceReportBuilder.name) # builder to execute
if app.config.confluence_publish:
process_ask_configs(app.config)
# extract configuration information
for k, v in app.config.values.items():
raw = getattr(app.config, k)
if raw is None:
continue
if callable(raw):
value = '(callable)'
else:
value = raw
if not args.full_config and not k.startswith('confluence_'):
continue
# always extract some known builder configurations
if args.full_config and k.startswith(IGNORE_BUILDER_CONFS):
continue
config[k] = value
# initialize the publisher (if needed later)
publisher.init(app.config)
except Exception as ex:
logger.error(ex)
if os.path.isfile(os.path.join(work_dir, 'conf.py')):
configuration_load_issue = 'unable to load configuration'
else:
configuration_load_issue = 'no documentation/missing configuration'
rv = 1
# attempt to fetch confluence instance version
confluence_publish = config.get('confluence_publish')
confluence_server_url = config.get('confluence_server_url')
if not args.offline and confluence_publish and confluence_server_url:
base_url = ConfluenceUtil.normalizeBaseUrl(confluence_server_url)
info = ''
session = None
try:
print('connecting to confluence instance...')
publisher.connect()
info += ' connected: yes\n'
session = publisher.rest_client.session
except Exception as ex:
logger.error(ex)
info += ' connected: no\n'
rv = 1
if session:
try:
# fetch
print('fetching confluence instance information...')
manifest_url = base_url + MANIFEST_PATH
rsp = session.get(manifest_url)
if rsp.status_code == 200:
info += ' fetched: yes\n'
# extract
print('decoding information...')
rsp.encoding = 'utf-8'
raw_data = rsp.text
info += ' decoded: yes\n'
# parse
print('parsing information...')
xml_data = ElementTree.fromstring(raw_data)
info += ' parsed: yes\n'
root = ElementTree.ElementTree(xml_data)
for o in root.findall('typeId'):
info += ' type: ' + o.text + '\n'
for o in root.findall('version'):
info += ' version: ' + o.text + '\n'
for o in root.findall('buildNumber'):
info += ' build: ' + o.text + '\n'
else:
logger.error('bad response from server ({})'.format(
rsp.status_code))
info += ' fetched: error ({})\n'.format(rsp.status_code)
rv = 1
except Exception as ex:
logger.error(ex)
info += 'failure to determine confluence data\n'
rv = 1
confluence_instance_info = info
def sensitive_config(key):
if key in config:
if config[key]:
config[key] = '(set)'
else:
config[key] = '(set; empty)'
# always sanitize out sensitive information
sensitive_config('confluence_client_cert_pass')
sensitive_config('confluence_server_pass')
# optional sanitization
if not args.no_sanitize:
sensitive_config('author')
sensitive_config('confluence_client_cert')
sensitive_config('confluence_global_labels')
sensitive_config('confluence_jira_servers')
sensitive_config('confluence_parent_page')
sensitive_config('confluence_parent_page_id_check')
sensitive_config('confluence_proxy')
sensitive_config('confluence_server_auth')
sensitive_config('confluence_server_cookies')
sensitive_config('confluence_server_user')
sensitive_config('project')
# remove confluence instance (attempt to keep scheme)
if 'confluence_server_url' in config:
value = config['confluence_server_url']
parsed = urlparse(value)
if parsed.scheme:
value = parsed.scheme + '://<removed>'
else:
value = '(set; no scheme)'
if parsed.netloc and parsed.netloc.endswith('atlassian.net'):
value += ' (cloud)'
config['confluence_server_url'] = value
# remove space name, but track casing
if 'confluence_space_name' in config:
value = config['confluence_space_name']
if value.isupper():
value = '(set; upper)'
elif value.islower():
value = '(set; upper)'
else:
value = '(set; mixed)'
config['confluence_space_name'] = value
print('')
print('Confluence builder report has been generated.')
print('Please copy the following text for the GitHub issue:')
print('')
logger.note('------------[ cut here ]------------')
print('(system)')
print(' platform:', single_line_version(platform.platform()))
print(' python:', single_line_version(sys.version))
print(' sphinx:', single_line_version(sphinx_version))
print(' requests:', single_line_version(requests_version))
print(' builder:', single_line_version(scb_version))
print('')
print('(configuration)')
if config:
for k, v in OrderedDict(sorted(config.items())).items():
print('{}: {}'.format(k, v))
else:
print('~default configuration~')
if configuration_load_issue:
print('')
print('(error loading configuration)')
print(configuration_load_issue)
if confluence_instance_info:
print('')
print('(confluence instance)')
print(confluence_instance_info.rstrip())
logger.note('------------[ cut here ]------------')
return rv
def correction(activity: dict, collection_id=None, **kwargs):
"""Celery task to deal with Surface Reflectance processors."""
execution = execution_from_collection(activity,
collection_id=collection_id,
activity_type=correction.__name__)
collection: Collection = execution.activity.collection
scene_id = activity['sceneid']
logging.info(
f'Starting Correction Task for {collection.name}(id={collection.id}, scene_id={scene_id})'
)
data_collection = get_provider_collection_from_activity(activity)
try:
output_path = data_collection.path(collection)
if collection._metadata and collection._metadata.get('processors'):
processor_name = collection._metadata['processors'][0]['name']
with TemporaryDirectory(prefix='correction_',
suffix=f'_{scene_id}') as tmp:
shutil.unpack_archive(activity['args']['compressed_file'], tmp)
# Process environment
env = dict(**os.environ,
INDIR=str(tmp),
OUTDIR=str(output_path))
entry = scene_id
entries = list(Path(tmp).iterdir())
if len(entries) == 1 and entries[0].suffix == '.SAFE':
entry = entries[0].name
if processor_name.lower() == 'sen2cor':
output_path.parent.mkdir(exist_ok=True, parents=True)
sen2cor_conf = Config.SEN2COR_CONFIG
logging.info(f'Using {entry} of sceneid {scene_id}')
# TODO: Use custom sen2cor version (2.5 or 2.8)
cmd = f'''docker run --rm -i \
-v $INDIR:/mnt/input-dir \
-v $OUTDIR:/mnt/output-dir \
-v {sen2cor_conf["SEN2COR_AUX_DIR"]}:/home/lib/python2.7/site-packages/sen2cor/aux_data \
-v {sen2cor_conf["SEN2COR_CONFIG_DIR"]}:/root/sen2cor/2.8 \
{sen2cor_conf["SEN2COR_DOCKER_IMAGE"]} {entry}'''
env['OUTDIR'] = str(Path(tmp) / 'output')
else:
output_path.mkdir(exist_ok=True, parents=True)
lasrc_conf = Config.LASRC_CONFIG
cmd = f'''docker run --rm -i \
-v $INDIR:/mnt/input-dir \
-v $OUTDIR:/mnt/output-dir \
--env INDIR=/mnt/input-dir \
--env OUTDIR=/mnt/output-dir \
-v {lasrc_conf["LASRC_AUX_DIR"]}:/mnt/lasrc-aux:ro \
-v {lasrc_conf["LEDAPS_AUX_DIR"]}:/mnt/ledaps-aux:ro \
{lasrc_conf["LASRC_DOCKER_IMAGE"]} {entry}'''
logging.debug(cmd)
# subprocess
process = subprocess.Popen(cmd,
shell=True,
env=env,
stdin=subprocess.PIPE)
process.wait()
assert process.returncode == 0
# TODO: We should be able to get output name from execution
if processor_name.lower() == 'sen2cor':
# Since sen2cor creates an custom directory name (based in scene_id) and changing processing date
# we create it inside "output" folder. After that, get first entry of that directory
output_tmp = list(Path(env['OUTDIR']).iterdir())[0]
output_path = output_path.parent / output_tmp.name
if execution.activity.args.get('file'):
last_processed_file = execution.activity.args['file']
if last_processed_file and os.path.exists(last_processed_file) and \
last_processed_file.endswith('.SAFE'):
# TODO: validate scene id (without processing_date)
if len(os.listdir(last_processed_file)) < 9:
shutil.rmtree(last_processed_file,
ignore_errors=True)
shutil.move(output_tmp, output_path)
refresh_execution_args(execution,
activity,
file=str(output_path))
else:
raise RuntimeError(
f'Processor not supported. Check collection {collection.name} metadata processors'
)
except Exception as e:
logging.error(f'Error in correction {scene_id} - {str(e)}',
exc_info=True)
raise e
return activity
def test_start_and_stop_schedule_cron_tab(restore_cron_tab, ): # pylint:disable=unused-argument,redefined-outer-name
with TemporaryDirectory() as tempdir:
instance = define_scheduler_instance(tempdir)
with get_test_external_repo() as external_repo:
# Start schedule
instance.start_schedule_and_update_storage_state(
external_repo.get_external_schedule(
"no_config_pipeline_every_min_schedule"))
cron_jobs = get_cron_jobs()
assert len(cron_jobs) == 1
# Try starting it again
with pytest.raises(DagsterSchedulerError):
instance.start_schedule_and_update_storage_state(
external_repo.get_external_schedule(
"no_config_pipeline_every_min_schedule"))
cron_jobs = get_cron_jobs()
assert len(cron_jobs) == 1
# Start another schedule
instance.start_schedule_and_update_storage_state(
external_repo.get_external_schedule(
"no_config_pipeline_daily_schedule"))
cron_jobs = get_cron_jobs()
assert len(cron_jobs) == 2
# Stop second schedule
instance.stop_schedule_and_update_storage_state(
external_repo.get_external_schedule(
"no_config_pipeline_daily_schedule").
get_external_origin_id())
cron_jobs = get_cron_jobs()
assert len(cron_jobs) == 1
# Try stopping second schedule again
instance.stop_schedule_and_update_storage_state(
external_repo.get_external_schedule(
"no_config_pipeline_daily_schedule").
get_external_origin_id())
cron_jobs = get_cron_jobs()
assert len(cron_jobs) == 1
# Start second schedule
instance.start_schedule_and_update_storage_state(
external_repo.get_external_schedule(
"no_config_pipeline_daily_schedule"))
cron_jobs = get_cron_jobs()
assert len(cron_jobs) == 2
# Reconcile schedule state, should be in the same state
instance.reconcile_scheduler_state(external_repo)
cron_jobs = get_cron_jobs()
assert len(cron_jobs) == 2
instance.start_schedule_and_update_storage_state(
external_repo.get_external_schedule(
"default_config_pipeline_every_min_schedule"))
cron_jobs = get_cron_jobs()
assert len(cron_jobs) == 3
# Reconcile schedule state, should be in the same state
instance.reconcile_scheduler_state(external_repo)
cron_jobs = get_cron_jobs()
assert len(cron_jobs) == 3
# Stop all schedules
instance.stop_schedule_and_update_storage_state(
external_repo.get_external_schedule(
"no_config_pipeline_every_min_schedule").
get_external_origin_id())
instance.stop_schedule_and_update_storage_state(
external_repo.get_external_schedule(
"no_config_pipeline_daily_schedule").
get_external_origin_id())
instance.stop_schedule_and_update_storage_state(
external_repo.get_external_schedule(
"default_config_pipeline_every_min_schedule").
get_external_origin_id())
cron_jobs = get_cron_jobs()
assert len(cron_jobs) == 0
# Reconcile schedule state, should be in the same state
instance.reconcile_scheduler_state(external_repo)
cron_jobs = get_cron_jobs()
assert len(cron_jobs) == 0
def test_start_non_existent_schedule(restore_cron_tab, ): # pylint:disable=unused-argument,redefined-outer-name
with TemporaryDirectory() as tempdir:
instance = define_scheduler_instance(tempdir)
with pytest.raises(DagsterScheduleDoesNotExist):
instance.stop_schedule_and_update_storage_state("asdf")
class GPGMailTests(unittest.TestCase):
def setUp(self):
self.temp_gpg_homedir = TemporaryDirectory()
gpg = gnupg.GPG(gnupghome=self.temp_gpg_homedir.name)
alice_input = gpg.gen_key_input(
name_real="Alice",
name_email="*****@*****.**",
key_type="RSA",
key_length=4096,
key_usage="",
subkey_type="RSA",
subkey_length=4096,
passphrase="test",
subkey_usage="encrypt,sign,auth",
)
self.alice_key = gpg.gen_key(alice_input)
self.assertIsNotNone(self.alice_key)
self.assertIsNotNone(self.alice_key.fingerprint)
def tearDown(self):
self.temp_gpg_homedir.cleanup()
def test_encrypt_decrypt(self):
mail = (
"Return-Path: <*****@*****.**>\nReceived: from example.com (example.com "
+
"[127.0.0.1])\n by example.com (Postfix) with ESMTPSA id E8DB612009F\n"
+
" for <*****@*****.**>; Tue, 7 Jan 2020 19:30:03 +0200 (CEST)\n"
+
'Content-Type: text/plain; charset="utf-8"\nMIME-Version: 1.0\n' +
"Content-Transfer-Encoding: 7bit\nSubject: Test\nFrom: " +
"[email protected]\nTo: [email protected]\nDate: Tue, 07 Jan 2020 "
+
"19:30:03 -0000\nMessage-ID:\n <*****@*****.**>"
+ "\n\nThis is a test message.")
msg = "This is a test message."
p = Popen(
[
"./gpgmail",
"-e",
"*****@*****.**",
"--gnupghome",
self.temp_gpg_homedir.name,
],
stdout=PIPE,
stdin=PIPE,
stderr=PIPE,
encoding="utf8",
)
encrypted = p.communicate(input=mail)[0]
self.assertTrue(msg not in encrypted)
p = Popen(
[
"./gpgmail",
"-p",
"test",
"-d",
"--gnupghome",
self.temp_gpg_homedir.name,
],
stdout=PIPE,
stdin=PIPE,
stderr=PIPE,
encoding="utf8",
)
decrypted = p.communicate(input=encrypted)[0]
self.assertTrue(msg in decrypted)
regex = (
r'Content-Type: multipart/mixed; protected-headers="v1";? boundary="'
+
r'===============\d+=="\nMIME-Version: 1\.0\nReturn-Path: <alice@example'
+
r"\.com>\nReceived: from example\.com \(example.com \[127\.0\.0\.1\]\)\n"
+
r" by example\.com \(Postfix\) with ESMTPSA id E8DB612009F\n for "
+
r"<alice@example\.com>; Tue, 7 Jan 2020 19:30:03 \+0200 \(CEST\)\n"
+
r"Subject: Test\nFrom: alice@example\.com\nTo: alice@example\.com\nDate: "
+
r"Tue, 07 Jan 2020 19:30:03 -0000\nMessage-ID: \n <123456789\.123456\."
+
r"123456789@example\.com>\n\n--===============\d+==\nContent-Type: "
+
r'text/rfc822-headers; protected-headers="v1"\nContent-Disposition: '
+
r"inline\n(Date: Tue, 07 Jan 2020 19:30:03 -0000\n|Subject: Test\n|From: "
+
r"alice@example\.com\n|To: alice@example\.com\n|Message-ID: \n <123456789"
+
r"\.123456\.123456789@example\.com>\n)+\n\n--===============\d+==\n"
+
r'Content-Type: text/plain; charset="utf-8"\nContent-Transfer-Encoding: '
+ r"7bit\n\nThis is a test message\.\n--===============\d+==--\n")
self.assertIsNotNone(re.fullmatch(regex, decrypted))
def test_sign(self):
mail = (
"Return-Path: <*****@*****.**>\nReceived: from example.com (example.com "
+
"[127.0.0.1])\n by example.com (Postfix) with ESMTPSA id E8DB612009F\n"
+
" for <*****@*****.**>; Tue, 7 Jan 2020 19:30:03 +0200 (CEST)\n"
+
'Content-Type: text/plain; charset="utf-8"\nMIME-Version: 1.0\n' +
"Content-Transfer-Encoding: 7bit\nSubject: Test\nFrom: " +
"[email protected]\nTo: [email protected]\nDate: Tue, 07 Jan 2020 "
+
"19:30:03 -0000\nMessage-ID:\n <*****@*****.**>"
+ "\n\nThis is a test message.")
msg = "This is a test message."
p = Popen(
[
"./gpgmail",
"-s",
"*****@*****.**",
"--gnupghome",
self.temp_gpg_homedir.name,
"-p",
"test",
],
stdout=PIPE,
stdin=PIPE,
stderr=PIPE,
encoding="utf8",
)
signed = p.communicate(input=mail)[0]
self.assertTrue(msg in signed)
regex = (
r"Content-Type: multipart/signed; micalg=\"pgp-sha512\"; " +
r"protocol=\"application/pgp-signature\"; " +
r"boundary=\"===============\d+==\"\nMIME-Version: 1\.0\n" +
r"Return-Path: <alice@example\.com>\nReceived: from " +
r"example\.com \(example\.com \[127\.0\.0\.1\]\)\n by " +
r"example\.com \(Postfix\) with ESMTPSA id E8DB612009F\n " +
r"for <alice@example\.com>; Tue, 7 Jan 2020 19:30:03 \+0200 " +
r"\(CEST\)\nSubject: Test\nFrom: alice@example\.com\nTo: " +
r"alice@example\.com\nDate: Tue, 07 Jan 2020 19:30:03 " +
r"-0000\nMessage-ID: \n <123456789\.123456\.123456789@example" +
r"\.com>\n\n--===============\d+==\nContent-Type: " +
r"multipart/mixed; protected-headers=\"v1\"; " +
r"boundary=\"===============\d+==\"\nMIME-Version: 1\.0\n" +
r"Return-Path: <alice@example\.com>\nReceived: from " +
r"example\.com \(example\.com \[127\.0\.0\.1\]\)\n by " +
r"example.com \(Postfix\) with ESMTPSA id E8DB612009F\n for" +
r" <alice@example\.com>; Tue, 7 Jan 2020 19:30:03 \+0200 " +
r"\(CEST\)\nSubject: Test\nFrom: [email protected]\nTo: " +
r"alice@example\.com\nDate: Tue, 07 Jan 2020 19:30:03 -0000\n" +
r"Message-ID: \n <123456789\.123456\.123456789@example\.com>\n" +
r"\n--===============\d+==\nContent-Type: text/rfc822-headers;" +
r" protected-headers=\"v1\"\nContent-Disposition: inline\n" +
r"(Date: Tue, 07 Jan 2020 19:30:03 -0000\n|Subject: Test\n|" +
r"From: alice@example\.com\n|To: alice@example\.com\n|" +
r"Message-ID: \n <123456789\.123456\.123456789@example\.com>\n" +
r")+\n\n--===============\d+==\nContent-Type: text/plain; " +
r"charset=\"utf-8\"\nContent-Transfer-Encoding: 7bit\n\nThis " +
r"is a test message\.\n--===============\d+==--\n\n" +
r"--===============\d+==\nContent-Type: application/pgp-" +
r"signature; name=\"signature\.asc\"\nContent-Description: " +
r"OpenPGP digital signature\nContent-Disposition: attachment; " +
r"filename=\"signature\.asc\"\n\n-----BEGIN PGP SIGNATURE-----" +
r"\n\n[\w\+/\n=]+-----END PGP SIGNATURE-----\n\n" +
r"--===============\d+==--\n")
self.assertIsNotNone(re.fullmatch(regex, signed))
def test_sign_encrypt_decrypt(self):
mail = (
"Return-Path: <*****@*****.**>\nReceived: from example.com (example.com "
+
"[127.0.0.1])\n by example.com (Postfix) with ESMTPSA id E8DB612009F\n"
+
" for <*****@*****.**>; Tue, 7 Jan 2020 19:30:03 +0200 (CEST)\n"
+
'Content-Type: text/plain; charset="utf-8"\n MIME-Version: 1.0\n' +
"Content-Transfer-Encoding: 7bit\nSubject: Test\nFrom: [email protected]"
+
"\nTo: [email protected]\nDate: Tue, 07 Jan 2020 19:30:03 -0000\n"
+
"Message-ID:\n <*****@*****.**>\n\nThis is a "
+ "test message.")
msg = "This is a test message."
p = Popen(
[
"./gpgmail",
"-E",
"*****@*****.**",
"--gnupghome",
self.temp_gpg_homedir.name,
"-p",
"test",
],
stdout=PIPE,
stdin=PIPE,
stderr=PIPE,
encoding="utf8",
)
encrypted = p.communicate(input=mail)[0]
self.assertTrue(msg not in encrypted)
p = Popen(
[
"./gpgmail",
"-p",
"test",
"-d",
"--gnupghome",
self.temp_gpg_homedir.name,
],
stdout=PIPE,
stdin=PIPE,
stderr=PIPE,
encoding="utf8",
)
decrypted = p.communicate(input=encrypted)[0]
self.assertTrue(msg in decrypted)
regex = (
r'Content-Type: multipart/mixed; protected-headers="v1"; boundary="'
+
r'===============\d+=="\nMIME-Version: 1\.0\nReturn-Path: <alice@example'
+
r"\.com>\nReceived: from example\.com \(example.com \[127\.0\.0\.1\]\)\n"
+
r" by example\.com \(Postfix\) with ESMTPSA id E8DB612009F\n for "
+
r"<alice@example\.com>; Tue, 7 Jan 2020 19:30:03 \+0200 \(CEST\)\n"
+
r"Subject: Test\nFrom: alice@example\.com\nTo: alice@example\.com\nDate: "
+
r"Tue, 07 Jan 2020 19:30:03 -0000\nMessage-ID: \n <123456789\.123456\."
+
r"123456789@example\.com>\n\n--===============\d+==\nContent-Type: text/"
+
r'rfc822-headers; protected-headers="v1"\nContent-Disposition: inline\n'
+
r"(Date: Tue, 07 Jan 2020 19:30:03 -0000\n|Subject: Test\n|From: "
+ r"alice@example\.com\n|To: alice@example\.com\n|Message-ID: \n" +
r" <123456789\.123456\.123456789@example\.com>\n)+\n\n--==============="
+
r'\d+==\nContent-Type: text/plain; charset="utf-8"\n MIME-Version: 1.0\n'
+
r"Content-Transfer-Encoding: 7bit\n\nThis is a test message\.\n--"
+ r"===============\d+==--\n")
self.assertIsNotNone(re.fullmatch(regex, decrypted))
def test_encryptheaders(self):
mail = (
"Return-Path: <*****@*****.**>\nReceived: from example.com (example.com "
+
"[127.0.0.1])\n by example.com (Postfix) with ESMTPSA id E8DB612009F\n"
+
" for <*****@*****.**>; Tue, 7 Jan 2020 19:30:03 +0200 (CEST)\n"
+
'Content-Type: text/plain; charset="utf-8"\nMIME-Version: 1.0\n' +
"Content-Transfer-Encoding: 7bit\nSubject: Test\nFrom: [email protected]"
+
"\nTo: [email protected]\nDate: Tue, 07 Jan 2020 19:30:03 -0000\n"
+
"Message-ID:\n <*****@*****.**>\n\nThis is a "
+ "test message.")
msg = "This is a test message."
p = Popen(
[
"./gpgmail",
"-e",
"*****@*****.**",
"--gnupghome",
self.temp_gpg_homedir.name,
"-H",
],
stdout=PIPE,
stdin=PIPE,
stderr=PIPE,
encoding="utf8",
)
encrypted, stderr = p.communicate(input=mail)
self.assertTrue(msg not in encrypted)
self.assertIn("Date: ...\n", encrypted)
self.assertIn("From: ...\n", encrypted)
self.assertIn("Message-ID: ...\n", encrypted)
self.assertIn("Subject: ...\n", encrypted)
self.assertIn("To: ...\n", encrypted)
def test_encryptfail(self):
mail = (
"Return-Path: <*****@*****.**>\nReceived: from example.com (example.com "
+
"[127.0.0.1])\n by example.com (Postfix) with ESMTPSA id E8DB612009F\n"
+
" for <*****@*****.**>; Tue, 7 Jan 2020 19:30:03 +0200 (CEST)\n"
+
'Content-Type: text/plain; charset="utf-8"\nMIME-Version: 1.0\n' +
"Content-Transfer-Encoding: 7bit\nSubject: Test\nFrom: [email protected]"
+
"\nTo: [email protected]\nDate: Tue, 07 Jan 2020 19:30:03 -0000\n"
+
"Message-ID:\n <*****@*****.**>\n\nThis is a "
+ "test message.")
p = Popen(
[
"./gpgmail",
"-e",
"*****@*****.**",
"--gnupghome",
self.temp_gpg_homedir.name,
"-H",
],
stdout=PIPE,
stdin=PIPE,
stderr=PIPE,
encoding="utf8",
)
encrypted, stderr = p.communicate(input=mail)
self.assertEqual(mail, encrypted)
def package_csar(self,
output_file: str,
service_template: str = None,
csar_format: str = "zip") -> str:
meta = self.parse_csar_meta()
try:
if not service_template:
root_yaml_files = self.get_root_yaml_files()
if meta is None and len(root_yaml_files) != 1:
raise ParseError(
f"You didn't specify the CSAR TOSCA entrypoint with '-t/--service-template' option. Therefore "
f"there should be one YAML file in the root of the CSAR to select it as the entrypoint. More "
f"than one YAML has been found: {list(map(str, root_yaml_files))}. Please select one of the "
f"files as the CSAR entrypoint using '-t/--service-template' flag or remove all the excessive "
f"YAML files.", self)
service_template = root_yaml_files[0].name
else:
if not self._member_exists(PurePath(service_template)):
raise ParseError(
f"The supplied TOSCA service template file '{service_template}' does not exist in folder "
f"'{self.csar_dir}'.", self)
meta = self.parse_csar_meta()
if meta is not None:
template_entry = meta.entry_definitions
if service_template and template_entry != service_template:
raise ParseError(
f"The file entry '{template_entry}' defined within 'Entry-Definitions' in "
f"'TOSCA-Metadata/TOSCA.meta' does not match with the file name '{service_template}' supplied "
f"in service_template CLI argument.", self)
# check if "Entry-Definitions" points to an existing
# template file in the CSAR
if template_entry is not None and not self._member_exists(
PurePath(template_entry)):
raise ParseError(
f"The file '{template_entry}' defined within 'Entry-Definitions' in "
f"'TOSCA-Metadata/TOSCA.meta' does not exist.", self)
return shutil.make_archive(output_file, csar_format,
self.csar_dir)
else:
with TemporaryDirectory(prefix="opera-package-") as tempdir:
extract_path = Path(tempdir) / "extract"
shutil.copytree(self.csar_dir, extract_path)
# create TOSCA-Metadata/TOSCA.meta file using the specified
# TOSCA service template or directory root YAML file
content = (f"TOSCA-Meta-File-Version: 1.1\n"
f"CSAR-Version: 1.1\n"
f"Created-By: xOpera TOSCA orchestrator\n"
f"Entry-Definitions: {service_template}\n")
meta_file_folder = extract_path / "TOSCA-Metadata"
meta_file = (meta_file_folder / "TOSCA.meta")
meta_file_folder.mkdir()
meta_file.touch()
meta_file.write_text(content, encoding="utf-8")
return shutil.make_archive(output_file, csar_format,
extract_path)
except Exception as e: # noqa: W0703
raise ParseError(f"Error creating CSAR:\n{traceback.format_exc()}",
self) from e
#
# Unit test cases for buku
#
import os
import re
import sqlite3
from genericpath import exists
from tempfile import TemporaryDirectory
import pytest
import unittest
from unittest import mock as mock
from buku import BukuDb, parse_tags, prompt
TEST_TEMP_DIR_OBJ = TemporaryDirectory(prefix='bukutest_')
TEST_TEMP_DIR_PATH = TEST_TEMP_DIR_OBJ.name
TEST_TEMP_DBDIR_PATH = os.path.join(TEST_TEMP_DIR_PATH, 'buku')
TEST_TEMP_DBFILE_PATH = os.path.join(TEST_TEMP_DBDIR_PATH, 'bookmarks.db')
TEST_BOOKMARKS = [
[
'http://slashdot.org', 'SLASHDOT',
parse_tags(['old,news']),
"News for old nerds, stuff that doesn't matter"
],
[
'http://www.zażółćgęśląjaźń.pl/', 'ZAŻÓŁĆ',
parse_tags(['zażółć,gęślą,jaźń']), "Testing UTF-8, zażółć gęślą jaźń."
],
[
def process_view(out_dir, subject, action, subaction, camera):
subj_dir = path.join('extracted', subject)
base_filename = metadata.get_base_filename(subject, action, subaction,
camera)
# Load joint position annotations
with pycdf.CDF(
path.join(subj_dir, 'Poses_D2_Positions',
base_filename + '.cdf')) as cdf:
poses_2d = np.array(cdf['Pose'])
poses_2d = poses_2d.reshape(poses_2d.shape[1], 32, 2)
with pycdf.CDF(
path.join(subj_dir, 'Poses_D3_Positions_mono_universal',
base_filename + '.cdf')) as cdf:
poses_3d_univ = np.array(cdf['Pose'])
poses_3d_univ = poses_3d_univ.reshape(poses_3d_univ.shape[1], 32, 3)
with pycdf.CDF(
path.join(subj_dir, 'Poses_D3_Positions_mono',
base_filename + '.cdf')) as cdf:
poses_3d = np.array(cdf['Pose'])
poses_3d = poses_3d.reshape(poses_3d.shape[1], 32, 3)
# Infer camera intrinsics
camera_int = infer_camera_intrinsics(poses_2d, poses_3d)
camera_int_univ = infer_camera_intrinsics(poses_2d, poses_3d_univ)
frame_indices = select_frame_indices_to_include(subject, poses_3d_univ)
frames = frame_indices + 1
video_file = path.join(subj_dir, 'Videos', base_filename + '.mp4')
frames_dir = path.join(out_dir, 'imageSequence', camera)
makedirs(frames_dir, exist_ok=True)
# Check to see whether the frame images have already been extracted previously
existing_files = {f for f in listdir(frames_dir)}
frames_are_extracted = True
for i in frames:
filename = 'img_%06d.jpg' % i
if filename not in existing_files:
frames_are_extracted = False
break
if not frames_are_extracted:
with TemporaryDirectory() as tmp_dir:
# Use ffmpeg to extract frames into a temporary directory
call([
'ffmpeg', '-nostats', '-loglevel', '0', '-i', video_file,
'-qscale:v', '3',
path.join(tmp_dir, 'img_%06d.jpg')
])
# Move included frame images into the output directory
for i in frames:
filename = 'img_%06d.jpg' % i
move(path.join(tmp_dir, filename),
path.join(frames_dir, filename))
return {
'pose/2d': poses_2d[frame_indices],
'pose/3d-univ': poses_3d_univ[frame_indices],
'pose/3d': poses_3d[frame_indices],
'intrinsics/' + camera: camera_int,
'intrinsics-univ/' + camera: camera_int_univ,
'frame': frames,
'camera': np.full(frames.shape, int(camera)),
'subject': np.full(frames.shape, int(included_subjects[subject])),
'action': np.full(frames.shape, int(action)),
'subaction': np.full(frames.shape, int(subaction)),
}
def load_df(
self,
df: pandas.DataFrame,
table: str,
field_dict: Optional[Dict[Any, Any]] = None,
delimiter: str = ',',
encoding: str = 'utf8',
pandas_kwargs: Any = None,
**kwargs: Any,
) -> None:
"""
Loads a pandas DataFrame into hive.
Hive data types will be inferred if not passed but column names will
not be sanitized.
:param df: DataFrame to load into a Hive table
:type df: pandas.DataFrame
:param table: target Hive table, use dot notation to target a
specific database
:type table: str
:param field_dict: mapping from column name to hive data type.
Note that it must be OrderedDict so as to keep columns' order.
:type field_dict: collections.OrderedDict
:param delimiter: field delimiter in the file
:type delimiter: str
:param encoding: str encoding to use when writing DataFrame to file
:type encoding: str
:param pandas_kwargs: passed to DataFrame.to_csv
:type pandas_kwargs: dict
:param kwargs: passed to self.load_file
"""
def _infer_field_types_from_df(df: pandas.DataFrame) -> Dict[Any, Any]:
dtype_kind_hive_type = {
'b': 'BOOLEAN', # boolean
'i': 'BIGINT', # signed integer
'u': 'BIGINT', # unsigned integer
'f': 'DOUBLE', # floating-point
'c': 'STRING', # complex floating-point
'M': 'TIMESTAMP', # datetime
'O': 'STRING', # object
'S': 'STRING', # (byte-)string
'U': 'STRING', # Unicode
'V': 'STRING', # void
}
order_type = OrderedDict()
for col, dtype in df.dtypes.iteritems():
order_type[col] = dtype_kind_hive_type[dtype.kind]
return order_type
if pandas_kwargs is None:
pandas_kwargs = {}
with TemporaryDirectory(prefix='airflow_hiveop_') as tmp_dir:
with NamedTemporaryFile(dir=tmp_dir, mode="w") as f:
if field_dict is None:
field_dict = _infer_field_types_from_df(df)
df.to_csv(
path_or_buf=f,
sep=delimiter,
header=False,
index=False,
encoding=encoding,
date_format="%Y-%m-%d %H:%M:%S",
**pandas_kwargs,
)
f.flush()
return self.load_file(filepath=f.name,
table=table,
delimiter=delimiter,
field_dict=field_dict,
**kwargs)
def setUp(self) -> None:
os.chdir(os.path.join(TEST_ROOT, "assets/nixpkgs"))
self.directory = TemporaryDirectory()
os.environ["HOME"] = self.directory.name
os.environ["GITHUB_TOKEN"] = "0000000000000000000000000000000000000000"
def harmonization(activity: dict, collection_id=None, **kwargs):
"""Harmonize Landsat and Sentinel-2 products."""
execution = execution_from_collection(activity,
collection_id=collection_id,
activity_type=harmonization.__name__)
collection = execution.activity.collection
from sensor_harm import landsat_harmonize, sentinel_harmonize
with TemporaryDirectory(prefix='harmonization',
suffix=activity['sceneid']) as tmp:
data_collection = get_provider_collection_from_activity(activity)
data_collection.path(collection)
target_dir = str(data_collection.path(collection))
target_tmp_dir = Path(tmp) / 'target'
target_tmp_dir.mkdir(exist_ok=True, parents=True)
reflectance_dir = Path(activity['args']['file'])
glob = list(
reflectance_dir.glob(f'**/{activity["sceneid"]}_Fmask4.tif'))
fmask = glob[0]
shutil.copy(str(fmask), target_tmp_dir)
if activity['sceneid'].startswith('S2'):
shutil.unpack_archive(activity['args']['compressed_file'], tmp)
entry = activity['sceneid']
entries = list(Path(tmp).glob('*.SAFE'))
if len(entries) == 1 and entries[0].suffix == '.SAFE':
entry = entries[0].name
l1 = Path(tmp) / entry
sentinel_harmonize(l1,
activity['args']['file'],
target_tmp_dir,
apply_bandpass=True)
else:
product_version = int(data_collection.parser.satellite())
sat_sensor = '{}{}'.format(data_collection.parser.source()[:2],
product_version)
landsat_harmonize(sat_sensor, activity["sceneid"],
activity['args']['file'], str(target_tmp_dir))
Path(target_dir).mkdir(exist_ok=True, parents=True)
for entry in Path(target_tmp_dir).iterdir():
entry_name = entry.name
target_entry = Path(target_dir) / entry_name
if target_entry.exists():
os.remove(str(target_entry))
shutil.move(str(entry), target_dir)
activity['args']['file'] = target_dir
return activity
def pdf_to_images(self) -> list:
with TemporaryDirectory() as path:
pil_images = convert_from_path(self.file_path, output_folder=path)
return pil_images
def download(activity: dict, **kwargs):
"""Celery tasks to deal with download data product from given providers."""
execution = create_execution(activity)
collector_extension = flask_app.extensions['bdc:collector']
collection = execution.activity.collection
scene_id = execution.activity.sceneid
logging.info(
f'Starting Download Task for {collection.name}(id={collection.id}, scene_id={scene_id})'
)
# Use parallel flag for providers which has number maximum of connections per client (Sentinel-Hub only)
download_order = collector_extension.get_provider_order(collection,
lazy=True,
parallel=True,
progress=False)
if len(download_order) == 0:
raise RuntimeError(
f'No provider set for collection {collection.id}({collection.name})'
)
data_collection = get_provider_collection_from_activity(activity)
download_file = data_collection.compressed_file(collection)
has_compressed_file = download_file is not None
# For files that does not have compressed file (Single file/folder), use native path
if download_file is None:
download_file = data_collection.path(collection)
is_valid_file = False
item = Item.query().filter(Item.collection_id == collection.id,
Item.name == scene_id).first()
if item:
# TODO: Get asset name of download file
item_path = item.assets['asset']['href']
item_path = item_path if not item_path.startswith(
'/') else item_path[1:]
item_path = Path(Config.DATA_DIR) / item_path
if item_path.exists():
logging.info(
f'Item {scene_id} exists. {str(item_path)} -> {str(download_file)}'
)
download_file = item_path
if download_file.exists() and has_compressed_file:
logging.info('File {} downloaded. Checking file integrity...'.format(
str(download_file)))
# TODO: Should we validate using Factory Provider.is_valid() ?
is_valid_file = is_valid_compressed_file(
str(download_file)) if download_file.is_file() else False
if not download_file.exists() or not is_valid_file:
# Ensure file is removed since it may be corrupted
if download_file.exists() and download_file.is_file():
download_file.unlink()
if not has_compressed_file:
download_file.mkdir(exist_ok=True, parents=True)
else:
download_file.parent.mkdir(exist_ok=True, parents=True)
with TemporaryDirectory(prefix='download_',
suffix=f'_{scene_id}') as tmp:
temp_file: Path = None
should_retry = False
for collector in download_order:
try:
logging.info(
f'Trying to download from {collector.provider_name}(id={collector.instance.id})'
)
temp_file = Path(
collector.download(
scene_id,
output=tmp,
dataset=activity['args']['dataset']))
activity['args']['provider_id'] = collector.instance.id
break
except DataOfflineError:
should_retry = True
except Exception as e:
logging.error(
f'Download error in provider {collector.provider_name} - {str(e)}'
)
if temp_file is None or not temp_file.exists():
if should_retry:
raise DataOfflineError(scene_id)
raise RuntimeError(f'Download fails {activity["sceneid"]}.')
shutil.move(str(temp_file), str(download_file))
refresh_execution_args(execution,
activity,
compressed_file=str(download_file))
return activity
def _delete_cluster(self, terraform_dir=None):
module_logger.debug(
'cluster._delete_cluster() entry, terraform_dir={} self.deploy_dir={} self.platform={}'
.format(terraform_dir, self.deploy_dir, self.platform))
if terraform_dir is None:
if self.name is None:
raise ClusterNotInitialized(
'Cannot delete if no cluster name is specified.')
terraform_dir = self.deploy_dir
if 'openshift4-aws' in self.platform:
deploy_dir_suffix = '/aws-ipi'
else:
deploy_dir_suffix = ''
tfstate_file = '{}{}/terraform.tfstate'.format(terraform_dir,
deploy_dir_suffix)
module_logger.debug('tfstate_file={} exists={}'.format(
tfstate_file, os.path.isfile(tfstate_file)))
self._modify_modules_json()
tfvars_file = '{}/terraform_inputs/.{}.tfvars'.format(
self.git_repo.git_dir, self.platform)
sh_env = os.environ.copy()
sh_env['TF_VAR_private_key_file'] = self.private_key_file
sh_env['TF_VAR_public_key_file'] = '{}.pub'.format(
self.private_key_file)
with with_cd(self.git_repo.git_dir):
with TemporaryDirectory() as temp_tfstate_dir:
temp_tfstate_dir = str(Path(temp_tfstate_dir).resolve())
if 'openshift4-aws' in self.platform:
run('sed -i="" -e "s|__W3_EMAIL__|$TF_VAR_user_name|g" -e "s|__AWS_ACCESS_KEY__|$TF_VAR_aws_access_key|g" -e "s|__AWS_SECRET_ACCESS_KEY__|$TF_VAR_aws_secret_key|g" -e "s|__RH_SUBSCRIPTION_USERNAME__|$TF_VAR_rhel_subscription_username|g" -e "s|__RH_SUBSCRIPTION_PASSWORD__|$TF_VAR_rhel_subscription_password|g" -e "s|__OCP__PULL_SECRET_FILE__|$OCP_PULL_SECRET_FILE|g" -e "s|__PRIVATE_KEY_FILE__|$PRIVATE_KEY_FILE|g" -e "s|__PUBLIC_KEY_FILE__|$PUBLIC_KEY_FILE|g" -e "s|__EDITION__|$EDITION|g" -e "s|__FIXPACK__|$FIXPACK|g" -e "s|__VERSION__|$VERSION|g" -e "s|__REPO__|$DEPLOY_REPO|g" -e "s|__ARTIFACTORY_USER__|$ARTIFACTORY_USER|g" -e "s|__ARTIFACTORY_API_KEY__|$ARTIFACTORY_TOKEN|g" {}'
.format(tfvars_file),
shell=True,
check=True,
env=sh_env)
if not os.path.isfile('pull-secret'):
run('openssl aes-256-cbc -K ${} -iv ${} -in pull-secret.enc -out pull-secret -d'
.format(os.environ['OCP_PULL_SECRET_FILE_VAR_KEY'],
os.environ['OCP_PULL_SECRET_FILE_VAR_IV']),
shell=True,
check=True,
env=sh_env)
if 'travis' in self.platform:
module_logger.debug(
'found openshift4-aws-travis in self.platform; running sed on __CLUSTER_NAME__ of {}.'
.format(tfvars_file))
run('sed -i="" -e "s|__CLUSTER_NAME__|{}|g" {}'.format(
self._get_terraform_output('cluster-name',
self.deploy_dir,
length_of_output=1),
tfvars_file),
shell=True,
check=True,
env=sh_env)
if os.path.isfile('{}.enc'.format(tfstate_file)):
decrypted_tfstate_file = '{}/terraform.tfstate'.format(
temp_tfstate_dir)
run_cmd = 'openssl enc -aes-256-cbc -d -in {}.enc -out {} -k {}'.format(
tfstate_file, decrypted_tfstate_file,
'afro-donkey-seldom-waterfall-compute')
run(run_cmd, shell=True)
else:
decrypted_tfstate_file = tfstate_file
# aws-ipi needs this abomination because of its directory hierarchy; else cluster destroys fail
destroyer_cmd = 'cd ..; make -s terraform:destroy TERRAFORM_DIR={}/aws-ipi TERRAFORM_VARS_FILE={} TERRAFORM_STATE_FILE={}'.format(
self.deploy_dir, tfvars_file, decrypted_tfstate_file)
else:
destroyer_cmd = 'make -s terraform:destroy TERRAFORM_DIR={} TERRAFORM_VARS_FILE={} TERRAFORM_STATE_FILE={}'.format(
self.deploy_dir, tfvars_file, tfstate_file)
module_logger.debug(
'terraform invocation:\n {}'.format(destroyer_cmd))
output = run(destroyer_cmd, shell=True, env=sh_env)
if output.returncode != 0:
module_logger.info(
'Destroy completed with errors, but everything is likely cleaned up.'
)
def install_frontend(self,
source,
oauth_key,
oauth_secret,
backend_url,
settings_file=None,
network='public',
hub_id=None):
""" Install connector-frontend in Odin Automation Hub, --source can be http(s):// or
filepath"""
with TemporaryDirectory() as tdir:
is_http_source = True if source.startswith('http://') or source.startswith('https://') \
else False
if is_http_source:
package_name = _download_file(source, target=tdir)
else:
package_name = os.path.basename(source)
copyfile(os.path.expanduser(source),
os.path.join(tdir, package_name))
package_path = os.path.join(tdir, package_name)
with zipfile.ZipFile(package_path, 'r') as zip_ref:
meta_path = zip_ref.extract('APP-META.xml', path=tdir)
tenant_schema_path = zip_ref.extract('schemas/tenant.schema',
tdir)
app_schema_path = zip_ref.extract('schemas/app.schema', tdir)
try:
zip_ref.extract('schemas/user.schema', tdir)
user_service = True
except KeyError:
user_service = False
tree = xml_et.ElementTree(file=meta_path)
namespace = '{http://aps-standard.org/ns/2}'
connector_id = tree.find('{}id'.format(namespace)).text
version = tree.find('{}version'.format(namespace)).text
release = tree.find('{}release'.format(namespace)).text
# Get connector name from id as <name> field may not be unique
url_path = urlparse(connector_id).path
connector_name = os.path.split(url_path)[-1]
if not settings_file:
settings_file = {}
else:
settings_file = json.load(open(settings_file))
if backend_url.startswith('http://'):
print(
"WARN: Make sure that the APS development mode enabled for http backend. "
"Run `apsconnect aps_devel_mode` command.")
elif backend_url.startswith('https://'):
pass
else:
print("Backend url must be URL http(s)://, got {}".format(
backend_url))
sys.exit(1)
cfg, hub = _get_cfg(), _get_hub()
with open(package_path, 'rb') as package_binary:
print("Importing connector {} {}-{}".format(
connector_id, version, release))
import_kwargs = {'package_url': source} if is_http_source \
else {'package_body': xmlrpclib.Binary(package_binary.read())}
response = hub.APS.importPackage(**import_kwargs)
_osaapi_raise_for_status(response)
application_id = str(response['result']['application_id'])
print("Connector {} imported with id={} [ok]".format(
connector_id, application_id))
payload = {
'aps': {
'package': {
'type': connector_id,
'version': version,
'release': release,
},
'endpoint': backend_url,
'network': network,
'auth': {
'oauth': {
'key': oauth_key,
'secret': oauth_secret,
},
},
},
}
# Get Unique OA id for using as hubId parameter while endpoint deploying
base_aps_url = _get_aps_url(
**{k: _get_cfg()[k]
for k in APS_CONNECT_PARAMS})
app_properties = _get_properties(app_schema_path)
if 'hubId' in app_properties:
url = '{}/{}'.format(
base_aps_url,
'aps/2/resources?implementing(http://parallels.com/aps/types/pa/poa/1.0)',
)
response = request(method='GET',
url=url,
headers=_get_user_token(hub, cfg['user']),
verify=False)
response.raise_for_status()
try:
data = json.loads(response.content.decode('utf-8'))
except ValueError:
print("APSController provided non-json format")
sys.exit(1)
if not data and not hub_id:
raise Exception(
"Core OA resource is not found\n"
"Use --hub-id={value} argument to specify the ID "
"manually or --hub-id=auto to generate it automatically"
)
elif data:
hub_id = data[0]['aps']['id']
elif hub_id == 'auto':
hub_id = str(uuid.uuid4())
payload.update({'app': {'hubId': hub_id}})
payload.update(settings_file)
response = request(method='POST',
url='{}/{}'.format(base_aps_url,
'aps/2/applications/'),
headers=_get_user_token(hub, cfg['user']),
verify=False,
json=payload)
try:
response.raise_for_status()
except Exception as e:
if 'error' in response.json():
err = "{} {}".format(response.json()['error'],
response.json()['message'])
else:
err = str(e)
print("Installation of connector {} FAILED.\n"
"Hub APS API response {} code.\n"
"Error: {}".format(connector_id, response.status_code,
err))
# Create app, tenant, users resource types
resource_uid = json.loads(
response.content.decode('utf-8'))['app']['aps']['id']
core_resource_types_payload = [
{
'resclass_name':
'rc.saas.service.link',
'name':
connector_name,
'act_params': [
{
'var_name': 'app_id',
'var_value': application_id
},
{
'var_name': 'resource_uid',
'var_value': resource_uid
},
]
},
{
'resclass_name':
'rc.saas.service',
'name':
'{} tenant'.format(connector_name),
'act_params': [
{
'var_name': 'app_id',
'var_value': application_id
},
{
'var_name': 'service_id',
'var_value': 'tenant'
},
{
'var_name': 'autoprovide_service',
'var_value': '1'
},
]
},
]
# Collect ids for service template creation
resource_types_ids = []
limited_resources = {}
for type in core_resource_types_payload:
response = hub.addResourceType(**type)
_osaapi_raise_for_status(response)
resource_types_ids.append(
response['result']['resource_type_id'])
for id in list(resource_types_ids):
limited_resources[id] = 1
if user_service:
user_resource_type_payload = {
'resclass_name':
'rc.saas.service',
'name':
'{} users'.format(connector_name),
'act_params': [
{
'var_name': 'app_id',
'var_value': application_id
},
{
'var_name': 'service_id',
'var_value': 'user'
},
{
'var_name': 'autoprovide_service',
'var_value': '0'
},
]
}
response = hub.addResourceType(**user_resource_type_payload)
_osaapi_raise_for_status(response)
resource_types_ids.append(
response['result']['resource_type_id'])
# Create counters resource types
counters = _get_counters(tenant_schema_path)
for counter in counters:
payload = {
'resclass_name':
"rc.saas.resource.unit",
'name':
'{} {}'.format(connector_name, counter),
'act_params': [
{
'var_name': 'app_id',
'var_value': application_id
},
{
'var_name': 'service_id',
'var_value': "tenant"
},
{
'var_name': 'resource_id',
'var_value': counter
},
]
}
response = hub.addResourceType(**payload)
_osaapi_raise_for_status(response)
resource_types_ids.append(
response['result']['resource_type_id'])
# Create parameters resource types
parameters = _get_parameters(tenant_schema_path)
for parameter in parameters:
payload = {
'resclass_name':
"rc.saas.resource.unit",
'name':
'{} {}'.format(connector_name, parameter),
'act_params': [
{
'var_name': 'app_id',
'var_value': application_id
},
{
'var_name': 'service_id',
'var_value': "tenant"
},
{
'var_name': 'resource_id',
'var_value': parameter
},
]
}
response = hub.addResourceType(**payload)
_osaapi_raise_for_status(response)
resource_types_ids.append(
response['result']['resource_type_id'])
limited_resources[response['result']['resource_type_id']] = 0
print("Resource types creation [ok]")
# Create service template
payload = {
'name': connector_name,
'owner_id': 1,
'resources': [],
}
for type_id in resource_types_ids:
payload['resources'].append({'resource_type_id': type_id})
response = hub.addServiceTemplate(**payload)
_osaapi_raise_for_status(response)
service_template_id = response['result']['st_id']
print("Service template \"{}\" created with id={} [ok]".format(
connector_name, service_template_id))
# Set up limits
payload = {
'st_id': service_template_id,
'limits': [],
}
for type_id, limit in limited_resources.items():
payload['limits'].append({
'resource_id': type_id,
'resource_limit64': str(limit)
})
response = hub.setSTRTLimits(**payload)
_osaapi_raise_for_status(response)
print("Limits for Service template \"{}\" are applied [ok]".format(
service_template_id))
def main():
print(f"\x1b[33m[{now()}]\x1b[0m")
base = Path(gettempdir()) / "texclip"
text = ClipBoard.read(CF_UNICODETEXT)
with TemporaryDirectory() as workspace:
workspace = Path(workspace)
template = tex_content(text)
texfile = workspace / "main.tex"
texfile.write_bytes(template.encode("utf-8"))
tex_cmd = ["xelatex", texfile.as_posix()]
try:
tex_status = run(tex_cmd,
shell=False,
cwd=workspace,
stdout=PIPE,
stderr=PIPE,
timeout=10)
tex_status.check_returncode()
except TimeoutExpired as e:
print(f"error in tex: {text}")
print("\x1b[31m")
print(e.stdout.decode("utf-8"))
print("\x1b[0m")
return
except CalledProcessError as e:
print(f"error in tex: {text}")
print("\x1b[31m")
print(e.stderr.decode("utf-8"))
print("\x1b[0m")
return
# main.pdf 已生成
pdffile = workspace / "main.pdf"
outimgfile = base / \
f"texclip-{strftime('%Y%m%d-%H%M%S', localtime())}.png"
magick_cmd = [
"magick", "-density", "300",
pdffile.as_posix(), "-quality", "100", "-trim",
outimgfile.absolute().as_posix()
]
try:
magick_status = run(magick_cmd,
shell=False,
cwd=workspace,
stdout=PIPE,
stderr=PIPE,
timeout=5)
magick_status.check_returncode()
except TimeoutExpired as e:
print(f"[{now()}] error in magick")
print("\x1b[31m")
print(e.stdout.decode("utf-8"))
print("\x1b[0m")
return
except CalledProcessError as e:
print(f"[{now()}] error in magick")
print("\x1b[31m")
print(e.stderr.decode("utf-8"))
print("\x1b[0m")
return
print(f"\x1b[32m{text[:10]}.. saved at {outimgfile.as_posix()}\x1b[0m")
ToastNotifier().show_toast("Task Completed",
f"{outimgfile.absolute().as_posix()}")
def generate(self, initialized):
with self.git_repo as r:
creating_dir = '{}/{}/{}'.format(r.working_dir,
self.cluster_directory,
creating_id())
creating_file = '{}/{}'.format(creating_dir, CREATING_NAME)
if not os.path.exists(creating_dir):
os.makedirs(creating_dir)
with open(creating_file, 'w') as f:
f.write(self.identifying_info)
r.index.add([creating_file])
# if we are not initialized, trigger another travis build
if initialized:
r.index.commit('cluster build started [skip ci]')
else:
r.index.commit('cluster build started')
tfvars_file = '{}/terraform_inputs/.{}.tfvars'.format(
r.working_dir, self.platform)
with TemporaryDirectory() as temp_dir:
temp_dir = str(Path(temp_dir).resolve())
self._temp_deploy_dir = temp_dir
sh_env = os.environ.copy()
sh_env['TF_VAR_private_key_file'] = self.private_key_file
sh_env['TF_VAR_public_key_file'] = '{}.pub'.format(
self.private_key_file)
sh_env['TF_VAR_keypair_override'] = 'icp-cicd-pipeline-keypair'
sh_env['TF_VAR_instance_name'] = INSTANCE_NAME
exception = None
tries = 1
try:
if 'openstack' in self.platform:
run('make -s deploy:openstack OPENSTACK_DEPLOY_DIR={} OPENSTACK_TERRAFORM_VARS_FILE={}'
.format(temp_dir, tfvars_file),
shell=True,
check=True,
env=sh_env)
elif 'openshift-aws' in self.platform:
run('make -s deploy:openshift:aws OPENSHIFT_AWS_DEPLOY_DIR={} OPENSHIFT_AWS_TERRAFORM_VARS_FILE={}'
.format(temp_dir, tfvars_file),
shell=True,
check=True,
env=sh_env)
elif 'openshift4-aws' in self.platform:
if 'travis' in self.platform:
run('sed -i="" -e "s|__CLUSTER_NAME__|pool-os4-{}|g" {}'
.format(creating_id(), tfvars_file),
shell=True,
check=True,
env=sh_env)
run('sed -i="" -e "s|__W3_EMAIL__|$TF_VAR_user_name|g" -e "s|__AWS_ACCESS_KEY__|$TF_VAR_aws_access_key|g" -e "s|__AWS_SECRET_ACCESS_KEY__|$TF_VAR_aws_secret_key|g" -e "s|__RH_SUBSCRIPTION_USERNAME__|$TF_VAR_rhel_subscription_username|g" -e "s|__RH_SUBSCRIPTION_PASSWORD__|$TF_VAR_rhel_subscription_password|g" -e "s|__OCP__PULL_SECRET_FILE__|$OCP_PULL_SECRET_FILE|g" -e "s|__PRIVATE_KEY_FILE__|$PRIVATE_KEY_FILE|g" -e "s|__PUBLIC_KEY_FILE__|$PUBLIC_KEY_FILE|g" -e "s|__EDITION__|$EDITION|g" -e "s|__FIXPACK__|$FIXPACK|g" -e "s|__VERSION__|$VERSION|g" -e "s|__REPO__|$DEPLOY_REPO|g" -e "s|__ARTIFACTORY_USER__|$ARTIFACTORY_USER|g" -e "s|__ARTIFACTORY_API_KEY__|$ARTIFACTORY_TOKEN|g" {}'
.format(tfvars_file),
shell=True,
check=True,
env=sh_env)
run('make -s deploy:openshift4:aws OPENSHIFT_4_AWS_DEPLOY_DIR={} OPENSHIFT_4_AWS_TERRAFORM_VARS_FILE={}'
.format(temp_dir, tfvars_file),
shell=True,
check=True,
env=sh_env)
elif 'aws' in self.platform:
run('make -s deploy:aws AWS_DEPLOY_DIR={} AWS_TERRAFORM_VARS_FILE={}'
.format(temp_dir, tfvars_file),
shell=True,
check=True,
env=sh_env)
else:
raise ClusterException(
'No rule in place to handle the tfvars file passed: {}. Ensure that you are trying to deploy a supported cluster.'
.format(tfvars_file))
exception = None
except CalledProcessError as e:
exception = e
module_logger.error(
'Cluster failed to create successfully; retries left: {}'.
format(self.max_tries - tries))
while tries < self.max_tries:
tries += 1
if exception is None:
# If we don't have an exception, there is no need to retry.
break
try:
run('make -s terraform:apply TERRAFORM_DIR={} TERRAFORM_VARS_FILE={}'
.format(temp_dir, tfvars_file),
shell=True,
check=True,
env=sh_env)
exception = None
except CalledProcessError as e:
exception = e
module_logger.error(
'Cluster failed to create successfully; retries left: {}'
.format(self.max_tries - tries))
if exception is not None:
module_logger.error(
'Cluster failed to create successfully; deleting cluster')
module_logger.error(exception)
if not self.allow_failed_create:
self.delete(terraform_dir=temp_dir, skip_ci=False)
return
else:
module_logger.info(
'Deletion skipped; we are allowing failed clusters')
self.name = self._get_terraform_output('cluster-name',
self.deploy_dir,
length_of_output=1)
if self.name == '':
self.name = None
self.delete(terraform_dir=temp_dir)
return
# If there is a duplicate directory, shutil.copytree will fail with FileExistsError
# TODO determine what to do when this happens
try:
with self.git_repo as r:
dest_repo_dir = '{}/{}'.format(self.cluster_dir,
DEPLOY_DIR_NAME)
if 'openshift4-aws' in self.platform:
src_repo_tfstate_dir = '{}/aws-ipi'.format(
self.deploy_dir, DEPLOY_DIR_NAME)
dest_repo_tfstate_dir = '{}/aws-ipi'.format(
dest_repo_dir, DEPLOY_DIR_NAME)
else:
src_repo_tfstate_dir = self.deploy_dir
dest_repo_tfstate_dir = dest_repo_dir
shutil.copytree(temp_dir,
'{}'.format(dest_repo_dir),
ignore=shutil.ignore_patterns(
'.git*', 'terraform.tfstate'))
# encrypt tfstate file, remove real one before pushing to github
run_cmd = 'openssl enc -aes-256-cbc -salt -in {}/terraform.tfstate -out {}/terraform.tfstate.enc -k {}'.format(
src_repo_tfstate_dir, dest_repo_tfstate_dir,
'afro-donkey-seldom-waterfall-compute')
run(run_cmd, shell=True)
touch('{}/{}'.format(self.cluster_dir, AVAILABLE_NAME))
self._modify_modules_json(key='Dir',
value='{}/{}/{}'.format(
self.cluster_directory,
self.name, DEPLOY_DIR_NAME))
r.index.add([self.cluster_dir])
r.index.move([
creating_file,
'{}/source-job.md'.format(self.cluster_dir)
])
r.index.commit('save cluster state {} [skip ci]'.format(
self.name))
with with_cd(self.git_repo.git_dir):
run('pwd; git status', shell=True)
except Exception as e:
module_logger.error(
'Error trying to save the cluster. Deleting.')
module_logger.error(e)
self.delete(terraform_dir=temp_dir)
raise e
self._temp_deploy_dir = None
def predict(self, dataset: Dataset, **kwargs) -> Iterator[Forecast]:
with TemporaryDirectory() as tempdir:
predictor_path = Path(tempdir)
self._base_predictor.serialize(predictor_path)
# TODO: Consider using shared memory for the data transfer.
self._input_queues = [mp.Queue() for _ in range(self._num_workers)]
self._output_queue = mp.Queue()
workers = []
for worker_id, in_q in enumerate(self._input_queues):
worker = mp.Process(
target=_worker_loop,
args=(predictor_path, in_q, self._output_queue, worker_id),
kwargs=kwargs,
)
worker.daemon = True
worker.start()
workers.append(worker)
self._num_running_workers += 1
self._workers = workers
chunked_data = self._grouper(dataset, self._chunk_size)
self._send_idx = 0
self._next_idx = 0
self._data_buffer = {}
worker_ids = list(range(self._num_workers))
def receive():
idx, worker_id, result = self._output_queue.get()
if isinstance(idx, WorkerError):
self._num_running_workers -= 1
self.terminate()
raise Exception(idx.msg)
if idx is not None:
self._data_buffer[idx] = result
return idx, worker_id, result
def get_next_from_buffer():
while self._next_idx in self._data_buffer:
result_batch = self._data_buffer.pop(self._next_idx)
self._next_idx += 1
yield from result_batch
def send(worker_id, chunk):
q = self._input_queues[worker_id]
q.put((self._send_idx, chunk))
self._send_idx += 1
try:
# prime the queues
for wid in worker_ids:
chunk = next(chunked_data)
send(wid, chunk)
while True:
idx, wid, result = receive()
yield from get_next_from_buffer()
chunk = next(chunked_data)
send(wid, chunk)
except StopIteration:
# signal workers end of data
for q in self._input_queues:
q.put((None, None))
# collect any outstanding results
while self._num_running_workers > 0:
idx, worker_id, result = receive()
if idx is None:
self._num_running_workers -= 1
continue
yield from get_next_from_buffer()
assert len(self._data_buffer) == 0
assert self._send_idx == self._next_idx
def temp_dir():
with TemporaryDirectory() as tempdir:
yield tempdir
# We check the kinetic energy to be sure that the particles are not overlapping and do not explode
def checkKineticEnergy(lowerBound, upperBound):
Ekin = kineticEnergy()
if (Ekin > upperBound or Ekin < lowerBound):
O.pause()
raise YadeCheckError(
"Kinetic energy %E is not within bounds from %E to %E! Error!" %
(Ekin, lowerBound, upperBound))
else:
print("Kinetic energy OK %E" % (Ekin))
#write some restart files
with TemporaryDirectory() as tmp_dir:
O.save(tmp_dir + '/restartMinWorkEx_' + partType + '_Initial')
# Run simulation to check the kinetic energy
O.run(10, True)
checkKineticEnergy(0.6e-9, 1.4e-9)
# Let the object settle
O.run(100000, True)
curIter = O.iter
checkKineticEnergy(0.07, 0.16)
# Save simulation
O.save(tmp_dir + '/restartMinWorkEx_%s_%d' % (partType, curIter))
time.sleep(1)
def handle(self):
output_directory = Path(self.args.output_directory)
# We simulate pybabel and sphinx-build commands. Variable names are chosen to match upstream code.
# For sphinx-build, the code path is:
#
# * bin/sphinx-build calls main() in sphinx, which calls build_main(), which calls main() in sphinx.cmdline
# * main() calls Sphinx(…).build(…) in sphinx.application
# sphinx-build -E -q …
kwargs = {
'confoverrides': {
'source_suffix': ['.rst', '.md'],
'source_parsers': {
'.md': CommonMarkParser,
},
},
'freshenv': True,
'parallel': 1,
}
if not self.args.verbose:
kwargs.update(status=None)
# For pybabel, the code path is:
#
# * bin/pybabel calls main() in babel.messages.frontend
# * main() calls CommandLineInterface().run(sys.argv)
# * CommandLineInterface() calls extract_messages(), which:
# 1. Reads the input path and method map from command-line options
# 2. Instantiates a catalog
# 3. Calls extract_from_dir() in babel.messages.extract to extract messages
# 4. extract_from_dir() calls check_and_call_extract_file() to find the method in the method map
# 5. check_and_call_extract_file() calls extract_from_file() to open a file for extraction
# 6. extract_from_file() calls extract() to extract messages
# 7. Adds the messages to the catalog
# 8. Writes a POT file
# 1. Reads the input path and method map from command-line options
arguments = [
# pybabel extract -F babel_ocds_codelist.cfg . -o $(POT_DIR)/$(DOMAIN_PREFIX)codelists.pot
('codelists.pot', [
('codelists/*.csv', extract_codelist),
]),
# pybabel extract -F babel_ocds_schema.cfg . -o $(POT_DIR)/$(DOMAIN_PREFIX)schema.pot
('schema.pot', [
('*-schema.json', extract_schema),
('extension.json', extract_extension_metadata),
]),
]
for version in self.versions():
if not version.download_url:
logger.warning('No Download URL for {}=={}'.format(
version.id, version.version))
outdir = output_directory / version.id / version.version
outdir.mkdir(parents=True, exist_ok=True)
# See the `files` method of `ExtensionVersion` for similar code.
response = requests.get(version.download_url, allow_redirects=True)
response.raise_for_status()
with closing(ZipFile(BytesIO(response.content))) as zipfile:
names = zipfile.namelist()
start = len(names[0])
for output_file, method_map in arguments:
# 2. Instantiates a catalog
catalog = Catalog()
# 3. Calls extract_from_dir() in babel.messages.extract to extract messages
for name in names[1:]:
filename = name[start:]
# 4. extract_from_dir() calls check_and_call_extract_file()
for pattern, method in method_map:
if not pathmatch(pattern, filename):
continue
# 5. check_and_call_extract_file() calls extract_from_file()
with zipfile.open(name) as fileobj:
# 6. extract_from_file() calls extract() to extract messages
for lineno, message, comments, context in extract(
method, fileobj):
# 7. Adds the messages to the catalog
catalog.add(message,
None, [(filename, lineno)],
auto_comments=comments,
context=context)
break
# 8. Writes a POT file
if catalog:
with open(outdir / output_file, 'wb') as outfile:
write_po(outfile, catalog)
with TemporaryDirectory() as srcdir:
for info in zipfile.infolist()[1:]:
filename = info.filename[start:]
if filename[-1] != '/' and filename.startswith(
'docs/') or filename == 'README.md':
info.filename = filename
zipfile.extract(info, srcdir)
with cd(srcdir):
# Eliminates a warning, without change to output.
with open('contents.rst', 'w') as f:
f.write(
'.. toctree::\n :glob:\n\n docs/*\n README'
)
# sphinx-build -b gettext $(DOCS_DIR) $(POT_DIR)
app = Sphinx('.', None, '.', '.', 'gettext', **kwargs)
app.build(True)
# https://stackoverflow.com/questions/15408348
content = subprocess.run(['msgcat', *glob('*.pot')],
check=True,
stdout=subprocess.PIPE).stdout
with open(outdir / 'docs.pot', 'wb') as f:
f.write(content)
def start_python_pipeline( # pylint: disable=too-many-arguments
self,
variables: dict,
py_file: str,
py_options: List[str],
py_interpreter: str = "python3",
py_requirements: Optional[List[str]] = None,
py_system_site_packages: bool = False,
process_line_callback: Optional[Callable[[str], None]] = None,
):
"""
Starts Apache Beam python pipeline.
:param variables: Variables passed to the pipeline.
:type variables: Dict
:param py_options: Additional options.
:type py_options: List[str]
:param py_interpreter: Python version of the Apache Beam pipeline.
If None, this defaults to the python3.
To track python versions supported by beam and related
issues check: https://issues.apache.org/jira/browse/BEAM-1251
:type py_interpreter: str
:param py_requirements: Additional python package(s) to install.
If a value is passed to this parameter, a new virtual environment has been created with
additional packages installed.
You could also install the apache-beam package if it is not installed on your system or you want
to use a different version.
:type py_requirements: List[str]
:param py_system_site_packages: Whether to include system_site_packages in your virtualenv.
See virtualenv documentation for more information.
This option is only relevant if the ``py_requirements`` parameter is not None.
:type py_system_site_packages: bool
:param on_new_job_id_callback: Callback called when the job ID is known.
:type on_new_job_id_callback: callable
"""
if "labels" in variables:
variables["labels"] = [
f"{key}={value}" for key, value in variables["labels"].items()
]
if py_requirements is not None:
if not py_requirements and not py_system_site_packages:
warning_invalid_environment = textwrap.dedent("""\
Invalid method invocation. You have disabled inclusion of system packages and empty list
required for installation, so it is not possible to create a valid virtual environment.
In the virtual environment, apache-beam package must be installed for your job to be \
executed. To fix this problem:
* install apache-beam on the system, then set parameter py_system_site_packages to True,
* add apache-beam to the list of required packages in parameter py_requirements.
""")
raise AirflowException(warning_invalid_environment)
with TemporaryDirectory(prefix="apache-beam-venv") as tmp_dir:
py_interpreter = prepare_virtualenv(
venv_directory=tmp_dir,
python_bin=py_interpreter,
system_site_packages=py_system_site_packages,
requirements=py_requirements,
)
command_prefix = [py_interpreter] + py_options + [py_file]
self._start_pipeline(
variables=variables,
command_prefix=command_prefix,
process_line_callback=process_line_callback,
)
else:
command_prefix = [py_interpreter] + py_options + [py_file]
self._start_pipeline(
variables=variables,
command_prefix=command_prefix,
process_line_callback=process_line_callback,
)
def test_command_line_interface():
"""Test the CLI."""
runner = CliRunner()
help_result = runner.invoke(cli.main, ['--help'])
assert help_result.exit_code == 0
assert re.search(r'--help\s+Show this message and exit.', help_result.output, flags=re.DOTALL) is not None
# Test replacing multi-char deletion and SNPs and setting sample name via command-line
with TemporaryDirectory(prefix='vcf_consensus_builder', dir='/tmp') as tempdir:
temppath = Path(tempdir)
full_fasta_output = temppath / OUTPUT_FASTA
sample_name = 'SAMPLE1'
result = runner.invoke(cli.main,
['-v', VCF_DEL,
'-d', DEPTHS,
'-r', REF_FASTA,
'-o', full_fasta_output,
'--sample-name', sample_name])
assert result.exit_code == 0
rec = SeqIO.read(full_fasta_output, 'fasta')
assert str(rec.seq) == 'NACCGTANACAATAN--', 'There must be a deletion of 3 characters in the middle of the seq'
assert rec.id == sample_name
# Test changing no coverage threshold
with TemporaryDirectory(prefix='vcf_consensus_builder', dir='/tmp') as tempdir:
temppath = Path(tempdir)
full_fasta_output = temppath / OUTPUT_FASTA
sample_name = 'SAMPLE1'
result = runner.invoke(cli.main,
['-v', VCF_DEL,
'-d', DEPTHS,
'-r', REF_FASTA,
'-o', full_fasta_output,
'--sample-name', sample_name,
'--no-coverage', 4])
assert result.exit_code == 0
rec = SeqIO.read(full_fasta_output, 'fasta')
assert str(rec.seq) == '-ACCGTA-ACAATA---', 'Positions below 5X coverage must be replaced with "-"'
assert rec.id == sample_name
# Test replacing low and no coverage characters with other characters than default N and - respectively
with TemporaryDirectory(prefix='vcf_consensus_builder', dir='/tmp') as tempdir:
temppath = Path(tempdir)
full_fasta_output = temppath / OUTPUT_FASTA
sample_name = 'SAMPLE1'
result = runner.invoke(cli.main,
['-v', VCF_DEL,
'-d', DEPTHS,
'-r', REF_FASTA,
'-o', full_fasta_output,
'--sample-name', sample_name,
'--no-cov-char', '=',
'--low-cov-char', '@'])
assert result.exit_code == 0
rec = SeqIO.read(full_fasta_output, 'fasta')
assert str(rec.seq) == '@ACCGTA@ACAATA@==', \
'No coverage positions must be replaced with "=". Low coverage (<5X) positions must be replaced with "@".'
assert rec.id == sample_name
# Test replacing multi-char insertion and SNPs
with TemporaryDirectory(prefix='vcf_consensus_builder', dir='/tmp') as tempdir:
temppath = Path(tempdir)
full_fasta_output = temppath / OUTPUT_FASTA
result = runner.invoke(cli.main,
['-v', VCF_INS,
'-d', DEPTHS,
'-r', REF_FASTA,
'-o', full_fasta_output])
assert result.exit_code == 0
rec = SeqIO.read(full_fasta_output, 'fasta')
assert str(rec.seq) == 'NACCGTATTTGTCNACAATAN--', 'There must be an insertion of "TTT" in the middle of the seq'
assert rec.id == 'sample1', \
'FASTA ID must be the first sample name in the VCF if not explicitly specified as a command-line arg'
mock_detector.save_attributes("FeatureExtraction")
attribute_dict = mock_detector.toolset["attributes"]
test_id = checkpoint_save_config["test_id"]
for attribute_name in attribute_dict:
attribute_val = attribute_dict[attribute_name][test_id]
dummy_val = dummy_attributes[attribute_name]
if isinstance(attribute_val, torch.Tensor):
assert torch.all(torch.eq(attribute_val, dummy_val))
else:
assert attribute_val == dummy_val
@pytest.mark.parametrize(
"checkpoint_restore_config",
[
TemporaryDirectory().name,
os.path.join(TemporaryDirectory().name, "attributes.pkl"),
],
indirect=True,
)
def test_restore_attribute(checkpoint_restore_config, checkpoint_save_config,
dummy_attributes):
"""
Test attributes restored by checkpointer.
Args:
checkpoint_save_config (dict): Dictionary with the config to save attributes
checkpoint_restore_config (dict): Dictionary with the config to save attributes
dummy_attributes (dict): Dictionary with value for the attributes
Return:
None
def write_pandas(
conn: 'SnowflakeConnection',
df: 'pandas.DataFrame',
table_name: str,
database: Optional[str] = None,
schema: Optional[str] = None,
chunk_size: Optional[int] = None,
compression: str = 'gzip',
on_error: str = 'abort_statement',
parallel: int = 4,
quote_identifiers: bool = True
) -> Tuple[bool, int, int,
Sequence[Tuple[str, str, int, int, int, int, Optional[str],
Optional[int], Optional[int], Optional[str]]]]:
"""Allows users to most efficiently write back a pandas DataFrame to Snowflake.
It works by dumping the DataFrame into Parquet files, uploading them and finally copying their data into the table.
Returns whether all files were ingested correctly, number of chunks uploaded, and number of rows ingested
with all of the COPY INTO command's output for debugging purposes.
Example usage:
import pandas
from snowflake.connector.pandas_tools import write_pandas
df = pandas.DataFrame([('Mark', 10), ('Luke', 20)], columns=['name', 'balance'])
success, nchunks, nrows, _ = write_pandas(cnx, df, 'customers')
Args:
conn: Connection to be used to communicate with Snowflake.
df: Dataframe we'd like to write back.
table_name: Table name where we want to insert into.
database: Database schema and table is in, if not provided the default one will be used (Default value = None).
schema: Schema table is in, if not provided the default one will be used (Default value = None).
chunk_size: Number of elements to be inserted once, if not provided all elements will be dumped once
(Default value = None).
compression: The compression used on the Parquet files, can only be gzip, or snappy. Gzip gives supposedly a
better compression, while snappy is faster. Use whichever is more appropriate (Default value = 'gzip').
on_error: Action to take when COPY INTO statements fail, default follows documentation at:
https://docs.snowflake.com/en/sql-reference/sql/copy-into-table.html#copy-options-copyoptions
(Default value = 'abort_statement').
parallel: Number of threads to be used when uploading chunks, default follows documentation at:
https://docs.snowflake.com/en/sql-reference/sql/put.html#optional-parameters (Default value = 4).
quote_identifiers: By default, identifiers, specifically database, schema, table and column names
(from df.columns) will be quoted. If set to False, identifiers are passed on to Snowflake without quoting.
I.e. identifiers will be coerced to uppercase by Snowflake. (Default value = True)
Returns:
Returns the COPY INTO command's results to verify ingestion in the form of a tuple of whether all chunks were
ingested correctly, # of chunks, # of ingested rows, and ingest's output.
"""
if database is not None and schema is None:
raise ProgrammingError(
"Schema has to be provided to write_pandas when a database is provided"
)
# This dictionary maps the compression algorithm to Snowflake put copy into command type
# https://docs.snowflake.com/en/sql-reference/sql/copy-into-table.html#type-parquet
compression_map = {'gzip': 'auto', 'snappy': 'snappy'}
if compression not in compression_map.keys():
raise ProgrammingError(
"Invalid compression '{}', only acceptable values are: {}".format(
compression, compression_map.keys()))
if quote_identifiers:
location = ((('"' + database + '".') if database else '') +
(('"' + schema + '".') if schema else '') +
('"' + table_name + '"'))
else:
location = ((database + '.' if database else '') +
(schema + '.' if schema else '') + (table_name))
if chunk_size is None:
chunk_size = len(df)
cursor = conn.cursor()
stage_name = None # Forward declaration
while True:
try:
stage_name = ''.join(
random.choice(string.ascii_lowercase) for _ in range(5))
create_stage_sql = (
'create temporary stage /* Python:snowflake.connector.pandas_tools.write_pandas() */ '
'"{stage_name}"').format(stage_name=stage_name)
logger.debug("creating stage with '{}'".format(create_stage_sql))
cursor.execute(create_stage_sql, _is_internal=True).fetchall()
break
except ProgrammingError as pe:
if pe.msg.endswith('already exists.'):
continue
raise
with TemporaryDirectory() as tmp_folder:
for i, chunk in chunk_helper(df, chunk_size):
chunk_path = os.path.join(tmp_folder, 'file{}.txt'.format(i))
# Dump chunk into parquet file
chunk.to_parquet(chunk_path, compression=compression)
# Upload parquet file
upload_sql = (
'PUT /* Python:snowflake.connector.pandas_tools.write_pandas() */ '
'\'file://{path}\' @"{stage_name}" PARALLEL={parallel}'
).format(path=chunk_path.replace('\\',
'\\\\').replace('\'', '\\\''),
stage_name=stage_name,
parallel=parallel)
logger.debug("uploading files with '{}'".format(upload_sql))
cursor.execute(upload_sql, _is_internal=True)
# Remove chunk file
os.remove(chunk_path)
if quote_identifiers:
columns = '"' + '","'.join(list(df.columns)) + '"'
else:
columns = ','.join(list(df.columns))
# in Snowflake, all parquet data is stored in a single column, $1, so we must select columns explicitly
# see (https://docs.snowflake.com/en/user-guide/script-data-load-transform-parquet.html)
parquet_columns = '$1:' + ',$1:'.join(df.columns)
copy_into_sql = (
'COPY INTO {location} /* Python:snowflake.connector.pandas_tools.write_pandas() */ '
'({columns}) '
'FROM (SELECT {parquet_columns} FROM @"{stage_name}") '
'FILE_FORMAT=(TYPE=PARQUET COMPRESSION={compression}) '
'PURGE=TRUE ON_ERROR={on_error}').format(
location=location,
columns=columns,
parquet_columns=parquet_columns,
stage_name=stage_name,
compression=compression_map[compression],
on_error=on_error)
logger.debug("copying into with '{}'".format(copy_into_sql))
copy_results = cursor.execute(copy_into_sql, _is_internal=True).fetchall()
cursor.close()
return (all(e[1] == 'LOADED' for e in copy_results), len(copy_results),
sum(e[3] for e in copy_results), copy_results)
def diff_it(role_dir: Path, rev1: str, rev2: str) -> Any:
g = Git(role_dir)
g.checkout(rev1)
role_v1 = Role.load(role_tmp_dir, role_base_dir)
g.checkout(rev2)
role_v2 = Role.load(role_tmp_dir, role_base_dir)
return [d for d in role_v1.diff(role_v2) if d]
if sys.argv[2] == 'all':
revs = [t.name for t in Repo(role_dir).tags][::-1]
else:
revs = sys.argv[2:]
with TemporaryDirectory() as tmpd:
role_tmp_dir = Path(tmpd) / role_name
copytree(role_dir, role_tmp_dir, symlinks=True)
for rev2, rev1 in zip(revs, revs[1:]):
header = f'{rev1} -> {rev2}'
print(header)
print('=' * len(header))
print()
try:
diffs = diff_it(role_tmp_dir, rev1, rev2)
for diff in sorted(diffs, key=lambda d: d.object_id):
print(diff)
print()
except AnsibleError as exc:
print(exc)
print()