class JobRecoveryAfterHandledIntegerationTestCase(
integration_util.IntegrationTestCase):
framework_tool_and_types = True
def setUp(self):
super(JobRecoveryAfterHandledIntegerationTestCase, self).setUp()
self.dataset_populator = DatasetPopulator(self.galaxy_interactor)
@classmethod
def handle_galaxy_config_kwds(cls, config):
config["job_config_file"] = DELAY_JOB_CONFIG_FILE
def handle_reconfigure_galaxy_config_kwds(self, config):
config["job_config_file"] = SIMPLE_JOB_CONFIG_FILE
def test_recovery(self):
history_id = self.dataset_populator.new_history()
self.dataset_populator.run_tool(
"exit_code_oom",
{},
history_id,
assert_ok=False,
).json()
self.restart(
handle_reconfig=self.handle_reconfigure_galaxy_config_kwds)
self.dataset_populator = DatasetPopulator(self.galaxy_interactor)
self.dataset_populator.wait_for_history(history_id, assert_ok=True)
class JobRecoveryAfterHandledIntegerationTestCase(integration_util.IntegrationTestCase):
framework_tool_and_types = True
def setUp(self):
super(JobRecoveryAfterHandledIntegerationTestCase, self).setUp()
self.dataset_populator = DatasetPopulator(self.galaxy_interactor)
@classmethod
def handle_galaxy_config_kwds(cls, config):
config["job_config_file"] = DELAY_JOB_CONFIG_FILE
def handle_reconfigure_galaxy_config_kwds(self, config):
config["job_config_file"] = SIMPLE_JOB_CONFIG_FILE
def test_recovery(self):
history_id = self.dataset_populator.new_history()
self.dataset_populator.run_tool(
"exit_code_oom",
{},
history_id,
assert_ok=False,
).json()
self.restart(handle_reconfig=self.handle_reconfigure_galaxy_config_kwds)
self.dataset_populator = DatasetPopulator(self.galaxy_interactor)
self.dataset_populator.wait_for_history(history_id, assert_ok=True)
class FailJobWhenToolUnavailableTestCase(integration_util.IntegrationTestCase):
require_admin_user = True
def setUp(self):
super(FailJobWhenToolUnavailableTestCase, self).setUp()
self.dataset_populator = DatasetPopulator(self.galaxy_interactor)
self.workflow_populator = WorkflowPopulator(self.galaxy_interactor)
self.history_id = self.dataset_populator.new_history()
@classmethod
def handle_galaxy_config_kwds(
cls,
config,
):
# config["jobs_directory"] = cls.jobs_directory
# Disable tool dependency resolution.
config["tool_dependency_dir"] = "none"
def test_fail_job_when_tool_unavailable(self):
self.workflow_populator.run_workflow("""
class: GalaxyWorkflow
steps:
- label: sleep
run:
class: GalaxyTool
command: sleep 20s && echo 'hello world 2' > '$output1'
outputs:
output1:
format: txt
- tool_id: cat1
state:
input1:
$link: sleep#output1
queries:
input2:
$link: sleep#output1
""",
history_id=self.history_id,
assert_ok=False,
wait=False)
# Wait until workflow is fully scheduled, otherwise can't test effect of removing tool from queued job
time.sleep(10)
self._app.toolbox.remove_tool_by_id('cat1')
self.dataset_populator.wait_for_history(self.history_id,
assert_ok=False)
state_details = self.galaxy_interactor.get(
'histories/%s' % self.history_id).json()['state_details']
assert state_details['running'] == 0
assert state_details['ok'] == 1
assert state_details['error'] == 1
failed_hda = self.dataset_populator.get_history_dataset_details(
history_id=self.history_id, assert_ok=False, details=True)
assert failed_hda['state'] == 'error'
job = self.galaxy_interactor.get("jobs/%s" %
failed_hda['creating_job']).json()
assert job['state'] == 'error'
class DockerizedJobsIntegrationTestCase(integration_util.IntegrationTestCase, RunsEnvironmentJobs):
framework_tool_and_types = True
@classmethod
def handle_galaxy_config_kwds(cls, config):
cls.jobs_directory = tempfile.mkdtemp()
config["jobs_directory"] = cls.jobs_directory
config["job_config_file"] = DOCKERIZED_JOB_CONFIG_FILE
# Disable tool dependency resolution.
config["tool_dependency_dir"] = "none"
config["enable_beta_mulled_containers"] = "true"
def setUp(self):
super(DockerizedJobsIntegrationTestCase, self).setUp()
self.dataset_populator = DatasetPopulator(self.galaxy_interactor)
self.history_id = self.dataset_populator.new_history()
def test_explicit(self):
self.dataset_populator.run_tool("mulled_example_explicit", {}, self.history_id)
self.dataset_populator.wait_for_history(self.history_id, assert_ok=True)
output = self.dataset_populator.get_history_dataset_content(self.history_id)
assert "0.7.15-r1140" in output
def test_mulled_simple(self):
self.dataset_populator.run_tool("mulled_example_simple", {}, self.history_id)
self.dataset_populator.wait_for_history(self.history_id, assert_ok=True)
output = self.dataset_populator.get_history_dataset_content(self.history_id)
assert "0.7.15-r1140" in output
def test_docker_job_enviornment(self):
job_env = self._run_and_get_environment_properties("job_environment_default")
euid = os.geteuid()
egid = os.getgid()
assert job_env.user_id == str(euid), job_env.user_id
assert job_env.group_id == str(egid), job_env.group_id
assert job_env.pwd.startswith(self.jobs_directory)
assert job_env.pwd.endswith("/working")
assert job_env.home.startswith(self.jobs_directory)
assert job_env.home.endswith("/home")
def test_docker_job_environment_legacy(self):
job_env = self._run_and_get_environment_properties("job_environment_default_legacy")
euid = os.geteuid()
egid = os.getgid()
assert job_env.user_id == str(euid), job_env.user_id
assert job_env.group_id == str(egid), job_env.group_id
assert job_env.pwd.startswith(self.jobs_directory)
assert job_env.pwd.endswith("/working")
# Should we change env_pass_through to just always include TMP and HOME for docker?
# I'm not sure, if yes this would change.
assert job_env.home == "/", job_env.home
class ObjectStoreJobsIntegrationTestCase(integration_util.IntegrationTestCase):
framework_tool_and_types = True
@classmethod
def handle_galaxy_config_kwds(cls, config):
temp_directory = cls._test_driver.mkdtemp()
cls.object_stores_parent = temp_directory
for disk_store_file_name in ["files1", "files2", "files3"]:
disk_store_path = os.path.join(temp_directory, disk_store_file_name)
os.makedirs(disk_store_path)
setattr(cls, "%s_path" % disk_store_file_name, disk_store_path)
config_path = os.path.join(temp_directory, "object_store_conf.xml")
with open(config_path, "w") as f:
f.write(DISTRIBUTED_OBJECT_STORE_CONFIG_TEMPLATE.safe_substitute({"temp_directory": temp_directory}))
config["object_store_config_file"] = config_path
def setUp(self):
super(ObjectStoreJobsIntegrationTestCase, self).setUp()
self.dataset_populator = DatasetPopulator(self.galaxy_interactor)
def test_tool_simple_constructs(self):
with self.dataset_populator.test_history() as history_id:
hda1 = self.dataset_populator.new_dataset(history_id, content="1 2 3")
create_10_inputs = {
"input1": {"src": "hda", "id": hda1["id"]},
"input2": {"src": "hda", "id": hda1["id"]},
}
self.dataset_populator.run_tool(
"create_10",
create_10_inputs,
history_id,
assert_ok=True,
)
self.dataset_populator.wait_for_history(history_id)
files_1_count = _files_count(self.files1_path)
files_2_count = _files_count(self.files2_path)
files_3_count = _files_count(self.files3_path)
# Ensure no files written to the secondary/inactive hierarchical disk store.
assert files_3_count == 0
# Ensure the 10 inputs were written to one of the distributed object store's disk
# stores (it will have either 10 or 11 depeending on whether the input was also
# written there. The other disk store may or may not have the input file so should
# have at most one file.
assert (files_1_count >= 10) or (files_2_count >= 10)
assert (files_1_count <= 1) or (files_2_count <= 1)
# Other sanity checks on the test - just make sure the test was setup as intended
# and not actually testing object store behavior.
assert (files_1_count <= 11) and (files_2_count <= 11)
assert (files_1_count >= 0) and (files_2_count >= 0)
def test_legacy_r_mapping(self):
"""
"""
tool_id = "legacy_R"
dataset_populator = DatasetPopulator(self.galaxy_interactor)
history_id = dataset_populator.new_history()
endpoint = "tools/%s/install_dependencies" % tool_id
data = {'id': tool_id}
create_response = self._post(endpoint, data=data, admin=True)
self._assert_status_code_is(create_response, 200)
payload = dataset_populator.run_tool_payload(
tool_id=tool_id,
inputs={},
history_id=history_id,
)
create_response = self._post("tools", data=payload)
self._assert_status_code_is(create_response, 200)
dataset_populator.wait_for_history(history_id, assert_ok=True)
def test_legacy_r_mapping(self):
"""
"""
tool_id = "legacy_R"
dataset_populator = DatasetPopulator(self.galaxy_interactor)
history_id = dataset_populator.new_history()
endpoint = "tools/%s/install_dependencies" % tool_id
data = {'id': tool_id}
create_response = self._post(endpoint, data=data, admin=True)
self._assert_status_code_is(create_response, 200)
payload = dataset_populator.run_tool_payload(
tool_id=tool_id,
inputs={},
history_id=history_id,
)
create_response = self._post("tools", data=payload)
self._assert_status_code_is(create_response, 200)
dataset_populator.wait_for_history(history_id, assert_ok=True)
class MaximumWorkflowJobsPerSchedulingIterationTestCase(integration_util.IntegrationTestCase):
framework_tool_and_types = True
def setUp(self):
super(MaximumWorkflowJobsPerSchedulingIterationTestCase, self).setUp()
self.dataset_populator = DatasetPopulator(self.galaxy_interactor)
self.workflow_populator = WorkflowPopulator(self.galaxy_interactor)
self.dataset_collection_populator = DatasetCollectionPopulator(self.galaxy_interactor)
@classmethod
def handle_galaxy_config_kwds(cls, config):
config["maximum_workflow_jobs_per_scheduling_iteration"] = 1
def do_test(self):
workflow_id = self.workflow_populator.upload_yaml_workflow("""
class: GalaxyWorkflow
steps:
- type: input_collection
- tool_id: collection_creates_pair
state:
input1:
$link: 0
- tool_id: collection_paired_test
state:
f1:
$link: 1#paired_output
- tool_id: cat_list
state:
input1:
$link: 2#out1
""")
with self.dataset_populator.test_history() as history_id:
hdca1 = self.dataset_collection_populator.create_list_in_history(history_id, contents=["a\nb\nc\nd\n", "e\nf\ng\nh\n"]).json()
self.dataset_populator.wait_for_history(history_id, assert_ok=True)
inputs = {
'0': {"src": "hdca", "id": hdca1["id"]},
}
invocation_id = self.workflow_populator.invoke_workflow(history_id, workflow_id, inputs)
self.workflow_populator.wait_for_workflow(history_id, workflow_id, invocation_id)
self.dataset_populator.wait_for_history(history_id, assert_ok=True)
self.assertEqual("a\nc\nb\nd\ne\ng\nf\nh\n", self.dataset_populator.get_history_dataset_content(history_id, hid=0))
class ToolsUploadTestCase(api.ApiTestCase):
def setUp(self):
super(ToolsUploadTestCase, self).setUp()
self.dataset_populator = DatasetPopulator(self.galaxy_interactor)
def test_upload1_paste(self):
with self.dataset_populator.test_history() as history_id:
payload = self.dataset_populator.upload_payload(
history_id, 'Hello World')
create_response = self._post("tools", data=payload)
self._assert_has_keys(create_response.json(), 'outputs')
def test_upload1_paste_bad_datatype(self):
# Check that you get a nice message if you upload an incorrect datatype
with self.dataset_populator.test_history() as history_id:
file_type = "johnsawesomebutfakedatatype"
payload = self.dataset_populator.upload_payload(
history_id, 'Hello World', file_type=file_type)
create = self._post("tools", data=payload).json()
self._assert_has_keys(create, 'err_msg')
assert file_type in create['err_msg']
def test_upload_posix_newline_fixes(self):
windows_content = ONE_TO_SIX_ON_WINDOWS
result_content = self._upload_and_get_content(windows_content)
self.assertEquals(result_content, ONE_TO_SIX_WITH_TABS)
def test_upload_disable_posix_fix(self):
windows_content = ONE_TO_SIX_ON_WINDOWS
result_content = self._upload_and_get_content(windows_content,
to_posix_lines=None)
self.assertEquals(result_content, windows_content)
def test_upload_tab_to_space(self):
table = ONE_TO_SIX_WITH_SPACES
result_content = self._upload_and_get_content(table,
space_to_tab="Yes")
self.assertEquals(result_content, ONE_TO_SIX_WITH_TABS)
def test_upload_tab_to_space_off_by_default(self):
table = ONE_TO_SIX_WITH_SPACES
result_content = self._upload_and_get_content(table)
self.assertEquals(result_content, table)
@skip_without_datatype("rdata")
def test_rdata_not_decompressed(self):
# Prevent regression of https://github.com/galaxyproject/galaxy/issues/753
rdata_path = TestDataResolver().get_filename("1.RData")
rdata_metadata = self._upload_and_get_details(open(rdata_path, "rb"),
file_type="auto")
self.assertEquals(rdata_metadata["file_ext"], "rdata")
@skip_without_datatype("velvet")
def test_composite_datatype(self):
with self.dataset_populator.test_history() as history_id:
dataset = self._velvet_upload(history_id,
extra_inputs={
"files_1|url_paste":
"roadmaps content",
"files_1|type": "upload_dataset",
"files_2|url_paste":
"log content",
"files_2|type": "upload_dataset",
})
roadmaps_content = self._get_roadmaps_content(history_id, dataset)
assert roadmaps_content.strip(
) == "roadmaps content", roadmaps_content
@skip_without_datatype("velvet")
def test_composite_datatype_space_to_tab(self):
# Like previous test but set one upload with space_to_tab to True to
# verify that works.
with self.dataset_populator.test_history() as history_id:
dataset = self._velvet_upload(history_id,
extra_inputs={
"files_1|url_paste":
"roadmaps content",
"files_1|type": "upload_dataset",
"files_1|space_to_tab": "Yes",
"files_2|url_paste":
"log content",
"files_2|type": "upload_dataset",
})
roadmaps_content = self._get_roadmaps_content(history_id, dataset)
assert roadmaps_content.strip(
) == "roadmaps\tcontent", roadmaps_content
@skip_without_datatype("velvet")
def test_composite_datatype_posix_lines(self):
# Like previous test but set one upload with space_to_tab to True to
# verify that works.
with self.dataset_populator.test_history() as history_id:
dataset = self._velvet_upload(history_id,
extra_inputs={
"files_1|url_paste":
"roadmaps\rcontent",
"files_1|type": "upload_dataset",
"files_1|space_to_tab": "Yes",
"files_2|url_paste":
"log\rcontent",
"files_2|type": "upload_dataset",
})
roadmaps_content = self._get_roadmaps_content(history_id, dataset)
assert roadmaps_content.strip(
) == "roadmaps\ncontent", roadmaps_content
def test_upload_dbkey(self):
with self.dataset_populator.test_history() as history_id:
payload = self.dataset_populator.upload_payload(history_id,
"Test123",
dbkey="hg19")
run_response = self.dataset_populator.tools_post(payload)
self.dataset_populator.wait_for_tool_run(history_id, run_response)
datasets = run_response.json()["outputs"]
assert datasets[0].get("genome_build") == "hg19", datasets[0]
def test_upload_multiple_files_1(self):
with self.dataset_populator.test_history() as history_id:
payload = self.dataset_populator.upload_payload(
history_id,
"Test123",
dbkey="hg19",
extra_inputs={
"files_1|url_paste": "SecondOutputContent",
"files_1|NAME": "SecondOutputName",
"files_1|file_type": "tabular",
"files_1|dbkey": "hg18",
"file_count": "2",
})
run_response = self.dataset_populator.tools_post(payload)
self.dataset_populator.wait_for_tool_run(history_id, run_response)
datasets = run_response.json()["outputs"]
assert len(datasets) == 2, datasets
content = self.dataset_populator.get_history_dataset_content(
history_id, dataset=datasets[0])
assert content.strip() == "Test123"
assert datasets[0]["file_ext"] == "txt"
assert datasets[0]["genome_build"] == "hg19", datasets
content = self.dataset_populator.get_history_dataset_content(
history_id, dataset=datasets[1])
assert content.strip() == "SecondOutputContent"
assert datasets[1]["file_ext"] == "tabular"
assert datasets[1]["genome_build"] == "hg18", datasets
def test_upload_multiple_files_2(self):
with self.dataset_populator.test_history() as history_id:
payload = self.dataset_populator.upload_payload(
history_id,
"Test123",
file_type="tabular",
dbkey="hg19",
extra_inputs={
"files_1|url_paste": "SecondOutputContent",
"files_1|NAME": "SecondOutputName",
"files_1|file_type": "txt",
"files_1|dbkey": "hg18",
"file_count": "2",
})
run_response = self.dataset_populator.tools_post(payload)
self.dataset_populator.wait_for_tool_run(history_id, run_response)
datasets = run_response.json()["outputs"]
assert len(datasets) == 2, datasets
content = self.dataset_populator.get_history_dataset_content(
history_id, dataset=datasets[0])
assert content.strip() == "Test123"
assert datasets[0]["file_ext"] == "tabular", datasets
assert datasets[0]["genome_build"] == "hg19", datasets
content = self.dataset_populator.get_history_dataset_content(
history_id, dataset=datasets[1])
assert content.strip() == "SecondOutputContent"
assert datasets[1]["file_ext"] == "txt"
assert datasets[1]["genome_build"] == "hg18", datasets
def test_upload_multiple_files_3(self):
with self.dataset_populator.test_history() as history_id:
payload = self.dataset_populator.upload_payload(
history_id,
"Test123",
file_type="tabular",
dbkey="hg19",
extra_inputs={
"files_0|file_type": "txt",
"files_0|dbkey": "hg18",
"files_1|url_paste": "SecondOutputContent",
"files_1|NAME": "SecondOutputName",
"files_1|file_type": "txt",
"files_1|dbkey": "hg18",
"file_count": "2",
})
run_response = self.dataset_populator.tools_post(payload)
self.dataset_populator.wait_for_tool_run(history_id, run_response)
datasets = run_response.json()["outputs"]
assert len(datasets) == 2, datasets
content = self.dataset_populator.get_history_dataset_content(
history_id, dataset=datasets[0])
assert content.strip() == "Test123"
assert datasets[0]["file_ext"] == "txt", datasets
assert datasets[0]["genome_build"] == "hg18", datasets
content = self.dataset_populator.get_history_dataset_content(
history_id, dataset=datasets[1])
assert content.strip() == "SecondOutputContent"
assert datasets[1]["file_ext"] == "txt"
assert datasets[1]["genome_build"] == "hg18", datasets
def test_upload_multiple_files_no_dbkey(self):
with self.dataset_populator.test_history() as history_id:
payload = self.dataset_populator.upload_payload(
history_id,
"Test123",
file_type="tabular",
dbkey=None,
extra_inputs={
"files_0|file_type": "txt",
"files_1|url_paste": "SecondOutputContent",
"files_1|NAME": "SecondOutputName",
"files_1|file_type": "txt",
"file_count": "2",
})
run_response = self.dataset_populator.tools_post(payload)
self.dataset_populator.wait_for_tool_run(history_id, run_response)
datasets = run_response.json()["outputs"]
assert len(datasets) == 2, datasets
content = self.dataset_populator.get_history_dataset_content(
history_id, dataset=datasets[0])
assert content.strip() == "Test123"
assert datasets[0]["file_ext"] == "txt", datasets
assert datasets[0]["genome_build"] == "?", datasets
content = self.dataset_populator.get_history_dataset_content(
history_id, dataset=datasets[1])
assert content.strip() == "SecondOutputContent"
assert datasets[1]["file_ext"] == "txt"
assert datasets[1]["genome_build"] == "?", datasets
def test_upload_multiple_files_space_to_tab(self):
with self.dataset_populator.test_history() as history_id:
payload = self.dataset_populator.upload_payload(
history_id,
content=ONE_TO_SIX_WITH_SPACES,
file_type="tabular",
dbkey="hg19",
extra_inputs={
"files_0|file_type": "txt",
"files_0|space_to_tab": "Yes",
"files_1|url_paste": ONE_TO_SIX_WITH_SPACES,
"files_1|NAME": "SecondOutputName",
"files_1|file_type": "txt",
"files_2|url_paste": ONE_TO_SIX_WITH_SPACES,
"files_2|NAME": "ThirdOutputName",
"files_2|file_type": "txt",
"files_2|space_to_tab": "Yes",
"file_count": "3",
})
run_response = self.dataset_populator.tools_post(payload)
self.dataset_populator.wait_for_tool_run(history_id, run_response)
datasets = run_response.json()["outputs"]
assert len(datasets) == 3, datasets
content = self.dataset_populator.get_history_dataset_content(
history_id, dataset=datasets[0])
assert content == ONE_TO_SIX_WITH_TABS
content = self.dataset_populator.get_history_dataset_content(
history_id, dataset=datasets[1])
assert content == ONE_TO_SIX_WITH_SPACES
content = self.dataset_populator.get_history_dataset_content(
history_id, dataset=datasets[2])
assert content == ONE_TO_SIX_WITH_TABS
def test_multiple_files_posix_lines(self):
with self.dataset_populator.test_history() as history_id:
payload = self.dataset_populator.upload_payload(
history_id,
content=ONE_TO_SIX_ON_WINDOWS,
file_type="tabular",
dbkey="hg19",
extra_inputs={
"files_0|file_type": "txt",
"files_0|to_posix_lines": "Yes",
"files_1|url_paste": ONE_TO_SIX_ON_WINDOWS,
"files_1|NAME": "SecondOutputName",
"files_1|file_type": "txt",
"files_1|to_posix_lines": None,
"files_2|url_paste": ONE_TO_SIX_ON_WINDOWS,
"files_2|NAME": "ThirdOutputName",
"files_2|file_type": "txt",
"file_count": "3",
})
run_response = self.dataset_populator.tools_post(payload)
self.dataset_populator.wait_for_tool_run(history_id, run_response)
datasets = run_response.json()["outputs"]
assert len(datasets) == 3, datasets
content = self.dataset_populator.get_history_dataset_content(
history_id, dataset=datasets[0])
assert content == ONE_TO_SIX_WITH_TABS
content = self.dataset_populator.get_history_dataset_content(
history_id, dataset=datasets[1])
assert content == ONE_TO_SIX_ON_WINDOWS
content = self.dataset_populator.get_history_dataset_content(
history_id, dataset=datasets[2])
assert content == ONE_TO_SIX_WITH_TABS
def _velvet_upload(self, history_id, extra_inputs):
payload = self.dataset_populator.upload_payload(
history_id,
"sequences content",
file_type="velvet",
extra_inputs=extra_inputs,
)
run_response = self.dataset_populator.tools_post(payload)
self.dataset_populator.wait_for_tool_run(history_id, run_response)
datasets = run_response.json()["outputs"]
assert len(datasets) == 1
dataset = datasets[0]
return dataset
def _get_roadmaps_content(self, history_id, dataset):
roadmaps_content = self.dataset_populator.get_history_dataset_content(
history_id, dataset=dataset, filename="Roadmaps")
return roadmaps_content
def _upload_and_get_content(self, content, **upload_kwds):
history_id, new_dataset = self._upload(content, **upload_kwds)
return self.dataset_populator.get_history_dataset_content(
history_id, dataset=new_dataset)
def _upload_and_get_details(self, content, **upload_kwds):
history_id, new_dataset = self._upload(content, **upload_kwds)
return self.dataset_populator.get_history_dataset_details(
history_id, dataset=new_dataset)
def _upload(self, content, **upload_kwds):
history_id = self.dataset_populator.new_history()
new_dataset = self.dataset_populator.new_dataset(history_id,
content=content,
**upload_kwds)
self.dataset_populator.wait_for_history(history_id, assert_ok=True)
return history_id, new_dataset
class ToolsUploadTestCase(api.ApiTestCase):
def setUp(self):
super(ToolsUploadTestCase, self).setUp()
self.dataset_populator = DatasetPopulator(self.galaxy_interactor)
def test_upload1_paste(self):
with self.dataset_populator.test_history() as history_id:
payload = self.dataset_populator.upload_payload(history_id, 'Hello World')
create_response = self._post("tools", data=payload)
self._assert_has_keys(create_response.json(), 'outputs')
def test_upload1_paste_bad_datatype(self):
# Check that you get a nice message if you upload an incorrect datatype
with self.dataset_populator.test_history() as history_id:
file_type = "johnsawesomebutfakedatatype"
payload = self.dataset_populator.upload_payload(history_id, 'Hello World', file_type=file_type)
create = self._post("tools", data=payload).json()
self._assert_has_keys(create, 'err_msg')
assert file_type in create['err_msg']
def test_upload_posix_newline_fixes(self):
windows_content = ONE_TO_SIX_ON_WINDOWS
result_content = self._upload_and_get_content(windows_content)
self.assertEquals(result_content, ONE_TO_SIX_WITH_TABS)
def test_upload_disable_posix_fix(self):
windows_content = ONE_TO_SIX_ON_WINDOWS
result_content = self._upload_and_get_content(windows_content, to_posix_lines=None)
self.assertEquals(result_content, windows_content)
def test_upload_tab_to_space(self):
table = ONE_TO_SIX_WITH_SPACES
result_content = self._upload_and_get_content(table, space_to_tab="Yes")
self.assertEquals(result_content, ONE_TO_SIX_WITH_TABS)
def test_upload_tab_to_space_off_by_default(self):
table = ONE_TO_SIX_WITH_SPACES
result_content = self._upload_and_get_content(table)
self.assertEquals(result_content, table)
@skip_without_datatype("rdata")
def test_rdata_not_decompressed(self):
# Prevent regression of https://github.com/galaxyproject/galaxy/issues/753
rdata_path = TestDataResolver().get_filename("1.RData")
rdata_metadata = self._upload_and_get_details(open(rdata_path, "rb"), file_type="auto")
self.assertEquals(rdata_metadata["file_ext"], "rdata")
@skip_without_datatype("velvet")
def test_composite_datatype(self):
with self.dataset_populator.test_history() as history_id:
dataset = self._velvet_upload(history_id, extra_inputs={
"files_1|url_paste": "roadmaps content",
"files_1|type": "upload_dataset",
"files_2|url_paste": "log content",
"files_2|type": "upload_dataset",
})
roadmaps_content = self._get_roadmaps_content(history_id, dataset)
assert roadmaps_content.strip() == "roadmaps content", roadmaps_content
@skip_without_datatype("velvet")
def test_composite_datatype_space_to_tab(self):
# Like previous test but set one upload with space_to_tab to True to
# verify that works.
with self.dataset_populator.test_history() as history_id:
dataset = self._velvet_upload(history_id, extra_inputs={
"files_1|url_paste": "roadmaps content",
"files_1|type": "upload_dataset",
"files_1|space_to_tab": "Yes",
"files_2|url_paste": "log content",
"files_2|type": "upload_dataset",
})
roadmaps_content = self._get_roadmaps_content(history_id, dataset)
assert roadmaps_content.strip() == "roadmaps\tcontent", roadmaps_content
@skip_without_datatype("velvet")
def test_composite_datatype_posix_lines(self):
# Like previous test but set one upload with space_to_tab to True to
# verify that works.
with self.dataset_populator.test_history() as history_id:
dataset = self._velvet_upload(history_id, extra_inputs={
"files_1|url_paste": "roadmaps\rcontent",
"files_1|type": "upload_dataset",
"files_1|space_to_tab": "Yes",
"files_2|url_paste": "log\rcontent",
"files_2|type": "upload_dataset",
})
roadmaps_content = self._get_roadmaps_content(history_id, dataset)
assert roadmaps_content.strip() == "roadmaps\ncontent", roadmaps_content
def test_upload_dbkey(self):
with self.dataset_populator.test_history() as history_id:
payload = self.dataset_populator.upload_payload(history_id, "Test123", dbkey="hg19")
run_response = self.dataset_populator.tools_post(payload)
self.dataset_populator.wait_for_tool_run(history_id, run_response)
datasets = run_response.json()["outputs"]
assert datasets[0].get("genome_build") == "hg19", datasets[0]
def test_upload_multiple_files_1(self):
with self.dataset_populator.test_history() as history_id:
payload = self.dataset_populator.upload_payload(history_id, "Test123",
dbkey="hg19",
extra_inputs={
"files_1|url_paste": "SecondOutputContent",
"files_1|NAME": "SecondOutputName",
"files_1|file_type": "tabular",
"files_1|dbkey": "hg18",
"file_count": "2",
}
)
run_response = self.dataset_populator.tools_post(payload)
self.dataset_populator.wait_for_tool_run(history_id, run_response)
datasets = run_response.json()["outputs"]
assert len(datasets) == 2, datasets
content = self.dataset_populator.get_history_dataset_content(history_id, dataset=datasets[0])
assert content.strip() == "Test123"
assert datasets[0]["file_ext"] == "txt"
assert datasets[0]["genome_build"] == "hg19", datasets
content = self.dataset_populator.get_history_dataset_content(history_id, dataset=datasets[1])
assert content.strip() == "SecondOutputContent"
assert datasets[1]["file_ext"] == "tabular"
assert datasets[1]["genome_build"] == "hg18", datasets
def test_upload_multiple_files_2(self):
with self.dataset_populator.test_history() as history_id:
payload = self.dataset_populator.upload_payload(history_id, "Test123",
file_type="tabular",
dbkey="hg19",
extra_inputs={
"files_1|url_paste": "SecondOutputContent",
"files_1|NAME": "SecondOutputName",
"files_1|file_type": "txt",
"files_1|dbkey": "hg18",
"file_count": "2",
}
)
run_response = self.dataset_populator.tools_post(payload)
self.dataset_populator.wait_for_tool_run(history_id, run_response)
datasets = run_response.json()["outputs"]
assert len(datasets) == 2, datasets
content = self.dataset_populator.get_history_dataset_content(history_id, dataset=datasets[0])
assert content.strip() == "Test123"
assert datasets[0]["file_ext"] == "tabular", datasets
assert datasets[0]["genome_build"] == "hg19", datasets
content = self.dataset_populator.get_history_dataset_content(history_id, dataset=datasets[1])
assert content.strip() == "SecondOutputContent"
assert datasets[1]["file_ext"] == "txt"
assert datasets[1]["genome_build"] == "hg18", datasets
def test_upload_multiple_files_3(self):
with self.dataset_populator.test_history() as history_id:
payload = self.dataset_populator.upload_payload(history_id, "Test123",
file_type="tabular",
dbkey="hg19",
extra_inputs={
"files_0|file_type": "txt",
"files_0|dbkey": "hg18",
"files_1|url_paste": "SecondOutputContent",
"files_1|NAME": "SecondOutputName",
"files_1|file_type": "txt",
"files_1|dbkey": "hg18",
"file_count": "2",
}
)
run_response = self.dataset_populator.tools_post(payload)
self.dataset_populator.wait_for_tool_run(history_id, run_response)
datasets = run_response.json()["outputs"]
assert len(datasets) == 2, datasets
content = self.dataset_populator.get_history_dataset_content(history_id, dataset=datasets[0])
assert content.strip() == "Test123"
assert datasets[0]["file_ext"] == "txt", datasets
assert datasets[0]["genome_build"] == "hg18", datasets
content = self.dataset_populator.get_history_dataset_content(history_id, dataset=datasets[1])
assert content.strip() == "SecondOutputContent"
assert datasets[1]["file_ext"] == "txt"
assert datasets[1]["genome_build"] == "hg18", datasets
def test_upload_multiple_files_no_dbkey(self):
with self.dataset_populator.test_history() as history_id:
payload = self.dataset_populator.upload_payload(history_id, "Test123",
file_type="tabular",
dbkey=None,
extra_inputs={
"files_0|file_type": "txt",
"files_1|url_paste": "SecondOutputContent",
"files_1|NAME": "SecondOutputName",
"files_1|file_type": "txt",
"file_count": "2",
}
)
run_response = self.dataset_populator.tools_post(payload)
self.dataset_populator.wait_for_tool_run(history_id, run_response)
datasets = run_response.json()["outputs"]
assert len(datasets) == 2, datasets
content = self.dataset_populator.get_history_dataset_content(history_id, dataset=datasets[0])
assert content.strip() == "Test123"
assert datasets[0]["file_ext"] == "txt", datasets
assert datasets[0]["genome_build"] == "?", datasets
content = self.dataset_populator.get_history_dataset_content(history_id, dataset=datasets[1])
assert content.strip() == "SecondOutputContent"
assert datasets[1]["file_ext"] == "txt"
assert datasets[1]["genome_build"] == "?", datasets
def test_upload_multiple_files_space_to_tab(self):
with self.dataset_populator.test_history() as history_id:
payload = self.dataset_populator.upload_payload(history_id,
content=ONE_TO_SIX_WITH_SPACES,
file_type="tabular",
dbkey="hg19",
extra_inputs={
"files_0|file_type": "txt",
"files_0|space_to_tab": "Yes",
"files_1|url_paste": ONE_TO_SIX_WITH_SPACES,
"files_1|NAME": "SecondOutputName",
"files_1|file_type": "txt",
"files_2|url_paste": ONE_TO_SIX_WITH_SPACES,
"files_2|NAME": "ThirdOutputName",
"files_2|file_type": "txt",
"files_2|space_to_tab": "Yes",
"file_count": "3",
}
)
run_response = self.dataset_populator.tools_post(payload)
self.dataset_populator.wait_for_tool_run(history_id, run_response)
datasets = run_response.json()["outputs"]
assert len(datasets) == 3, datasets
content = self.dataset_populator.get_history_dataset_content(history_id, dataset=datasets[0])
assert content == ONE_TO_SIX_WITH_TABS
content = self.dataset_populator.get_history_dataset_content(history_id, dataset=datasets[1])
assert content == ONE_TO_SIX_WITH_SPACES
content = self.dataset_populator.get_history_dataset_content(history_id, dataset=datasets[2])
assert content == ONE_TO_SIX_WITH_TABS
def test_multiple_files_posix_lines(self):
with self.dataset_populator.test_history() as history_id:
payload = self.dataset_populator.upload_payload(history_id,
content=ONE_TO_SIX_ON_WINDOWS,
file_type="tabular",
dbkey="hg19",
extra_inputs={
"files_0|file_type": "txt",
"files_0|to_posix_lines": "Yes",
"files_1|url_paste": ONE_TO_SIX_ON_WINDOWS,
"files_1|NAME": "SecondOutputName",
"files_1|file_type": "txt",
"files_1|to_posix_lines": None,
"files_2|url_paste": ONE_TO_SIX_ON_WINDOWS,
"files_2|NAME": "ThirdOutputName",
"files_2|file_type": "txt",
"file_count": "3",
}
)
run_response = self.dataset_populator.tools_post(payload)
self.dataset_populator.wait_for_tool_run(history_id, run_response)
datasets = run_response.json()["outputs"]
assert len(datasets) == 3, datasets
content = self.dataset_populator.get_history_dataset_content(history_id, dataset=datasets[0])
assert content == ONE_TO_SIX_WITH_TABS
content = self.dataset_populator.get_history_dataset_content(history_id, dataset=datasets[1])
assert content == ONE_TO_SIX_ON_WINDOWS
content = self.dataset_populator.get_history_dataset_content(history_id, dataset=datasets[2])
assert content == ONE_TO_SIX_WITH_TABS
def test_upload_from_invalid_url(self):
history_id, new_dataset = self._upload('https://usegalaxy.org/bla123', assert_ok=False)
dataset_details = self.dataset_populator.get_history_dataset_details(history_id, dataset_id=new_dataset["id"], assert_ok=False)
assert dataset_details['state'] == 'error', "expected dataset state to be 'error', but got '%s'" % dataset_details['state']
def test_upload_from_valid_url(self):
history_id, new_dataset = self._upload('https://usegalaxy.org/api/version')
self.dataset_populator.get_history_dataset_details(history_id, dataset_id=new_dataset["id"], assert_ok=True)
def _velvet_upload(self, history_id, extra_inputs):
payload = self.dataset_populator.upload_payload(
history_id,
"sequences content",
file_type="velvet",
extra_inputs=extra_inputs,
)
run_response = self.dataset_populator.tools_post(payload)
self.dataset_populator.wait_for_tool_run(history_id, run_response)
datasets = run_response.json()["outputs"]
assert len(datasets) == 1
dataset = datasets[0]
return dataset
def _get_roadmaps_content(self, history_id, dataset):
roadmaps_content = self.dataset_populator.get_history_dataset_content(history_id, dataset=dataset, filename="Roadmaps")
return roadmaps_content
def _upload_and_get_content(self, content, **upload_kwds):
history_id, new_dataset = self._upload(content, **upload_kwds)
return self.dataset_populator.get_history_dataset_content(history_id, dataset=new_dataset)
def _upload_and_get_details(self, content, **upload_kwds):
history_id, new_dataset = self._upload(content, **upload_kwds)
return self.dataset_populator.get_history_dataset_details(history_id, dataset=new_dataset)
def _upload(self, content, **upload_kwds):
history_id = self.dataset_populator.new_history()
new_dataset = self.dataset_populator.new_dataset(history_id, content=content, **upload_kwds)
self.dataset_populator.wait_for_history(history_id, assert_ok=upload_kwds.get("assert_ok", True))
return history_id, new_dataset
class ToolsUploadTestCase(api.ApiTestCase):
def setUp(self):
super(ToolsUploadTestCase, self).setUp()
self.dataset_populator = DatasetPopulator(self.galaxy_interactor)
def test_upload1_paste(self):
with self.dataset_populator.test_history() as history_id:
payload = self.dataset_populator.upload_payload(
history_id, 'Hello World')
create_response = self._post("tools", data=payload)
self._assert_has_keys(create_response.json(), 'outputs')
def test_upload1_paste_bad_datatype(self):
# Check that you get a nice message if you upload an incorrect datatype
with self.dataset_populator.test_history() as history_id:
file_type = "johnsawesomebutfakedatatype"
payload = self.dataset_populator.upload_payload(
history_id, 'Hello World', file_type=file_type)
create = self._post("tools", data=payload).json()
self._assert_has_keys(create, 'err_msg')
assert file_type in create['err_msg']
# upload1 rewrites content with posix lines by default but this can be disabled by setting
# to_posix_lines=None in the request. Newer fetch API does not do this by default prefering
# to keep content unaltered if possible but it can be enabled with a simple JSON boolean switch
# of the same name (to_posix_lines).
def test_upload_posix_newline_fixes_by_default(self):
windows_content = ONE_TO_SIX_ON_WINDOWS
result_content = self._upload_and_get_content(windows_content)
self.assertEquals(result_content, ONE_TO_SIX_WITH_TABS)
def test_fetch_posix_unaltered(self):
windows_content = ONE_TO_SIX_ON_WINDOWS
result_content = self._upload_and_get_content(windows_content,
api="fetch")
self.assertEquals(result_content, ONE_TO_SIX_ON_WINDOWS)
def test_upload_disable_posix_fix(self):
windows_content = ONE_TO_SIX_ON_WINDOWS
result_content = self._upload_and_get_content(windows_content,
to_posix_lines=None)
self.assertEquals(result_content, windows_content)
def test_fetch_post_lines_option(self):
windows_content = ONE_TO_SIX_ON_WINDOWS
result_content = self._upload_and_get_content(windows_content,
api="fetch",
to_posix_lines=True)
self.assertEquals(result_content, ONE_TO_SIX_WITH_TABS)
def test_upload_tab_to_space_off_by_default(self):
table = ONE_TO_SIX_WITH_SPACES
result_content = self._upload_and_get_content(table)
self.assertEquals(result_content, table)
def test_fetch_tab_to_space_off_by_default(self):
table = ONE_TO_SIX_WITH_SPACES
result_content = self._upload_and_get_content(table, api='fetch')
self.assertEquals(result_content, table)
def test_upload_tab_to_space(self):
table = ONE_TO_SIX_WITH_SPACES
result_content = self._upload_and_get_content(table,
space_to_tab="Yes")
self.assertEquals(result_content, ONE_TO_SIX_WITH_TABS)
def test_fetch_tab_to_space(self):
table = ONE_TO_SIX_WITH_SPACES
result_content = self._upload_and_get_content(table,
api="fetch",
space_to_tab=True)
self.assertEquals(result_content, ONE_TO_SIX_WITH_TABS)
def test_fetch_compressed_with_explicit_type(self):
fastqgz_path = TestDataResolver().get_filename("1.fastqsanger.gz")
with open(fastqgz_path, "rb") as fh:
details = self._upload_and_get_details(fh,
api="fetch",
ext="fastqsanger.gz")
assert details["state"] == "ok"
assert details["file_ext"] == "fastqsanger.gz"
def test_fetch_compressed_default(self):
fastqgz_path = TestDataResolver().get_filename("1.fastqsanger.gz")
with open(fastqgz_path, "rb") as fh:
details = self._upload_and_get_details(fh,
api="fetch",
assert_ok=False)
assert details["state"] == "ok"
assert details["file_ext"] == "fastqsanger.gz", details
@uses_test_history(require_new=True)
def test_fetch_compressed_auto_decompress_target(self, history_id):
# TODO: this should definitely be fixed to allow auto decompression via that API.
fastqgz_path = TestDataResolver().get_filename("1.fastqsanger.gz")
with open(fastqgz_path, "rb") as fh:
details = self._upload_and_get_details(fh,
api="fetch",
history_id=history_id,
assert_ok=False,
auto_decompress=True)
assert details["state"] == "ok"
assert details["file_ext"] == "fastqsanger.gz", details
def test_upload_decompress_off_with_auto_by_default(self):
# UNSTABLE_FLAG: This might default to a bed.gz datatype in the future.
bedgz_path = TestDataResolver().get_filename("4.bed.gz")
with open(bedgz_path, "rb") as fh:
details = self._upload_and_get_details(fh, file_type="auto")
assert details["state"] == "ok"
assert details["file_ext"] == "bed", details
def test_upload_decompresses_if_uncompressed_type_selected(self):
fastqgz_path = TestDataResolver().get_filename("1.fastqsanger.gz")
with open(fastqgz_path, "rb") as fh:
details = self._upload_and_get_details(fh, file_type="fastqsanger")
assert details["state"] == "ok"
assert details["file_ext"] == "fastqsanger", details
assert details["file_size"] == 178, details
def test_upload_decompress_off_if_compressed_type_selected(self):
fastqgz_path = TestDataResolver().get_filename("1.fastqsanger.gz")
with open(fastqgz_path, "rb") as fh:
details = self._upload_and_get_details(fh,
file_type="fastqsanger.gz")
assert details["state"] == "ok"
assert details["file_ext"] == "fastqsanger.gz", details
assert details["file_size"] == 161, details
def test_upload_auto_decompress_off(self):
# UNSTABLE_FLAG: This might default to a bed.gz datatype in the future.
bedgz_path = TestDataResolver().get_filename("4.bed.gz")
with open(bedgz_path, "rb") as fh:
details = self._upload_and_get_details(fh,
file_type="auto",
assert_ok=False,
auto_decompress=False)
assert details["file_ext"] == "binary", details
@uses_test_history(require_new=True)
def test_fetch_compressed_with_auto(self, history_id):
# UNSTABLE_FLAG: This might default to a bed.gz datatype in the future.
# TODO: this should definitely be fixed to allow auto decompression via that API.
bedgz_path = TestDataResolver().get_filename("4.bed.gz")
with open(bedgz_path, "rb") as fh:
details = self._upload_and_get_details(fh,
api="fetch",
history_id=history_id,
auto_decompress=True,
assert_ok=False)
assert details["state"] == "ok"
assert details["file_ext"] == "bed"
@skip_without_datatype("rdata")
def test_rdata_not_decompressed(self):
# Prevent regression of https://github.com/galaxyproject/galaxy/issues/753
rdata_path = TestDataResolver().get_filename("1.RData")
with open(rdata_path, "rb") as fh:
rdata_metadata = self._upload_and_get_details(fh, file_type="auto")
self.assertEquals(rdata_metadata["file_ext"], "rdata")
@skip_without_datatype("csv")
def test_csv_upload(self):
csv_path = TestDataResolver().get_filename("1.csv")
with open(csv_path, "rb") as fh:
csv_metadata = self._upload_and_get_details(fh, file_type="csv")
self.assertEquals(csv_metadata["file_ext"], "csv")
@skip_without_datatype("csv")
def test_csv_upload_auto(self):
csv_path = TestDataResolver().get_filename("1.csv")
with open(csv_path, "rb") as fh:
csv_metadata = self._upload_and_get_details(fh, file_type="auto")
self.assertEquals(csv_metadata["file_ext"], "csv")
@skip_without_datatype("csv")
def test_csv_fetch(self):
csv_path = TestDataResolver().get_filename("1.csv")
with open(csv_path, "rb") as fh:
csv_metadata = self._upload_and_get_details(fh,
api="fetch",
ext="csv",
to_posix_lines=True)
self.assertEquals(csv_metadata["file_ext"], "csv")
@skip_without_datatype("csv")
def test_csv_sniff_fetch(self):
csv_path = TestDataResolver().get_filename("1.csv")
with open(csv_path, "rb") as fh:
csv_metadata = self._upload_and_get_details(fh,
api="fetch",
ext="auto",
to_posix_lines=True)
self.assertEquals(csv_metadata["file_ext"], "csv")
@skip_without_datatype("tiff")
def test_image_upload_auto(self):
tiff_path = TestDataResolver().get_filename("1.tiff")
with open(tiff_path, "rb") as fh:
tiff_metadata = self._upload_and_get_details(fh, file_type="auto")
self.assertEquals(tiff_metadata["file_ext"], "tiff")
@skip_without_datatype("velvet")
def test_composite_datatype(self):
with self.dataset_populator.test_history() as history_id:
dataset = self._velvet_upload(history_id,
extra_inputs={
"files_1|url_paste":
"roadmaps content",
"files_1|type": "upload_dataset",
"files_2|url_paste":
"log content",
"files_2|type": "upload_dataset",
})
roadmaps_content = self._get_roadmaps_content(history_id, dataset)
assert roadmaps_content.strip(
) == "roadmaps content", roadmaps_content
@skip_without_datatype("velvet")
def test_composite_datatype_space_to_tab(self):
# Like previous test but set one upload with space_to_tab to True to
# verify that works.
with self.dataset_populator.test_history() as history_id:
dataset = self._velvet_upload(history_id,
extra_inputs={
"files_1|url_paste":
"roadmaps content",
"files_1|type": "upload_dataset",
"files_1|space_to_tab": "Yes",
"files_2|url_paste":
"log content",
"files_2|type": "upload_dataset",
})
roadmaps_content = self._get_roadmaps_content(history_id, dataset)
assert roadmaps_content.strip(
) == "roadmaps\tcontent", roadmaps_content
@skip_without_datatype("velvet")
def test_composite_datatype_posix_lines(self):
# Like previous test but set one upload with space_to_tab to True to
# verify that works.
with self.dataset_populator.test_history() as history_id:
dataset = self._velvet_upload(history_id,
extra_inputs={
"files_1|url_paste":
"roadmaps\rcontent",
"files_1|type": "upload_dataset",
"files_1|space_to_tab": "Yes",
"files_2|url_paste":
"log\rcontent",
"files_2|type": "upload_dataset",
})
roadmaps_content = self._get_roadmaps_content(history_id, dataset)
assert roadmaps_content.strip(
) == "roadmaps\ncontent", roadmaps_content
def test_upload_dbkey(self):
with self.dataset_populator.test_history() as history_id:
payload = self.dataset_populator.upload_payload(history_id,
"Test123",
dbkey="hg19")
run_response = self.dataset_populator.tools_post(payload)
self.dataset_populator.wait_for_tool_run(history_id, run_response)
datasets = run_response.json()["outputs"]
assert datasets[0].get("genome_build") == "hg19", datasets[0]
@uses_test_history(require_new=False)
def test_fetch_bam_file(self, history_id):
bam_path = TestDataResolver().get_filename("1.bam")
with open(bam_path, "rb") as fh:
details = self._upload_and_get_details(fh,
api="fetch",
history_id=history_id,
assert_ok=False)
assert details["state"] == "ok"
assert details["file_ext"] == "bam", details
def test_upload_bam_file(self):
bam_path = TestDataResolver().get_filename("1.bam")
with open(bam_path, "rb") as fh:
details = self._upload_and_get_details(fh, file_type="auto")
assert details["state"] == "ok"
assert details["file_ext"] == "bam", details
def test_fetch_metadata(self):
table = ONE_TO_SIX_WITH_SPACES
details = self._upload_and_get_details(
table,
api='fetch',
dbkey="hg19",
info="cool upload",
tags=["name:data", "group:type:paired-end"])
assert details.get("genome_build") == "hg19"
assert details.get("misc_info") == "cool upload", details
tags = details.get("tags")
assert len(tags) == 2, details
assert "group:type:paired-end" in tags
assert "name:data" in tags
def test_upload_multiple_files_1(self):
with self.dataset_populator.test_history() as history_id:
payload = self.dataset_populator.upload_payload(
history_id,
"Test123",
dbkey="hg19",
extra_inputs={
"files_1|url_paste": "SecondOutputContent",
"files_1|NAME": "SecondOutputName",
"files_1|file_type": "tabular",
"files_1|dbkey": "hg18",
"file_count": "2",
})
run_response = self.dataset_populator.tools_post(payload)
self.dataset_populator.wait_for_tool_run(history_id, run_response)
datasets = run_response.json()["outputs"]
assert len(datasets) == 2, datasets
content = self.dataset_populator.get_history_dataset_content(
history_id, dataset=datasets[0])
assert content.strip() == "Test123"
assert datasets[0]["file_ext"] == "txt"
assert datasets[0]["genome_build"] == "hg19", datasets
content = self.dataset_populator.get_history_dataset_content(
history_id, dataset=datasets[1])
assert content.strip() == "SecondOutputContent"
assert datasets[1]["file_ext"] == "tabular"
assert datasets[1]["genome_build"] == "hg18", datasets
def test_upload_multiple_files_2(self):
with self.dataset_populator.test_history() as history_id:
payload = self.dataset_populator.upload_payload(
history_id,
"Test123",
file_type="tabular",
dbkey="hg19",
extra_inputs={
"files_1|url_paste": "SecondOutputContent",
"files_1|NAME": "SecondOutputName",
"files_1|file_type": "txt",
"files_1|dbkey": "hg18",
"file_count": "2",
})
run_response = self.dataset_populator.tools_post(payload)
self.dataset_populator.wait_for_tool_run(history_id, run_response)
datasets = run_response.json()["outputs"]
assert len(datasets) == 2, datasets
content = self.dataset_populator.get_history_dataset_content(
history_id, dataset=datasets[0])
assert content.strip() == "Test123"
assert datasets[0]["file_ext"] == "tabular", datasets
assert datasets[0]["genome_build"] == "hg19", datasets
content = self.dataset_populator.get_history_dataset_content(
history_id, dataset=datasets[1])
assert content.strip() == "SecondOutputContent"
assert datasets[1]["file_ext"] == "txt"
assert datasets[1]["genome_build"] == "hg18", datasets
def test_upload_multiple_files_3(self):
with self.dataset_populator.test_history() as history_id:
payload = self.dataset_populator.upload_payload(
history_id,
"Test123",
file_type="tabular",
dbkey="hg19",
extra_inputs={
"files_0|file_type": "txt",
"files_0|dbkey": "hg18",
"files_1|url_paste": "SecondOutputContent",
"files_1|NAME": "SecondOutputName",
"files_1|file_type": "txt",
"files_1|dbkey": "hg18",
"file_count": "2",
})
run_response = self.dataset_populator.tools_post(payload)
self.dataset_populator.wait_for_tool_run(history_id, run_response)
datasets = run_response.json()["outputs"]
assert len(datasets) == 2, datasets
content = self.dataset_populator.get_history_dataset_content(
history_id, dataset=datasets[0])
assert content.strip() == "Test123"
assert datasets[0]["file_ext"] == "txt", datasets
assert datasets[0]["genome_build"] == "hg18", datasets
content = self.dataset_populator.get_history_dataset_content(
history_id, dataset=datasets[1])
assert content.strip() == "SecondOutputContent"
assert datasets[1]["file_ext"] == "txt"
assert datasets[1]["genome_build"] == "hg18", datasets
def test_upload_multiple_files_no_dbkey(self):
with self.dataset_populator.test_history() as history_id:
payload = self.dataset_populator.upload_payload(
history_id,
"Test123",
file_type="tabular",
dbkey=None,
extra_inputs={
"files_0|file_type": "txt",
"files_1|url_paste": "SecondOutputContent",
"files_1|NAME": "SecondOutputName",
"files_1|file_type": "txt",
"file_count": "2",
})
run_response = self.dataset_populator.tools_post(payload)
self.dataset_populator.wait_for_tool_run(history_id, run_response)
datasets = run_response.json()["outputs"]
assert len(datasets) == 2, datasets
content = self.dataset_populator.get_history_dataset_content(
history_id, dataset=datasets[0])
assert content.strip() == "Test123"
assert datasets[0]["file_ext"] == "txt", datasets
assert datasets[0]["genome_build"] == "?", datasets
content = self.dataset_populator.get_history_dataset_content(
history_id, dataset=datasets[1])
assert content.strip() == "SecondOutputContent"
assert datasets[1]["file_ext"] == "txt"
assert datasets[1]["genome_build"] == "?", datasets
def test_upload_multiple_files_space_to_tab(self):
with self.dataset_populator.test_history() as history_id:
payload = self.dataset_populator.upload_payload(
history_id,
content=ONE_TO_SIX_WITH_SPACES,
file_type="tabular",
dbkey="hg19",
extra_inputs={
"files_0|file_type": "txt",
"files_0|space_to_tab": "Yes",
"files_1|url_paste": ONE_TO_SIX_WITH_SPACES,
"files_1|NAME": "SecondOutputName",
"files_1|file_type": "txt",
"files_2|url_paste": ONE_TO_SIX_WITH_SPACES,
"files_2|NAME": "ThirdOutputName",
"files_2|file_type": "txt",
"files_2|space_to_tab": "Yes",
"file_count": "3",
})
run_response = self.dataset_populator.tools_post(payload)
self.dataset_populator.wait_for_tool_run(history_id, run_response)
datasets = run_response.json()["outputs"]
assert len(datasets) == 3, datasets
content = self.dataset_populator.get_history_dataset_content(
history_id, dataset=datasets[0])
assert content == ONE_TO_SIX_WITH_TABS
content = self.dataset_populator.get_history_dataset_content(
history_id, dataset=datasets[1])
assert content == ONE_TO_SIX_WITH_SPACES
content = self.dataset_populator.get_history_dataset_content(
history_id, dataset=datasets[2])
assert content == ONE_TO_SIX_WITH_TABS
def test_multiple_files_posix_lines(self):
with self.dataset_populator.test_history() as history_id:
payload = self.dataset_populator.upload_payload(
history_id,
content=ONE_TO_SIX_ON_WINDOWS,
file_type="tabular",
dbkey="hg19",
extra_inputs={
"files_0|file_type": "txt",
"files_0|to_posix_lines": "Yes",
"files_1|url_paste": ONE_TO_SIX_ON_WINDOWS,
"files_1|NAME": "SecondOutputName",
"files_1|file_type": "txt",
"files_1|to_posix_lines": None,
"files_2|url_paste": ONE_TO_SIX_ON_WINDOWS,
"files_2|NAME": "ThirdOutputName",
"files_2|file_type": "txt",
"file_count": "3",
})
run_response = self.dataset_populator.tools_post(payload)
self.dataset_populator.wait_for_tool_run(history_id, run_response)
datasets = run_response.json()["outputs"]
assert len(datasets) == 3, datasets
content = self.dataset_populator.get_history_dataset_content(
history_id, dataset=datasets[0])
assert content == ONE_TO_SIX_WITH_TABS
content = self.dataset_populator.get_history_dataset_content(
history_id, dataset=datasets[1])
assert content == ONE_TO_SIX_ON_WINDOWS
content = self.dataset_populator.get_history_dataset_content(
history_id, dataset=datasets[2])
assert content == ONE_TO_SIX_WITH_TABS
def test_upload_from_invalid_url(self):
history_id, new_dataset = self._upload('https://usegalaxy.org/bla123',
assert_ok=False)
dataset_details = self.dataset_populator.get_history_dataset_details(
history_id, dataset_id=new_dataset["id"], assert_ok=False)
assert dataset_details[
'state'] == 'error', "expected dataset state to be 'error', but got '%s'" % dataset_details[
'state']
def test_upload_from_valid_url(self):
history_id, new_dataset = self._upload(
'https://usegalaxy.org/api/version')
self.dataset_populator.get_history_dataset_details(
history_id, dataset_id=new_dataset["id"], assert_ok=True)
def _velvet_upload(self, history_id, extra_inputs):
payload = self.dataset_populator.upload_payload(
history_id,
"sequences content",
file_type="velvet",
extra_inputs=extra_inputs,
)
run_response = self.dataset_populator.tools_post(payload)
self.dataset_populator.wait_for_tool_run(history_id, run_response)
datasets = run_response.json()["outputs"]
assert len(datasets) == 1
dataset = datasets[0]
return dataset
def _get_roadmaps_content(self, history_id, dataset):
roadmaps_content = self.dataset_populator.get_history_dataset_content(
history_id, dataset=dataset, filename="Roadmaps")
return roadmaps_content
def _upload_and_get_content(self, content, **upload_kwds):
history_id, new_dataset = self._upload(content, **upload_kwds)
return self.dataset_populator.get_history_dataset_content(
history_id, dataset=new_dataset)
def _upload_and_get_details(self, content, **upload_kwds):
history_id, new_dataset = self._upload(content, **upload_kwds)
assert_ok = upload_kwds.get("assert_ok", True)
return self.dataset_populator.get_history_dataset_details(
history_id, dataset=new_dataset, assert_ok=assert_ok)
def _upload(self, content, api="upload1", history_id=None, **upload_kwds):
assert_ok = upload_kwds.get("assert_ok", True)
history_id = history_id or self.dataset_populator.new_history()
if api == "upload1":
new_dataset = self.dataset_populator.new_dataset(history_id,
content=content,
**upload_kwds)
else:
assert api == "fetch"
element = dict(src="files", **upload_kwds)
target = {
"destination": {
"type": "hdas"
},
"elements": [element],
}
targets = json.dumps([target])
payload = {
"history_id": history_id,
"targets": targets,
"__files": {
"files_0|file_data": content
}
}
new_dataset = self.dataset_populator.fetch(
payload, assert_ok=assert_ok).json()["outputs"][0]
self.dataset_populator.wait_for_history(history_id,
assert_ok=assert_ok)
return history_id, new_dataset
class DatasetsApiTestCase(api.ApiTestCase):
def setUp(self):
super(DatasetsApiTestCase, self).setUp()
self.dataset_populator = DatasetPopulator(self.galaxy_interactor)
self.dataset_collection_populator = DatasetCollectionPopulator(self.galaxy_interactor)
self.history_id = self.dataset_populator.new_history()
def test_index(self):
index_response = self._get("datasets")
self._assert_status_code_is(index_response, 200)
def test_search_datasets(self):
hda_id = self.dataset_populator.new_dataset(self.history_id)['id']
payload = {'limit': 1, 'offset': 0}
index_response = self._get("datasets", payload).json()
assert len(index_response) == 1
assert index_response[0]['id'] == hda_id
hdca_id = self.dataset_collection_populator.create_list_in_history(self.history_id,
contents=["1\n2\n3"]).json()['id']
payload = {'limit': 3, 'offset': 0}
index_response = self._get("datasets", payload).json()
assert len(index_response) == 3
assert index_response[0]['id'] == hdca_id
assert index_response[0]['history_content_type'] == 'dataset_collection'
assert index_response[2]['id'] == hda_id
assert index_response[2]['history_content_type'] == 'dataset'
payload = {'limit': 2, 'offset': 0, 'q': ['history_content_type'], 'qv': ['dataset']}
index_response = self._get("datasets", payload).json()
assert index_response[1]['id'] == hda_id
def test_search_by_tag(self):
hda_id = self.dataset_populator.new_dataset(self.history_id)['id']
update_payload = {
'tags': ['cool:new_tag', 'cool:another_tag'],
}
updated_hda = self._put(
"histories/{history_id}/contents/{hda_id}".format(history_id=self.history_id, hda_id=hda_id),
update_payload).json()
assert 'cool:new_tag' in updated_hda['tags']
assert 'cool:another_tag' in updated_hda['tags']
payload = {'limit': 10, 'offset': 0, 'q': ['history_content_type', 'tag'], 'qv': ['dataset', 'cool:new_tag']}
index_response = self._get("datasets", payload).json()
assert len(index_response) == 1
payload = {'limit': 10, 'offset': 0, 'q': ['history_content_type', 'tag-contains'],
'qv': ['dataset', 'new_tag']}
index_response = self._get("datasets", payload).json()
assert len(index_response) == 1
payload = {'limit': 10, 'offset': 0, 'q': ['history_content_type', 'tag-contains'], 'qv': ['dataset', 'notag']}
index_response = self._get("datasets", payload).json()
assert len(index_response) == 0
def test_search_by_tool_id(self):
self.dataset_populator.new_dataset(self.history_id)
payload = {'limit': 1, 'offset': 0, 'q': ['history_content_type', 'tool_id'], 'qv': ['dataset', 'upload1']}
assert len(self._get("datasets", payload).json()) == 1
payload = {'limit': 1, 'offset': 0, 'q': ['history_content_type', 'tool_id'], 'qv': ['dataset', 'uploadX']}
assert len(self._get("datasets", payload).json()) == 0
payload = {'limit': 1, 'offset': 0, 'q': ['history_content_type', 'tool_id-contains'], 'qv': ['dataset', 'pload1']}
assert len(self._get("datasets", payload).json()) == 1
self.dataset_collection_populator.create_list_in_history(self.history_id,
name="search by tool id",
contents=["1\n2\n3"]).json()
self.dataset_populator.wait_for_history(self.history_id)
payload = {'limit': 10, 'offset': 0, 'history_id': self.history_id, 'q': ['name', 'tool_id'],
'qv': ['search by tool id', 'upload1']}
result = self._get("datasets", payload).json()
assert result[0]['name'] == 'search by tool id', result
payload = {'limit': 1, 'offset': 0, 'q': ['history_content_type', 'tool_id'],
'qv': ['dataset_collection', 'uploadX']}
result = self._get("datasets", payload).json()
assert len(result) == 0
def test_invalid_search(self):
payload = {'limit': 10, 'offset': 0, 'q': ['history_content_type', 'tag-invalid_op'], 'qv': ['dataset', 'notag']}
index_response = self._get("datasets", payload)
self._assert_status_code_is(index_response, 400)
assert index_response.json()['err_msg'] == 'bad op in filter'
def test_search_returns_only_accessible(self):
hda_id = self.dataset_populator.new_dataset(self.history_id)['id']
with self._different_user():
payload = {'limit': 10, 'offset': 0, 'q': ['history_content_type'], 'qv': ['dataset']}
index_response = self._get("datasets", payload).json()
for item in index_response:
assert hda_id != item['id']
def test_show(self):
hda1 = self.dataset_populator.new_dataset(self.history_id)
show_response = self._get("datasets/%s" % (hda1["id"]))
self._assert_status_code_is(show_response, 200)
self.__assert_matches_hda(hda1, show_response.json())
def __assert_matches_hda(self, input_hda, query_hda):
self._assert_has_keys(query_hda, "id", "name")
assert input_hda["name"] == query_hda["name"]
assert input_hda["id"] == query_hda["id"]
def test_display(self):
contents = textwrap.dedent("""\
1 2 3 4
A B C D
10 20 30 40
""")
hda1 = self.dataset_populator.new_dataset(self.history_id, content=contents)
self.dataset_populator.wait_for_history(self.history_id)
display_response = self._get("histories/%s/contents/%s/display" % (self.history_id, hda1["id"]), {
'raw': 'True'
})
self._assert_status_code_is(display_response, 200)
assert display_response.text == contents
class ObjectStoreJobsIntegrationTestCase(integration_util.IntegrationTestCase):
framework_tool_and_types = True
@classmethod
def handle_galaxy_config_kwds(cls, config):
temp_directory = cls._test_driver.mkdtemp()
cls.object_stores_parent = temp_directory
for disk_store_file_name in [
"files_default", "files_static", "files_dynamic_ebs",
"files_dynamic_s3"
]:
disk_store_path = os.path.join(temp_directory,
disk_store_file_name)
os.makedirs(disk_store_path)
setattr(cls, "%s_path" % disk_store_file_name, disk_store_path)
config_path = os.path.join(temp_directory, "object_store_conf.xml")
with open(config_path, "w") as f:
f.write(
DISTRIBUTED_OBJECT_STORE_CONFIG_TEMPLATE.safe_substitute(
{"temp_directory": temp_directory}))
config["object_store_config_file"] = config_path
config["job_config_file"] = JOB_CONFIG_FILE
config[
"job_resource_params_file"] = JOB_RESOURCE_PARAMETERS_CONFIG_FILE
def setUp(self):
super(ObjectStoreJobsIntegrationTestCase, self).setUp()
self.dataset_populator = DatasetPopulator(self.galaxy_interactor)
def _object_store_counts(self):
files_default_count = _files_count(self.files_default_path)
files_static_count = _files_count(self.files_static_path)
files_dynamic_count = _files_count(self.files_dynamic_path)
return files_default_count, files_static_count, files_dynamic_count
def _assert_file_counts(self, default, static, dynamic_ebs, dynamic_s3):
files_default_count = _files_count(self.files_default_path)
files_static_count = _files_count(self.files_static_path)
files_dynamic_ebs_count = _files_count(self.files_dynamic_ebs_path)
files_dynamic_s3_count = _files_count(self.files_dynamic_s3_path)
assert default == files_default_count
assert static == files_static_count
assert dynamic_ebs == files_dynamic_ebs_count
assert dynamic_s3 == files_dynamic_s3_count
def test_tool_simple_constructs(self):
with self.dataset_populator.test_history() as history_id:
def _run_tool(tool_id, inputs):
self.dataset_populator.run_tool(
tool_id,
inputs,
history_id,
assert_ok=True,
)
self.dataset_populator.wait_for_history(history_id)
self._assert_file_counts(0, 0, 0, 0)
hda1 = self.dataset_populator.new_dataset(history_id,
content="1 2 3")
self.dataset_populator.wait_for_history(history_id)
hda1_input = {"src": "hda", "id": hda1["id"]}
# One file uploaded, added to default object store ID.
self._assert_file_counts(1, 0, 0, 0)
# should create two files in static object store.
_run_tool("multi_data_param", {"f1": hda1_input, "f2": hda1_input})
self._assert_file_counts(1, 2, 0, 0)
# should create two files in ebs object store.
create_10_inputs = {
"input1": hda1_input,
"input2": hda1_input,
}
_run_tool("create_10", create_10_inputs)
self._assert_file_counts(1, 2, 10, 0)
# should create 10 files in S3 object store.
create_10_inputs = {
"__job_resource|__job_resource__select": "yes",
"__job_resource|how_store": "slow",
"input1": hda1_input,
"input2": hda1_input,
}
_run_tool("create_10", create_10_inputs)
self._assert_file_counts(1, 2, 10, 10)
class ObjectStoreJobsIntegrationTestCase(integration_util.IntegrationTestCase):
framework_tool_and_types = True
@classmethod
def handle_galaxy_config_kwds(cls, config):
temp_directory = cls._test_driver.mkdtemp()
cls.object_stores_parent = temp_directory
for disk_store_file_name in ["files_default", "files_static", "files_dynamic_ebs", "files_dynamic_s3"]:
disk_store_path = os.path.join(temp_directory, disk_store_file_name)
os.makedirs(disk_store_path)
setattr(cls, "%s_path" % disk_store_file_name, disk_store_path)
config_path = os.path.join(temp_directory, "object_store_conf.xml")
with open(config_path, "w") as f:
f.write(DISTRIBUTED_OBJECT_STORE_CONFIG_TEMPLATE.safe_substitute({"temp_directory": temp_directory}))
config["object_store_config_file"] = config_path
config["job_config_file"] = JOB_CONFIG_FILE
config["job_resource_params_file"] = JOB_RESOURCE_PARAMETERS_CONFIG_FILE
def setUp(self):
super(ObjectStoreJobsIntegrationTestCase, self).setUp()
self.dataset_populator = DatasetPopulator(self.galaxy_interactor)
def _object_store_counts(self):
files_default_count = _files_count(self.files_default_path)
files_static_count = _files_count(self.files_static_path)
files_dynamic_count = _files_count(self.files_dynamic_path)
return files_default_count, files_static_count, files_dynamic_count
def _assert_file_counts(self, default, static, dynamic_ebs, dynamic_s3):
files_default_count = _files_count(self.files_default_path)
files_static_count = _files_count(self.files_static_path)
files_dynamic_ebs_count = _files_count(self.files_dynamic_ebs_path)
files_dynamic_s3_count = _files_count(self.files_dynamic_s3_path)
assert default == files_default_count
assert static == files_static_count
assert dynamic_ebs == files_dynamic_ebs_count
assert dynamic_s3 == files_dynamic_s3_count
def test_tool_simple_constructs(self):
with self.dataset_populator.test_history() as history_id:
def _run_tool(tool_id, inputs):
self.dataset_populator.run_tool(
tool_id,
inputs,
history_id,
assert_ok=True,
)
self.dataset_populator.wait_for_history(history_id)
self._assert_file_counts(0, 0, 0, 0)
hda1 = self.dataset_populator.new_dataset(history_id, content="1 2 3")
self.dataset_populator.wait_for_history(history_id)
hda1_input = {"src": "hda", "id": hda1["id"]}
# One file uploaded, added to default object store ID.
self._assert_file_counts(1, 0, 0, 0)
# should create two files in static object store.
_run_tool("multi_data_param", {"f1": hda1_input, "f2": hda1_input})
self._assert_file_counts(1, 2, 0, 0)
# should create two files in ebs object store.
create_10_inputs = {
"input1": hda1_input,
"input2": hda1_input,
}
_run_tool("create_10", create_10_inputs)
self._assert_file_counts(1, 2, 10, 0)
# should create 10 files in S3 object store.
create_10_inputs = {
"__job_resource|__job_resource__select": "yes",
"__job_resource|how_store": "slow",
"input1": hda1_input,
"input2": hda1_input,
}
_run_tool("create_10", create_10_inputs)
self._assert_file_counts(1, 2, 10, 10)
class HistoriesApiTestCase(api.ApiTestCase):
def setUp(self):
super(HistoriesApiTestCase, self).setUp()
self.dataset_populator = DatasetPopulator(self.galaxy_interactor)
self.dataset_collection_populator = DatasetCollectionPopulator(self.galaxy_interactor)
def test_create_history(self):
# Create a history.
create_response = self._create_history("TestHistory1")
created_id = create_response["id"]
# Make sure new history appears in index of user's histories.
index_response = self._get("histories").json()
indexed_history = [h for h in index_response if h["id"] == created_id][0]
self.assertEqual(indexed_history["name"], "TestHistory1")
def test_show_history(self):
history_id = self._create_history("TestHistoryForShow")["id"]
show_response = self._show(history_id)
self._assert_has_key(
show_response,
'id', 'name', 'annotation', 'size', 'contents_url',
'state', 'state_details', 'state_ids'
)
state_details = show_response["state_details"]
state_ids = show_response["state_ids"]
states = [
'discarded', 'empty', 'error', 'failed_metadata', 'new',
'ok', 'paused', 'queued', 'running', 'setting_metadata', 'upload'
]
assert isinstance(state_details, dict)
assert isinstance(state_ids, dict)
self._assert_has_keys(state_details, *states)
self._assert_has_keys(state_ids, *states)
def test_show_most_recently_used(self):
history_id = self._create_history("TestHistoryRecent")["id"]
show_response = self._get("histories/most_recently_used").json()
assert show_response["id"] == history_id
def test_index_order(self):
slightly_older_history_id = self._create_history("TestHistorySlightlyOlder")["id"]
newer_history_id = self._create_history("TestHistoryNewer")["id"]
index_response = self._get("histories").json()
assert index_response[0]["id"] == newer_history_id
assert index_response[1]["id"] == slightly_older_history_id
def test_delete(self):
# Setup a history and ensure it is in the index
history_id = self._create_history("TestHistoryForDelete")["id"]
index_response = self._get("histories").json()
assert index_response[0]["id"] == history_id
show_response = self._show(history_id)
assert not show_response["deleted"]
# Delete the history
self._delete("histories/%s" % history_id)
# Check can view it - but it is deleted
show_response = self._show(history_id)
assert show_response["deleted"]
# Verify it is dropped from history index
index_response = self._get("histories").json()
assert len(index_response) == 0 or index_response[0]["id"] != history_id
# Add deleted filter to index to view it
index_response = self._get("histories", {"deleted": "true"}).json()
assert index_response[0]["id"] == history_id
def test_purge(self):
history_id = self._create_history("TestHistoryForPurge")["id"]
data = {'purge': True}
self._delete("histories/%s" % history_id, data=data)
show_response = self._show(history_id)
assert show_response["deleted"]
assert show_response["purged"]
def test_undelete(self):
history_id = self._create_history("TestHistoryForDeleteAndUndelete")["id"]
self._delete("histories/%s" % history_id)
self._post("histories/deleted/%s/undelete" % history_id)
show_response = self._show(history_id)
assert not show_response["deleted"]
def test_update(self):
history_id = self._create_history("TestHistoryForUpdating")["id"]
self._update(history_id, {"name": "New Name"})
show_response = self._show(history_id)
assert show_response["name"] == "New Name"
unicode_name = u'桜ゲノム'
self._update(history_id, {"name": unicode_name})
show_response = self._show(history_id)
assert show_response["name"] == unicode_name, show_response
quoted_name = "'MooCow'"
self._update(history_id, {"name": quoted_name})
show_response = self._show(history_id)
assert show_response["name"] == quoted_name
self._update(history_id, {"deleted": True})
show_response = self._show(history_id)
assert show_response["deleted"], show_response
self._update(history_id, {"deleted": False})
show_response = self._show(history_id)
assert not show_response["deleted"]
self._update(history_id, {"published": True})
show_response = self._show(history_id)
assert show_response["published"]
self._update(history_id, {"genome_build": "hg18"})
show_response = self._show(history_id)
assert show_response["genome_build"] == "hg18"
self._update(history_id, {"annotation": "The annotation is cool"})
show_response = self._show(history_id)
assert show_response["annotation"] == "The annotation is cool"
self._update(history_id, {"annotation": unicode_name})
show_response = self._show(history_id)
assert show_response["annotation"] == unicode_name, show_response
self._update(history_id, {"annotation": quoted_name})
show_response = self._show(history_id)
assert show_response["annotation"] == quoted_name
def test_update_invalid_attribute(self):
history_id = self._create_history("TestHistoryForInvalidUpdating")["id"]
put_response = self._update(history_id, {"invalidkey": "moo"})
assert "invalidkey" not in put_response.json()
def test_update_invalid_types(self):
history_id = self._create_history("TestHistoryForUpdatingInvalidTypes")["id"]
for str_key in ["name", "annotation"]:
assert self._update(history_id, {str_key: False}).status_code == 400
for bool_key in ['deleted', 'importable', 'published']:
assert self._update(history_id, {bool_key: "a string"}).status_code == 400
assert self._update(history_id, {"tags": "a simple string"}).status_code == 400
assert self._update(history_id, {"tags": [True]}).status_code == 400
def test_invalid_keys(self):
invalid_history_id = "1234123412341234"
assert self._get("histories/%s" % invalid_history_id).status_code == 400
assert self._update(invalid_history_id, {"name": "new name"}).status_code == 400
assert self._delete("histories/%s" % invalid_history_id).status_code == 400
assert self._post("histories/deleted/%s/undelete" % invalid_history_id).status_code == 400
def test_create_anonymous_fails(self):
post_data = dict(name="CannotCreate")
# Using lower-level _api_url will cause key to not be injected.
histories_url = self._api_url("histories")
create_response = post(url=histories_url, data=post_data)
self._assert_status_code_is(create_response, 403)
def test_import_export(self):
history_name = "for_export_default"
history_id = self.dataset_populator.new_history(name=history_name)
self.dataset_populator.new_dataset(history_id, content="1 2 3")
deleted_hda = self.dataset_populator.new_dataset(history_id, content="1 2 3")
self.dataset_populator.delete_dataset(history_id, deleted_hda["id"])
deleted_details = self.dataset_populator.get_history_dataset_details(history_id, id=deleted_hda["id"])
assert deleted_details["deleted"]
imported_history_id = self._reimport_history(history_id, history_name, wait_on_history_length=2)
def upload_job_check(job):
assert job["tool_id"] == "upload1"
def check_discarded(hda):
assert hda["deleted"]
assert hda["state"] == "discarded", hda
assert hda["purged"] is True
self._check_imported_dataset(history_id=imported_history_id, hid=1, job_checker=upload_job_check)
self._check_imported_dataset(history_id=imported_history_id, hid=2, has_job=False, hda_checker=check_discarded, job_checker=upload_job_check)
imported_content = self.dataset_populator.get_history_dataset_content(
history_id=imported_history_id,
hid=1,
)
assert imported_content == "1 2 3\n"
def test_import_1901_histories(self):
f = open(self.test_data_resolver.get_filename("exports/1901_two_datasets.tgz"), 'rb')
import_data = dict(archive_source='', archive_file=f)
self._import_history_and_wait(import_data, "API Test History", wait_on_history_length=2)
def test_import_export_include_deleted(self):
history_name = "for_export_include_deleted"
history_id = self.dataset_populator.new_history(name=history_name)
self.dataset_populator.new_dataset(history_id, content="1 2 3")
deleted_hda = self.dataset_populator.new_dataset(history_id, content="1 2 3")
self.dataset_populator.delete_dataset(history_id, deleted_hda["id"])
imported_history_id = self._reimport_history(history_id, history_name, wait_on_history_length=2, export_kwds={"include_deleted": "True"})
self._assert_history_length(imported_history_id, 2)
def upload_job_check(job):
assert job["tool_id"] == "upload1"
def check_ok(hda):
assert hda["state"] == "ok", hda
assert hda["deleted"] is True, hda
self._check_imported_dataset(history_id=imported_history_id, hid=1, job_checker=upload_job_check)
self._check_imported_dataset(history_id=imported_history_id, hid=2, hda_checker=check_ok, job_checker=upload_job_check)
imported_content = self.dataset_populator.get_history_dataset_content(
history_id=imported_history_id,
hid=1,
)
assert imported_content == "1 2 3\n"
def test_import_metadata_regeneration(self):
history_name = "for_import_metadata_regeneration"
history_id = self.dataset_populator.new_history(name=history_name)
self.dataset_populator.new_dataset(history_id, content=open(self.test_data_resolver.get_filename("1.bam"), 'rb'), file_type='bam', wait=True)
imported_history_id = self._reimport_history(history_id, history_name)
self._assert_history_length(imported_history_id, 1)
self._check_imported_dataset(history_id=imported_history_id, hid=1)
import_bam_metadata = self.dataset_populator.get_history_dataset_details(
history_id=imported_history_id,
hid=1,
)
# The cleanup() method of the __IMPORT_HISTORY__ job (which is executed
# after the job has entered its final state):
# - creates a new dataset with 'ok' state and adds it to the history
# - starts a __SET_METADATA__ job to regenerate the dataset metadata, if
# needed
# We need to wait a bit for the creation of the __SET_METADATA__ job.
time.sleep(1)
self.dataset_populator.wait_for_history_jobs(imported_history_id, assert_ok=True)
bai_metadata = import_bam_metadata["meta_files"][0]
assert bai_metadata["file_type"] == "bam_index"
api_url = bai_metadata["download_url"].split("api/", 1)[1]
bai_response = self._get(api_url)
self._assert_status_code_is(bai_response, 200)
assert len(bai_response.content) > 4
def test_import_export_collection(self):
history_name = "for_export_with_collections"
history_id = self.dataset_populator.new_history(name=history_name)
self.dataset_collection_populator.create_list_in_history(history_id, contents=["Hello", "World"], direct_upload=True)
imported_history_id = self._reimport_history(history_id, history_name, wait_on_history_length=3)
self._assert_history_length(imported_history_id, 3)
def check_elements(elements):
assert len(elements) == 2
element0 = elements[0]["object"]
element1 = elements[1]["object"]
for element in [element0, element1]:
assert not element["visible"]
assert not element["deleted"]
assert element["state"] == "ok"
assert element0["hid"] == 2
assert element1["hid"] == 3
self._check_imported_collection(imported_history_id, hid=1, collection_type="list", elements_checker=check_elements)
def test_import_export_nested_collection(self):
history_name = "for_export_with_nested_collections"
history_id = self.dataset_populator.new_history(name=history_name)
self.dataset_collection_populator.create_list_of_pairs_in_history(history_id)
imported_history_id = self._reimport_history(history_id, history_name, wait_on_history_length=3)
self._assert_history_length(imported_history_id, 3)
def check_elements(elements):
assert len(elements) == 1
element0 = elements[0]["object"]
self._assert_has_keys(element0, "elements", "collection_type")
child_elements = element0["elements"]
assert len(child_elements) == 2
assert element0["collection_type"] == "paired"
self._check_imported_collection(imported_history_id, hid=1, collection_type="list:paired", elements_checker=check_elements)
def _reimport_history(self, history_id, history_name, wait_on_history_length=None, export_kwds={}):
# Ensure the history is ready to go...
self.dataset_populator.wait_for_history(history_id, assert_ok=True)
return self.dataset_populator.reimport_history(
history_id, history_name, wait_on_history_length=wait_on_history_length, export_kwds=export_kwds, url=self.url, api_key=self.galaxy_interactor.api_key
)
def _import_history_and_wait(self, import_data, history_name, wait_on_history_length=None):
imported_history_id = self.dataset_populator.import_history_and_wait_for_name(import_data, history_name)
if wait_on_history_length:
self.dataset_populator.wait_on_history_length(imported_history_id, wait_on_history_length)
return imported_history_id
def _assert_history_length(self, history_id, n):
contents_response = self._get("histories/%s/contents" % history_id)
self._assert_status_code_is(contents_response, 200)
contents = contents_response.json()
assert len(contents) == n, contents
def _check_imported_dataset(self, history_id, hid, has_job=True, hda_checker=None, job_checker=None):
imported_dataset_metadata = self.dataset_populator.get_history_dataset_details(
history_id=history_id,
hid=hid,
)
assert imported_dataset_metadata["history_content_type"] == "dataset"
assert imported_dataset_metadata["history_id"] == history_id
if hda_checker is not None:
hda_checker(imported_dataset_metadata)
assert "creating_job" in imported_dataset_metadata
job_id = imported_dataset_metadata["creating_job"]
if has_job:
assert job_id
job_details = self.dataset_populator.get_job_details(job_id, full=True)
assert job_details.status_code == 200, job_details.content
job = job_details.json()
assert 'history_id' in job, job
assert job['history_id'] == history_id, job
if job_checker is not None:
job_checker(job)
def _check_imported_collection(self, history_id, hid, collection_type=None, elements_checker=None):
imported_collection_metadata = self.dataset_populator.get_history_collection_details(
history_id=history_id,
hid=hid,
)
assert imported_collection_metadata["history_content_type"] == "dataset_collection"
assert imported_collection_metadata["history_id"] == history_id
assert "collection_type" in imported_collection_metadata
assert "elements" in imported_collection_metadata
if collection_type is not None:
assert imported_collection_metadata["collection_type"] == collection_type, imported_collection_metadata
if elements_checker is not None:
elements_checker(imported_collection_metadata["elements"])
def test_create_tag(self):
post_data = dict(name="TestHistoryForTag")
history_id = self._post("histories", data=post_data).json()["id"]
tag_data = dict(value="awesometagvalue")
tag_url = "histories/%s/tags/awesometagname" % history_id
tag_create_response = self._post(tag_url, data=tag_data)
self._assert_status_code_is(tag_create_response, 200)
def _show(self, history_id):
return self._get("histories/%s" % history_id).json()
def _update(self, history_id, data):
update_url = self._api_url("histories/%s" % history_id, use_key=True)
put_response = put(update_url, json=data)
return put_response
def _create_history(self, name):
post_data = dict(name=name)
create_response = self._post("histories", data=post_data).json()
self._assert_has_keys(create_response, "name", "id")
self.assertEqual(create_response["name"], name)
return create_response
class ObjectStoreJobsIntegrationTestCase(integration_util.IntegrationTestCase):
framework_tool_and_types = True
@classmethod
def handle_galaxy_config_kwds(cls, config):
temp_directory = cls._test_driver.mkdtemp()
cls.object_stores_parent = temp_directory
for disk_store_file_name in ["files1", "files2", "files3"]:
disk_store_path = os.path.join(temp_directory,
disk_store_file_name)
os.makedirs(disk_store_path)
setattr(cls, "%s_path" % disk_store_file_name, disk_store_path)
config_path = os.path.join(temp_directory, "object_store_conf.xml")
with open(config_path, "w") as f:
f.write(
DISTRIBUTED_OBJECT_STORE_CONFIG_TEMPLATE.safe_substitute(
{"temp_directory": temp_directory}))
config["object_store_config_file"] = config_path
def setUp(self):
super(ObjectStoreJobsIntegrationTestCase, self).setUp()
self.dataset_populator = DatasetPopulator(self.galaxy_interactor)
def test_tool_simple_constructs(self):
with self.dataset_populator.test_history() as history_id:
hda1 = self.dataset_populator.new_dataset(history_id,
content="1 2 3")
create_10_inputs = {
"input1": {
"src": "hda",
"id": hda1["id"]
},
"input2": {
"src": "hda",
"id": hda1["id"]
},
}
self.dataset_populator.run_tool(
"create_10",
create_10_inputs,
history_id,
assert_ok=True,
)
self.dataset_populator.wait_for_history(history_id)
files_1_count = _files_count(self.files1_path)
files_2_count = _files_count(self.files2_path)
files_3_count = _files_count(self.files3_path)
# Ensure no files written to the secondary/inactive hierarchical disk store.
assert files_3_count == 0
# Ensure the 10 inputs were written to one of the distributed object store's disk
# stores (it will have either 10 or 11 depeending on whether the input was also
# written there. The other disk store may or may not have the input file so should
# have at most one file.
assert (files_1_count >= 10) or (files_2_count >= 10)
assert (files_1_count <= 1) or (files_2_count <= 1)
# Other sanity checks on the test - just make sure the test was setup as intended
# and not actually testing object store behavior.
assert (files_1_count <= 11) and (files_2_count <= 11)
assert (files_1_count >= 0) and (files_2_count >= 0)
class JobsApiTestCase(api.ApiTestCase):
def setUp(self):
super(JobsApiTestCase, self).setUp()
self.dataset_populator = DatasetPopulator(self.galaxy_interactor)
def test_index(self):
# Create HDA to ensure at least one job exists...
self.__history_with_new_dataset()
jobs = self.__jobs_index()
assert "upload1" in map(itemgetter("tool_id"), jobs)
def test_system_details_admin_only(self):
self.__history_with_new_dataset()
jobs = self.__jobs_index(admin=False)
job = jobs[0]
self._assert_not_has_keys(job, "command_line", "external_id")
jobs = self.__jobs_index(admin=True)
job = jobs[0]
self._assert_has_keys(job, "command_line", "external_id")
def test_index_state_filter(self):
# Initial number of ok jobs
original_count = len(self.__uploads_with_state("ok"))
# Run through dataset upload to ensure num uplaods at least greater
# by 1.
self.__history_with_ok_dataset()
# Verify number of ok jobs is actually greater.
count_increased = False
for i in range(10):
new_count = len(self.__uploads_with_state("ok"))
if original_count < new_count:
count_increased = True
break
time.sleep(.1)
if not count_increased:
template = "Jobs in ok state did not increase (was %d, now %d)"
message = template % (original_count, new_count)
raise AssertionError(message)
def test_index_date_filter(self):
self.__history_with_new_dataset()
two_weeks_ago = (datetime.datetime.utcnow() - datetime.timedelta(7)).isoformat()
last_week = (datetime.datetime.utcnow() - datetime.timedelta(7)).isoformat()
next_week = (datetime.datetime.utcnow() + datetime.timedelta(7)).isoformat()
today = datetime.datetime.utcnow().isoformat()
tomorrow = (datetime.datetime.utcnow() + datetime.timedelta(1)).isoformat()
jobs = self.__jobs_index(data={"date_range_min": today[0:10], "date_range_max": tomorrow[0:10]})
assert len(jobs) > 0
today_job_id = jobs[0]["id"]
jobs = self.__jobs_index(data={"date_range_min": two_weeks_ago, "date_range_max": last_week})
assert today_job_id not in map(itemgetter("id"), jobs)
jobs = self.__jobs_index(data={"date_range_min": last_week, "date_range_max": next_week})
assert today_job_id in map(itemgetter("id"), jobs)
def test_index_history(self):
history_id, _ = self.__history_with_new_dataset()
jobs = self.__jobs_index(data={"history_id": history_id})
assert len(jobs) > 0
history_id = self.dataset_populator.new_history()
jobs = self.__jobs_index(data={"history_id": history_id})
assert len(jobs) == 0
def test_index_multiple_states_filter(self):
# Initial number of ok jobs
original_count = len(self.__uploads_with_state("ok", "new"))
# Run through dataset upload to ensure num uplaods at least greater
# by 1.
self.__history_with_ok_dataset()
# Verify number of ok jobs is actually greater.
new_count = len(self.__uploads_with_state("new", "ok"))
assert original_count < new_count, new_count
def test_show(self):
# Create HDA to ensure at least one job exists...
self.__history_with_new_dataset()
jobs_response = self._get("jobs")
first_job = jobs_response.json()[0]
self._assert_has_key(first_job, 'id', 'state', 'exit_code', 'update_time', 'create_time')
job_id = first_job["id"]
show_jobs_response = self._get("jobs/%s" % job_id)
self._assert_status_code_is(show_jobs_response, 200)
job_details = show_jobs_response.json()
self._assert_has_key(job_details, 'id', 'state', 'exit_code', 'update_time', 'create_time')
def test_show_security(self):
history_id, _ = self.__history_with_new_dataset()
jobs_response = self._get("jobs", data={"history_id": history_id})
job = jobs_response.json()[0]
job_id = job["id"]
show_jobs_response = self._get("jobs/%s" % job_id, admin=False)
self._assert_not_has_keys(show_jobs_response.json(), "command_line", "external_id")
# TODO: Re-activate test case when API accepts privacy settings
# with self._different_user():
# show_jobs_response = self._get( "jobs/%s" % job_id, admin=False )
# self._assert_status_code_is( show_jobs_response, 200 )
show_jobs_response = self._get("jobs/%s" % job_id, admin=True)
self._assert_has_keys(show_jobs_response.json(), "command_line", "external_id")
def test_deleting_output_keep_running_until_all_deleted(self):
history_id, job_state, outputs = self._setup_running_two_output_job(120)
self._hack_to_skip_test_if_state_ok(job_state)
# Delete one of the two outputs and make sure the job is still running.
self._raw_update_history_item(history_id, outputs[0]["id"], {"deleted": True})
self._hack_to_skip_test_if_state_ok(job_state)
time.sleep(1)
self._hack_to_skip_test_if_state_ok(job_state)
state = job_state().json()["state"]
assert state == "running", state
# Delete the second output and make sure the job is cancelled.
self._raw_update_history_item(history_id, outputs[1]["id"], {"deleted": True})
final_state = wait_on_state(job_state, assert_ok=False, timeout=15)
assert final_state in ["deleted_new", "deleted"], final_state
def test_purging_output_keep_running_until_all_purged(self):
history_id, job_state, outputs = self._setup_running_two_output_job(120)
# Pretty much right away after the job is running, these paths should be populated -
# if they are grab them and make sure they are deleted at the end of the job.
dataset_1 = self._get_history_item_as_admin(history_id, outputs[0]["id"])
dataset_2 = self._get_history_item_as_admin(history_id, outputs[1]["id"])
if "file_name" in dataset_1:
output_dataset_paths = [dataset_1["file_name"], dataset_2["file_name"]]
# This may or may not exist depending on if the test is local or not.
output_dataset_paths_exist = os.path.exists(output_dataset_paths[0])
else:
output_dataset_paths = []
output_dataset_paths_exist = False
self._hack_to_skip_test_if_state_ok(job_state)
current_state = job_state().json()["state"]
assert current_state == "running", current_state
# Purge one of the two outputs and make sure the job is still running.
self._raw_update_history_item(history_id, outputs[0]["id"], {"purged": True})
time.sleep(1)
self._hack_to_skip_test_if_state_ok(job_state)
current_state = job_state().json()["state"]
assert current_state == "running", current_state
# Purge the second output and make sure the job is cancelled.
self._raw_update_history_item(history_id, outputs[1]["id"], {"purged": True})
final_state = wait_on_state(job_state, assert_ok=False, timeout=15)
assert final_state in ["deleted_new", "deleted"], final_state
def paths_deleted():
if not os.path.exists(output_dataset_paths[0]) and not os.path.exists(output_dataset_paths[1]):
return True
if output_dataset_paths_exist:
wait_on(paths_deleted, "path deletion")
def test_purging_output_cleaned_after_ok_run(self):
history_id, job_state, outputs = self._setup_running_two_output_job(10)
# Pretty much right away after the job is running, these paths should be populated -
# if they are grab them and make sure they are deleted at the end of the job.
dataset_1 = self._get_history_item_as_admin(history_id, outputs[0]["id"])
dataset_2 = self._get_history_item_as_admin(history_id, outputs[1]["id"])
if "file_name" in dataset_1:
output_dataset_paths = [dataset_1["file_name"], dataset_2["file_name"]]
# This may or may not exist depending on if the test is local or not.
output_dataset_paths_exist = os.path.exists(output_dataset_paths[0])
else:
output_dataset_paths = []
output_dataset_paths_exist = False
if not output_dataset_paths_exist:
# Given this Galaxy configuration - there is nothing more to be tested here.
# Consider throwing a skip instead.
return
# Purge one of the two outputs and wait for the job to complete.
self._raw_update_history_item(history_id, outputs[0]["id"], {"purged": True})
wait_on_state(job_state, assert_ok=True)
if output_dataset_paths_exist:
time.sleep(.5)
# Make sure the non-purged dataset is on disk and the purged one is not.
assert os.path.exists(output_dataset_paths[1])
assert not os.path.exists(output_dataset_paths[0])
def _hack_to_skip_test_if_state_ok(self, job_state):
from nose.plugins.skip import SkipTest
if job_state().json()["state"] == "ok":
message = "Job state switch from running to ok too quickly - the rest of the test requires the job to be in a running state. Skipping test."
raise SkipTest(message)
def _setup_running_two_output_job(self, sleep_time):
history_id = self.dataset_populator.new_history()
payload = self.dataset_populator.run_tool_payload(
tool_id='create_2',
inputs=dict(
sleep_time=sleep_time,
),
history_id=history_id,
)
run_response = self._post("tools", data=payload).json()
outputs = run_response["outputs"]
jobs = run_response["jobs"]
assert len(outputs) == 2
assert len(jobs) == 1
def job_state():
jobs_response = self._get("jobs/%s" % jobs[0]["id"])
return jobs_response
# Give job some time to get up and running.
time.sleep(2)
running_state = wait_on_state(job_state, skip_states=["queued", "new"], assert_ok=False, timeout=15)
assert running_state == "running", running_state
def job_state():
jobs_response = self._get("jobs/%s" % jobs[0]["id"])
return jobs_response
return history_id, job_state, outputs
def _raw_update_history_item(self, history_id, item_id, data):
update_url = self._api_url("histories/%s/contents/%s" % (history_id, item_id), use_key=True)
update_response = put(update_url, json=data)
assert_status_code_is_ok(update_response)
return update_response
def _get_history_item_as_admin(self, history_id, item_id):
response = self._get("histories/%s/contents/%s?view=detailed" % (history_id, item_id), admin=True)
assert_status_code_is_ok(response)
return response.json()
def test_search(self):
history_id, dataset_id = self.__history_with_ok_dataset()
inputs = json.dumps(
dict(
input1=dict(
src='hda',
id=dataset_id,
)
)
)
search_payload = dict(
tool_id="cat1",
inputs=inputs,
state="ok",
)
empty_search_response = self._post("jobs/search", data=search_payload)
self._assert_status_code_is(empty_search_response, 200)
self.assertEquals(len(empty_search_response.json()), 0)
self.__run_cat_tool(history_id, dataset_id)
self.dataset_populator.wait_for_history(history_id, assert_ok=True)
search_count = -1
# in case job and history aren't updated at exactly the same
# time give time to wait
for i in range(5):
search_count = self._search_count(search_payload)
if search_count == 1:
break
time.sleep(.1)
self.assertEquals(search_count, 1)
def _search_count(self, search_payload):
search_response = self._post("jobs/search", data=search_payload)
self._assert_status_code_is(search_response, 200)
search_json = search_response.json()
return len(search_json)
def __run_cat_tool(self, history_id, dataset_id):
# Code duplication with test_jobs.py, eliminate
payload = self.dataset_populator.run_tool_payload(
tool_id='cat1',
inputs=dict(
input1=dict(
src='hda',
id=dataset_id
),
),
history_id=history_id,
)
self._post("tools", data=payload)
def __run_randomlines_tool(self, lines, history_id, dataset_id):
payload = self.dataset_populator.run_tool_payload(
tool_id="random_lines1",
inputs=dict(
num_lines=lines,
input=dict(
src='hda',
id=dataset_id,
),
),
history_id=history_id,
)
self._post("tools", data=payload)
def __uploads_with_state(self, *states):
jobs_response = self._get("jobs", data=dict(state=states))
self._assert_status_code_is(jobs_response, 200)
jobs = jobs_response.json()
assert not filter(lambda j: j["state"] not in states, jobs)
return filter(lambda j: j["tool_id"] == "upload1", jobs)
def __history_with_new_dataset(self):
history_id = self.dataset_populator.new_history()
dataset_id = self.dataset_populator.new_dataset(history_id)["id"]
return history_id, dataset_id
def __history_with_ok_dataset(self):
history_id = self.dataset_populator.new_history()
dataset_id = self.dataset_populator.new_dataset(history_id, wait=True)["id"]
return history_id, dataset_id
def __jobs_index(self, **kwds):
jobs_response = self._get("jobs", **kwds)
self._assert_status_code_is(jobs_response, 200)
jobs = jobs_response.json()
assert isinstance(jobs, list)
return jobs
class DockerizedJobsIntegrationTestCase(integration_util.IntegrationTestCase,
RunsEnvironmentJobs):
framework_tool_and_types = True
@classmethod
def handle_galaxy_config_kwds(cls, config):
cls.jobs_directory = tempfile.mkdtemp()
config["jobs_directory"] = cls.jobs_directory
config["job_config_file"] = DOCKERIZED_JOB_CONFIG_FILE
# Disable tool dependency resolution.
config["tool_dependency_dir"] = "none"
config["enable_beta_mulled_containers"] = "true"
def setUp(self):
super(DockerizedJobsIntegrationTestCase, self).setUp()
self.dataset_populator = DatasetPopulator(self.galaxy_interactor)
self.history_id = self.dataset_populator.new_history()
def test_explicit(self):
self.dataset_populator.run_tool("mulled_example_explicit", {},
self.history_id)
self.dataset_populator.wait_for_history(self.history_id,
assert_ok=True)
output = self.dataset_populator.get_history_dataset_content(
self.history_id)
assert "0.7.15-r1140" in output
def test_mulled_simple(self):
self.dataset_populator.run_tool("mulled_example_simple", {},
self.history_id)
self.dataset_populator.wait_for_history(self.history_id,
assert_ok=True)
output = self.dataset_populator.get_history_dataset_content(
self.history_id)
assert "0.7.15-r1140" in output
def test_docker_job_enviornment(self):
job_env = self._run_and_get_environment_properties(
"job_environment_default")
euid = os.geteuid()
egid = os.getgid()
assert job_env.user_id == str(euid), job_env.user_id
assert job_env.group_id == str(egid), job_env.group_id
assert job_env.pwd.startswith(self.jobs_directory)
assert job_env.pwd.endswith("/working")
assert job_env.home.startswith(self.jobs_directory)
assert job_env.home.endswith("/home")
def test_docker_job_environment_legacy(self):
job_env = self._run_and_get_environment_properties(
"job_environment_default_legacy")
euid = os.geteuid()
egid = os.getgid()
assert job_env.user_id == str(euid), job_env.user_id
assert job_env.group_id == str(egid), job_env.group_id
assert job_env.pwd.startswith(self.jobs_directory)
assert job_env.pwd.endswith("/working")
# Should we change env_pass_through to just always include TMP and HOME for docker?
# I'm not sure, if yes this would change.
assert job_env.home == "/", job_env.home
class ToolsUploadTestCase(api.ApiTestCase):
def setUp(self):
super(ToolsUploadTestCase, self).setUp()
self.dataset_populator = DatasetPopulator(self.galaxy_interactor)
def test_upload1_paste(self):
with self.dataset_populator.test_history() as history_id:
payload = self.dataset_populator.upload_payload(history_id, 'Hello World')
create_response = self._post("tools", data=payload)
self._assert_has_keys(create_response.json(), 'outputs')
def test_upload1_paste_bad_datatype(self):
# Check that you get a nice message if you upload an incorrect datatype
with self.dataset_populator.test_history() as history_id:
file_type = "johnsawesomebutfakedatatype"
payload = self.dataset_populator.upload_payload(history_id, 'Hello World', file_type=file_type)
create = self._post("tools", data=payload).json()
self._assert_has_keys(create, 'err_msg')
assert file_type in create['err_msg']
# upload1 rewrites content with posix lines by default but this can be disabled by setting
# to_posix_lines=None in the request. Newer fetch API does not do this by default prefering
# to keep content unaltered if possible but it can be enabled with a simple JSON boolean switch
# of the same name (to_posix_lines).
def test_upload_posix_newline_fixes_by_default(self):
windows_content = ONE_TO_SIX_ON_WINDOWS
result_content = self._upload_and_get_content(windows_content)
self.assertEquals(result_content, ONE_TO_SIX_WITH_TABS)
def test_fetch_posix_unaltered(self):
windows_content = ONE_TO_SIX_ON_WINDOWS
result_content = self._upload_and_get_content(windows_content, api="fetch")
self.assertEquals(result_content, ONE_TO_SIX_ON_WINDOWS)
def test_upload_disable_posix_fix(self):
windows_content = ONE_TO_SIX_ON_WINDOWS
result_content = self._upload_and_get_content(windows_content, to_posix_lines=None)
self.assertEquals(result_content, windows_content)
def test_fetch_post_lines_option(self):
windows_content = ONE_TO_SIX_ON_WINDOWS
result_content = self._upload_and_get_content(windows_content, api="fetch", to_posix_lines=True)
self.assertEquals(result_content, ONE_TO_SIX_WITH_TABS)
def test_upload_tab_to_space_off_by_default(self):
table = ONE_TO_SIX_WITH_SPACES
result_content = self._upload_and_get_content(table)
self.assertEquals(result_content, table)
def test_fetch_tab_to_space_off_by_default(self):
table = ONE_TO_SIX_WITH_SPACES
result_content = self._upload_and_get_content(table, api='fetch')
self.assertEquals(result_content, table)
def test_upload_tab_to_space(self):
table = ONE_TO_SIX_WITH_SPACES
result_content = self._upload_and_get_content(table, space_to_tab="Yes")
self.assertEquals(result_content, ONE_TO_SIX_WITH_TABS)
def test_fetch_tab_to_space(self):
table = ONE_TO_SIX_WITH_SPACES
result_content = self._upload_and_get_content(table, api="fetch", space_to_tab=True)
self.assertEquals(result_content, ONE_TO_SIX_WITH_TABS)
def test_fetch_compressed_with_explicit_type(self):
fastqgz_path = TestDataResolver().get_filename("1.fastqsanger.gz")
with open(fastqgz_path, "rb") as fh:
details = self._upload_and_get_details(fh, api="fetch", ext="fastqsanger.gz")
assert details["state"] == "ok"
assert details["file_ext"] == "fastqsanger.gz"
def test_fetch_compressed_default(self):
fastqgz_path = TestDataResolver().get_filename("1.fastqsanger.gz")
with open(fastqgz_path, "rb") as fh:
details = self._upload_and_get_details(fh, api="fetch", assert_ok=False)
assert details["state"] == "ok"
assert details["file_ext"] == "fastqsanger.gz", details
@uses_test_history(require_new=True)
def test_fetch_compressed_auto_decompress_target(self, history_id):
# TODO: this should definitely be fixed to allow auto decompression via that API.
fastqgz_path = TestDataResolver().get_filename("1.fastqsanger.gz")
with open(fastqgz_path, "rb") as fh:
details = self._upload_and_get_details(fh,
api="fetch",
history_id=history_id,
assert_ok=False,
auto_decompress=True)
assert details["state"] == "ok"
assert details["file_ext"] == "fastqsanger.gz", details
def test_upload_decompress_off_with_auto_by_default(self):
# UNSTABLE_FLAG: This might default to a bed.gz datatype in the future.
bedgz_path = TestDataResolver().get_filename("4.bed.gz")
with open(bedgz_path, "rb") as fh:
details = self._upload_and_get_details(fh, file_type="auto")
assert details["state"] == "ok"
assert details["file_ext"] == "bed", details
def test_upload_decompresses_if_uncompressed_type_selected(self):
fastqgz_path = TestDataResolver().get_filename("1.fastqsanger.gz")
with open(fastqgz_path, "rb") as fh:
details = self._upload_and_get_details(fh, file_type="fastqsanger")
assert details["state"] == "ok"
assert details["file_ext"] == "fastqsanger", details
assert details["file_size"] == 178, details
def test_upload_decompress_off_if_compressed_type_selected(self):
fastqgz_path = TestDataResolver().get_filename("1.fastqsanger.gz")
with open(fastqgz_path, "rb") as fh:
details = self._upload_and_get_details(fh, file_type="fastqsanger.gz")
assert details["state"] == "ok"
assert details["file_ext"] == "fastqsanger.gz", details
assert details["file_size"] == 161, details
def test_upload_auto_decompress_off(self):
# UNSTABLE_FLAG: This might default to a bed.gz datatype in the future.
bedgz_path = TestDataResolver().get_filename("4.bed.gz")
with open(bedgz_path, "rb") as fh:
details = self._upload_and_get_details(fh, file_type="auto", assert_ok=False, auto_decompress=False)
assert details["file_ext"] == "binary", details
@uses_test_history(require_new=True)
def test_fetch_compressed_with_auto(self, history_id):
# UNSTABLE_FLAG: This might default to a bed.gz datatype in the future.
# TODO: this should definitely be fixed to allow auto decompression via that API.
bedgz_path = TestDataResolver().get_filename("4.bed.gz")
with open(bedgz_path, "rb") as fh:
details = self._upload_and_get_details(fh,
api="fetch",
history_id=history_id,
auto_decompress=True,
assert_ok=False)
assert details["state"] == "ok"
assert details["file_ext"] == "bed"
@skip_without_datatype("rdata")
def test_rdata_not_decompressed(self):
# Prevent regression of https://github.com/galaxyproject/galaxy/issues/753
rdata_path = TestDataResolver().get_filename("1.RData")
with open(rdata_path, "rb") as fh:
rdata_metadata = self._upload_and_get_details(fh, file_type="auto")
self.assertEquals(rdata_metadata["file_ext"], "rdata")
@skip_without_datatype("csv")
def test_csv_upload(self):
csv_path = TestDataResolver().get_filename("1.csv")
with open(csv_path, "rb") as fh:
csv_metadata = self._upload_and_get_details(fh, file_type="csv")
self.assertEquals(csv_metadata["file_ext"], "csv")
@skip_without_datatype("csv")
def test_csv_upload_auto(self):
csv_path = TestDataResolver().get_filename("1.csv")
with open(csv_path, "rb") as fh:
csv_metadata = self._upload_and_get_details(fh, file_type="auto")
self.assertEquals(csv_metadata["file_ext"], "csv")
@skip_without_datatype("csv")
def test_csv_fetch(self):
csv_path = TestDataResolver().get_filename("1.csv")
with open(csv_path, "rb") as fh:
csv_metadata = self._upload_and_get_details(fh, api="fetch", ext="csv", to_posix_lines=True)
self.assertEquals(csv_metadata["file_ext"], "csv")
@skip_without_datatype("csv")
def test_csv_sniff_fetch(self):
csv_path = TestDataResolver().get_filename("1.csv")
with open(csv_path, "rb") as fh:
csv_metadata = self._upload_and_get_details(fh, api="fetch", ext="auto", to_posix_lines=True)
self.assertEquals(csv_metadata["file_ext"], "csv")
@skip_without_datatype("tiff")
def test_image_upload_auto(self):
tiff_path = TestDataResolver().get_filename("1.tiff")
with open(tiff_path, "rb") as fh:
tiff_metadata = self._upload_and_get_details(fh, file_type="auto")
self.assertEquals(tiff_metadata["file_ext"], "tiff")
@skip_without_datatype("velvet")
def test_composite_datatype(self):
with self.dataset_populator.test_history() as history_id:
dataset = self._velvet_upload(history_id, extra_inputs={
"files_1|url_paste": "roadmaps content",
"files_1|type": "upload_dataset",
"files_2|url_paste": "log content",
"files_2|type": "upload_dataset",
})
roadmaps_content = self._get_roadmaps_content(history_id, dataset)
assert roadmaps_content.strip() == "roadmaps content", roadmaps_content
@skip_without_datatype("velvet")
def test_composite_datatype_space_to_tab(self):
# Like previous test but set one upload with space_to_tab to True to
# verify that works.
with self.dataset_populator.test_history() as history_id:
dataset = self._velvet_upload(history_id, extra_inputs={
"files_1|url_paste": "roadmaps content",
"files_1|type": "upload_dataset",
"files_1|space_to_tab": "Yes",
"files_2|url_paste": "log content",
"files_2|type": "upload_dataset",
})
roadmaps_content = self._get_roadmaps_content(history_id, dataset)
assert roadmaps_content.strip() == "roadmaps\tcontent", roadmaps_content
@skip_without_datatype("velvet")
def test_composite_datatype_posix_lines(self):
# Like previous test but set one upload with space_to_tab to True to
# verify that works.
with self.dataset_populator.test_history() as history_id:
dataset = self._velvet_upload(history_id, extra_inputs={
"files_1|url_paste": "roadmaps\rcontent",
"files_1|type": "upload_dataset",
"files_1|space_to_tab": "Yes",
"files_2|url_paste": "log\rcontent",
"files_2|type": "upload_dataset",
})
roadmaps_content = self._get_roadmaps_content(history_id, dataset)
assert roadmaps_content.strip() == "roadmaps\ncontent", roadmaps_content
def test_upload_dbkey(self):
with self.dataset_populator.test_history() as history_id:
payload = self.dataset_populator.upload_payload(history_id, "Test123", dbkey="hg19")
run_response = self.dataset_populator.tools_post(payload)
self.dataset_populator.wait_for_tool_run(history_id, run_response)
datasets = run_response.json()["outputs"]
assert datasets[0].get("genome_build") == "hg19", datasets[0]
@uses_test_history(require_new=False)
def test_fetch_bam_file(self, history_id):
bam_path = TestDataResolver().get_filename("1.bam")
with open(bam_path, "rb") as fh:
details = self._upload_and_get_details(fh,
api="fetch",
history_id=history_id,
assert_ok=False)
assert details["state"] == "ok"
assert details["file_ext"] == "bam", details
def test_upload_bam_file(self):
bam_path = TestDataResolver().get_filename("1.bam")
with open(bam_path, "rb") as fh:
details = self._upload_and_get_details(fh, file_type="auto")
assert details["state"] == "ok"
assert details["file_ext"] == "bam", details
def test_fetch_metadata(self):
table = ONE_TO_SIX_WITH_SPACES
details = self._upload_and_get_details(table, api='fetch', dbkey="hg19", info="cool upload", tags=["name:data", "group:type:paired-end"])
assert details.get("genome_build") == "hg19"
assert details.get("misc_info") == "cool upload", details
tags = details.get("tags")
assert len(tags) == 2, details
assert "group:type:paired-end" in tags
assert "name:data" in tags
def test_upload_multiple_files_1(self):
with self.dataset_populator.test_history() as history_id:
payload = self.dataset_populator.upload_payload(history_id, "Test123",
dbkey="hg19",
extra_inputs={
"files_1|url_paste": "SecondOutputContent",
"files_1|NAME": "SecondOutputName",
"files_1|file_type": "tabular",
"files_1|dbkey": "hg18",
"file_count": "2",
}
)
run_response = self.dataset_populator.tools_post(payload)
self.dataset_populator.wait_for_tool_run(history_id, run_response)
datasets = run_response.json()["outputs"]
assert len(datasets) == 2, datasets
content = self.dataset_populator.get_history_dataset_content(history_id, dataset=datasets[0])
assert content.strip() == "Test123"
assert datasets[0]["file_ext"] == "txt"
assert datasets[0]["genome_build"] == "hg19", datasets
content = self.dataset_populator.get_history_dataset_content(history_id, dataset=datasets[1])
assert content.strip() == "SecondOutputContent"
assert datasets[1]["file_ext"] == "tabular"
assert datasets[1]["genome_build"] == "hg18", datasets
def test_upload_multiple_files_2(self):
with self.dataset_populator.test_history() as history_id:
payload = self.dataset_populator.upload_payload(history_id, "Test123",
file_type="tabular",
dbkey="hg19",
extra_inputs={
"files_1|url_paste": "SecondOutputContent",
"files_1|NAME": "SecondOutputName",
"files_1|file_type": "txt",
"files_1|dbkey": "hg18",
"file_count": "2",
}
)
run_response = self.dataset_populator.tools_post(payload)
self.dataset_populator.wait_for_tool_run(history_id, run_response)
datasets = run_response.json()["outputs"]
assert len(datasets) == 2, datasets
content = self.dataset_populator.get_history_dataset_content(history_id, dataset=datasets[0])
assert content.strip() == "Test123"
assert datasets[0]["file_ext"] == "tabular", datasets
assert datasets[0]["genome_build"] == "hg19", datasets
content = self.dataset_populator.get_history_dataset_content(history_id, dataset=datasets[1])
assert content.strip() == "SecondOutputContent"
assert datasets[1]["file_ext"] == "txt"
assert datasets[1]["genome_build"] == "hg18", datasets
def test_upload_multiple_files_3(self):
with self.dataset_populator.test_history() as history_id:
payload = self.dataset_populator.upload_payload(history_id, "Test123",
file_type="tabular",
dbkey="hg19",
extra_inputs={
"files_0|file_type": "txt",
"files_0|dbkey": "hg18",
"files_1|url_paste": "SecondOutputContent",
"files_1|NAME": "SecondOutputName",
"files_1|file_type": "txt",
"files_1|dbkey": "hg18",
"file_count": "2",
}
)
run_response = self.dataset_populator.tools_post(payload)
self.dataset_populator.wait_for_tool_run(history_id, run_response)
datasets = run_response.json()["outputs"]
assert len(datasets) == 2, datasets
content = self.dataset_populator.get_history_dataset_content(history_id, dataset=datasets[0])
assert content.strip() == "Test123"
assert datasets[0]["file_ext"] == "txt", datasets
assert datasets[0]["genome_build"] == "hg18", datasets
content = self.dataset_populator.get_history_dataset_content(history_id, dataset=datasets[1])
assert content.strip() == "SecondOutputContent"
assert datasets[1]["file_ext"] == "txt"
assert datasets[1]["genome_build"] == "hg18", datasets
def test_upload_multiple_files_no_dbkey(self):
with self.dataset_populator.test_history() as history_id:
payload = self.dataset_populator.upload_payload(history_id, "Test123",
file_type="tabular",
dbkey=None,
extra_inputs={
"files_0|file_type": "txt",
"files_1|url_paste": "SecondOutputContent",
"files_1|NAME": "SecondOutputName",
"files_1|file_type": "txt",
"file_count": "2",
}
)
run_response = self.dataset_populator.tools_post(payload)
self.dataset_populator.wait_for_tool_run(history_id, run_response)
datasets = run_response.json()["outputs"]
assert len(datasets) == 2, datasets
content = self.dataset_populator.get_history_dataset_content(history_id, dataset=datasets[0])
assert content.strip() == "Test123"
assert datasets[0]["file_ext"] == "txt", datasets
assert datasets[0]["genome_build"] == "?", datasets
content = self.dataset_populator.get_history_dataset_content(history_id, dataset=datasets[1])
assert content.strip() == "SecondOutputContent"
assert datasets[1]["file_ext"] == "txt"
assert datasets[1]["genome_build"] == "?", datasets
def test_upload_multiple_files_space_to_tab(self):
with self.dataset_populator.test_history() as history_id:
payload = self.dataset_populator.upload_payload(history_id,
content=ONE_TO_SIX_WITH_SPACES,
file_type="tabular",
dbkey="hg19",
extra_inputs={
"files_0|file_type": "txt",
"files_0|space_to_tab": "Yes",
"files_1|url_paste": ONE_TO_SIX_WITH_SPACES,
"files_1|NAME": "SecondOutputName",
"files_1|file_type": "txt",
"files_2|url_paste": ONE_TO_SIX_WITH_SPACES,
"files_2|NAME": "ThirdOutputName",
"files_2|file_type": "txt",
"files_2|space_to_tab": "Yes",
"file_count": "3",
}
)
run_response = self.dataset_populator.tools_post(payload)
self.dataset_populator.wait_for_tool_run(history_id, run_response)
datasets = run_response.json()["outputs"]
assert len(datasets) == 3, datasets
content = self.dataset_populator.get_history_dataset_content(history_id, dataset=datasets[0])
assert content == ONE_TO_SIX_WITH_TABS
content = self.dataset_populator.get_history_dataset_content(history_id, dataset=datasets[1])
assert content == ONE_TO_SIX_WITH_SPACES
content = self.dataset_populator.get_history_dataset_content(history_id, dataset=datasets[2])
assert content == ONE_TO_SIX_WITH_TABS
def test_multiple_files_posix_lines(self):
with self.dataset_populator.test_history() as history_id:
payload = self.dataset_populator.upload_payload(history_id,
content=ONE_TO_SIX_ON_WINDOWS,
file_type="tabular",
dbkey="hg19",
extra_inputs={
"files_0|file_type": "txt",
"files_0|to_posix_lines": "Yes",
"files_1|url_paste": ONE_TO_SIX_ON_WINDOWS,
"files_1|NAME": "SecondOutputName",
"files_1|file_type": "txt",
"files_1|to_posix_lines": None,
"files_2|url_paste": ONE_TO_SIX_ON_WINDOWS,
"files_2|NAME": "ThirdOutputName",
"files_2|file_type": "txt",
"file_count": "3",
}
)
run_response = self.dataset_populator.tools_post(payload)
self.dataset_populator.wait_for_tool_run(history_id, run_response)
datasets = run_response.json()["outputs"]
assert len(datasets) == 3, datasets
content = self.dataset_populator.get_history_dataset_content(history_id, dataset=datasets[0])
assert content == ONE_TO_SIX_WITH_TABS
content = self.dataset_populator.get_history_dataset_content(history_id, dataset=datasets[1])
assert content == ONE_TO_SIX_ON_WINDOWS
content = self.dataset_populator.get_history_dataset_content(history_id, dataset=datasets[2])
assert content == ONE_TO_SIX_WITH_TABS
def test_upload_from_invalid_url(self):
history_id, new_dataset = self._upload('https://usegalaxy.org/bla123', assert_ok=False)
dataset_details = self.dataset_populator.get_history_dataset_details(history_id, dataset_id=new_dataset["id"], assert_ok=False)
assert dataset_details['state'] == 'error', "expected dataset state to be 'error', but got '%s'" % dataset_details['state']
def test_upload_from_valid_url(self):
history_id, new_dataset = self._upload('https://usegalaxy.org/api/version')
self.dataset_populator.get_history_dataset_details(history_id, dataset_id=new_dataset["id"], assert_ok=True)
def _velvet_upload(self, history_id, extra_inputs):
payload = self.dataset_populator.upload_payload(
history_id,
"sequences content",
file_type="velvet",
extra_inputs=extra_inputs,
)
run_response = self.dataset_populator.tools_post(payload)
self.dataset_populator.wait_for_tool_run(history_id, run_response)
datasets = run_response.json()["outputs"]
assert len(datasets) == 1
dataset = datasets[0]
return dataset
def _get_roadmaps_content(self, history_id, dataset):
roadmaps_content = self.dataset_populator.get_history_dataset_content(history_id, dataset=dataset, filename="Roadmaps")
return roadmaps_content
def _upload_and_get_content(self, content, **upload_kwds):
history_id, new_dataset = self._upload(content, **upload_kwds)
return self.dataset_populator.get_history_dataset_content(history_id, dataset=new_dataset)
def _upload_and_get_details(self, content, **upload_kwds):
history_id, new_dataset = self._upload(content, **upload_kwds)
assert_ok = upload_kwds.get("assert_ok", True)
return self.dataset_populator.get_history_dataset_details(history_id, dataset=new_dataset, assert_ok=assert_ok)
def _upload(self, content, api="upload1", history_id=None, **upload_kwds):
assert_ok = upload_kwds.get("assert_ok", True)
history_id = history_id or self.dataset_populator.new_history()
if api == "upload1":
new_dataset = self.dataset_populator.new_dataset(history_id, content=content, **upload_kwds)
else:
assert api == "fetch"
element = dict(src="files", **upload_kwds)
target = {
"destination": {"type": "hdas"},
"elements": [element],
}
targets = json.dumps([target])
payload = {
"history_id": history_id,
"targets": targets,
"__files": {"files_0|file_data": content}
}
new_dataset = self.dataset_populator.fetch(payload, assert_ok=assert_ok).json()["outputs"][0]
self.dataset_populator.wait_for_history(history_id, assert_ok=assert_ok)
return history_id, new_dataset
class ObjectStoreJobsIntegrationTestCase(integration_util.IntegrationTestCase):
framework_tool_and_types = True
@classmethod
def handle_galaxy_config_kwds(cls, config):
temp_directory = cls._test_driver.mkdtemp()
cls.object_stores_parent = temp_directory
for disk_store_file_name in ["files1", "files2", "files3"]:
disk_store_path = os.path.join(temp_directory,
disk_store_file_name)
os.makedirs(disk_store_path)
setattr(cls, "%s_path" % disk_store_file_name, disk_store_path)
config_path = os.path.join(temp_directory, "object_store_conf.xml")
with open(config_path, "w") as f:
f.write(
DISTRIBUTED_OBJECT_STORE_CONFIG_TEMPLATE.safe_substitute(
{"temp_directory": temp_directory}))
config["object_store_config_file"] = config_path
def setUp(self):
super(ObjectStoreJobsIntegrationTestCase, self).setUp()
self.dataset_populator = DatasetPopulator(self.galaxy_interactor)
with self.dataset_populator.test_history() as history_id:
hda1 = self.dataset_populator.new_dataset(
history_id, content=TEST_INPUT_FILES_CONTENT)
create_10_inputs = {
"input1": {
"src": "hda",
"id": hda1["id"]
},
"input2": {
"src": "hda",
"id": hda1["id"]
},
}
self.dataset_populator.run_tool(
"create_10",
create_10_inputs,
history_id,
assert_ok=True,
)
self.dataset_populator.wait_for_history(history_id)
def test_files_count_and_content_in_each_objectstore_backend(self):
"""
According to the ObjectStore configuration given in the
`DISTRIBUTED_OBJECT_STORE_CONFIG_TEMPLATE` variable, datasets
can be stored on three backends, named:
- primary/files1;
- primary/files2;
- secondary/files3.
Objectstore _randomly_ distributes tools outputs on
`primary/files1` and `primary/files2`, and will use
`secondary/files3` if both `primary` backends fail.
This test runs a tools that creates ten dummy datasets,
and asserts if ObjectStore correctly creates ten files
in `primary/files1` and `primary/files2`, and none in
`secondary/files3`, assuming it will not fail persisting
data in `primary` backend.
"""
files_1_count = _files_count(self.files1_path)
files_2_count = _files_count(self.files2_path)
files_3_count = _files_count(self.files3_path)
# Ensure no files written to the secondary/inactive hierarchical disk store.
assert files_3_count == 0
# Ensure the 10 inputs were written to one of the distributed object store's disk
# stores (it will have either 10 or 11 depending on whether the input was also
# written there. The other disk store may or may not have the input file so should
# have at most one file.
assert (files_1_count + files_2_count
== 10) or (files_1_count + files_2_count == 11)
# Other sanity checks on the test - just make sure the test was setup as intended
# and not actually testing object store behavior.
assert (files_1_count <= 11) and (files_2_count <= 11)
assert (files_1_count >= 0) and (files_2_count >= 0)
# TODO: ideally the following assertion should be separated in a different test method.
contents = []
path1_files = _get_datasets_files_in_path(self.files1_path)
path2_files = _get_datasets_files_in_path(self.files2_path)
path3_files = _get_datasets_files_in_path(self.files3_path)
for filename in path1_files + path2_files + path3_files:
with open(filename) as f:
content = f.read().strip()
if content != TEST_INPUT_FILES_CONTENT:
contents.append(content)
for expected_content in range(1, 10):
assert str(expected_content) in contents
class HistoriesApiTestCase(api.ApiTestCase):
def setUp(self):
super(HistoriesApiTestCase, self).setUp()
self.dataset_populator = DatasetPopulator(self.galaxy_interactor)
self.dataset_collection_populator = DatasetCollectionPopulator(
self.galaxy_interactor)
def test_create_history(self):
# Create a history.
create_response = self._create_history("TestHistory1")
created_id = create_response["id"]
# Make sure new history appears in index of user's histories.
index_response = self._get("histories").json()
indexed_history = [h for h in index_response
if h["id"] == created_id][0]
self.assertEquals(indexed_history["name"], "TestHistory1")
def test_show_history(self):
history_id = self._create_history("TestHistoryForShow")["id"]
show_response = self._show(history_id)
self._assert_has_key(show_response, 'id', 'name', 'annotation', 'size',
'contents_url', 'state', 'state_details',
'state_ids')
state_details = show_response["state_details"]
state_ids = show_response["state_ids"]
states = [
'discarded', 'empty', 'error', 'failed_metadata', 'new', 'ok',
'paused', 'queued', 'running', 'setting_metadata', 'upload'
]
assert isinstance(state_details, dict)
assert isinstance(state_ids, dict)
self._assert_has_keys(state_details, *states)
self._assert_has_keys(state_ids, *states)
def test_show_most_recently_used(self):
history_id = self._create_history("TestHistoryRecent")["id"]
show_response = self._get("histories/most_recently_used").json()
assert show_response["id"] == history_id
def test_index_order(self):
slightly_older_history_id = self._create_history(
"TestHistorySlightlyOlder")["id"]
newer_history_id = self._create_history("TestHistoryNewer")["id"]
index_response = self._get("histories").json()
assert index_response[0]["id"] == newer_history_id
assert index_response[1]["id"] == slightly_older_history_id
def test_delete(self):
# Setup a history and ensure it is in the index
history_id = self._create_history("TestHistoryForDelete")["id"]
index_response = self._get("histories").json()
assert index_response[0]["id"] == history_id
show_response = self._show(history_id)
assert not show_response["deleted"]
# Delete the history
self._delete("histories/%s" % history_id)
# Check can view it - but it is deleted
show_response = self._show(history_id)
assert show_response["deleted"]
# Verify it is dropped from history index
index_response = self._get("histories").json()
assert len(
index_response) == 0 or index_response[0]["id"] != history_id
# Add deleted filter to index to view it
index_response = self._get("histories", {"deleted": "true"}).json()
assert index_response[0]["id"] == history_id
def test_purge(self):
history_id = self._create_history("TestHistoryForPurge")["id"]
data = {'purge': True}
self._delete("histories/%s" % history_id, data=data)
show_response = self._show(history_id)
assert show_response["deleted"]
assert show_response["purged"]
def test_undelete(self):
history_id = self._create_history(
"TestHistoryForDeleteAndUndelete")["id"]
self._delete("histories/%s" % history_id)
self._post("histories/deleted/%s/undelete" % history_id)
show_response = self._show(history_id)
assert not show_response["deleted"]
def test_update(self):
history_id = self._create_history("TestHistoryForUpdating")["id"]
self._update(history_id, {"name": "New Name"})
show_response = self._show(history_id)
assert show_response["name"] == "New Name"
unicode_name = u'桜ゲノム'
self._update(history_id, {"name": unicode_name})
show_response = self._show(history_id)
assert show_response["name"] == unicode_name, show_response
quoted_name = "'MooCow'"
self._update(history_id, {"name": quoted_name})
show_response = self._show(history_id)
assert show_response["name"] == quoted_name
self._update(history_id, {"deleted": True})
show_response = self._show(history_id)
assert show_response["deleted"], show_response
self._update(history_id, {"deleted": False})
show_response = self._show(history_id)
assert not show_response["deleted"]
self._update(history_id, {"published": True})
show_response = self._show(history_id)
assert show_response["published"]
self._update(history_id, {"genome_build": "hg18"})
show_response = self._show(history_id)
assert show_response["genome_build"] == "hg18"
self._update(history_id, {"annotation": "The annotation is cool"})
show_response = self._show(history_id)
assert show_response["annotation"] == "The annotation is cool"
self._update(history_id, {"annotation": unicode_name})
show_response = self._show(history_id)
assert show_response["annotation"] == unicode_name, show_response
self._update(history_id, {"annotation": quoted_name})
show_response = self._show(history_id)
assert show_response["annotation"] == quoted_name
def test_update_invalid_attribute(self):
history_id = self._create_history(
"TestHistoryForInvalidUpdating")["id"]
put_response = self._update(history_id, {"invalidkey": "moo"})
assert "invalidkey" not in put_response.json()
def test_update_invalid_types(self):
history_id = self._create_history(
"TestHistoryForUpdatingInvalidTypes")["id"]
for str_key in ["name", "annotation"]:
assert self._update(history_id, {
str_key: False
}).status_code == 400
for bool_key in ['deleted', 'importable', 'published']:
assert self._update(history_id, {
bool_key: "a string"
}).status_code == 400
assert self._update(history_id, {
"tags": "a simple string"
}).status_code == 400
assert self._update(history_id, {"tags": [True]}).status_code == 400
def test_invalid_keys(self):
invalid_history_id = "1234123412341234"
assert self._get("histories/%s" %
invalid_history_id).status_code == 400
assert self._update(invalid_history_id, {
"name": "new name"
}).status_code == 400
assert self._delete("histories/%s" %
invalid_history_id).status_code == 400
assert self._post("histories/deleted/%s/undelete" %
invalid_history_id).status_code == 400
def test_create_anonymous_fails(self):
post_data = dict(name="CannotCreate")
# Using lower-level _api_url will cause key to not be injected.
histories_url = self._api_url("histories")
create_response = post(url=histories_url, data=post_data)
self._assert_status_code_is(create_response, 403)
def test_import_export(self):
history_name = "for_export"
history_id = self.dataset_populator.new_history(name=history_name)
self.dataset_populator.new_dataset(history_id, content="1 2 3")
imported_history_id = self._reimport_history(history_id, history_name)
contents_response = self._get("histories/%s/contents" %
imported_history_id)
self._assert_status_code_is(contents_response, 200)
contents = contents_response.json()
assert len(contents) == 1
imported_content = self.dataset_populator.get_history_dataset_content(
history_id=imported_history_id, dataset_id=contents[0]["id"])
assert imported_content == "1 2 3\n"
def test_import_export_collection(self):
from nose.plugins.skip import SkipTest
raise SkipTest("Collection import/export not yet implemented")
history_name = "for_export_with_collections"
history_id = self.dataset_populator.new_history(name=history_name)
self.dataset_collection_populator.create_list_in_history(
history_id, contents=["Hello", "World"])
imported_history_id = self._reimport_history(history_id, history_name)
contents_response = self._get("histories/%s/contents" %
imported_history_id)
self._assert_status_code_is(contents_response, 200)
contents = contents_response.json()
assert len(contents) == 3
def _reimport_history(self, history_id, history_name):
# Ensure the history is ready to go...
self.dataset_populator.wait_for_history(history_id, assert_ok=True)
# Export the history.
download_path = self._export(history_id)
# Create download for history
full_download_url = "%s%s?key=%s" % (self.url, download_path,
self.galaxy_interactor.api_key)
download_response = get(full_download_url)
self._assert_status_code_is(download_response, 200)
def history_names():
history_index = self._get("histories")
return dict((h["name"], h) for h in history_index.json())
import_name = "imported from archive: %s" % history_name
assert import_name not in history_names()
import_data = dict(archive_source=full_download_url,
archive_type="url")
import_response = self._post("histories", data=import_data)
self._assert_status_code_is(import_response, 200)
def has_history_with_name():
histories = history_names()
return histories.get(import_name, None)
imported_history = wait_on(has_history_with_name,
desc="import history")
imported_history_id = imported_history["id"]
self.dataset_populator.wait_for_history(imported_history_id)
return imported_history_id
def test_create_tag(self):
post_data = dict(name="TestHistoryForTag")
history_id = self._post("histories", data=post_data).json()["id"]
tag_data = dict(value="awesometagvalue")
tag_url = "histories/%s/tags/awesometagname" % history_id
tag_create_response = self._post(tag_url, data=tag_data)
self._assert_status_code_is(tag_create_response, 200)
def _export(self, history_id):
export_url = self._api_url("histories/%s/exports" % history_id,
use_key=True)
put_response = put(export_url)
self._assert_status_code_is(put_response, 202)
def export_ready_response():
put_response = put(export_url)
if put_response.status_code == 202:
return None
return put_response
put_response = wait_on(export_ready_response, desc="export ready")
self._assert_status_code_is(put_response, 200)
response = put_response.json()
self._assert_has_keys(response, "download_url")
download_path = response["download_url"]
return download_path
def _show(self, history_id):
return self._get("histories/%s" % history_id).json()
def _update(self, history_id, data):
update_url = self._api_url("histories/%s" % history_id, use_key=True)
put_response = put(update_url, json=data)
return put_response
def _create_history(self, name):
post_data = dict(name=name)
create_response = self._post("histories", data=post_data).json()
self._assert_has_keys(create_response, "name", "id")
self.assertEquals(create_response["name"], name)
return create_response
class JobsApiTestCase(api.ApiTestCase):
def setUp(self):
super(JobsApiTestCase, self).setUp()
self.dataset_populator = DatasetPopulator(self.galaxy_interactor)
def test_index(self):
# Create HDA to ensure at least one job exists...
self.__history_with_new_dataset()
jobs = self.__jobs_index()
assert "upload1" in map(itemgetter("tool_id"), jobs)
def test_system_details_admin_only(self):
self.__history_with_new_dataset()
jobs = self.__jobs_index(admin=False)
job = jobs[0]
self._assert_not_has_keys(job, "command_line", "external_id")
jobs = self.__jobs_index(admin=True)
job = jobs[0]
self._assert_has_keys(job, "command_line", "external_id")
def test_index_state_filter(self):
# Initial number of ok jobs
original_count = len(self.__uploads_with_state("ok"))
# Run through dataset upload to ensure num uplaods at least greater
# by 1.
self.__history_with_ok_dataset()
# Verify number of ok jobs is actually greater.
count_increased = False
for i in range(10):
new_count = len(self.__uploads_with_state("ok"))
if original_count < new_count:
count_increased = True
break
time.sleep(.1)
if not count_increased:
template = "Jobs in ok state did not increase (was %d, now %d)"
message = template % (original_count, new_count)
raise AssertionError(message)
def test_index_date_filter(self):
self.__history_with_new_dataset()
two_weeks_ago = (datetime.datetime.utcnow() -
datetime.timedelta(7)).isoformat()
last_week = (datetime.datetime.utcnow() -
datetime.timedelta(7)).isoformat()
next_week = (datetime.datetime.utcnow() +
datetime.timedelta(7)).isoformat()
today = datetime.datetime.utcnow().isoformat()
tomorrow = (datetime.datetime.utcnow() +
datetime.timedelta(1)).isoformat()
jobs = self.__jobs_index(data={
"date_range_min": today[0:10],
"date_range_max": tomorrow[0:10]
})
assert len(jobs) > 0
today_job_id = jobs[0]["id"]
jobs = self.__jobs_index(data={
"date_range_min": two_weeks_ago,
"date_range_max": last_week
})
assert today_job_id not in map(itemgetter("id"), jobs)
jobs = self.__jobs_index(data={
"date_range_min": last_week,
"date_range_max": next_week
})
assert today_job_id in map(itemgetter("id"), jobs)
def test_index_history(self):
history_id, _ = self.__history_with_new_dataset()
jobs = self.__jobs_index(data={"history_id": history_id})
assert len(jobs) > 0
history_id = self.dataset_populator.new_history()
jobs = self.__jobs_index(data={"history_id": history_id})
assert len(jobs) == 0
def test_index_multiple_states_filter(self):
# Initial number of ok jobs
original_count = len(self.__uploads_with_state("ok", "new"))
# Run through dataset upload to ensure num uplaods at least greater
# by 1.
self.__history_with_ok_dataset()
# Verify number of ok jobs is actually greater.
new_count = len(self.__uploads_with_state("new", "ok"))
assert original_count < new_count, new_count
def test_show(self):
# Create HDA to ensure at least one job exists...
self.__history_with_new_dataset()
jobs_response = self._get("jobs")
first_job = jobs_response.json()[0]
self._assert_has_key(first_job, 'id', 'state', 'exit_code',
'update_time', 'create_time')
job_id = first_job["id"]
show_jobs_response = self._get("jobs/%s" % job_id)
self._assert_status_code_is(show_jobs_response, 200)
job_details = show_jobs_response.json()
self._assert_has_key(job_details, 'id', 'state', 'exit_code',
'update_time', 'create_time')
def test_show_security(self):
history_id, _ = self.__history_with_new_dataset()
jobs_response = self._get("jobs", data={"history_id": history_id})
job = jobs_response.json()[0]
job_id = job["id"]
show_jobs_response = self._get("jobs/%s" % job_id, admin=False)
self._assert_not_has_keys(show_jobs_response.json(), "command_line",
"external_id")
# TODO: Re-activate test case when API accepts privacy settings
# with self._different_user():
# show_jobs_response = self._get( "jobs/%s" % job_id, admin=False )
# self._assert_status_code_is( show_jobs_response, 200 )
show_jobs_response = self._get("jobs/%s" % job_id, admin=True)
self._assert_has_keys(show_jobs_response.json(), "command_line",
"external_id")
def test_deleting_output_keep_running_until_all_deleted(self):
history_id, job_state, outputs = self._setup_running_two_output_job(60)
# Delete one of the two outputs and make sure the job is still running.
self._raw_update_history_item(history_id, outputs[0]["id"],
{"deleted": True})
time.sleep(1)
state = job_state().json()["state"]
assert state == "running", state
# Delete the second output and make sure the job is cancelled.
self._raw_update_history_item(history_id, outputs[1]["id"],
{"deleted": True})
final_state = wait_on_state(job_state, assert_ok=False, timeout=15)
assert final_state in ["deleted_new", "deleted"], final_state
def test_purging_output_keep_running_until_all_purged(self):
history_id, job_state, outputs = self._setup_running_two_output_job(60)
# Pretty much right away after the job is running, these paths should be populated -
# if they are grab them and make sure they are deleted at the end of the job.
dataset_1 = self._get_history_item_as_admin(history_id,
outputs[0]["id"])
dataset_2 = self._get_history_item_as_admin(history_id,
outputs[1]["id"])
if "file_name" in dataset_1:
output_dataset_paths = [
dataset_1["file_name"], dataset_2["file_name"]
]
# This may or may not exist depending on if the test is local or not.
output_dataset_paths_exist = os.path.exists(
output_dataset_paths[0])
else:
output_dataset_paths = []
output_dataset_paths_exist = False
current_state = job_state().json()["state"]
assert current_state == "running", current_state
# Purge one of the two outputs and make sure the job is still running.
self._raw_update_history_item(history_id, outputs[0]["id"],
{"purged": True})
time.sleep(1)
current_state = job_state().json()["state"]
assert current_state == "running", current_state
# Purge the second output and make sure the job is cancelled.
self._raw_update_history_item(history_id, outputs[1]["id"],
{"purged": True})
final_state = wait_on_state(job_state, assert_ok=False, timeout=15)
assert final_state in ["deleted_new", "deleted"], final_state
def paths_deleted():
if not os.path.exists(
output_dataset_paths[0]) and not os.path.exists(
output_dataset_paths[1]):
return True
if output_dataset_paths_exist:
wait_on(paths_deleted, "path deletion")
def test_purging_output_cleaned_after_ok_run(self):
history_id, job_state, outputs = self._setup_running_two_output_job(10)
# Pretty much right away after the job is running, these paths should be populated -
# if they are grab them and make sure they are deleted at the end of the job.
dataset_1 = self._get_history_item_as_admin(history_id,
outputs[0]["id"])
dataset_2 = self._get_history_item_as_admin(history_id,
outputs[1]["id"])
if "file_name" in dataset_1:
output_dataset_paths = [
dataset_1["file_name"], dataset_2["file_name"]
]
# This may or may not exist depending on if the test is local or not.
output_dataset_paths_exist = os.path.exists(
output_dataset_paths[0])
else:
output_dataset_paths = []
output_dataset_paths_exist = False
if not output_dataset_paths_exist:
# Given this Galaxy configuration - there is nothing more to be tested here.
# Consider throwing a skip instead.
return
# Purge one of the two outputs and wait for the job to complete.
self._raw_update_history_item(history_id, outputs[0]["id"],
{"purged": True})
wait_on_state(job_state, assert_ok=True)
if output_dataset_paths_exist:
time.sleep(.5)
# Make sure the non-purged dataset is on disk and the purged one is not.
assert os.path.exists(output_dataset_paths[1])
assert not os.path.exists(output_dataset_paths[0])
def _setup_running_two_output_job(self, sleep_time):
history_id = self.dataset_populator.new_history()
payload = self.dataset_populator.run_tool_payload(
tool_id='create_2',
inputs=dict(sleep_time=sleep_time, ),
history_id=history_id,
)
run_response = self._post("tools", data=payload).json()
outputs = run_response["outputs"]
jobs = run_response["jobs"]
assert len(outputs) == 2
assert len(jobs) == 1
def job_state():
jobs_response = self._get("jobs/%s" % jobs[0]["id"])
return jobs_response
# Give job some time to get up and running.
time.sleep(2)
running_state = wait_on_state(job_state,
skip_states=["queued", "new"],
assert_ok=False,
timeout=15)
assert running_state == "running", running_state
def job_state():
jobs_response = self._get("jobs/%s" % jobs[0]["id"])
return jobs_response
return history_id, job_state, outputs
def _raw_update_history_item(self, history_id, item_id, data):
update_url = self._api_url("histories/%s/contents/%s" %
(history_id, item_id),
use_key=True)
update_response = put(update_url, json=data)
assert_status_code_is_ok(update_response)
return update_response
def _get_history_item_as_admin(self, history_id, item_id):
response = self._get("histories/%s/contents/%s?view=detailed" %
(history_id, item_id),
admin=True)
assert_status_code_is_ok(response)
return response.json()
def test_search(self):
history_id, dataset_id = self.__history_with_ok_dataset()
inputs = json.dumps(dict(input1=dict(
src='hda',
id=dataset_id,
)))
search_payload = dict(
tool_id="cat1",
inputs=inputs,
state="ok",
)
empty_search_response = self._post("jobs/search", data=search_payload)
self._assert_status_code_is(empty_search_response, 200)
self.assertEquals(len(empty_search_response.json()), 0)
self.__run_cat_tool(history_id, dataset_id)
self.dataset_populator.wait_for_history(history_id, assert_ok=True)
search_count = -1
# in case job and history aren't updated at exactly the same
# time give time to wait
for i in range(5):
search_count = self._search_count(search_payload)
if search_count == 1:
break
time.sleep(.1)
self.assertEquals(search_count, 1)
def _search_count(self, search_payload):
search_response = self._post("jobs/search", data=search_payload)
self._assert_status_code_is(search_response, 200)
search_json = search_response.json()
return len(search_json)
def __run_cat_tool(self, history_id, dataset_id):
# Code duplication with test_jobs.py, eliminate
payload = self.dataset_populator.run_tool_payload(
tool_id='cat1',
inputs=dict(input1=dict(src='hda', id=dataset_id), ),
history_id=history_id,
)
self._post("tools", data=payload)
def __run_randomlines_tool(self, lines, history_id, dataset_id):
payload = self.dataset_populator.run_tool_payload(
tool_id="random_lines1",
inputs=dict(
num_lines=lines,
input=dict(
src='hda',
id=dataset_id,
),
),
history_id=history_id,
)
self._post("tools", data=payload)
def __uploads_with_state(self, *states):
jobs_response = self._get("jobs", data=dict(state=states))
self._assert_status_code_is(jobs_response, 200)
jobs = jobs_response.json()
assert not filter(lambda j: j["state"] not in states, jobs)
return filter(lambda j: j["tool_id"] == "upload1", jobs)
def __history_with_new_dataset(self):
history_id = self.dataset_populator.new_history()
dataset_id = self.dataset_populator.new_dataset(history_id)["id"]
return history_id, dataset_id
def __history_with_ok_dataset(self):
history_id = self.dataset_populator.new_history()
dataset_id = self.dataset_populator.new_dataset(history_id,
wait=True)["id"]
return history_id, dataset_id
def __jobs_index(self, **kwds):
jobs_response = self._get("jobs", **kwds)
self._assert_status_code_is(jobs_response, 200)
jobs = jobs_response.json()
assert isinstance(jobs, list)
return jobs
class HistoriesApiTestCase(api.ApiTestCase):
def setUp(self):
super(HistoriesApiTestCase, self).setUp()
self.dataset_populator = DatasetPopulator(self.galaxy_interactor)
self.dataset_collection_populator = DatasetCollectionPopulator(self.galaxy_interactor)
def test_create_history(self):
# Create a history.
create_response = self._create_history("TestHistory1")
created_id = create_response["id"]
# Make sure new history appears in index of user's histories.
index_response = self._get("histories").json()
indexed_history = [h for h in index_response if h["id"] == created_id][0]
self.assertEquals(indexed_history["name"], "TestHistory1")
def test_show_history(self):
history_id = self._create_history("TestHistoryForShow")["id"]
show_response = self._show(history_id)
self._assert_has_key(
show_response,
'id', 'name', 'annotation', 'size', 'contents_url',
'state', 'state_details', 'state_ids'
)
state_details = show_response["state_details"]
state_ids = show_response["state_ids"]
states = [
'discarded', 'empty', 'error', 'failed_metadata', 'new',
'ok', 'paused', 'queued', 'running', 'setting_metadata', 'upload'
]
assert isinstance(state_details, dict)
assert isinstance(state_ids, dict)
self._assert_has_keys(state_details, *states)
self._assert_has_keys(state_ids, *states)
def test_show_most_recently_used(self):
history_id = self._create_history("TestHistoryRecent")["id"]
show_response = self._get("histories/most_recently_used").json()
assert show_response["id"] == history_id
def test_index_order(self):
slightly_older_history_id = self._create_history("TestHistorySlightlyOlder")["id"]
newer_history_id = self._create_history("TestHistoryNewer")["id"]
index_response = self._get("histories").json()
assert index_response[0]["id"] == newer_history_id
assert index_response[1]["id"] == slightly_older_history_id
def test_delete(self):
# Setup a history and ensure it is in the index
history_id = self._create_history("TestHistoryForDelete")["id"]
index_response = self._get("histories").json()
assert index_response[0]["id"] == history_id
show_response = self._show(history_id)
assert not show_response["deleted"]
# Delete the history
self._delete("histories/%s" % history_id)
# Check can view it - but it is deleted
show_response = self._show(history_id)
assert show_response["deleted"]
# Verify it is dropped from history index
index_response = self._get("histories").json()
assert len(index_response) == 0 or index_response[0]["id"] != history_id
# Add deleted filter to index to view it
index_response = self._get("histories", {"deleted": "true"}).json()
assert index_response[0]["id"] == history_id
def test_purge(self):
history_id = self._create_history("TestHistoryForPurge")["id"]
data = {'purge': True}
self._delete("histories/%s" % history_id, data=data)
show_response = self._show(history_id)
assert show_response["deleted"]
assert show_response["purged"]
def test_undelete(self):
history_id = self._create_history("TestHistoryForDeleteAndUndelete")["id"]
self._delete("histories/%s" % history_id)
self._post("histories/deleted/%s/undelete" % history_id)
show_response = self._show(history_id)
assert not show_response["deleted"]
def test_update(self):
history_id = self._create_history("TestHistoryForUpdating")["id"]
self._update(history_id, {"name": "New Name"})
show_response = self._show(history_id)
assert show_response["name"] == "New Name"
unicode_name = u'桜ゲノム'
self._update(history_id, {"name": unicode_name})
show_response = self._show(history_id)
assert show_response["name"] == unicode_name, show_response
quoted_name = "'MooCow'"
self._update(history_id, {"name": quoted_name})
show_response = self._show(history_id)
assert show_response["name"] == quoted_name
self._update(history_id, {"deleted": True})
show_response = self._show(history_id)
assert show_response["deleted"], show_response
self._update(history_id, {"deleted": False})
show_response = self._show(history_id)
assert not show_response["deleted"]
self._update(history_id, {"published": True})
show_response = self._show(history_id)
assert show_response["published"]
self._update(history_id, {"genome_build": "hg18"})
show_response = self._show(history_id)
assert show_response["genome_build"] == "hg18"
self._update(history_id, {"annotation": "The annotation is cool"})
show_response = self._show(history_id)
assert show_response["annotation"] == "The annotation is cool"
self._update(history_id, {"annotation": unicode_name})
show_response = self._show(history_id)
assert show_response["annotation"] == unicode_name, show_response
self._update(history_id, {"annotation": quoted_name})
show_response = self._show(history_id)
assert show_response["annotation"] == quoted_name
def test_update_invalid_attribute(self):
history_id = self._create_history("TestHistoryForInvalidUpdating")["id"]
put_response = self._update(history_id, {"invalidkey": "moo"})
assert "invalidkey" not in put_response.json()
def test_update_invalid_types(self):
history_id = self._create_history("TestHistoryForUpdatingInvalidTypes")["id"]
for str_key in ["name", "annotation"]:
assert self._update(history_id, {str_key: False}).status_code == 400
for bool_key in ['deleted', 'importable', 'published']:
assert self._update(history_id, {bool_key: "a string"}).status_code == 400
assert self._update(history_id, {"tags": "a simple string"}).status_code == 400
assert self._update(history_id, {"tags": [True]}).status_code == 400
def test_invalid_keys(self):
invalid_history_id = "1234123412341234"
assert self._get("histories/%s" % invalid_history_id).status_code == 400
assert self._update(invalid_history_id, {"name": "new name"}).status_code == 400
assert self._delete("histories/%s" % invalid_history_id).status_code == 400
assert self._post("histories/deleted/%s/undelete" % invalid_history_id).status_code == 400
def test_create_anonymous_fails(self):
post_data = dict(name="CannotCreate")
# Using lower-level _api_url will cause key to not be injected.
histories_url = self._api_url("histories")
create_response = post(url=histories_url, data=post_data)
self._assert_status_code_is(create_response, 403)
def test_import_export(self):
history_name = "for_export"
history_id = self.dataset_populator.new_history(name=history_name)
self.dataset_populator.new_dataset(history_id, content="1 2 3")
imported_history_id = self._reimport_history(history_id, history_name)
contents_response = self._get("histories/%s/contents" % imported_history_id)
self._assert_status_code_is(contents_response, 200)
contents = contents_response.json()
assert len(contents) == 1
imported_content = self.dataset_populator.get_history_dataset_content(
history_id=imported_history_id,
dataset_id=contents[0]["id"]
)
assert imported_content == "1 2 3\n"
def test_import_export_collection(self):
from nose.plugins.skip import SkipTest
raise SkipTest("Collection import/export not yet implemented")
history_name = "for_export_with_collections"
history_id = self.dataset_populator.new_history(name=history_name)
self.dataset_collection_populator.create_list_in_history(history_id, contents=["Hello", "World"])
imported_history_id = self._reimport_history(history_id, history_name)
contents_response = self._get("histories/%s/contents" % imported_history_id)
self._assert_status_code_is(contents_response, 200)
contents = contents_response.json()
assert len(contents) == 3
def _reimport_history(self, history_id, history_name):
# Ensure the history is ready to go...
self.dataset_populator.wait_for_history(history_id, assert_ok=True)
# Export the history.
download_path = self._export(history_id)
# Create download for history
full_download_url = "%s%s?key=%s" % (self.url, download_path, self.galaxy_interactor.api_key)
download_response = get(full_download_url)
self._assert_status_code_is(download_response, 200)
def history_names():
history_index = self._get("histories")
return dict((h["name"], h) for h in history_index.json())
import_name = "imported from archive: %s" % history_name
assert import_name not in history_names()
import_data = dict(archive_source=full_download_url, archive_type="url")
import_response = self._post("histories", data=import_data)
self._assert_status_code_is(import_response, 200)
def has_history_with_name():
histories = history_names()
return histories.get(import_name, None)
imported_history = wait_on(has_history_with_name, desc="import history")
imported_history_id = imported_history["id"]
self.dataset_populator.wait_for_history(imported_history_id)
return imported_history_id
def test_create_tag(self):
post_data = dict(name="TestHistoryForTag")
history_id = self._post("histories", data=post_data).json()["id"]
tag_data = dict(value="awesometagvalue")
tag_url = "histories/%s/tags/awesometagname" % history_id
tag_create_response = self._post(tag_url, data=tag_data)
self._assert_status_code_is(tag_create_response, 200)
def _export(self, history_id):
export_url = self._api_url("histories/%s/exports" % history_id, use_key=True)
put_response = put(export_url)
self._assert_status_code_is(put_response, 202)
def export_ready_response():
put_response = put(export_url)
if put_response.status_code == 202:
return None
return put_response
put_response = wait_on(export_ready_response, desc="export ready")
self._assert_status_code_is(put_response, 200)
response = put_response.json()
self._assert_has_keys(response, "download_url")
download_path = response["download_url"]
return download_path
def _show(self, history_id):
return self._get("histories/%s" % history_id).json()
def _update(self, history_id, data):
update_url = self._api_url("histories/%s" % history_id, use_key=True)
put_response = put(update_url, json=data)
return put_response
def _create_history(self, name):
post_data = dict(name=name)
create_response = self._post("histories", data=post_data).json()
self._assert_has_keys(create_response, "name", "id")
self.assertEquals(create_response["name"], name)
return create_response
class DockerizedJobsIntegrationTestCase(integration_util.IntegrationTestCase, RunsEnvironmentJobs):
framework_tool_and_types = True
job_config_file = DOCKERIZED_JOB_CONFIG_FILE
build_mulled_resolver = 'build_mulled'
container_type = 'docker'
default_container_home_dir = '/'
@classmethod
def handle_galaxy_config_kwds(cls, config):
cls.jobs_directory = cls._test_driver.mkdtemp()
config["jobs_directory"] = cls.jobs_directory
config["job_config_file"] = cls.job_config_file
# Disable tool dependency resolution.
config["tool_dependency_dir"] = "none"
config["conda_auto_init"] = False
config["conda_auto_install"] = False
config["enable_beta_mulled_containers"] = "true"
@classmethod
def setUpClass(cls):
if not which(cls.container_type):
raise unittest.SkipTest("Executable '%s' not found on PATH" % cls.container_type)
super(DockerizedJobsIntegrationTestCase, cls).setUpClass()
def setUp(self):
super(DockerizedJobsIntegrationTestCase, self).setUp()
self.dataset_populator = DatasetPopulator(self.galaxy_interactor)
self.history_id = self.dataset_populator.new_history()
def test_explicit(self):
self.dataset_populator.run_tool("mulled_example_explicit", {}, self.history_id)
self.dataset_populator.wait_for_history(self.history_id, assert_ok=True)
output = self.dataset_populator.get_history_dataset_content(self.history_id, timeout=EXTENDED_TIMEOUT)
assert "0.7.15-r1140" in output
def test_mulled_simple(self):
self.dataset_populator.run_tool("mulled_example_simple", {}, self.history_id)
self.dataset_populator.wait_for_history(self.history_id, assert_ok=True)
output = self.dataset_populator.get_history_dataset_content(self.history_id, timeout=EXTENDED_TIMEOUT)
assert "0.7.15-r1140" in output
def test_container_job_enviornment(self):
job_env = self._run_and_get_environment_properties("job_environment_default")
euid = os.geteuid()
egid = os.getgid()
assert job_env.user_id == str(euid), job_env.user_id
assert job_env.group_id == str(egid), job_env.group_id
assert job_env.pwd.startswith(self.jobs_directory)
assert job_env.pwd.endswith("/working")
assert job_env.home.startswith(self.jobs_directory)
assert job_env.home.endswith("/home")
def test_container_job_environment_legacy(self):
job_env = self._run_and_get_environment_properties("job_environment_default_legacy")
euid = os.geteuid()
egid = os.getgid()
assert job_env.user_id == str(euid), job_env.user_id
assert job_env.group_id == str(egid), job_env.group_id
assert job_env.pwd.startswith(self.jobs_directory)
assert job_env.pwd.endswith("/working")
# Should we change env_pass_through to just always include TMP and HOME for docker?
# I'm not sure, if yes this would change.
assert job_env.home == self.default_container_home_dir, job_env.home
def test_build_mulled(self):
if not which('docker'):
raise unittest.SkipTest("Docker not found on PATH, required for building images via involucro")
resolver_type = self.build_mulled_resolver
tool_id = 'mulled_example_multi_1'
endpoint = "tools/%s/dependencies" % tool_id
data = {'id': tool_id, 'resolver_type': resolver_type}
create_response = self._post(endpoint, data=data, admin=True)
self._assert_status_code_is(create_response, 200)
response = create_response.json()
assert any([True for d in response if d['dependency_type'] == self.container_type])
