def index(self):
""" Create new environment, fill it, and create an IndexAPI
"""
from docido_sdk.index.config import YamlPullCrawlersIndexingConfig
config_yaml = osp.splitext(__file__)[0] + '.yml'
with docido_config:
docido_config.clear()
docido_config.update(Configuration.from_file(config_yaml))
env = Environment()
env.temp_dir = tempfile.mkdtemp()
test_components = self._setup_test_components(env)
pipeline = env[IndexPipelineProvider]
env[Elasticsearch]
try:
# build and provide an IndexAPI
env[YamlPullCrawlersIndexingConfig]
yield pipeline.get_index_api(None, None, None, None)
finally:
# Hide from Environment the Component classes defined
# for this test only.
for test_component in test_components:
test_component.unregister()
# Remove temporary directory previously created
if osp.isdir(env.temp_dir):
shutil.rmtree(env.temp_dir)
def index(self):
""" Create new environment, fill it, and create an IndexAPI
"""
from docido_sdk.index.config import YamlPullCrawlersIndexingConfig
config_yaml = osp.splitext(__file__)[0] + '.yml'
with docido_config:
docido_config.clear()
docido_config.update(Configuration.from_file(config_yaml))
env = Environment()
env.temp_dir = tempfile.mkdtemp()
test_components = self._setup_test_components(env)
pipeline = env[IndexPipelineProvider]
env[Elasticsearch]
try:
# build and provide an IndexAPI
env[YamlPullCrawlersIndexingConfig]
yield pipeline.get_index_api(None, None, None, None)
finally:
# Hide from Environment the Component classes defined
# for this test only.
for test_component in test_components:
test_component.unregister()
# Remove temporary directory previously created
if osp.isdir(env.temp_dir):
shutil.rmtree(env.temp_dir)
def run_crawl(self, cls, *args, **kwargs):
with restore_dict_kv(os.environ, 'DOCIDO_CC_RUNS'), \
docido_config, \
self.crawler(cls, *args, **kwargs), \
self.check_crawl(*args, **kwargs):
config_prefix = osp.splitext(__file__)[0]
os.environ['DOCIDO_CC_RUNS'] = config_prefix + '-runs.yml'
config_settings = config_prefix + '-settings.yml'
docido_config.update(Configuration.from_file(config_settings))
for c in dcc_run.run([], environment=Environment()):
shutil.rmtree(c['crawl_path'])
def run_crawl(self, cls, *args, **kwargs):
with restore_dict_kv(os.environ, 'DOCIDO_CC_RUNS'), \
docido_config, \
self.crawler(cls, *args, **kwargs), \
self.check_crawl(*args, **kwargs):
config_prefix = osp.splitext(__file__)[0]
os.environ['DOCIDO_CC_RUNS'] = config_prefix + '-runs.yml'
config_settings = config_prefix + '-settings.yml'
docido_config.update(Configuration.from_file(config_settings))
for c in dcc_run.run([], environment=Environment()):
shutil.rmtree(c['crawl_path'])
def index(self):
from docido_sdk.index.config import YamlPullCrawlersIndexingConfig
config_yaml = osp.splitext(__file__)[0] + '.yml'
with docido_config:
docido_config.clear()
docido_config.update(Configuration.from_file(config_yaml))
env = Environment()
test_components = self._setup_test_components(env)
env[IndexPipelineProvider]
env[LocalDumbIndex]
env[processor.CheckProcessor]
try:
env[YamlPullCrawlersIndexingConfig]
index_builder = env[IndexPipelineProvider]
yield index_builder.get_index_api('check-processor-test',
'user2', 'account3', None)
finally:
for test_component in test_components:
test_component.unregister()
def index(self):
from docido_sdk.index.config import YamlPullCrawlersIndexingConfig
config_yaml = osp.splitext(__file__)[0] + '.yml'
with docido_config:
docido_config.clear()
docido_config.update(Configuration.from_file(config_yaml))
env = Environment()
test_components = self._setup_test_components(env)
env[IndexPipelineProvider]
env[LocalDumbIndex]
env[processor.CheckProcessor]
try:
env[YamlPullCrawlersIndexingConfig]
index_builder = env[IndexPipelineProvider]
yield index_builder.get_index_api(
'check-processor-test', 'user2', 'account3'
)
finally:
for test_component in test_components:
test_component.unregister()
def run(self, logger, config, crawler):
logger.info("starting crawl")
self.prepare_crawl_path()
logger.info('pushed data will be stored in {}'.format(self.crawl_path))
index_provider = env[IndexPipelineProvider]
with docido_config:
if config.environment is not None:
docido_config.clear()
new_config = Configuration.from_file(config.environment)
docido_config.update(new_config)
index_api = index_provider.get_index_api(
self.service, None, None, config.get('config') or {}
)
runner = TasksRunner(crawler, index_api, config, logger)
self._check_pickle(runner.tasks)
runner.execute()
return {
'service': self.service,
'name': self.launch,
'crawl_path': self.crawl_path,
}
def run(self, logger, config, crawler):
logger.info("starting crawl")
self.prepare_crawl_path()
logger.info('pushed data will be stored in {}'.format(self.crawl_path))
index_provider = env[IndexPipelineProvider]
with docido_config:
if config.environment is not None:
docido_config.clear()
new_config = Configuration.from_file(config.environment)
docido_config.update(new_config)
index_api = index_provider.get_index_api(
self.service, None, None,
config.get('config') or {})
runner = TasksRunner(crawler, index_api, config, logger)
self._check_pickle(runner.tasks)
runner.execute()
return {
'service': self.service,
'name': self.launch,
'crawl_path': self.crawl_path,
}
def run(self, logger, config, crawler):
logger.info("starting crawl")
self.prepare_crawl_path()
logger.info('pushed data will be stored in {}'.format(self.crawl_path))
index_provider = env[IndexPipelineProvider]
with docido_config:
if config.config is not None:
docido_config.clear()
new_config = Configuration.from_file(config.config)
docido_config.update(new_config)
index_api = index_provider.get_index_api(
self.service, None, None
)
attempt = 1
while True:
try:
tasks = crawler.iter_crawl_tasks(
index_api, config.token,
logger, config.get('full', False)
)
break
except Retry as e:
try:
wait_or_raise(logger, e, attempt)
except:
logger.exception('Max retries reached')
raise
else:
attempt += 1
except Exception:
logger.exception('Unexpected exception was raised')
raise
self._check_pickle(tasks)
tasks, epilogue, concurrency = reorg_crawl_tasks(
tasks,
int(config.get('max_concurrent_tasks', 2))
)
tasks = split_crawl_tasks(tasks, concurrency)
def _runtask(task, prev_result):
attempt = 1
result = None
kwargs = dict()
while True:
try:
result = task(index_api, config.token,
prev_result, logger, **kwargs)
break
except Retry as e:
try:
wait_or_raise(logger, e, attempt)
except:
logger.exception('Max retries reached')
result = e
break
else:
attempt += 1
kwargs = e.kwargs
except Exception as e:
logger.exception('Unexpected exception was raised')
result = e
break
return result
results = []
for seq in tasks:
previous_result = None
for task in seq:
previous_result = _runtask(task, previous_result)
results.append(previous_result)
if epilogue is not None:
_runtask(epilogue, results)
return {
'service': self.service,
'name': self.launch,
'crawl_path': self.crawl_path,
}
