def test_download_proxy():
import urllib.request
# first test no proxy
args = set_hw_parser().parse_args([])
opener = urllib.request.build_opener()
if args.download_proxy:
proxy = urllib.request.ProxyHandler({'http': args.download_proxy, 'https': args.download_proxy})
opener.add_handler(proxy)
urllib.request.install_opener(opener)
# head check
req = urllib.request.Request(args.index_data_url, method="HEAD")
response = urllib.request.urlopen(req, timeout=5)
assert response.status == 200
# test with proxy
args = set_hw_parser().parse_args(["--download-proxy", os.getenv("HTTP_PROXY")])
opener = urllib.request.build_opener()
if args.download_proxy:
proxy = urllib.request.ProxyHandler({'http': args.download_proxy, 'https': args.download_proxy})
opener.add_handler(proxy)
urllib.request.install_opener(opener)
# head check
req = urllib.request.Request(args.index_data_url, method="HEAD")
response = urllib.request.urlopen(req, timeout=5)
assert response.status == 200
def test_helloworld_flow(tmpdir):
args = set_hw_parser().parse_args([])
os.environ['RESOURCE_DIR'] = resource_filename('jina', 'resources')
os.environ['SHARDS'] = str(args.shards)
os.environ['PARALLEL'] = str(args.parallel)
os.environ['HW_WORKDIR'] = str(tmpdir)
f = Flow.load_config(resource_filename('jina', '/'.join(('resources', 'helloworld.flow.index.yml'))))
targets = {
'index': {
'url': args.index_data_url,
'filename': os.path.join(tmpdir, 'index-original')
},
'query': {
'url': args.query_data_url,
'filename': os.path.join(tmpdir, 'query-original')
}
}
# download the data
Path(tmpdir).mkdir(parents=True, exist_ok=True)
download_data(targets)
# run it!
with f:
f.index(_input_ndarray(targets['index']['data']), request_size=args.index_request_size)
def test_helloworld_flow_dry_run(tmpdir):
args = set_hw_parser().parse_args([])
os.environ['RESOURCE_DIR'] = resource_filename('jina', 'resources')
os.environ['SHARDS'] = str(args.shards)
os.environ['PARALLEL'] = str(args.parallel)
os.environ['HW_WORKDIR'] = str(tmpdir)
# run it!
with Flow.load_config(resource_filename('jina', '/'.join(('resources', 'helloworld.flow.index.yml')))):
pass
# run it!
with Flow.load_config(resource_filename('jina', '/'.join(('resources', 'helloworld.flow.query.yml')))):
pass
def test_helloworld_flow_dry_run(tmpdir):
args = set_hw_parser().parse_args([])
os.environ['PATH'] += os.pathsep + resource_filename(
'jina', 'resources/fashion')
os.environ['SHARDS'] = str(args.shards)
os.environ['PARALLEL'] = str(args.parallel)
os.environ['HW_WORKDIR'] = str(tmpdir)
# run it!
with Flow.load_config('helloworld.flow.index.yml'):
pass
# run it!
with Flow.load_config('helloworld.flow.query.yml'):
pass
def test_helloworld_py(tmpdir):
from jina.helloworld import hello_world
hello_world(set_hw_parser().parse_args(['--workdir', str(tmpdir)]))
check_hello_world_results(os.path.join(str(tmpdir), 'hello-world.html'))
)
# f.search(
# query_generator(
# num_docs=args.num_query, target=targets, with_groundtruth=True
# ),
# shuffle=True,
# on_done=print_result,
# request_size=args.request_size,
# parameters={'top_k': args.top_k},
# )
f.post(
'/eval',
query_generator(num_docs=args.num_query,
target=targets,
with_groundtruth=True),
shuffle=True,
on_done=print_result,
request_size=args.request_size,
parameters={'top_k': args.top_k},
)
# write result to html
write_html(os.path.join(args.workdir, 'demo.html'))
if __name__ == '__main__':
args = set_hw_parser().parse_args()
hello_world(args)
def test_helloworld_py(tmpdir):
from jina.helloworld import hello_world
hello_world(set_hw_parser().parse_args(['--workdir', str(tmpdir)]))
def helloworld_args(tmpdir):
return set_hw_parser().parse_args(['--workdir', str(tmpdir)])
def _benchmark_qps() -> Dict[str, float]:
"""Benchmark Jina Core Indexing and Query.
Returns:
A dict mapping keys
"""
args = set_hw_parser().parse_args()
args.workdir = os.path.join(os.getcwd(), 'original')
args.num_query = 4096
targets = {
'index-labels': {
'url': args.index_labels_url,
'filename': os.path.join(args.workdir, 'index-labels'),
},
'query-labels': {
'url': args.query_labels_url,
'filename': os.path.join(args.workdir, 'query-labels'),
},
'index': {
'url': args.index_data_url,
'filename': os.path.join(args.workdir, 'index-original'),
},
'query': {
'url': args.query_data_url,
'filename': os.path.join(args.workdir, 'query-original'),
},
}
# download the data
Path(args.workdir).mkdir(parents=True, exist_ok=True)
download_data(targets, args.download_proxy)
try:
f = Flow().add(uses=MyEncoder).add(workspace='./', uses=MyIndexer)
with f:
# do index
log.info('Benchmarking index')
st = time.perf_counter()
f.index(
index_generator(num_docs=targets['index']['data'].shape[0],
target=targets),
show_progress=True,
)
index_time = time.perf_counter() - st
log.info(
'Indexed %d docs within %d seconds',
targets['index']['data'].shape[0],
index_time,
)
# do query
log.info('Benchmarking query')
st = time.perf_counter()
f.search(
query_generator(num_docs=args.num_query, target=targets),
shuffle=True,
parameters={'top_k': args.top_k},
show_progress=True,
)
query_time = time.perf_counter() - st
log.info('%d query within %d seconds', args.num_query, query_time)
except Exception as e:
log.error(e)
sys.exit(1)
return {
'index_time': index_time,
'query_time': query_time,
'index_qps': targets['index']['data'].shape[0] / index_time,
'query_qps': args.num_query / query_time,
}
