def _install(args, use_cache, debug):
"""Install datasets for retriever."""
engine = choose_engine(args)
engine.use_cache = use_cache
script_list = SCRIPT_LIST()
if not (script_list or os.listdir(SCRIPT_WRITE_PATH)):
check_for_updates()
script_list = SCRIPT_LIST()
data_sets_scripts = name_matches(script_list, args['dataset'])
if data_sets_scripts:
for data_sets_script in data_sets_scripts:
try:
engine.script_table_registry = OrderedDict()
data_sets_script.download(engine, debug=debug)
data_sets_script.engine.final_cleanup()
except Exception as e:
print(e)
if debug:
raise
else:
message = "Run retriever.datasets()to list the currently available " \
"datasets."
raise ValueError(message)
return engine
def _install(args, use_cache, debug):
"""Install datasets for retriever."""
engine = choose_engine(args)
engine.use_cache = use_cache
script_list = SCRIPT_LIST()
if not script_list or not os.listdir(SCRIPT_WRITE_PATH):
check_for_updates()
script_list = SCRIPT_LIST(force_compile=False)
data_sets_scripts = name_matches(script_list, args['dataset'])
if data_sets_scripts:
for data_sets_script in data_sets_scripts:
print("=> Installing", data_sets_script.name)
try:
data_sets_script.download(engine, debug=debug)
data_sets_script.engine.final_cleanup()
except Exception as e:
print(e)
if debug:
raise
else:
message = "The dataset \"{}\" isn't available in the Retriever. " \
"Run retriever.datasets()to list the currently available " \
"datasets".format(args['dataset'])
raise ValueError(message)
def download(dataset, path='./', quiet=False, sub_dir='', debug=False, use_cache=True):
"""Download scripts for retriever."""
args = {
'dataset': dataset,
'command': 'download',
'path': path,
'sub_dir': sub_dir,
'quiet': quiet
}
engine = choose_engine(args)
engine.use_cache = use_cache
script_list = SCRIPT_LIST()
if not script_list or not os.listdir(SCRIPT_WRITE_PATH):
check_for_updates()
script_list = SCRIPT_LIST()
scripts = name_matches(script_list, args['dataset'])
if scripts:
for script in scripts:
print("=> Downloading", script.name)
try:
script.download(engine, debug=debug)
script.engine.final_cleanup()
except Exception as e:
print(e)
if debug:
raise
else:
message = "Run retriever.datasets() to see the list of currently " \
"available datasets."
raise ValueError(message)
return engine
def download(dataset, path='./', quiet=False, subdir=False, debug=False):
"""Download scripts for retriever."""
args = {
'dataset': dataset,
'command': 'download',
'path': path,
'subdir': subdir,
'quiet': quiet
}
engine = choose_engine(args)
script_list = SCRIPT_LIST()
if not script_list or not os.listdir(SCRIPT_WRITE_PATH):
check_for_updates()
script_list = SCRIPT_LIST()
scripts = name_matches(script_list, args['dataset'])
if scripts:
for script in scripts:
print("=> Downloading", script.name)
try:
script.download(engine, debug=debug)
script.engine.final_cleanup()
except Exception as e:
print(e)
if debug:
raise
else:
message = "Run retriever.datasets() to see a list of currently " \
"available datasets."
raise ValueError(message)
def _install(args, use_cache, debug):
"""Install datasets for retriever."""
engine = choose_engine(args)
engine.use_cache = use_cache
script_list = SCRIPT_LIST()
if not (script_list or os.listdir(SCRIPT_WRITE_PATH)):
check_for_updates()
script_list = SCRIPT_LIST()
data_sets_scripts = name_matches(script_list, args['dataset'])
if data_sets_scripts:
for data_sets_script in data_sets_scripts:
try:
engine.script_table_registry = OrderedDict()
data_sets_script.download(engine, debug=debug)
data_sets_script.engine.final_cleanup()
except Exception as e:
print(e)
if debug:
raise
else:
message = "Run retriever.datasets()to list the currently available " \
"datasets."
raise ValueError(message)
return engine
def main():
"""This function launches the Data Retriever."""
if len(sys.argv) == 1:
# if no command line args are passed, show the help options
parser.parse_args(['-h'])
else:
# otherwise, parse them
args = parser.parse_args()
if args.command not in ['reset', 'update'] \
and not os.path.isdir(SCRIPT_SEARCH_PATHS[1]) \
and not [f for f in os.listdir(SCRIPT_SEARCH_PATHS[-1])
if os.path.exists(SCRIPT_SEARCH_PATHS[-1])]:
check_for_updates()
reload_scripts()
script_list = SCRIPT_LIST()
if args.command == "install" and not args.engine:
parser.parse_args(['install', '-h'])
if args.quiet:
sys.stdout = open(os.devnull, 'w')
if args.command == 'help':
parser.parse_args(['-h'])
if hasattr(args, 'compile') and args.compile:
script_list = reload_scripts()
if args.command == 'defaults':
for engine_item in engine_list:
print("Default options for engine ", engine_item.name)
for default_opts in engine_item.required_opts:
print(default_opts[0], " ", default_opts[2])
print()
return
if args.command == 'update':
check_for_updates()
reload_scripts()
return
elif args.command == 'citation':
if args.dataset is None:
print("\nCitation for retriever:\n")
print(CITATION)
else:
scripts = name_matches(script_list, args.dataset)
for dataset in scripts:
print("\nDataset: {}".format(dataset.name))
print("Citation: {}".format(dataset.citation))
print("Description: {}\n".format(dataset.description))
return
elif args.command == 'license':
if args.dataset is None:
print(LICENSE)
else:
dataset_license = license(args.dataset)
if dataset_license:
print(dataset_license)
else:
print("There is no license information for {}".format(
args.dataset))
return
elif args.command == 'new':
f = open(args.filename, 'w')
f.write(sample_script)
f.close()
return
elif args.command == 'reset':
reset_retriever(args.scope)
return
elif args.command == 'new_json':
# create new JSON script
create_json()
return
elif args.command == 'edit_json':
# edit existing JSON script
json_file = get_script_filename(args.dataset.lower())
edit_json(json_file)
return
elif args.command == 'autocreate':
if sum([args.f, args.d]) == 1:
file_flag = True if args.f else False
create_package(args.path, args.dt, file_flag, args.o,
args.skip_lines)
else:
print('Please use one and only one of the flags -f -d')
return
elif args.command == 'delete_json':
# delete existing JSON script from home directory and or script directory if exists in current dir
confirm = input("Really remove " + args.dataset.lower() +
" and all its contents? (y/N): ")
if confirm.lower().strip() in ['y', 'yes']:
json_file = get_script_filename(args.dataset.lower())
delete_json(json_file)
return
if args.command == 'ls':
# scripts should never be empty because check_for_updates is run on SCRIPT_LIST init
if not (args.l or args.k or isinstance(args.v, list)):
all_scripts = dataset_names()
print("Available datasets : {}\n".format(len(all_scripts)))
from retriever import lscolumns
lscolumns.printls(all_scripts)
elif isinstance(args.v, list):
if args.v:
try:
all_scripts = [
get_script(dataset) for dataset in args.v
]
except KeyError:
all_scripts = []
print("Dataset(s) is not found.")
else:
all_scripts = datasets()
count = 1
for script in all_scripts:
print("{count}. {title}\n {name}\n"
"{keywords}\n{description}\n"
"{licenses}\n{citation}\n"
"".format(
count=count,
title=script.title,
name=script.name,
keywords=script.keywords,
description=script.description,
licenses=str(script.licenses[0]['name']),
citation=script.citation,
))
count += 1
else:
param_licenses = args.l if args.l else None
keywords = args.k if args.k else None
# search
searched_scripts = datasets(keywords, param_licenses)
if not searched_scripts:
print("No available datasets found")
else:
print("Available datasets : {}\n".format(
len(searched_scripts)))
count = 1
for script in searched_scripts:
print("{count}. {title}\n{name}\n"
"{keywords}\n{licenses}\n".format(
count=count,
title=script.title,
name=script.name,
keywords=script.keywords,
licenses=str(script.licenses[0]['name']),
))
count += 1
return
engine = choose_engine(args.__dict__)
if hasattr(args, 'debug') and args.debug:
debug = True
else:
debug = False
sys.tracebacklimit = 0
if hasattr(args, 'debug') and args.not_cached:
engine.use_cache = False
else:
engine.use_cache = True
if args.dataset is not None:
scripts = name_matches(script_list, args.dataset)
else:
raise Exception("no dataset specified.")
if scripts:
for dataset in scripts:
print("=> Installing", dataset.name)
try:
dataset.download(engine, debug=debug)
dataset.engine.final_cleanup()
except KeyboardInterrupt:
pass
except Exception as e:
print(e)
if debug:
raise
print("Done!")
else:
print(
"Run 'retriever ls' to see a list of currently available datasets."
)
def main():
"""This function launches the Data Retriever."""
if len(sys.argv) == 1:
# if no command line args are passed, show the help options
parser.parse_args(['-h'])
else:
# otherwise, parse them
args = parser.parse_args()
if args.command not in ['reset', 'update'] \
and not os.path.isdir(SCRIPT_SEARCH_PATHS[1]) \
and not [f for f in os.listdir(SCRIPT_SEARCH_PATHS[-1])
if os.path.exists(SCRIPT_SEARCH_PATHS[-1])]:
check_for_updates()
reload_scripts()
script_list = SCRIPT_LIST()
if args.command == "install" and not args.engine:
parser.parse_args(['install', '-h'])
if args.quiet:
sys.stdout = open(os.devnull, 'w')
if args.command == 'help':
parser.parse_args(['-h'])
if hasattr(args, 'compile') and args.compile:
script_list = reload_scripts()
if args.command == 'defaults':
for engine_item in engine_list:
print("Default options for engine ", engine_item.name)
for default_opts in engine_item.required_opts:
print(default_opts[0], " ", default_opts[2])
print()
return
if args.command == 'update':
check_for_updates()
reload_scripts()
return
elif args.command == 'citation':
if args.dataset is None:
print("\nCitation for retriever:\n")
print(CITATION)
else:
scripts = name_matches(script_list, args.dataset)
for dataset in scripts:
print("\nDataset: {}".format(dataset.name))
print("Citation: {}".format(dataset.citation))
print("Description: {}\n".format(dataset.description))
return
elif args.command == 'license':
if args.dataset is None:
print(LICENSE)
else:
dataset_license = license(args.dataset)
if dataset_license:
print(dataset_license)
else:
print("There is no license information for {}".format(args.dataset))
return
elif args.command == 'new':
f = open(args.filename, 'w')
f.write(sample_script)
f.close()
return
elif args.command == 'reset':
reset_retriever(args.scope)
return
elif args.command == 'new_json':
# create new JSON script
create_json()
return
elif args.command == 'edit_json':
# edit existing JSON script
json_file = get_script_filename(args.dataset.lower())
edit_json(json_file)
return
elif args.command == 'autocreate':
if sum([args.f, args.d]) == 1:
file_flag = True if args.f else False
create_package(args.path, args.dt, file_flag, args.o, args.skip_lines)
else:
print('Please use one and only one of the flags -f -d')
return
elif args.command == 'delete_json':
# delete existing JSON script from home directory and or script directory if exists in current dir
confirm = input("Really remove " + args.dataset.lower() +
" and all its contents? (y/N): ")
if confirm.lower().strip() in ['y', 'yes']:
json_file = get_script_filename(args.dataset.lower())
delete_json(json_file)
return
if args.command == 'ls':
# scripts should never be empty because check_for_updates is run on SCRIPT_LIST init
if not (args.l or args.k or isinstance(args.v, list)):
all_scripts = dataset_names()
print("Available datasets : {}\n".format(len(all_scripts)))
from retriever import lscolumns
lscolumns.printls(all_scripts)
elif isinstance(args.v, list):
if args.v:
try:
all_scripts = [get_script(dataset) for dataset in args.v]
except KeyError:
all_scripts = []
print("Dataset(s) is not found.")
else:
all_scripts = datasets()
count = 1
for script in all_scripts:
print(
"{count}. {title}\n {name}\n"
"{keywords}\n{description}\n"
"{licenses}\n{citation}\n"
"".format(
count=count,
title=script.title,
name=script.name,
keywords=script.keywords,
description=script.description,
licenses=str(script.licenses[0]['name']),
citation=script.citation,
)
)
count += 1
else:
param_licenses = args.l if args.l else None
keywords = args.k if args.k else None
# search
searched_scripts = datasets(keywords, param_licenses)
if not searched_scripts:
print("No available datasets found")
else:
print("Available datasets : {}\n".format(len(searched_scripts)))
count = 1
for script in searched_scripts:
print(
"{count}. {title}\n{name}\n"
"{keywords}\n{licenses}\n".format(
count=count,
title=script.title,
name=script.name,
keywords=script.keywords,
licenses=str(script.licenses[0]['name']),
)
)
count += 1
return
engine = choose_engine(args.__dict__)
if hasattr(args, 'debug') and args.debug:
debug = True
else:
debug = False
sys.tracebacklimit = 0
if hasattr(args, 'debug') and args.not_cached:
engine.use_cache = False
else:
engine.use_cache = True
if args.dataset is not None:
scripts = name_matches(script_list, args.dataset)
else:
raise Exception("no dataset specified.")
if scripts:
for dataset in scripts:
print("=> Installing", dataset.name)
try:
dataset.download(engine, debug=debug)
dataset.engine.final_cleanup()
except KeyboardInterrupt:
pass
except Exception as e:
print(e)
if debug:
raise
print("Done!")
else:
print("Run 'retriever ls' to see a list of currently available datasets.")
def main():
"""This function launches the Data Retriever."""
sys.argv[1:] = [arg.lower() for arg in sys.argv[1:]]
if len(sys.argv) == 1:
# if no command line args are passed, show the help options
parser.parse_args(['-h'])
else:
# otherwise, parse them
if not os.path.isdir(SCRIPT_SEARCH_PATHS[1]) and not \
[f for f in os.listdir(SCRIPT_SEARCH_PATHS[-1])
if os.path.exists(SCRIPT_SEARCH_PATHS[-1])]:
check_for_updates()
script_list = SCRIPT_LIST()
args = parser.parse_args()
if args.command == "install" and not args.engine:
parser.parse_args(['install', '-h'])
if args.quiet:
sys.stdout = open(os.devnull, 'w')
if args.command == 'help':
parser.parse_args(['-h'])
if hasattr(args, 'compile') and args.compile:
script_list = SCRIPT_LIST(force_compile=True)
if args.command == 'defaults':
for engine_item in engine_list:
print("Default options for engine ", engine_item.name)
for default_opts in engine_item.required_opts:
print(default_opts[0], " ", default_opts[2])
print()
return
if args.command == 'update':
check_for_updates(False)
script_list = SCRIPT_LIST()
return
elif args.command == 'citation':
if args.dataset is None:
print("\nCitation for retriever:\n")
print(CITATION)
else:
scripts = name_matches(script_list, args.dataset)
for dataset in scripts:
print("\nDataset: {}".format(dataset.name))
print("Citation: {}".format(dataset.citation))
print("Description: {}\n".format(dataset.description))
return
elif args.command == 'license':
dataset_license = license(args.dataset)
if dataset_license:
print(dataset_license)
else:
print("There is no license information for {}".format(args.dataset))
return
elif args.command == 'new':
f = open(args.filename, 'w')
f.write(sample_script)
f.close()
return
elif args.command == 'reset':
reset_retriever(args.scope)
return
elif args.command == 'new_json':
# create new JSON script
create_json()
return
elif args.command == 'edit_json':
# edit existing JSON script
json_file = get_script_filename(args.dataset.lower())
edit_json(json_file)
return
elif args.command == 'delete_json':
# delete existing JSON script from home directory and or script directory if exists in current dir
confirm = input("Really remove " + args.dataset.lower() +
" and all its contents? (y/N): ")
if confirm.lower().strip() in ['y', 'yes']:
json_file = get_script_filename(args.dataset.lower())
delete_json(json_file)
return
if args.command == 'ls':
# If scripts have never been downloaded there is nothing to list
if not script_list:
print("No scripts are currently available. Updating scripts now...")
check_for_updates(False)
print("\n\nScripts downloaded.\n")
if args.l is None:
all_scripts = datasets()
print("Available datasets : {}\n".format(len(all_scripts)))
from retriever import lscolumns
lscolumns.printls(dataset_names())
else:
all_scripts = datasets(args.l[0])
print("Available datasets : {}\n".format(len(all_scripts)))
count = 1
for script in all_scripts:
print("{}. {}".format(count, script.title))
print(script.name)
print(script.keywords)
print()
count += 1
return
engine = choose_engine(args.__dict__)
if hasattr(args, 'debug') and args.debug:
debug = True
else:
debug = False
sys.tracebacklimit = 0
if hasattr(args, 'debug') and args.not_cached:
engine.use_cache = False
else:
engine.use_cache = True
if args.dataset is not None:
scripts = name_matches(script_list, args.dataset)
else:
raise Exception("no dataset specified.")
if scripts:
for dataset in scripts:
print("=> Installing", dataset.name)
try:
dataset.download(engine, debug=debug)
dataset.engine.final_cleanup()
except KeyboardInterrupt:
pass
except Exception as e:
print(e)
if debug:
raise
print("Done!")
else:
print("The dataset {} isn't currently available in the Retriever".format(
args.dataset))
print("Run 'retriever ls to see a list of currently available datasets")