def api_get_project_data(project_id): # noqa: F401
"""Get info on the article"""
if not is_project(project_id):
response = jsonify(message="Project not found.")
return response, 404
try:
filename = get_data_file_path(project_id).stem
# get statistics of the dataset
statistics = get_data_statistics(project_id)
statistics["filename"] = filename
except FileNotFoundError as err:
print(err)
statistics = {"filename": None}
except Exception as err:
print(err)
message = f"Failed to get file. {err}"
return jsonify(message=message), 400
response = jsonify(statistics)
response.headers.add('Access-Control-Allow-Origin', '*')
return response
def api_upload_data_project(project_id): # noqa: F401
"""Get info on the article"""
if not is_project(project_id):
response = jsonify(message="project-not-found")
return response, 404
if request.form.get('demo_data', None):
# download file and save to folder
demo_data = get_dataset(request.form['demo_data'])
if demo_data.dataset_id in ["hall", "ace", "ptsd"]:
download_url = demo_data.url_demo
else:
download_url = demo_data.url
url_parts = urllib.parse.urlparse(download_url)
filename = url_parts.path.rsplit('/', 1)[-1]
urlretrieve(download_url, get_data_path(project_id) / filename)
add_dataset_to_project(project_id, filename)
elif 'file' in request.files:
data_file = request.files['file']
# check the file is file is in a correct format
check_dataset(data_file) # TODO{qubixes}: implement val strategy
try:
filename = secure_filename(data_file.filename)
fp_data = get_data_path(project_id) / filename
# save the file
data_file.save(str(fp_data))
add_dataset_to_project(project_id, filename)
except Exception as err:
logging.error(err)
response = jsonify(message="project-upload-failure")
return response, 500
else:
response = jsonify(message="no-file-found")
return response, 500
return jsonify({"success": True})
def api_import_project():
"""Import uploaded project"""
import_project_id = None
if 'file' in request.files:
project_file = request.files['file']
filename = secure_filename(project_file.filename)
try:
with zipfile.ZipFile(project_file, "r") as zipObj:
FileNames = zipObj.namelist()
# check if the zip file contains a ASReview project
if sum([fn.endswith("project.json") for fn in FileNames]) == 1:
# extract all files to a temporary folder
tmpdir = tempfile.TemporaryDirectory()
zipObj.extractall(path=tmpdir.name)
for fn in FileNames:
if fn.endswith("project.json"):
fp = Path(tmpdir.name, fn)
with open(fp, "r+") as f:
project = json.load(f)
# if the uploaded project already exists,
# then make a copy
if is_project(project["id"]):
project["id"] += " copy"
project["name"] += " copy"
f.seek(0)
json.dump(project, f)
f.truncate()
else:
response = jsonify(
message="No project found within the chosen file.")
return response, 404
try:
# check if a copy of a project already exists
os.rename(tmpdir.name, asreview_path() / f"{project['id']}")
import_project_id = project['id']
except Exception as err:
logging.error(err)
response = jsonify(
message=f"A copy of {project['id'][:-5]} already exists.")
return response, 400
except Exception as err:
logging.error(err)
response = jsonify(message=f"Failed to upload file '{filename}'.")
return response, 400
else:
response = jsonify(message="No file found to upload.")
return response, 400
# return the project info in the same format as project_info
return api_get_project_info(import_project_id)
def api_upload_data_to_project(project_id): # noqa: F401
"""Get info on the article"""
if not is_project(project_id):
response = jsonify(message="Project not found.")
return response, 404
if request.form.get('demo_data', None):
# download file and save to folder
demo_data = DatasetManager().find(request.form['demo_data'])
if demo_data.dataset_id in ["hall", "ace", "ptsd"]:
download_url = demo_data.url_demo
else:
download_url = demo_data.url
url_parts = urllib.parse.urlparse(download_url)
filename = secure_filename(url_parts.path.rsplit('/', 1)[-1])
urlretrieve(download_url, get_data_path(project_id) / filename)
elif request.form.get('url', None):
# download file and save to folder
download_url = request.form['url']
try:
url_parts = urllib.parse.urlparse(download_url)
filename = secure_filename(url_parts.path.rsplit('/', 1)[-1])
urlretrieve(download_url, get_data_path(project_id) / filename)
except ValueError as err:
logging.error(err)
message = f"Invalid URL '{download_url}'."
if isinstance(download_url, str) \
and not download_url.startswith("http"):
message += " Usually, the URL starts with 'http' or 'https'."
return jsonify(message=message), 400
except Exception as err:
logging.error(err)
message = f"Can't retrieve data from URL {download_url}."
return jsonify(message=message), 400
elif 'file' in request.files:
data_file = request.files['file']
# check the file is file is in a correct format
check_dataset(data_file) # TODO{qubixes}: implement val strategy
try:
filename = secure_filename(data_file.filename)
fp_data = get_data_path(project_id) / filename
# save the file
data_file.save(str(fp_data))
except Exception as err:
logging.error(err)
response = jsonify(
message=f"Failed to upload file '{filename}'. {err}")
return response, 400
else:
response = jsonify(message="No file or dataset found to upload.")
return response, 400
try:
# add the file to the project
add_dataset_to_project(project_id, filename)
# Bad format. TODO{Jonathan} Return informative message with link.
except BadFileFormatError as err:
message = f"Failed to upload file '{filename}'. {err}"
return jsonify(message=message), 400
response = jsonify({'success': True})
response.headers.add('Access-Control-Allow-Origin', '*')
return response
def api_import_project():
"""Import uploaded project"""
# raise error if file not given
if 'file' not in request.files:
response = jsonify(message="No file found to upload.")
return response, 400
# import project id
import_project = None
# set the project file
project_file = request.files['file']
try:
with zipfile.ZipFile(project_file, "r") as zip_obj:
zip_filenames = zip_obj.namelist()
# raise error if no ASReview project file
if "project.json" not in zip_filenames:
response = jsonify(
message="File doesn't contain valid project format.")
return response, 404
# extract all files to a temporary folder
tmpdir = tempfile.TemporaryDirectory()
zip_obj.extractall(path=tmpdir.name)
# Open the project file and check the id. The id needs to be
# unique, otherwise it is exended with -copy.
fp = Path(tmpdir.name, "project.json")
with open(fp, "r+") as f:
# load the project info in scope of function
import_project = json.load(f)
# If the uploaded project already exists,
# then overwrite project.json with a copy suffix.
while is_project(import_project["id"]):
# project update
import_project["id"] = f"{import_project['id']}-copy"
import_project["name"] = f"{import_project['name']} copy"
else:
# write to file
f.seek(0)
json.dump(import_project, f)
f.truncate()
except Exception as err:
# Unknown error.
logging.error(err)
response = jsonify(message="Unknown error when uploading project "
f"'{project_file.filename}'.")
return response, 400
# location to copy file to
fp_copy = get_project_path(import_project["id"])
try:
# Move the project from the temp folder to the projects folder.
os.rename(tmpdir.name, fp_copy)
except Exception as err:
logging.error(err)
response = jsonify(message=f"Failed to copy project to {fp_copy}.")
return response, 400
# return the project info in the same format as project_info
return api_get_project_info(import_project["id"])
def api_upload_data_to_project(project_id): # noqa: F401
"""Get info on the article"""
if not is_project(project_id):
response = jsonify(message="Project not found.")
return response, 404
if request.form.get('plugin', None):
plugin_data = DatasetManager().find(request.form['plugin'])
url_parts = urllib.parse.urlparse(plugin_data.url)
filename = secure_filename(url_parts.path.rsplit('/', 1)[-1])
urlretrieve(plugin_data.url, get_data_path(project_id) / filename)
elif request.form.get('benchmark', None):
benchmark_dataset_id = DatasetManager().find(request.form['benchmark'])
# read dataset
df = pd.read_csv(benchmark_dataset_id.url)
# rename label column
df.rename({"label_included": "debug_label"}, axis=1, inplace=True)
# define export filepath
url_parts = urllib.parse.urlparse(benchmark_dataset_id.url)
filename = secure_filename(url_parts.path.rsplit('/', 1)[-1])
export_fp = get_data_path(project_id) / filename
# export file
df.to_csv(export_fp, index=False)
elif request.form.get('url', None):
# download file and save to folder
download_url = request.form['url']
try:
url_parts = urllib.parse.urlparse(download_url)
filename = secure_filename(url_parts.path.rsplit('/', 1)[-1])
urlretrieve(download_url, get_data_path(project_id) / filename)
except ValueError as err:
logging.error(err)
message = f"Invalid URL '{download_url}'."
if isinstance(download_url, str) \
and not download_url.startswith("http"):
message += " Usually, the URL starts with 'http' or 'https'."
return jsonify(message=message), 400
except Exception as err:
logging.error(err)
message = f"Can't retrieve data from URL {download_url}."
return jsonify(message=message), 400
elif 'file' in request.files:
data_file = request.files['file']
# check the file is file is in a correct format
check_dataset(data_file) # TODO{qubixes}: implement val strategy
try:
filename = secure_filename(data_file.filename)
fp_data = get_data_path(project_id) / filename
# save the file
data_file.save(str(fp_data))
except Exception as err:
logging.error(err)
response = jsonify(
message=f"Failed to upload file '{filename}'. {err}")
return response, 400
else:
response = jsonify(message="No file or dataset found to upload.")
return response, 400
try:
# add the file to the project
add_dataset_to_project(project_id, filename)
# Bad format. TODO{Jonathan} Return informative message with link.
except BadFileFormatError as err:
message = f"Failed to upload file '{filename}'. {err}"
return jsonify(message=message), 400
response = jsonify({'success': True})
response.headers.add('Access-Control-Allow-Origin', '*')
return response
def api_import_project():
"""Import uploaded project"""
if 'file' in request.files:
project_file = request.files['file']
filename = secure_filename(project_file.filename)
# check the file format
if not os.path.splitext(filename)[1] == ".asreview":
response = jsonify(message="Incorrect file format.")
return response, 400
try:
with zipfile.ZipFile(project_file, "r") as zipObj:
FileNames = zipObj.namelist()
# check if the zip file contains a ASReview project
if sum([fn.endswith("project.json") for fn in FileNames]) == 1:
# extract all files to a temporary folder
tmpdir = tempfile.TemporaryDirectory()
zipObj.extractall(path=tmpdir.name)
for fn in FileNames:
if fn.endswith("project.json"):
fp = Path(tmpdir.name, fn)
with open(fp, "r+") as f:
project = json.load(f)
# if the uploaded project already exists, then make a copy
if is_project(project["id"]):
project["id"] += " copy"
project["name"] += " copy"
f.seek(0)
json.dump(project, f)
f.truncate()
else:
response = jsonify(message="No project found within the chosen file.")
return response, 404
try:
# check if a copy of a project already exists
os.rename(tmpdir.name, asreview_path() / f"{project['id']}")
except Exception as err:
logging.error(err)
response = jsonify(message=f"A copy of {project['id'][:-5]} already exists.")
return response, 400
except Exception as err:
logging.error(err)
response = jsonify(message=f"Failed to upload file '{filename}'. {err}")
return response, 400
else:
response = jsonify(message="No file found to upload.")
return response, 400
response = jsonify({'success': True})
response.headers.add('Access-Control-Allow-Origin', '*')
return response
