def find_qualtrics_readme(self, dn: DepositorName):
"""Get Response ID based on a article_id,curation_id search"""
dn_dict = dn.name_dict
qualtrics_df = self.get_survey_responses(self.readme_survey_id)
# First perform search via article_id or curation_id
self.log.info("Attempting to identify using article_id or curation_id ...")
article_id = str(dn_dict['article_id'])
curation_id = str(dn_dict['curation_id'])
try:
response_df = qualtrics_df[(qualtrics_df['article_id'] == article_id) |
(qualtrics_df['curation_id'] == curation_id)]
except KeyError:
self.log.warn("article_id and curation_id not in qualtrics survey !")
response_df = pd.DataFrame()
if response_df.empty:
self.log.warn("Empty DataFrame")
raise ValueError
else:
self.log.info("Unique match based on article_id or curation_id !")
self.pandas_write_buffer(response_df[readme_cols_order])
if response_df.shape[0] == 1:
response_dict = df_to_dict_single(response_df)
self.log.info("Only one entry found!")
self.log.info(f"Survey completed on {response_dict['date_completed']}")
self.log.info(f" ... for {response_dict['article_id']}")
return response_dict['ResponseId'], response_df
else:
self.log.warn("Multiple entries found")
raise ValueError
def find_deposit_agreement(self, dn_dict):
"""Get Response ID based on a match search for depositor name"""
qualtrics_df = self.get_survey_responses()
# First perform search via article_id or curation_id
print("Attempting to identify using article_id or curation_id")
article_id = str(dn_dict['article_id'])
curation_id = str(dn_dict['curation_id'])
try:
response_df = qualtrics_df[
(qualtrics_df['article_id'] == article_id) |
(qualtrics_df['curation_id'] == curation_id)]
except KeyError:
print("article_id and curation_id not in qualtrics survey")
response_df = pd.DataFrame()
if not response_df.empty:
print("Unique match based on article_id or curation_id !")
if response_df.shape[0] != 1:
print("More than one entries found !!!")
print(response_df[cols_order].to_markdown())
else:
print("Unable to identify based article_id or curation_id.")
print("Attempting to identify with name")
response_df = qualtrics_df[
(qualtrics_df['Q4_1'] == dn_dict['fullName']) |
(qualtrics_df['Q4_1'] == dn_dict['simplify_fullName']) |
(qualtrics_df['Q4_2'] == dn_dict['depositor_email'])]
# Identify corresponding author cases if different from depositor name
if not dn_dict['self_deposit'] and not response_df.empty:
print(
"Not self-deposit. Identifying based on corresponding author as well"
)
df_select = response_df[(
response_df['Q6_1'] == dn_dict['authors'][0])]
if df_select.empty:
print("Unable to identify based on corresponding author")
print("Listing all deposit agreements based on Depositor")
print(response_df[cols_order].to_markdown())
else:
response_df = df_select
if response_df.empty:
print("Empty DataFrame")
raise ValueError
else:
if response_df.shape[0] == 1:
response_dict = df_to_dict_single(response_df)
print("Only one entry found!")
print("Survey completed on {} for {}".format(
response_dict['Date Completed'], response_dict['Q7']))
return response_dict['ResponseId']
else:
print("Multiple entries found")
print(response_df[cols_order].to_markdown())
raise ValueError
def get_curation_id(self):
# This retrieves basic curation information for article (this includes all curation)
cur_df = self.fs_admin.get_curation_list(article_id=self.article_id)
# By default it retrieves the most recent one
cur_loc_dict = df_to_dict_single(cur_df)
return cur_loc_dict['id']
def get_name_dict(self):
if self.verbose:
print("Retrieving depositor_name for {} ... ".format(
self.article_id))
account_id = self.curation_dict['account_id']
acct_df = self.fs_admin.get_account_list()
temp_dict = df_to_dict_single(acct_df.loc[acct_df['id'] == account_id])
surName = temp_dict['last_name'] # full last name
firstName = temp_dict['first_name'] # full first name
simplify_firstName = firstName.split(' ')[0]
simplify_surName = surName.split(' ')[0]
fullName = "{} {}".format(firstName, surName)
simplify_fullName = "{} {}".format(simplify_firstName,
simplify_surName)
name_dict = dict()
name_dict['surName'] = surName
name_dict['firstName'] = firstName
name_dict['simplify_firstName'] = simplify_firstName
name_dict['simplify_surName'] = simplify_surName
name_dict['fullName'] = fullName
name_dict['simplify_fullName'] = simplify_fullName
authors = [
d['full_name'] for d in self.curation_dict['item']['authors']
]
name_dict['authors'] = authors
if fullName in authors or simplify_fullName in authors:
name_dict['self_deposit'] = True
else:
name_dict['self_deposit'] = False
# Add additional information about deposit, such as article and
# curation IDs, email, and title
name_dict['article_id'] = self.article_id
name_dict['curation_id'] = self.curation_id
name_dict['depositor_email'] = temp_dict['email']
name_dict['title'] = self.curation_dict['item']['title']
return name_dict
def retrieve_qualtrics_readme(self, dn=None, ResponseId='', browser=True,
save_metadata: bool = False):
"""Retrieve response to Qualtrics README form"""
if ResponseId:
response_df = self.get_survey_response(self.readme_survey_id, ResponseId)
else:
try:
ResponseId, response_df = self.find_qualtrics_readme(dn)
self.log.info(f"Qualtrics README ResponseID : {ResponseId}")
except ValueError:
self.log.warn("Error with retrieving ResponseId")
self.log.info("PROMPT: If you wish, you can manually enter ResponseId to retrieve.")
if self.interactive:
ResponseId = input("PROMPT: An EMPTY RETURN will generate a custom Qualtrics link to provide ... ")
self.log.info(f"RESPONSE: {ResponseId}")
else:
self.log.info("Interactive mode disabled. Skipping manual input")
ResponseId = ''
if ResponseId:
response_df = self.get_survey_response(self.readme_survey_id, ResponseId)
else:
response_df = pd.DataFrame()
readme_url = self.generate_readme_url(dn)
self.log.info(f"README URL: {readme_url}")
if response_df.empty:
self.log.warn("Empty DataFrame")
self.log.info("Filling with empty content")
qualtrics_dict = {}
for field in readme_custom_content:
qualtrics_dict[field] = 'nan'
qualtrics_dict['references'] = []
else:
qualtrics_dict = df_to_dict_single(response_df[readme_custom_content])
for key in qualtrics_dict.keys():
if isinstance(qualtrics_dict[key], float):
qualtrics_dict[key] = str(qualtrics_dict[key])
# Separate cite, contrib for list style
for field in ['cite', 'contrib']:
if qualtrics_dict[field] != 'nan':
qualtrics_dict[field] = qualtrics_dict[field].split('\n')
# Markdown files, materials
for field in ['files', 'materials']:
if qualtrics_dict[field] != 'nan':
if qualtrics_dict[field][0] == "'":
qualtrics_dict[field] = qualtrics_dict[field][1:]
self.log.debug(f"Removing extra single quote in {field} entry")
# Retrieve corresponding author info and append
self.log.info("Appending Deposit Agreement's Corresponding Author metadata")
if not self.da_response_id:
self.log.info("NO METADATA - Retrieving Deposit Agreement metadata")
self.find_deposit_agreement(dn)
else:
self.log.info(f"Parsed ResponseId : {self.da_response_id}")
self.log.info(f"Parsed SurveyID : {self.da_survey_id}")
DA_response_df = self.get_survey_response(self.da_survey_id, self.da_response_id)
DA_dict = df_to_dict_single(DA_response_df)
qualtrics_dict['corr_author_fullname'] = DA_dict['Q6_1']
qualtrics_dict['corr_author_email'] = DA_dict['Q6_2']
qualtrics_dict['corr_author_affil'] = DA_dict['Q6_3']
# Save Qualtrics README metadata
if save_metadata:
out_file_prefix = "qualtrics_readme_original_" + \
f"{dn.name_dict['article_id']}"
self.save_metadata(qualtrics_dict, dn,
out_file_prefix=out_file_prefix)
return qualtrics_dict
def find_deposit_agreement(self, dn: DepositorName):
"""Get Response ID based on a match search for depositor name"""
merged_df = self.merge_survey()
dn_dict = dn.name_dict
# First perform search via article_id or curation_id
self.log.info("Attempting to identify using article_id or curation_id ...")
article_id = str(dn_dict['article_id'])
curation_id = str(dn_dict['curation_id'])
try:
response_df = merged_df[(merged_df['article_id'] == article_id) |
(merged_df['curation_id'] == curation_id)]
except KeyError:
self.log.warn("article_id and curation_id not in qualtrics survey !")
response_df = pd.DataFrame()
if not response_df.empty:
self.log.info("Unique match based on article_id or curation_id !")
if response_df.shape[0] != 1:
self.log.warn("More than one entries found !!!")
else:
self.log.info("Unable to identify based on article_id or curation_id ...")
self.log.info("Attempting to identify with name ...")
response_df = merged_df[(merged_df['Q4_1'] == dn_dict['fullName']) |
(merged_df['Q4_1'] == dn_dict['simplify_fullName']) |
(merged_df['Q4_2'] == dn_dict['depositor_email'])]
# Identify corresponding author cases if different from depositor name
if not dn_dict['self_deposit'] and not response_df.empty:
self.log.info("Not self-deposit. Identifying based on corresponding author as well ...")
df_select = response_df[(response_df['Q6_1'] == dn_dict['authors'][0])]
if df_select.empty:
self.log.warn("Unable to identify based on corresponding author")
self.log.info("Listing all deposit agreements based on Depositor")
self.pandas_write_buffer(response_df[cols_order])
else:
response_df = df_select
if response_df.empty:
self.log.warn("Empty DataFrame")
raise ValueError
else:
if response_df.shape[0] == 1:
response_dict = df_to_dict_single(response_df)
self.save_metadata(response_dict, dn, out_file_prefix=
f'deposit_agreement_original_{article_id}')
self.pandas_write_buffer(response_df[cols_order])
self.log.info("Only one entry found!")
self.log.info(f"Survey completed on {response_dict['Date Completed']}")
self.log.info(f" ... for {response_dict['Q7']}")
survey_shortname = \
self.lookup_survey_shortname(response_dict['SurveyID'])
self.log.info(f"Survey name: {survey_shortname}")
self.da_response_id = response_dict['ResponseId']
self.da_survey_id = response_dict['SurveyID']
return response_dict['ResponseId'], response_dict['SurveyID']
else:
self.log.warn("Multiple entries found")
self.pandas_write_buffer(response_df[cols_order])
raise ValueError