def add_unique_id_to_net_sup_edges(path_net_sup_edges):
"""Input: File path of network supplement edges.
Output: Adds a unique ID column to the network supplement edges file."""
pids_from_subs_ids = w.get_data_from_one_col_as_list(
c.SUBJECTS_IDS_PATH, c.LABEL_PID)
unique_ids_from_subs_ids = w.get_data_from_one_col_as_list(
c.SUBJECTS_IDS_PATH, c.LABEL_UNIQUE_ID)
net_sup_edges = w.get_csv_as_list(path_net_sup_edges)
net_sup_edges_data = net_sup_edges[1:]
net_sup_edges_cols = net_sup_edges[:1][0]
net_sup_edges_cols.append(c.LABEL_UNIQUE_ID)
# make temp file to add unique ID column
w.create_csv_add_column_labels(c.TEMP_FILE_PATH, net_sup_edges_cols)
pid_index_net_sup_edges = w.get_index_of_file_col(path_net_sup_edges,
c.LABEL_SENDER_PID)
for row in net_sup_edges_data:
# get index of this row's project ID in the subjects_ids.csv
subs_ids_index = pids_from_subs_ids.index(row[pid_index_net_sup_edges])
unique_id = unique_ids_from_subs_ids[subs_ids_index]
row.append(unique_id)
# append the matching unique ID to the row and to the temp file
w.append_row_to_csv(c.TEMP_FILE_PATH, row)
w.rename_csv(
c.TEMP_FILE_PATH,
path_net_sup_edges) # replace the previous file with the temp file
def create_p1_p2_overlaps():
"""Input: None.
Output: Create the overlap file with entries for every participant who was present
in both p1_screenings and p2_network_interviews; populates it with unique id, project id, rds id."""
path_screenings = c.P1_SCREENINGS_PATH
path_network = c.P2_NET_SUPS_EXTRACT_PATH
path_nodes = c.OLD_NODES_PATH
path_overlap = c.P1_P2_OVERLAPS_PATH
pid, rds, unique = c.LABEL_PID, c.LABEL_RDS_ID, c.LABEL_UNIQUE_ID
w.create_csv_add_column_labels(path_overlap, [pid, rds, unique])
# get rds_ids from p1_screenings and unique_ids from p2_network_interviews
screening_rds_id_data = w.get_data_from_one_col_as_list(path_screenings, rds)
net_unique_id_data = w.get_data_from_one_col_as_list(path_network, unique)
# makes dictionaries from unique_ids to rds_ids and from unique_ids to pids, extracted from the old nodes files
unique_to_rds_dict = w.get_no_null_entries_dict_from_csv(path_nodes, unique, rds)
unique_to_pid_dict = w.get_no_null_entries_dict_from_csv(path_nodes, unique, pid)
for unique_id, rds_id in unique_to_rds_dict.items(): # for every unique_id/rds_id pair that exists in whole dataset
try: # try to remove the unique_id from p2_network_interviews
net_unique_id_data.remove(unique_id)
except ValueError: # if it does not exist in p2_netowrk_interviews,
continue # do not add it to the overlap file and try the next unique_id/rds_id pair
try: # try to remove the rds_id from p1_screenings
screening_rds_id_data.remove(rds_id)
except ValueError: # if it does not exist in p1_screenings,
continue # do not add it to the overlap file and try the next unique_id/rds_id pair
p_id = unique_to_pid_dict[unique_id] # if unique_id/rds_id exist in p2/p1 respectively, get the associated pid
row = [p_id, rds_id, unique_id]
w.append_row_to_csv(path_overlap, row) # add this p1/p2 overlap row of ids to the overlap file
def create_rds_edges_csv(path_rds_edges, old_edge_data, sender_index, receiver_index, rds_index):
"""Input: rds_edges.csv file path; data from the old edge file;
sender, receiver, and rds edge index in that file; edge column label.
Output: Creates the rds_edges.csv file and populates it with all rds edge connections."""
w.create_csv_add_column_labels(path_rds_edges, [c.LABEL_EDGE_ID])
for row in old_edge_data: # for each row from the old edge file,
if row[rds_index] == 'Yes': # if the column RDS Edge reads Yes,
edge_id = row[sender_index] + row[receiver_index]
w.append_row_to_csv(path_rds_edges, [edge_id]) # and add edge_id to the rds_edges table
def create_network_edges_csv(path_net_edges, old_edge_data, sender_index, receiver_index, net_index):
"""Input: net_edges.csv file path; data from the old edge file; sender_receiver to edge_id dictionary;
sender, receiver, and network supplement index from old edges file; edge column label.
Output: Creates the network_edges.csv file and populates it with all network supplement edge connections."""
w.create_csv_add_column_labels(path_net_edges, [c.LABEL_EDGE_ID])
for row in old_edge_data: # for each row from the old edge file,
if row[net_index] == 'Yes': # if the column Net Supplement reads Yes,
edge_id = row[sender_index] + row[receiver_index]
w.append_row_to_csv(path_net_edges, [edge_id]) # and add edge_id to the network_edges table
def create_edges_csv(path_edges, old_edge_data, sender_index, receiver_index):
"""Input: edges.csv file path; data from the old edge file;
sender and receiver index in that file.
Output: Creates edges.csv file and populates it with all edge connections."""
w.create_csv_add_column_labels(path_edges,
[c.LABEL_EDGE_ID, c.LABEL_SENDER_PID, c.LABEL_RECEIVER_PID])
for i in range(len(old_edge_data)):
row = old_edge_data[i] # get each row from the old edge file
w.append_row_to_csv(path_edges, [row[sender_index] + row[receiver_index], # and append the relevant data
row[sender_index], row[receiver_index]]) # to the new file
def make_net_sup_extract_file(path_old_net, path_net_sup_extract):
"""Input: File path of deprecated network supplement and new net sup extract.
Output: Extracts P2SF and P2NS1 columns from deprecated net sup, gets rds ids, and makes net sup extract file."""
# make unique id column in deprecated net sup file
w.add_merged_col_to_csv(path_old_net, c.LABEL_UNIQUE_ID,
c.NETWORK_HCV_HIV_COLS_TO_MERGE)
unique_ids_from_subs_ids = w.get_data_from_one_col_as_list(
c.SUBJECTS_IDS_PATH, c.LABEL_UNIQUE_ID)
unique_ids_from_depr_net_sup = w.get_data_from_one_col_as_list(
path_old_net, c.LABEL_UNIQUE_ID)
subs_ids_data = w.get_csv_as_list(c.SUBJECTS_IDS_PATH)[1:]
depr_net_sup_data = w.get_csv_as_list(path_old_net)[1:]
w.create_csv_add_column_labels(
path_net_sup_extract,
[c.LABEL_PID, c.LABEL_P2SF, c.LABEL_P2NS1, c.LABEL_UNIQUE_ID])
pid_index_subs_ids = w.get_index_of_file_col(c.SUBJECTS_IDS_PATH,
c.LABEL_PID)
P2SF_index_depr_net_sup = w.get_index_of_file_col(path_old_net,
c.LABEL_P2SF)
P2NS1_index_depr_net_sup = w.get_index_of_file_col(path_old_net,
c.LABEL_P2NS1)
for i in range(len(unique_ids_from_depr_net_sup)):
unique_id = unique_ids_from_depr_net_sup[i]
try:
# try to get the row index of this unique id in the subjects ids file
subs_ids_row_index = unique_ids_from_subs_ids.index(unique_id)
# get that row from subjects ids and get its rds id column entry
pid = subs_ids_data[subs_ids_row_index][pid_index_subs_ids]
# get current row from depr_net_sup to fetch its P2SF and P2NS1 columns entries
P2SF = depr_net_sup_data[i][P2SF_index_depr_net_sup]
P2NS1 = depr_net_sup_data[i][P2NS1_index_depr_net_sup]
# append the new row to the net_sup_extract file
w.append_row_to_csv(path_net_sup_extract,
[pid, P2SF, P2NS1, unique_id])
except ValueError:
print(unique_id + " not found in subjects_ids.csv")
pass
def create_sub_ids_csv():
"""Input: None.
Output: Creates subjects_ids.csv file and populates it with project_ids, rds_ids, and unique_ids."""
path_sub_ids, path_nodes = c.SUBJECTS_IDS_PATH, c.OLD_NODES_PATH
nodes_data = w.get_csv_as_list(path_nodes)[1:] # data of nodes csv as list
nodes_pid_index = w.get_index_of_file_col(path_nodes, c.LABEL_PID)
nodes_rds_index = w.get_index_of_file_col(path_nodes, c.LABEL_RDS_ID)
nodes_unique_index = w.get_index_of_file_col(path_nodes, c.LABEL_UNIQUE_ID)
w.create_csv_add_column_labels(
path_sub_ids, [c.LABEL_PID, c.LABEL_RDS_ID, c.LABEL_UNIQUE_ID])
# make sure that the data is put into the correct column in subjects_ids
# put project_id, rds_id, and unique_id from the old nodes table into the new subjects_ids table
for row in nodes_data:
for i in range(len(row)):
if row[i] in c.NO_ENTRIES:
row[i] = ''
add_row = [
row[nodes_pid_index], row[nodes_rds_index], row[nodes_unique_index]
]
w.append_row_to_csv(path_sub_ids, add_row)