goodByeHida/04_importSourceReferenceAssignment.py
2025-09-09 10:16:31 +02:00

73 lines
3 KiB
Python

import uuid # For UUID creation
from initDb import initDb # For database initialization
from wisski.api import Api, Pathbuilder, Entity # For WissKI API
import os # For environment variable loading
from dotenv import load_dotenv # For environment variable loading
import pandas as pd # For dataframe handling
def importSourceReferenceAssignment(api, engine):
print('Importing source reference assignments...')
tableName = "c__8130_que_kurzt_"
bundleId = 'b3c4232e84c2f39795bd602f152ed6f0' # Source reference assignment
try:
processedRows = pd.read_csv(f'./logs/{tableName}.csv')
except FileNotFoundError:
processedRows = pd.DataFrame(columns=[ 'id', 'uuid', 'uri'])
# Load sources table
sqlTable = pd.read_sql_table(tableName, con=engine)
# Create entities
for index, row in sqlTable.iterrows():
# For every row in table...
if index < len(processedRows) and sqlTable.loc[index, 'id'] == processedRows.loc[index, 'id']:
# skip if already processed
print(f'Skipping already processed entity {sqlTable.loc[index, "id"]}')
continue
# Create Entity property dicts
entityValues = {}
for key, value in row.items():
# For every column in row...
if (value is None) or (value == ''):
# skip if cell has no value
continue
# Properties of an entity have to be an array, so...
value = str(value).replace('&###{{new_line}}###'.format(), '&')
value = str(value).replace('###{{new_line}}###', '&')
value = str(value).replace(' & ', '&')
if '&' in str(value):
# ...Explode "&"-separated values to array items
value = [x.strip() for x in str(value).split('&')]
else:
# ...Or parse to array
value = [value]
# Map columns to fields. We use assignments for reification.
docId = ''
match key:
case 'id':
continue
case 'f__uuid':
entityValues['fbe74fcb0ab0ce5a0181467b9b07e12e'] = value # UUID
fUuid = value[0]
case 'f__8134_stelle':
entityValues['f769795b4fd628d01692dd4516322db4'] = value # Source reference
case 'f__8130_que_kurzt_':
entityValues['f3e841bf3b4e91716d1ff5b83bf293d9'] = value # Source short title
case _:
print(f'{key} is not a valid field, skipping.')
# Create Material
entity = Entity(api=api, fields=entityValues, bundle_id=bundleId)
api.save(entity)
print(f'Created source reference assignment {index}: {entity.uri} of {len(tableName)}')
# Write log
processedRows = processedRows._append({'id': row['id'], 'uuid': fUuid, 'uri': entity.uri}, ignore_index=True)
processedRows.to_csv(f'./logs/{tableName}.csv', index=False)
print('finished importing source reference assignments')