rnsrk/goodByeHida
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

new commit

Browse source
This commit is contained in:
rnsrk 2025-09-09 10:16:31 +02:00
parent da296f8a64
commit e46a9fd4ec
69 changed files with 4199 additions and 4805 deletions
Download patch file Download diff file Expand all files Collapse all files

138
15_importGoldsmithRelation.py
View file

@ -5,89 +5,75 @@ import os # For environment variable loading
from dotenv import load_dotenv # For environment variable loading
import pandas as pd # For dataframe handling
# Initialize the database
print('Initializing the database...')
engine, metadata, Session = initDb(True, './schemas/')
if engine == False:
print('Database initialization failed.')
exit()
def importGoldsmithRelation(api, engine):
print('Importing goldsmith relation...')
# Load the environment variables
load_dotenv()
test = False
# Initialize the WissKI API
print('Initializing the WissKI API...')
api_url = os.getenv('API_URL')
auth = (os.getenv('API_USERNAME'), os.getenv('API_PASSWORD'))
headers = {"Cache-Control": "no-cache"}
api = Api(api_url, auth, headers)
api.pathbuilder = api.get_pathbuilder('default')
tableName = "c__3007_bezieh__zu_gs"
bundleId = 'bef43e8a958e6a9bee04534b3841f6a0'
test = False
try:
processedRows = pd.read_csv(f'./logs/{tableName}.csv')
except FileNotFoundError:
processedRows = pd.DataFrame(columns=[ 'id', 'uuid', 'uri'])
tableName = "c__3007_bezieh__zu_gs"
bundleId = 'bef43e8a958e6a9bee04534b3841f6a0'
# Load sources table
sqlTable = pd.read_sql_table(tableName, con=engine)
try:
processedRows = pd.read_csv(f'./logs/processed-{tableName}.csv')
except FileNotFoundError:
processedRows = pd.DataFrame(columns=[ 'docId', 'uuid', 'uri'])
# Load sources table
sqlTable = pd.read_sql_table(tableName, con=engine)
entityValues = {}
# Create entities
for index, row in sqlTable.iterrows():
# For every row in table...
if index < len(processedRows) and sqlTable.iloc[index, 0] == processedRows.iloc[index, 0]:
# skip if already processed
print(f'Skipping already processed entity {sqlTable.iloc[index, 0]}')
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
# 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
# Properties of an entity have to be an array, so...
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':
docId = value[0]
case 'f__uuid':
entityValues['f588ff2629e3758ae18ec28c02270d27'] = value # UUID
fUuid = value[0]
case 'f__3011_verw__art':
entityValues['f2de276528d6b020306b8c7784008e5c'] = value # Actor relation type
case 'f__3010_name_gs':
entityValues['fc16719402aff4a1afec3387bf2bbc34'] = value # Goldsmith
case 'f__30bm_bem_beziehung':
entityValues['f7de6b267146070fa38ea5dc45150fa4'] = value # Note
case 'f__3007_bezieh__zu_gs':
entityValues['f8a46491ebad0ba670384a049402d697'] = value # Relation
case _:
print(f'{key} is not a valid field, skipping.')
# 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':
docId = value[0]
case 'f__uuid':
entityValues['f588ff2629e3758ae18ec28c02270d27'] = value # UUID
fUuid = value[0]
case 'f__3011_verw__art':
entityValues['f2de276528d6b020306b8c7784008e5c'] = value # Actor relation type
case 'f__3010_name_gs':
entityValues['fc16719402aff4a1afec3387bf2bbc34'] = value # Goldsmith
case 'f__30bm_bem_beziehung':
entityValues['f7de6b267146070fa38ea5dc45150fa4'] = value # Note
case 'f__3007_bezieh__zu_gs':
entityValues['f8a46491ebad0ba670384a049402d697'] = value # Relation
case _:
print(f'{key} is not a valid field, skipping.')
# Create Material
entity = Entity(api=api, fields=entityValues, bundle_id=bundleId)
api.save(entity)
# Create Material
entity = Entity(api=api, fields=entityValues, bundle_id=bundleId)
api.save(entity)
print(f'Created entity {index}: {entity.uri} of {len(sqlTable)}')
print(f'Created goldsmith relation {index}: {entity.uri} of {len(sqlTable)}')
# Write log
processedRows = processedRows._append({'docId': docId, 'uuid': fUuid, 'uri': entity.uri}, ignore_index=True)
processedRows.to_csv(f'./logs/processed-{tableName}.csv', index=False)
if test:
exit()
# 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)
if test:
exit()
print('finish')
print('finished importing goldsmith relation')