new commit

01_importMaterialsAndTechnique.py

@@ -0,0 +1,79 @@
+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 importMaterialsAndTechnique(api, engine):
+    print('Importing materials and technique...')
+
+    tableName = 'c__5280_material'
+    bundleId = 'b45978f2b073ff3c73b3c7220ebb3b89'
+
+    try:
+        processedRows = pd.read_csv(f'./logs/{tableName}.csv')
+    except FileNotFoundError:
+        processedRows = pd.DataFrame(columns=['id', 'uuid', 'uri'])
+
+    # Load materials table
+    sqlTable = pd.read_sql_table(tableName, con=engine)
+
+    # Create materials
+    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 material {sqlTable.loc[index, "id"]}')
+            continue
+        # Create Entity property dicts
+        materialValues = {}
+        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]
+            # If value is a list of comma-separated strings, split each item by ',' and flatten.
+            if isinstance(value, list):
+                new_value = []
+                for v in value:
+                    if isinstance(v, str) and ',' in v:
+                        new_value.extend([x.strip() for x in v.split(',') if x.strip()])
+                    else:
+                        new_value.append(v)
+                value = new_value
+
+            # Map columns to fields. We use assignments for reification.
+            match key:
+                case 'id':
+                    continue
+                case 'f__uuid':
+                    materialValues['fedfe553c2332bd4902c887813f29ed8'] = value # UUID
+                case 'f__5280_material':
+                    materialValues['f5f4251312f54c0d104ea87761b94bde'] = value # Material
+                case 'f__5300_technik':
+                    materialValues['f231e08850022f091ebd5055d8aad30f'] = value # Technique
+                case _:
+                    print(f'{key} is not a valid field, skipping.')
+
+        # Create Material
+        material = Entity(api=api, fields=materialValues, bundle_id=bundleId)
+        api.save(material)
+
+        print(f'Created material {index}: {material.uri} of {len(sqlTable)}')
+
+        # Write log
+        processedRows = processedRows._append({'id': row['id'], 'uuid': materialValues['fedfe553c2332bd4902c887813f29ed8'][0], 'uri': material.uri}, ignore_index=True)
+        processedRows.to_csv(f'./logs/{tableName}.csv', index=False)
+
+    print('finish')