django-cfg 1.1.81__py3-none-any.whl → 1.2.0__py3-none-any.whl

django_cfg/apps/knowbase/migrations/0002_archiveitem_archiveitemchunk_documentarchive_and_more.py

@@ -0,0 +1,434 @@
+# Generated by Django 5.2.6 on 2025-09-20 16:57
+
+import django.db.models.deletion
+import pgvector.django.vector
+import uuid
+from django.conf import settings
+from django.db import migrations, models
+
+
+class Migration(migrations.Migration):
+
+    dependencies = [
+        ('django_cfg_knowbase', '0001_initial'),
+        migrations.swappable_dependency(settings.AUTH_USER_MODEL),
+    ]
+
+    operations = [
+        migrations.CreateModel(
+            name='ArchiveItem',
+            fields=[
+                ('id', models.UUIDField(default=uuid.uuid4, editable=False, primary_key=True, serialize=False)),
+                ('created_at', models.DateTimeField(auto_now_add=True, db_index=True)),
+                ('updated_at', models.DateTimeField(auto_now=True)),
+                ('relative_path', models.CharField(help_text='Path within archive', max_length=1024)),
+                ('item_name', models.CharField(help_text='Item name', max_length=255)),
+                ('item_type', models.CharField(help_text='MIME type', max_length=100)),
+                ('content_type', models.CharField(choices=[('document', 'Document'), ('code', 'Code'), ('image', 'Image'), ('data', 'Data'), ('archive', 'Archive'), ('unknown', 'Unknown')], default='unknown', help_text='Content classification', max_length=20)),
+                ('file_size', models.PositiveIntegerField(default=0, help_text='Item size in bytes')),
+                ('content_hash', models.CharField(help_text='SHA-256 hash of item content', max_length=64)),
+                ('raw_content', models.TextField(blank=True, help_text='Extracted text content')),
+                ('is_processable', models.BooleanField(default=False, help_text='Whether item can be processed for chunks')),
+                ('language', models.CharField(blank=True, help_text='Programming language or document language', max_length=50)),
+                ('encoding', models.CharField(default='utf-8', help_text='Character encoding', max_length=50)),
+                ('chunks_count', models.PositiveIntegerField(default=0, help_text='Number of chunks created')),
+                ('total_tokens', models.PositiveIntegerField(default=0, help_text='Total tokens in all chunks')),
+                ('processing_cost', models.FloatField(default=0.0, help_text='Processing cost for this item')),
+                ('metadata', models.JSONField(blank=True, default=dict, help_text='Item-specific metadata', null=True)),
+            ],
+            options={
+                'verbose_name': 'Archive Item',
+                'verbose_name_plural': 'Archive Items',
+                'db_table': 'django_cfg_knowbase_archive_items',
+                'ordering': ['archive', 'relative_path'],
+            },
+        ),
+        migrations.CreateModel(
+            name='ArchiveItemChunk',
+            fields=[
+                ('id', models.UUIDField(default=uuid.uuid4, editable=False, primary_key=True, serialize=False)),
+                ('created_at', models.DateTimeField(auto_now_add=True, db_index=True)),
+                ('updated_at', models.DateTimeField(auto_now=True)),
+                ('content', models.TextField(help_text='Chunk text content')),
+                ('chunk_index', models.PositiveIntegerField(help_text='Sequential chunk number within item')),
+                ('chunk_type', models.CharField(choices=[('text', 'Text'), ('code', 'Code'), ('heading', 'Heading'), ('metadata', 'Metadata'), ('table', 'Table'), ('list', 'List')], default='text', help_text='Type of content in chunk', max_length=20)),
+                ('context_metadata', models.JSONField(default=dict, help_text='Rich context information for AI processing')),
+                ('embedding', pgvector.django.vector.VectorField(dimensions=1536, help_text='Vector embedding for semantic search', null=True)),
+                ('token_count', models.PositiveIntegerField(default=0, help_text='Number of tokens in chunk')),
+                ('character_count', models.PositiveIntegerField(default=0, help_text='Number of characters in chunk')),
+                ('embedding_model', models.CharField(default='text-embedding-ada-002', help_text='Model used for embedding generation', max_length=100)),
+                ('embedding_cost', models.FloatField(default=0.0, help_text='Cost in USD for embedding generation')),
+            ],
+            options={
+                'verbose_name': 'Archive Item Chunk',
+                'verbose_name_plural': 'Archive Item Chunks',
+                'db_table': 'django_cfg_knowbase_archive_item_chunks',
+                'ordering': ['item', 'chunk_index'],
+            },
+        ),
+        migrations.CreateModel(
+            name='DocumentArchive',
+            fields=[
+                ('id', models.UUIDField(default=uuid.uuid4, editable=False, primary_key=True, serialize=False)),
+                ('created_at', models.DateTimeField(auto_now_add=True, db_index=True)),
+                ('updated_at', models.DateTimeField(auto_now=True)),
+                ('title', models.CharField(help_text='Archive title', max_length=512)),
+                ('description', models.TextField(blank=True, help_text='Archive description')),
+                ('is_public', models.BooleanField(default=True, help_text='Whether this archive is publicly accessible')),
+                ('archive_file', models.FileField(help_text='Uploaded archive file', upload_to='archives/%Y/%m/%d/')),
+                ('original_filename', models.CharField(help_text='Original uploaded filename', max_length=255)),
+                ('file_size', models.PositiveIntegerField(default=0, help_text='Archive size in bytes')),
+                ('archive_type', models.CharField(choices=[('zip', 'ZIP'), ('tar', 'TAR'), ('tar.gz', 'TAR.GZ'), ('tar.bz2', 'TAR.BZ2')], help_text='Archive format', max_length=20)),
+                ('content_hash', models.CharField(db_index=True, help_text='SHA-256 hash for duplicate detection', max_length=64)),
+                ('processing_status', models.CharField(choices=[('pending', 'Pending'), ('processing', 'Processing'), ('completed', 'Completed'), ('failed', 'Failed'), ('cancelled', 'Cancelled')], db_index=True, default='pending', max_length=20)),
+                ('processed_at', models.DateTimeField(blank=True, help_text='When processing completed', null=True)),
+                ('processing_error', models.TextField(blank=True, default='', help_text='Error message if processing failed')),
+                ('processing_duration_ms', models.PositiveIntegerField(default=0, help_text='Processing time in milliseconds')),
+                ('total_items', models.PositiveIntegerField(default=0, help_text='Total items in archive')),
+                ('processed_items', models.PositiveIntegerField(default=0, help_text='Successfully processed items')),
+                ('total_chunks', models.PositiveIntegerField(default=0, help_text='Total chunks created')),
+                ('vectorized_chunks', models.PositiveIntegerField(default=0, help_text='Chunks with embeddings')),
+                ('total_tokens', models.PositiveIntegerField(default=0, help_text='Total tokens across all chunks')),
+                ('total_cost_usd', models.FloatField(default=0.0, help_text='Total processing cost in USD')),
+                ('metadata', models.JSONField(blank=True, default=dict, help_text='Additional archive metadata', null=True)),
+            ],
+            options={
+                'verbose_name': 'Document Archive',
+                'verbose_name_plural': 'Document Archives',
+                'db_table': 'django_cfg_knowbase_document_archives',
+                'ordering': ['-created_at'],
+            },
+        ),
+        migrations.CreateModel(
+            name='DocumentCategory',
+            fields=[
+                ('id', models.UUIDField(default=uuid.uuid4, editable=False, primary_key=True, serialize=False)),
+                ('created_at', models.DateTimeField(auto_now_add=True, db_index=True)),
+                ('updated_at', models.DateTimeField(auto_now=True)),
+                ('name', models.CharField(help_text='Category name', max_length=255, unique=True)),
+                ('description', models.TextField(blank=True, help_text='Category description')),
+                ('is_public', models.BooleanField(default=True, help_text='Whether documents in this category are publicly accessible')),
+            ],
+            options={
+                'verbose_name': 'Document Category',
+                'verbose_name_plural': 'Document Categories',
+                'db_table': 'django_cfg_knowbase_document_categories',
+                'ordering': ['name'],
+            },
+        ),
+        migrations.CreateModel(
+            name='ExternalData',
+            fields=[
+                ('id', models.UUIDField(default=uuid.uuid4, editable=False, primary_key=True, serialize=False)),
+                ('created_at', models.DateTimeField(auto_now_add=True, db_index=True)),
+                ('updated_at', models.DateTimeField(auto_now=True)),
+                ('title', models.CharField(help_text='Human-readable title for this external data source', max_length=512)),
+                ('description', models.TextField(blank=True, help_text='Description of what this external data contains')),
+                ('source_type', models.CharField(choices=[('model', 'Django Model'), ('api', 'API Endpoint'), ('database', 'Database Query'), ('file', 'File System'), ('custom', 'Custom Source')], default='model', help_text='Type of external data source', max_length=20)),
+                ('source_identifier', models.CharField(blank=True, help_text="Unique identifier for the data source (e.g., 'vehicles_data.Vehicle')", max_length=255)),
+                ('source_config', models.JSONField(blank=True, default=dict, help_text='Configuration for data extraction (fields, filters, etc.)')),
+                ('content', models.TextField(blank=True, help_text='Extracted text content for vectorization')),
+                ('content_hash', models.CharField(blank=True, db_index=True, help_text='SHA256 hash of content for change detection', max_length=64)),
+                ('metadata', models.JSONField(blank=True, default=dict, help_text='Additional metadata from the source')),
+                ('status', models.CharField(choices=[('pending', 'Pending'), ('processing', 'Processing'), ('completed', 'Completed'), ('failed', 'Failed'), ('outdated', 'Outdated')], default='pending', help_text='Current processing status', max_length=20)),
+                ('processing_error', models.TextField(blank=True, help_text='Error message if processing failed')),
+                ('chunk_size', models.PositiveIntegerField(default=1000, help_text='Size of text chunks for vectorization')),
+                ('overlap_size', models.PositiveIntegerField(default=200, help_text='Overlap between chunks')),
+                ('embedding_model', models.CharField(default='text-embedding-ada-002', help_text='Embedding model used for vectorization', max_length=100)),
+                ('similarity_threshold', models.FloatField(default=0.5, help_text='Similarity threshold for this external data (0.0-1.0). Lower = more results, higher = more precise')),
+                ('processed_at', models.DateTimeField(blank=True, help_text='When the data was last processed', null=True)),
+                ('source_updated_at', models.DateTimeField(blank=True, help_text='When the source data was last updated', null=True)),
+                ('total_chunks', models.PositiveIntegerField(default=0, help_text='Total number of chunks created')),
+                ('total_tokens', models.PositiveIntegerField(default=0, help_text='Total tokens processed')),
+                ('processing_cost', models.FloatField(default=0.0, help_text='Total cost for processing this data (USD)')),
+                ('tags', models.JSONField(blank=True, default=list, help_text='Tags for categorization and filtering')),
+                ('is_active', models.BooleanField(default=True, help_text='Whether this data source is active for search')),
+                ('is_public', models.BooleanField(default=False, help_text='Whether this data is publicly searchable')),
+            ],
+            options={
+                'verbose_name': 'External Data',
+                'verbose_name_plural': 'External Data',
+                'db_table': 'django_cfg_knowbase_external_data',
+                'ordering': ['-processed_at', '-created_at'],
+            },
+        ),
+        migrations.CreateModel(
+            name='ExternalDataChunk',
+            fields=[
+                ('id', models.UUIDField(default=uuid.uuid4, editable=False, primary_key=True, serialize=False)),
+                ('created_at', models.DateTimeField(auto_now_add=True, db_index=True)),
+                ('updated_at', models.DateTimeField(auto_now=True)),
+                ('content', models.TextField(blank=True, help_text='Text content of the chunk')),
+                ('chunk_index', models.PositiveIntegerField(default=0, help_text='Sequential index of this chunk within the external data')),
+                ('embedding', pgvector.django.vector.VectorField(blank=True, dimensions=1536, help_text='Vector embedding for semantic search', null=True)),
+                ('embedding_model', models.CharField(default='text-embedding-ada-002', help_text='Model used for embedding generation', max_length=100)),
+                ('token_count', models.PositiveIntegerField(default=0, help_text='Number of tokens in this chunk')),
+                ('character_count', models.PositiveIntegerField(default=0, help_text='Number of characters in this chunk')),
+                ('embedding_cost', models.FloatField(default=0.0, help_text='Cost for generating this embedding (USD)')),
+                ('chunk_metadata', models.JSONField(blank=True, default=dict, help_text='Additional metadata for this specific chunk')),
+            ],
+            options={
+                'verbose_name': 'External Data Chunk',
+                'verbose_name_plural': 'External Data Chunks',
+                'db_table': 'django_cfg_knowbase_external_data_chunk',
+                'ordering': ['external_data', 'chunk_index'],
+            },
+        ),
+        migrations.RenameIndex(
+            model_name='chatmessage',
+            new_name='django_cfg__session_564129_idx',
+            old_name='django_cfg_knowbase_ch_session_782e67_idx',
+        ),
+        migrations.RenameIndex(
+            model_name='chatmessage',
+            new_name='django_cfg__role_a6e8b7_idx',
+            old_name='django_cfg_knowbase_ch_role_e0d53e_idx',
+        ),
+        migrations.RenameIndex(
+            model_name='chatmessage',
+            new_name='django_cfg__created_875295_idx',
+            old_name='django_cfg_knowbase_ch_created_49bad9_idx',
+        ),
+        migrations.RenameIndex(
+            model_name='chatsession',
+            new_name='django_cfg__user_id_6bf22b_idx',
+            old_name='django_cfg_knowbase_ch_user_id_227a9f_idx',
+        ),
+        migrations.RenameIndex(
+            model_name='chatsession',
+            new_name='django_cfg__is_acti_12448e_idx',
+            old_name='django_cfg_knowbase_ch_is_acti_6fcdc8_idx',
+        ),
+        migrations.RenameIndex(
+            model_name='document',
+            new_name='django_cfg__user_id_ea1b86_idx',
+            old_name='django_cfg_knowbase_do_user_id_3f12df_idx',
+        ),
+        migrations.RenameIndex(
+            model_name='document',
+            new_name='django_cfg__content_ebac19_idx',
+            old_name='django_cfg_knowbase_do_content_0655c8_idx',
+        ),
+        migrations.RenameIndex(
+            model_name='document',
+            new_name='django_cfg__process_5f6075_idx',
+            old_name='django_cfg_knowbase_do_process_f2c2a8_idx',
+        ),
+        migrations.RenameIndex(
+            model_name='documentchunk',
+            new_name='django_cfg__user_id_ddd3b5_idx',
+            old_name='django_cfg_knowbase_do_user_id_a4e4de_idx',
+        ),
+        migrations.RenameIndex(
+            model_name='documentchunk',
+            new_name='django_cfg__documen_394c50_idx',
+            old_name='django_cfg_knowbase_do_documen_64aaaa_idx',
+        ),
+        migrations.AddField(
+            model_name='document',
+            name='is_public',
+            field=models.BooleanField(default=True, help_text='Whether this document is publicly accessible'),
+        ),
+        migrations.AlterField(
+            model_name='chatmessage',
+            name='cost_usd',
+            field=models.FloatField(default=0.0, help_text='Cost in USD for this message'),
+        ),
+        migrations.AlterField(
+            model_name='chatsession',
+            name='total_cost_usd',
+            field=models.FloatField(default=0.0, help_text='Total session cost for monitoring'),
+        ),
+        migrations.AlterField(
+            model_name='document',
+            name='metadata',
+            field=models.JSONField(blank=True, default=dict, help_text='Additional document metadata', null=True),
+        ),
+        migrations.AlterField(
+            model_name='document',
+            name='total_cost_usd',
+            field=models.FloatField(default=0.0, help_text='Total processing cost in USD'),
+        ),
+        migrations.AlterField(
+            model_name='documentchunk',
+            name='embedding_cost',
+            field=models.FloatField(default=0.0, help_text='Cost in USD for embedding generation'),
+        ),
+        migrations.AlterField(
+            model_name='documentchunk',
+            name='metadata',
+            field=models.JSONField(blank=True, default=dict, help_text='Chunk-specific metadata', null=True),
+        ),
+        migrations.AddField(
+            model_name='archiveitem',
+            name='user',
+            field=models.ForeignKey(help_text='Owner of this record', on_delete=django.db.models.deletion.CASCADE, to=settings.AUTH_USER_MODEL),
+        ),
+        migrations.AddField(
+            model_name='archiveitemchunk',
+            name='item',
+            field=models.ForeignKey(help_text='Parent item', on_delete=django.db.models.deletion.CASCADE, related_name='chunks', to='django_cfg_knowbase.archiveitem'),
+        ),
+        migrations.AddField(
+            model_name='archiveitemchunk',
+            name='user',
+            field=models.ForeignKey(help_text='Owner of this record', on_delete=django.db.models.deletion.CASCADE, to=settings.AUTH_USER_MODEL),
+        ),
+        migrations.AddField(
+            model_name='documentarchive',
+            name='user',
+            field=models.ForeignKey(help_text='Owner of this record', on_delete=django.db.models.deletion.CASCADE, to=settings.AUTH_USER_MODEL),
+        ),
+        migrations.AddField(
+            model_name='archiveitemchunk',
+            name='archive',
+            field=models.ForeignKey(help_text='Parent archive', on_delete=django.db.models.deletion.CASCADE, related_name='chunks', to='django_cfg_knowbase.documentarchive'),
+        ),
+        migrations.AddField(
+            model_name='archiveitem',
+            name='archive',
+            field=models.ForeignKey(help_text='Parent archive', on_delete=django.db.models.deletion.CASCADE, related_name='items', to='django_cfg_knowbase.documentarchive'),
+        ),
+        migrations.AddField(
+            model_name='documentarchive',
+            name='categories',
+            field=models.ManyToManyField(blank=True, help_text='Archive categories (supports multiple)', related_name='archives', to='django_cfg_knowbase.documentcategory'),
+        ),
+        migrations.AddField(
+            model_name='document',
+            name='categories',
+            field=models.ManyToManyField(blank=True, help_text='Document categories (supports multiple)', related_name='documents', to='django_cfg_knowbase.documentcategory'),
+        ),
+        migrations.AddIndex(
+            model_name='document',
+            index=models.Index(fields=['is_public', '-created_at'], name='django_cfg__is_publ_ebcc98_idx'),
+        ),
+        migrations.AddField(
+            model_name='externaldata',
+            name='category',
+            field=models.ForeignKey(blank=True, help_text='Category for organization', null=True, on_delete=django.db.models.deletion.SET_NULL, to='django_cfg_knowbase.documentcategory'),
+        ),
+        migrations.AddField(
+            model_name='externaldata',
+            name='user',
+            field=models.ForeignKey(help_text='Owner of this record', on_delete=django.db.models.deletion.CASCADE, to=settings.AUTH_USER_MODEL),
+        ),
+        migrations.AddField(
+            model_name='externaldatachunk',
+            name='external_data',
+            field=models.ForeignKey(help_text='External data this chunk belongs to', on_delete=django.db.models.deletion.CASCADE, related_name='chunks', to='django_cfg_knowbase.externaldata'),
+        ),
+        migrations.AddField(
+            model_name='externaldatachunk',
+            name='user',
+            field=models.ForeignKey(help_text='Owner of this chunk', on_delete=django.db.models.deletion.CASCADE, to=settings.AUTH_USER_MODEL),
+        ),
+        migrations.AddIndex(
+            model_name='archiveitemchunk',
+            index=models.Index(fields=['user'], name='django_cfg__user_id_a41505_idx'),
+        ),
+        migrations.AddIndex(
+            model_name='archiveitemchunk',
+            index=models.Index(fields=['archive'], name='django_cfg__archive_176a8a_idx'),
+        ),
+        migrations.AddIndex(
+            model_name='archiveitemchunk',
+            index=models.Index(fields=['item', 'chunk_index'], name='django_cfg__item_id_ec8c08_idx'),
+        ),
+        migrations.AddIndex(
+            model_name='archiveitemchunk',
+            index=models.Index(fields=['chunk_type'], name='django_cfg__chunk_t_2a4993_idx'),
+        ),
+        migrations.AddConstraint(
+            model_name='archiveitemchunk',
+            constraint=models.UniqueConstraint(fields=('item', 'chunk_index'), name='unique_item_chunk'),
+        ),
+        migrations.AddIndex(
+            model_name='archiveitem',
+            index=models.Index(fields=['user'], name='django_cfg__user_id_73874f_idx'),
+        ),
+        migrations.AddIndex(
+            model_name='archiveitem',
+            index=models.Index(fields=['archive', 'relative_path'], name='django_cfg__archive_42a267_idx'),
+        ),
+        migrations.AddIndex(
+            model_name='archiveitem',
+            index=models.Index(fields=['content_type', 'is_processable'], name='django_cfg__content_ea7d23_idx'),
+        ),
+        migrations.AddIndex(
+            model_name='archiveitem',
+            index=models.Index(fields=['language'], name='django_cfg__languag_f02bdb_idx'),
+        ),
+        migrations.AddConstraint(
+            model_name='archiveitem',
+            constraint=models.UniqueConstraint(fields=('archive', 'relative_path'), name='unique_archive_item_path'),
+        ),
+        migrations.AddIndex(
+            model_name='documentarchive',
+            index=models.Index(fields=['user', 'processing_status'], name='django_cfg__user_id_e4fa6e_idx'),
+        ),
+        migrations.AddIndex(
+            model_name='documentarchive',
+            index=models.Index(fields=['content_hash'], name='django_cfg__content_22fdac_idx'),
+        ),
+        migrations.AddIndex(
+            model_name='documentarchive',
+            index=models.Index(fields=['-processed_at'], name='django_cfg__process_c78bf9_idx'),
+        ),
+        migrations.AddIndex(
+            model_name='documentarchive',
+            index=models.Index(fields=['is_public', '-created_at'], name='django_cfg__is_publ_3c6f28_idx'),
+        ),
+        migrations.AddConstraint(
+            model_name='documentarchive',
+            constraint=models.UniqueConstraint(fields=('user', 'content_hash'), name='unique_user_archive'),
+        ),
+        migrations.AddIndex(
+            model_name='externaldata',
+            index=models.Index(fields=['user', 'source_type'], name='django_cfg__user_id_d556c1_idx'),
+        ),
+        migrations.AddIndex(
+            model_name='externaldata',
+            index=models.Index(fields=['status'], name='django_cfg__status_7b7d15_idx'),
+        ),
+        migrations.AddIndex(
+            model_name='externaldata',
+            index=models.Index(fields=['is_active'], name='django_cfg__is_acti_9d3ce2_idx'),
+        ),
+        migrations.AddIndex(
+            model_name='externaldata',
+            index=models.Index(fields=['processed_at'], name='django_cfg__process_fa1f31_idx'),
+        ),
+        migrations.AddIndex(
+            model_name='externaldata',
+            index=models.Index(fields=['source_identifier'], name='django_cfg__source__e5ed3f_idx'),
+        ),
+        migrations.AddIndex(
+            model_name='externaldata',
+            index=models.Index(fields=['content_hash'], name='django_cfg__content_f05aea_idx'),
+        ),
+        migrations.AlterUniqueTogether(
+            name='externaldata',
+            unique_together={('user', 'source_identifier')},
+        ),
+        migrations.AddIndex(
+            model_name='externaldatachunk',
+            index=models.Index(fields=['user', 'external_data'], name='django_cfg__user_id_7247ba_idx'),
+        ),
+        migrations.AddIndex(
+            model_name='externaldatachunk',
+            index=models.Index(fields=['embedding_model'], name='django_cfg__embeddi_5947c8_idx'),
+        ),
+        migrations.AddIndex(
+            model_name='externaldatachunk',
+            index=models.Index(fields=['token_count'], name='django_cfg__token_c_f8fa52_idx'),
+        ),
+        migrations.AddIndex(
+            model_name='externaldatachunk',
+            index=models.Index(fields=['chunk_index'], name='django_cfg__chunk_i_8ecf40_idx'),
+        ),
+        migrations.AlterUniqueTogether(
+            name='externaldatachunk',
+            unique_together={('external_data', 'chunk_index')},
+        ),
+    ]

django_cfg/apps/knowbase/migrations/__init__.py

@@ -0,0 +1,5 @@
+"""
+Knowledge Base Migrations
+
+Database migrations for the knowledge base application.
+"""

django_cfg/apps/knowbase/mixins/__init__.py

@@ -0,0 +1,15 @@
+"""
+Mixins for knowbase integration.
+"""
+
+from .external_data_mixin import ExternalDataMixin
+from .config import ExternalDataConfig
+from .creator import ExternalDataCreator
+from .service import ExternalDataService
+
+__all__ = [
+    'ExternalDataMixin',
+    'ExternalDataConfig',
+    'ExternalDataCreator',
+    'ExternalDataService',
+]

django_cfg/apps/knowbase/mixins/config.py

@@ -0,0 +1,108 @@
+"""
+Configuration classes for ExternalDataMixin.
+"""
+
+from typing import Optional, Dict, Any, List
+from pydantic import BaseModel, Field, field_validator, ConfigDict
+
+from ..models.external_data import ExternalDataType
+
+
+class ExternalDataConfig(BaseModel):
+    """Configuration for ExternalData creation with Pydantic2 validation."""
+    
+    model_config = ConfigDict(
+        validate_assignment=True,
+        extra="forbid",
+        str_strip_whitespace=True,
+        validate_default=True
+    )
+    
+    # Basic information
+    title: str = Field(
+        ...,
+        min_length=3,
+        max_length=512,
+        description="Human-readable title for the external data source"
+    )
+    
+    description: Optional[str] = Field(
+        default="",
+        max_length=2000,
+        description="Description of what this external data contains"
+    )
+    
+    # Source configuration
+    source_type: ExternalDataType = Field(
+        default=ExternalDataType.MODEL,
+        description="Type of external data source"
+    )
+    
+    source_identifier: str = Field(
+        ...,
+        min_length=1,
+        max_length=255,
+        description="Unique identifier for the data source"
+    )
+    
+    # Content
+    content: str = Field(
+        default="",
+        description="Extracted text content for vectorization"
+    )
+    
+    # Search and processing settings
+    similarity_threshold: float = Field(
+        default=0.5,
+        ge=0.0,
+        le=1.0,
+        description="Similarity threshold for search (0.0-1.0)"
+    )
+    
+    # Visibility settings
+    is_active: bool = Field(
+        default=True,
+        description="Whether this data source is active for search"
+    )
+    
+    is_public: bool = Field(
+        default=False,
+        description="Whether this data is publicly searchable"
+    )
+    
+    # Additional data
+    metadata: Dict[str, Any] = Field(
+        default_factory=dict,
+        description="Additional metadata from the source"
+    )
+    
+    source_config: Dict[str, Any] = Field(
+        default_factory=dict,
+        description="Configuration for data extraction"
+    )
+    
+    tags: List[str] = Field(
+        default_factory=list,
+        description="Tags for categorization and filtering"
+    )
+    
+    @field_validator('tags')
+    @classmethod
+    def validate_tags(cls, v):
+        """Validate tags are non-empty strings."""
+        if not isinstance(v, list):
+            raise ValueError('Tags must be a list of strings')
+        
+        for tag in v:
+            if not isinstance(tag, str) or not tag.strip():
+                raise ValueError('Each tag must be a non-empty string')
+        
+        return [tag.strip() for tag in v]
+    
+    @field_validator('metadata', 'source_config')
+    @classmethod
+    def validate_json_fields(cls, v):
+        """Validate JSON fields are serializable."""
+        if not isinstance(v, dict):
+            raise ValueError('Must be a dictionary')
+        return v

django_cfg/apps/knowbase/mixins/creator.py

@@ -0,0 +1,81 @@
+"""
+ExternalData creator for the mixin.
+"""
+
+import logging
+from typing import Optional, Dict, Any
+from django.contrib.auth import get_user_model
+from django.db import transaction
+from django.utils import timezone
+
+from .config import ExternalDataConfig
+from ..models.external_data import ExternalData, ExternalDataStatus
+
+logger = logging.getLogger(__name__)
+
+
+class ExternalDataCreator:
+    """
+    Creator class for ExternalData objects with validation.
+    """
+    
+    def __init__(self, user=None):
+        if user is None:
+            self.user = self._get_default_user()
+        else:
+            self.user = user
+    
+    def create_from_config(self, config: ExternalDataConfig) -> Dict[str, Any]:
+        """
+        Create an ExternalData object from a Pydantic configuration.
+        
+        Args:
+            config: An instance of ExternalDataConfig.
+            
+        Returns:
+            dict: Result with success status, external_data object, and message/error.
+        """
+        try:
+            with transaction.atomic():
+                external_data = ExternalData.objects.create(
+                    user=self.user,
+                    title=config.title,
+                    description=config.description,
+                    source_type=config.source_type,
+                    source_identifier=config.source_identifier,
+                    content=config.content,
+                    similarity_threshold=config.similarity_threshold,
+                    is_active=config.is_active,
+                    is_public=config.is_public,
+                    metadata=config.metadata,
+                    source_config=config.source_config,
+                    tags=config.tags,
+                    status=ExternalDataStatus.PENDING,  # Always set to pending on creation
+                    processed_at=None,
+                    processing_error="",
+                )
+                
+                logger.info(f"Created ExternalData: {external_data.title} (ID: {external_data.id})")
+                return {
+                    'success': True,
+                    'message': f"ExternalData '{external_data.title}' created successfully.",
+                    'external_data': external_data
+                }
+        except Exception as e:
+            logger.error(f"Failed to create ExternalData from config: {e}")
+            return {
+                'success': False,
+                'error': f"Failed to create ExternalData: {e}",
+                'external_data': None
+            }
+    
+    def _get_default_user(self):
+        """Get default user for ExternalData ownership."""
+        User = get_user_model()
+        
+        # Try to find a superuser
+        superuser = User.objects.filter(is_superuser=True).first()
+        if superuser:
+            return superuser
+        
+        raise ValueError("No user provided and no superuser found for ExternalData ownership")