// DataIngestionEngine – Autonomous data collection and enrichment
import { projectMemory } from '../project/ProjectMemory.js';
import { eventBus } from '../../mcp/EventBus.js';
import { dataSourceConfig } from './DataSourceConfigManager.js';

export interface DataSourceAdapter {
    name: string;
    type: 'local_files' | 'outlook_mail' | 'browser_history' | 'google_drive' | 'other';

    /** Fetch raw data from the source */
    fetch(): Promise<any[]>;

    /** Transform raw data into normalized entities */
    transform(raw: any[]): Promise<IngestedEntity[]>;

    /** Health check */
    isAvailable(): Promise<boolean>;
}

export interface IngestedEntity {
    id: string;
    type: string;
    source: string;
    title?: string;
    content?: string;
    metadata: Record<string, any>;
    timestamp: Date;
}

export class DataIngestionEngine {
    private adapters: Map<string, DataSourceAdapter> = new Map();
    private isRunning: boolean = false;
    private ingestedCount: number = 0;

    /** Register a data source adapter */
    async registerAdapter(adapter: DataSourceAdapter, description: string, requiresApproval: boolean = false): Promise<void> {
        this.adapters.set(adapter.name, adapter);
        const canUse = await dataSourceConfig.registerSource(adapter.name, description, requiresApproval);

        console.log(`📥 Registered data adapter: ${adapter.name} (${adapter.type}) - ${canUse ? 'Ready' : 'Awaiting approval'}`);
    }

    /** Get a registered adapter */
    getAdapter(name: string): DataSourceAdapter | undefined {
        return this.adapters.get(name);
    }


    /** Start ingestion from all registered adapters */
    async ingestAll(): Promise<void> {
        if (this.isRunning) {
            console.warn('⚠️ Ingestion already running');
            return;
        }

        // Load config
        await dataSourceConfig.load();

        this.isRunning = true;
        this.ingestedCount = 0;

        console.log(`🚀 Starting data ingestion from ${this.adapters.size} sources...`);

        projectMemory.logLifecycleEvent({
            eventType: 'other',
            status: 'in_progress',
            details: {
                type: 'data_ingestion_started',
                sources: Array.from(this.adapters.keys())
            }
        });

        const results: any[] = [];

        for (const [name, adapter] of this.adapters) {
            try {
                // Check if source is enabled
                if (!dataSourceConfig.isEnabled(name)) {
                    console.log(`⏭️ ${name} is disabled, skipping`);
                    results.push({ source: name, status: 'skipped', reason: 'disabled' });
                    continue;
                }

                console.log(`📂 Ingesting from: ${name}...`);

                // Check availability
                const available = await adapter.isAvailable();
                if (!available) {
                    console.warn(`⚠️ ${name} not available, skipping`);
                    results.push({ source: name, status: 'skipped', reason: 'not_available' });
                    continue;
                }

                // Fetch raw data
                const rawData = await adapter.fetch();
                console.log(`  → Fetched ${rawData.length} items from ${name}`);

                // Transform to normalized entities
                const entities = await adapter.transform(rawData);
                console.log(`  → Transformed ${entities.length} entities`);

                // Store entities (for now, just log - later we'll save to memory/database)
                this.ingestedCount += entities.length;

                // Auto-add to Vidensarkiv (Knowledge Archive) for continuous learning
                try {
                    const { getVectorStore } = await import('../../platform/vector/index.js');
                    const vectorStore = await getVectorStore();

                    // Batch add entities to vidensarkiv
                    const vectorRecords = entities.map(entity => ({
                        id: entity.id,
                        content: entity.content || entity.title || JSON.stringify(entity.metadata),
                        embedding: [], // Will be generated automatically
                        metadata: {
                            ...entity.metadata,
                            datasetType: 'new',
                            source: name,
                            type: entity.type,
                            ingestedAt: new Date().toISOString()
                        },
                        namespace: `org:default:user:system` // TODO: Get from context
                    }));

                    if (vectorRecords.length > 0) {
                        await vectorStore.batchUpsert({
                            records: vectorRecords,
                            namespace: `org:default:user:system`
                        });
                        console.log(`  → Added ${vectorRecords.length} entities to vidensarkiv`);
                    }
                } catch (err) {
                    console.warn(`⚠️ Failed to add to vidensarkiv:`, err);
                    // Non-critical, continue ingestion
                }

                results.push({
                    source: name,
                    status: 'success',
                    items: entities.length
                });

                // Mark as used
                await dataSourceConfig.markUsed(name);

                // Emit event for real-time updates
                // Emit event for real-time updates
                eventBus.emit('data:ingested', {
                    source: name,
                    count: entities.length,
                    entities: entities // Send ALL entities so IngestionPipeline can process them
                });

            } catch (error: any) {
                console.error(`❌ Failed to ingest from ${name}:`, error.message);
                results.push({
                    source: name,
                    status: 'error',
                    error: error.message
                });
            }
        }

        this.isRunning = false;

        // Log completion
        projectMemory.logLifecycleEvent({
            eventType: 'other',
            status: 'success',
            details: {
                type: 'data_ingestion_completed',
                totalIngested: this.ingestedCount,
                results
            }
        });

        console.log(`✅ Data ingestion complete! Total entities: ${this.ingestedCount}`);
    }

    /** Ingest from a specific source */
    async ingestFrom(sourceName: string): Promise<number> {
        const adapter = this.adapters.get(sourceName);

        if (!adapter) {
            throw new Error(`Unknown data source: ${sourceName}`);
        }

        // Check if enabled
        if (!dataSourceConfig.isEnabled(sourceName)) {
            throw new Error(`Data source ${sourceName} is not enabled`);
        }

        const available = await adapter.isAvailable();
        if (!available) {
            throw new Error(`Source ${sourceName} is not available`);
        }

        const rawData = await adapter.fetch();
        const entities = await adapter.transform(rawData);

        await dataSourceConfig.markUsed(sourceName);

        eventBus.emit('data:ingested', {
            source: sourceName,
            count: entities.length,
            entities: entities.slice(0, 5)
        });

        return entities.length;
    }

    /** Get ingestion status */
    getStatus() {
        return {
            running: this.isRunning,
            totalIngested: this.ingestedCount,
            adapters: Array.from(this.adapters.keys()),
            enabled: dataSourceConfig.getEnabledSources(),
            pendingApprovals: dataSourceConfig.getPendingApprovals()
        };
    }
}

// Singleton instance
export const dataIngestionEngine = new DataIngestionEngine();