95 lines
2.7 KiB
Markdown
95 lines
2.7 KiB
Markdown
|
# 🚨 CRITICAL API FIXES NEEDED FOR BRAINY 2.0
|
||
|
|
|
||
|
|
## 1. ❌ WRONG: MongoDB Operators (Legal Risk!)
|
||
|
|
We accidentally introduced MongoDB-style operators which we specifically avoided for legal reasons!
|
||
|
|
|
||
|
|
### MUST REPLACE:
|
||
|
|
```typescript
|
||
|
|
// ❌ WRONG (MongoDB style)
|
||
|
|
where: {
|
||
|
|
field: {$in: [values]},
|
||
|
|
field: {$gt: value},
|
||
|
|
field: {$regex: pattern}
|
||
|
|
}
|
||
|
|
|
||
|
|
// ✅ CORRECT (Brainy style)
|
||
|
|
where: {
|
||
|
|
field: {oneOf: [values]},
|
||
|
|
field: {greaterThan: value},
|
||
|
|
field: {matches: pattern}
|
||
|
|
}
|
||
|
|
```
|
||
|
|
|
||
|
|
### Complete Operator Mapping:
|
||
|
|
- `$eq` → `equals` or `is`
|
||
|
|
- `$ne` → `notEquals`
|
||
|
|
- `$gt` → `greaterThan`
|
||
|
|
- `$gte` → `greaterEqual`
|
||
|
|
- `$lt` → `lessThan`
|
||
|
|
- `$lte` → `lessEqual`
|
||
|
|
- `$in` → `oneOf`
|
||
|
|
- `$nin` → `notOneOf`
|
||
|
|
- `$regex` → `matches`
|
||
|
|
- `$contains` → `contains`
|
||
|
|
- (NEW) → `startsWith`
|
||
|
|
- (NEW) → `endsWith`
|
||
|
|
- (NEW) → `between`
|
||
|
|
|
||
|
|
## 2. 📊 Missing Neural API Features
|
||
|
|
The backup shows we had extensive neural capabilities:
|
||
|
|
|
||
|
|
```typescript
|
||
|
|
brain.neural.similar(a, b) // Semantic similarity
|
||
|
|
brain.neural.clusters() // Auto-clustering
|
||
|
|
brain.neural.hierarchy(id) // Semantic hierarchy
|
||
|
|
brain.neural.neighbors(id) // Neighbor graph
|
||
|
|
brain.neural.outliers() // Outlier detection
|
||
|
|
brain.neural.semanticPath(a, b) // Path finding
|
||
|
|
brain.neural.visualize() // Visualization data
|
||
|
|
```
|
||
|
|
|
||
|
|
## 3. 🔄 Missing Triple Intelligence Features
|
||
|
|
We need to ensure Triple Intelligence has all its features:
|
||
|
|
- Query optimization
|
||
|
|
- Progressive filtering
|
||
|
|
- Parallel execution
|
||
|
|
- Query learning/caching
|
||
|
|
- Explanations
|
||
|
|
|
||
|
|
## 4. 🧩 Missing Augmentation Features
|
||
|
|
- Synapses (Notion, Slack, Salesforce connectors)
|
||
|
|
- Conduits (Brainy-to-Brainy sync)
|
||
|
|
- Real-time bidirectional sync
|
||
|
|
|
||
|
|
## 5. 📥 Missing Import/Export Features
|
||
|
|
- Neural import with entity extraction
|
||
|
|
- CSV import with AI parsing
|
||
|
|
- JSON flattening and structure detection
|
||
|
|
- Batch neural processing
|
||
|
|
|
||
|
|
## 6. 🎯 API Simplification Issues
|
||
|
|
While simplifying, we may have lost:
|
||
|
|
- Verb scoring intelligence
|
||
|
|
- Clustering management
|
||
|
|
- Performance monitoring
|
||
|
|
- Health checks
|
||
|
|
- Statistics collection
|
||
|
|
|
||
|
|
## 7. 🔍 Search Method Confusion
|
||
|
|
Need to clarify:
|
||
|
|
- `search(query)` = simple convenience for `find({like: query})`
|
||
|
|
- `find(query)` = full Triple Intelligence
|
||
|
|
- Remove duplicate methods like `searchByNounTypes`, etc.
|
||
|
|
|
||
|
|
## IMMEDIATE ACTIONS:
|
||
|
|
1. Replace ALL MongoDB operators with Brainy operators
|
||
|
|
2. Restore neural API with all methods
|
||
|
|
3. Ensure Triple Intelligence is complete
|
||
|
|
4. Verify augmentation system works
|
||
|
|
5. Test import/export capabilities
|
||
|
|
6. Document the clean API properly
|
||
|
|
|
||
|
|
## BACKUP LOCATIONS:
|
||
|
|
- `/home/dpsifr/Projects/BACKUP/brainy (Copy)` - Full backup
|
||
|
|
- `/home/dpsifr/Projects/BACKUP/brainy-clean` - Clean version
|
||
|
|
- `/home/dpsifr/Projects/brainy (Copy)` - Another backup
|