JS query mode
NoSQL projects now show a Query editor in place of the SQL editor. Write queries exactly like you would with the SDK:
collection('orders')
.find({ status: 'paid', total: { $gte: 100 } })
.sort({ _createdAt: -1 })
.limit(50)
- Works against both Data Engine entities and NoSQL collections — the editor resolves the target automatically
- Full filter operators:
$eq $ne $gt $gte $lt $lte $in $nin $contains $containsAny $exists $regex $iregex $like $ilikeplus$and / $or / $not .sort(),.limit(),.skip(),.select(),.count()chain methods- Queries are parsed, never evaluated — a closed grammar with precise line/column errors, no code execution
- Multi-tab editor with rename, saved queries, table/JSON result views, Markdown/JSON copy, and Excel export
Aggregation mode
On data engines that support pipelines, the editor gains an Aggregation mode toggle:
[
{ "$match": { "status": "active" } },
{ "$group": { "_id": "customer.city", "orders": { "$count": null }, "revenue": { "$sum": "total" } } },
{ "$sort": { "revenue": -1 } },
{ "$limit": 10 }
]
- Stages:
$match $project $unwind $group $sort $limit $skip - Accumulators:
$count $sum $avg $min $max - Pipelines are validated server-side; dangerous stages (
$lookup,$out,$merge,$function,$where) are blocked - Every pipeline is automatically scoped to your project — tenant isolation is enforced server-side
The mode toggle appears automatically when the project's data engine reports pipeline support — no configuration needed.
MongoDB engine support
The Data Engine gains a MongoDB provider (DATA_ENGINE_PROVIDER=mongodb):
- Native aggregation pipeline execution — the first engine with full Aggregation mode support
- Same document envelope and tenant isolation model as the other engines
- The
mongodbdriver is an optional dependency — deployments not using it need nothing new
Saved queries
Saved queries now remember their dialect — reopening an aggregation restores Aggregation mode with the target entity preselected.