Filter on the Server, Not in Your App ($eq, $gt, $in)

Why this matters

MongoDB's query language is built around composable operator objects. Understanding the operator taxonomy — comparison ($eq, $ne, $gt, $lt, $in, $nin), logical ($and, $or, $not), element ($exists, $type) — lets you express complex filters without pulling data into application code and filtering there, which is a common performance anti-pattern. Every filter that runs server-side saves a network round-trip and lets the query planner use an index. The $regex operator deserves special attention because unanchored regex patterns always do a full collection scan regardless of indexes.

Demo

Query a products collection using comparison, logical, and regex operators.

Try it yourself

1. Query for products where price is NOT in [9.99, 49.99] using $nin. Verify the count matches documents outside those exact prices.
2. Add a field inStock: false to the Doohickey. Query { inStock: { $ne: true } } — confirm it returns both documents missing the field AND the one explicitly set to false. Explain why.
3. Create an index on name. Run { name: /widget/i } (unanchored) and { name: /^Widget/ } (anchored) through explain('executionStats'). Note which uses the index and which does a COLLSCAN.

References

• Query and Projection Operators — MongoDB docs
• Regex and indexes in MongoDB