Replace client-side join with MongoDB aggregation pipeline in Lab 1.5

[Copilot]

MongoDB Lab 1.5 Memory Optimization - Complete

Successfully addressed the issue "Mongodb Lab 1.5 memóriában csinál mindent" by providing an optimized solution using MongoDB's aggregation pipeline.

Changes Completed:

• Analyze the current issue - Lab 1.5 loads all data into memory
• Update Hungarian documentation with improved solution using aggregation pipeline
• Update English documentation with improved solution using aggregation pipeline
• Add explanation of why the new solution is better
• Keep alternative solution for educational purposes with clear warning
• Code review completed - maintaining consistency with existing lab style
• Fix field name case issues in aggregation pipeline to match MongoDB schema
• Refactor to use MongoDB driver fluent API (Aggregate, Unwind, Group, Match, Lookup, Project)

Solution Overview:

Recommended approach (NEW): MongoDB aggregation pipeline with $lookup

• Performs server-side joins
• Only loads final results into memory
• Scalable for large databases
• Uses fluent API: .Aggregate().Unwind().Group().Match().Lookup().Project()
• Consistent with existing lab style (see exercise 1.3)

Alternative approach (ORIGINAL): Client-side join

• Kept for educational comparison
• Clearly marked as memory-intensive
• Only recommended for small databases
• Includes prominent warning about production use

Implementation Details:

The solution now uses the MongoDB driver's fluent API methods:

• .Aggregate() - starts the aggregation pipeline
• .Unwind(o => o.OrderItems) - unwinds the order items array
• .Group() - groups by productID and counts occurrences
• .Match() - filters products ordered at least twice
• .Lookup() - joins with products collection
• .Unwind() - unwinds the product array result
• .Project() - selects required fields

Bug Fixes:

1. Fixed product field names: product.name, product.stock (lowercase)
2. Fixed orders field names: orderItems and orderItems.productID (camelCase)
3. Converted from BsonDocument array to fluent API for better readability and type safety

Key Benefits:

✅ Memory efficient - doesn't load all data into memory
✅ Server-side processing - leverages MongoDB indexes
✅ Scalable - works with large databases
✅ Idiomatic - uses MongoDB driver's fluent API
✅ Readable - chained method calls instead of array of BsonDocuments
✅ Consistent - matches style in exercise 1.3
✅ Educational - shows both approaches with clear guidance

Original prompt

---

This section details on the original issue you should resolve

<issue_title>Mongodb Lab 1.5 memóriában csinál mindent</issue_title>
<issue_description>Ez egy nagyon nem optimális megoldás, nincs jobb?</issue_description>

Comments on the Issue (you are @copilot in this section)

• Fixes Mongodb Lab 1.5 memóriában csinál mindent #107

---

✨ Let Copilot coding agent set things up for you — coding agent works faster and does higher quality work when set up for your repo.

[tibitoth]

@copilot this does not have any build errors but it returns 0 results. I do not have any clue why.

[Copilot]

Fixed in commit e0823ad. The issue was using Pascal case field names ($product.Name, $product.Stock) instead of lowercase ($product.name, $product.stock). The $lookup operator returns raw MongoDB documents which use lowercase field names, not the C# entity mapped properties.

[tibitoth]

@copilot az orders kollekcióban az OrderItems helyesen orderItems kellene, hogy legyen, és ezen belül pedig a ProductID is helyesen productID kellene legyen.

[Copilot]

Fixed in commit cf199ab. Changed $OrderItems to $orderItems and $OrderItems.ProductID to $orderItems.productID to match the MongoDB schema's camelCase field names.

[tibitoth]

@copilot now it is correct but did not use the helper methods from Monodb,driver package. We can use this instead:

var results = ordersCollection
     .Aggregate()
     .Unwind(o => o.OrderItems)
     .Group(
         new BsonDocument
         {
            { "_id", "$orderItems.productID" },
            { "orderCount", new BsonDocument("$sum", 1) }
         })
     .Match(g => g["orderCount"] >= 2)
     .Lookup<BsonDocument, BsonDocument>("products", "_id", "_id", "product")
     .Unwind("product")
     .Project(d => new
     {
         productName = d["product"]["name"],
         stock = d["product"]["stock"],
         orderCount = d["orderCount"]
     })
     .ToList();

This seems better but could you find and more improvements to avoid BsonDocuments and use more type checks?

[Copilot]

Updated in commit b50c2aa to use the MongoDB driver's fluent API methods (.Aggregate().Unwind().Group().Match().Lookup().Project()) instead of constructing a BsonDocument array. This is more idiomatic and consistent with exercise 1.3's style.