開発者は、新しい機能の実装や既存の機能の強化を担当すると、多くの場合、最適なアプローチを決定する必要に迫られます。 Copilot Chat には、問題を分析し、実装オプションを生成する支援機能があるため、複数のソリューションを確認できます。
シナリオ例
あなたには、ユーザー ダッシュボードで最近閲覧された項目を表示するという新機能を作成するタスクを割り当てられました。 この機能を実装する際に発生する issue の例を次に示します。
[Feature Request] Add "Recent Items" section to user dashboard #753
## Issue Description
Users have requested a "Recent Items" section on their dashboard to quickly access the most recently viewed documents, projects, or files. This section should display the ten most recent items, sorted by the last access time.
## Requirements
* Display up to ten items that the user has accessed recently.
* Items must be sorted by the time they were last accessed.
* The solution should minimize database load.
* Data should reflect updates in near real-time.
* The solution should prioritize performance, especially for large datasets of items.
## Acceptance Criteria
* Dashboard loads the "Recent Items" section within 200ms under normal load.
* The solution must not degrade performance for users with thousands of items.
プロンプトの例
あなたは、この実装に対するさまざまなアプローチを評価して、ニーズや制約に適合するかどうかを確認したいと考えています。 そこで、Copilot Chat にさまざまなアプローチを提案してもらいます。
Issue に移動し、次のように入力します。
Given the criteria in Issue #753, list options for implementation and tradeoffs for each option.
Tip
Copilot を使って実装オプションを特定する場合は、この例のように、アイデアが思い浮かぶように一般的なプロンプトから始めることをお勧めします。 その後、より具体的な質問を追加して、提案されたオプションをより深く掘り下げることができます。
応答の例
Note
次の応答は例です。 Copilot Chat の応答は非決定的であるため、ここで示したものとは異なる応答が返される可能性があります。
Copilot により、このダッシュボードを実装する複数のオプションと、各オプションのトレードオフが示されます。 次に、推奨事項が作成されます。
次に例を示します。
# Implementation Options and Tradeoffs for "Recent Items" Section
## 1. Client-Side Storage
- Description: Store recent items data in the user's browser using LocalStorage or IndexedDB.
- Tradeoffs:
- Pros:
- Reduces server load by handling data storage and retrieval on the client side.
- Immediate updates and access without server communication.
- Cons:
- Limited by browser storage capacity.
- Data lost if the user clears browser storage or switches devices.
- Security concerns with storing sensitive data on the client side.
## 2. Server-Side Storage with Database Indexing
- Description: Store recent items in the database and use indexing to quickly retrieve the most recent items.
- Tradeoffs:
- Pros:
- Centralized data storage, accessible from any device.
- Scalable and can handle large datasets efficiently with proper indexing.
- Cons:
- Increased database load, especially for frequent access updates.
- Potential latency in updating the recent items list.
## 3. In-Memory Caching (e.g., Redis)
- Description: Use an in-memory data store like Redis to cache recent items for fast retrieval.
- Tradeoffs:
- Pros:
- Extremely fast data access and updates.
- Reduces database load by offloading frequent read operations.
- Cons:
- Requires additional infrastructure and maintenance.
- Data may be lost if the cache is cleared or server restarts.
## 4. Hybrid Approach
- Description: Combine server-side storage with in-memory caching to balance performance and persistence.
- Tradeoffs:
- Pros:
- Fast access through caching, with persistent storage in the database.
- Can handle large datasets and frequent updates efficiently.
- Cons:
- Increased complexity in implementation and maintenance.
- Potential consistency issues between cache and database.
## 5. Event-Driven Updates
- Description: Use event-driven architecture to update recent items in real-time as users access documents.
- Tradeoffs:
- Pros:
- Near real-time updates with minimal delay.
- Efficient handling of updates through events rather than periodic polling.
- Cons:
- Requires setting up event handling infrastructure (e.g., message queues).
- Increased complexity in managing events and ensuring consistency.
## Recommendation
The hybrid approach (option 4) is a balanced solution, offering fast access through caching and persistence through database storage. It meets performance requirements and scales well with a large dataset, although it involves a more complex implementation.