Firestore大文档存取疑问：活动销售统计应用数据存储咨询

Hey there! Let's walk through your scenario since you're building an event sales tracking app and have doubts about your Firestore implementation.

First off, you made the right call skipping Firebase Realtime Database here. For your use case—2,000 to 40,000 sales records per event—Realtime DB's structure would force you into heavy denormalization (the redundancy you mentioned) to keep queries performant, which gets messy fast. Firestore's document-collection model is far better suited for large, queryable datasets like sales records.

Common Firestore Missteps to Audit #

Since you suspect you're using Firestore incorrectly, here are the most frequent pitfalls with event/sales data to check:

• Over-nesting collections: If you’re nesting sales directly under each event document, that’s a problem. This structure limits your ability to query sales across multiple events or apply complex filters efficiently. Instead, use a top-level sales collection with an eventId field to link each sale to its event. This makes queries like "get all sales for Event X where purchase amount > $50" straightforward and fast.
• Missing composite indexes: Firestore requires composite indexes for queries that combine filters and sorting (e.g., filtering by eventId and sorting sales by timestamp). If you’re getting query errors or slow load times, head to your Firestore console’s Indexes tab—you probably need to create a missing index.
• Fetching full documents unnecessarily: If you only need specific fields (like amount or purchaseTime) for stats calculations, use field masks in your SDK calls (e.g., db.collection('sales').select('amount', 'purchaseTime')) to reduce bandwidth usage and speed up queries.
• Not using batch operations: When importing bulk sales data (like post-event), avoid individual write calls. Use Firestore batch writes or bulk import tools—this avoids hitting rate limits and is way more efficient.

Questions to Refine Your Setup (Since Core Needs Are Incomplete) #

Even without all your core requirements laid out, clarifying these points will help you optimize your Firestore structure:

• Do you need real-time sales updates during events, or just post-event statistics? Real-time needs might require different listeners than static reporting.
• Will you frequently need aggregated data (total sales, average ticket price, top-selling items)? If yes, precompute these aggregates in a separate eventStats document (updated via Cloud Functions as sales are added) instead of calculating them on the fly—this cuts down on read costs and speeds up stats displays.
• Are multiple users accessing the data (e.g., event managers viewing stats, staff entering sales)? If so, set up Firestore security rules to restrict access by eventId and user roles to keep data secure.
• Do you need to archive old event data? Firestore doesn’t auto-archive, so consider a Cloud Function to move old sales records to a separate archivedSales collection or Cloud Storage to keep your active data lean.

Firestore is a great fit for your use case—chances are the issue is in data modeling or query patterns rather than the database itself. Once you flesh out those core requirements, you can tweak the setup even more!

内容的提问来源于stack exchange，提问作者Vinícius Pacheco Vieira