afv-library/skills/querying-soql/assets/optimization-patterns.soql

/**
 * SOQL OPTIMIZATION PATTERNS
 *
 * These patterns demonstrate how to write efficient, selective queries
 * that perform well within Salesforce governor limits.
 */

// ═══════════════════════════════════════════════════════════════════════════
// SELECTIVE vs NON-SELECTIVE QUERIES
// ═══════════════════════════════════════════════════════════════════════════

-- ❌ NON-SELECTIVE: Status alone is not indexed
SELECT Id, Name FROM Lead
WHERE Status = 'Open'
-- Scans ALL Lead records

-- ✅ SELECTIVE: Add indexed field filter
SELECT Id, Name FROM Lead
WHERE Status = 'Open'
AND CreatedDate = LAST_N_DAYS:30
-- CreatedDate is indexed, reduces scan

-- ❌ NON-SELECTIVE: Leading wildcard prevents index use
SELECT Id, Name FROM Account
WHERE Name LIKE '%corporation'

-- ✅ SELECTIVE: Trailing wildcard uses index
SELECT Id, Name FROM Account
WHERE Name LIKE 'Acme%'

-- ❌ NON-SELECTIVE: OR can prevent index optimization
SELECT Id FROM Contact
WHERE Email = 'test@test.com'
OR Phone = '555-1234'

-- ✅ SELECTIVE: Use UNION-style in Apex or multiple queries
-- Query 1: SELECT Id FROM Contact WHERE Email = 'test@test.com'
-- Query 2: SELECT Id FROM Contact WHERE Phone = '555-1234'
-- Combine results in Apex

// ═══════════════════════════════════════════════════════════════════════════
// INDEXED FIELDS
// ═══════════════════════════════════════════════════════════════════════════

-- Always indexed fields (use these in WHERE)
-- • Id (primary key)
-- • Name (standard Name field)
-- • OwnerId
-- • CreatedDate
-- • LastModifiedDate
-- • RecordTypeId
-- • External ID fields (custom)
-- • Master-Detail relationship fields
-- • Lookup fields (selective when <100k parent records)

-- ✅ GOOD: Uses indexed OwnerId
SELECT Id, Name FROM Account
WHERE OwnerId = '005xx000000XXXX'

-- ✅ GOOD: Uses indexed CreatedDate
SELECT Id, Name FROM Contact
WHERE CreatedDate >= 2024-01-01T00:00:00Z

-- ✅ GOOD: Uses indexed External ID
SELECT Id, Name FROM Account
WHERE External_Id__c = 'EXT-12345'

// ═══════════════════════════════════════════════════════════════════════════
// FIELD SELECTION OPTIMIZATION
// ═══════════════════════════════════════════════════════════════════════════

-- ❌ BAD: Selecting unnecessary fields
SELECT Id, Name, Description, BillingStreet, BillingCity, BillingState,
       BillingPostalCode, BillingCountry, ShippingStreet, ShippingCity,
       ShippingState, ShippingPostalCode, ShippingCountry, Phone, Fax,
       AccountNumber, Website, Industry, AnnualRevenue, NumberOfEmployees,
       Ownership, TickerSymbol, Description, Rating, Site, AccountSource
FROM Account

-- ✅ GOOD: Select only needed fields
SELECT Id, Name, Industry, AnnualRevenue
FROM Account

-- For display only: Use specific fields
SELECT Id, Name FROM Account

-- For processing: Query fields you'll update
SELECT Id, Status__c, Last_Processed__c FROM Account

// ═══════════════════════════════════════════════════════════════════════════
// LIMIT USAGE
// ═══════════════════════════════════════════════════════════════════════════

-- ❌ BAD: No limit (could return 50,000 rows)
SELECT Id, Name FROM Account

-- ✅ GOOD: Appropriate limit
SELECT Id, Name FROM Account LIMIT 200

-- For pagination
SELECT Id, Name FROM Account ORDER BY Name LIMIT 100 OFFSET 0

-- For single record lookup
SELECT Id, Name FROM Account WHERE Name = 'Acme Corp' LIMIT 1

-- For existence check
SELECT Id FROM Account WHERE Name = 'Acme Corp' LIMIT 1
-- In Apex: if (!results.isEmpty()) { /* exists */ }

// ═══════════════════════════════════════════════════════════════════════════
// RELATIONSHIP QUERY OPTIMIZATION
// ═══════════════════════════════════════════════════════════════════════════

-- ❌ BAD: Deep nesting (5 levels is max, avoid if possible)
SELECT Id, Account.Owner.Manager.Department.Name FROM Contact

-- ✅ GOOD: Flatten when possible
SELECT Id, Account.Name, Account.OwnerId FROM Contact
-- Then query Owner separately if needed

-- ❌ BAD: Too many child subqueries
SELECT Id, Name,
       (SELECT Id FROM Contacts),
       (SELECT Id FROM Opportunities),
       (SELECT Id FROM Cases),
       (SELECT Id FROM Tasks),
       (SELECT Id FROM Events),
       (SELECT Id FROM Notes)
FROM Account

-- ✅ GOOD: Query what you need
SELECT Id, Name,
       (SELECT Id, Name FROM Contacts LIMIT 5)
FROM Account

-- ❌ BAD: Unfiltered subquery
SELECT Id, (SELECT Id FROM Opportunities) FROM Account

-- ✅ GOOD: Filtered subquery
SELECT Id, (SELECT Id FROM Opportunities WHERE StageName != 'Closed Lost' LIMIT 10)
FROM Account

// ═══════════════════════════════════════════════════════════════════════════
// BULK QUERY PATTERNS
// ═══════════════════════════════════════════════════════════════════════════

-- ❌ BAD: Query in loop
/*
for (Contact c : contacts) {
    Account a = [SELECT Name FROM Account WHERE Id = :c.AccountId];
}
*/

-- ✅ GOOD: Bulk query with Map
/*
Set<Id> accountIds = new Set<Id>();
for (Contact c : contacts) {
    accountIds.add(c.AccountId);
}

Map<Id, Account> accountMap = new Map<Id, Account>(
    [SELECT Id, Name FROM Account WHERE Id IN :accountIds]
);

for (Contact c : contacts) {
    Account a = accountMap.get(c.AccountId);
}
*/

-- Use IN clause for bulk operations
SELECT Id, Name FROM Account WHERE Id IN :accountIdSet

-- Bind variable with Set (most efficient)
SELECT Id, Name FROM Contact WHERE AccountId IN :accountIds

// ═══════════════════════════════════════════════════════════════════════════
// FOR UPDATE (Record Locking)
// ═══════════════════════════════════════════════════════════════════════════

-- Lock records for update (prevents concurrent modification)
SELECT Id, Status__c FROM Account WHERE Id = :accountId FOR UPDATE

-- Note: Only use FOR UPDATE when you need to prevent race conditions
-- Adds overhead and can cause lock contention

// ═══════════════════════════════════════════════════════════════════════════
// SECURITY ENFORCEMENT
// ═══════════════════════════════════════════════════════════════════════════

-- ✅ Recommended: WITH SECURITY_ENFORCED
SELECT Id, Name, Phone FROM Account
WITH SECURITY_ENFORCED
-- Throws exception if user lacks field access

-- Alternative: USER_MODE (API 54.0+)
SELECT Id, Name FROM Account
WITH USER_MODE
-- Respects sharing rules and FLS

-- For admin operations: SYSTEM_MODE
SELECT Id, Name FROM Account
WITH SYSTEM_MODE
-- Bypasses sharing (use with caution)

// ═══════════════════════════════════════════════════════════════════════════
// QUERY PLAN ANALYSIS
// ═══════════════════════════════════════════════════════════════════════════

/*
Use Developer Console or CLI to analyze query plans:

1. Developer Console:
   - Query Editor > Check "Use Tooling API"
   - Click "Query Plan" button

2. CLI (uses REST API explain endpoint):
   sf api request rest '/query/?explain=SELECT+Id+FROM+Account+WHERE+Name='\''Test'\''' --target-org my-org --json

Plan Output:
- Cardinality: Estimated rows returned
- Fields: Index fields used
- LeadingOperationType:
  - "Index" = Good (using index)
  - "TableScan" = Bad (scanning all records)
- Cost: Relative cost (lower is better)
- sObjectCardinality: Total records in object
*/

// ═══════════════════════════════════════════════════════════════════════════
// LARGE DATA VOLUME PATTERNS
// ═══════════════════════════════════════════════════════════════════════════

-- For tables with millions of records:

-- 1. Always use indexed fields
SELECT Id FROM Account
WHERE External_Id__c = 'EXT-123'

-- 2. Use date range with limit
SELECT Id, Name FROM Contact
WHERE CreatedDate >= 2024-01-01T00:00:00Z
AND CreatedDate < 2024-02-01T00:00:00Z
LIMIT 10000

-- 3. Use skinny tables (configured by Salesforce Support)
-- Query fields included in skinny table for best performance

-- 4. Consider archiving old records
SELECT Id FROM Account
WHERE LastActivityDate < LAST_N_YEARS:2
AND CreatedDate < LAST_N_YEARS:3

-- 5. Use Database.query() with query locator for batch
/*
Database.QueryLocator ql = Database.getQueryLocator(
    'SELECT Id, Name FROM Account WHERE CreatedDate = THIS_YEAR'
);
// Can process up to 50 million records in batch
*/