Apex Cursor Demo

Lightning demos for Apex Cursors (Database.Cursor and pagination cursors) with LWC: infinite scroll, reporting drill-down, and adaptive async chunking. Blog posts: infinite scroll and pagination cursors. Also see associated Londons Calling presentation here and Apex Hours presentation here.

Shared: ApexCursorDemo app and permission set.

• Salesforce CLI (sf) installed and authenticated
• A Dev Hub org configured
• Basic knowledge of Lightning Web Components and Apex

# Create a fresh scratch org
sf org create scratch --definition-file config/project-scratch-def.json --alias apex-cursor-demo --set-default --duration-days 7

# Deploy all metadata to the scratch org
sf project deploy start

# Run the Apex script to create dummy Account records
sf apex run --file apex/CreateDummyAccounts.apex

# Assign the permission set
sf org assign permset --name ApexCursorDemo

# Generate ~8 years of daily RAM/CPU unit prices (CSV)
node scripts/generate-price-observations.mjs

# Upsert into UnitPriceObservation__c
sf data upsert bulk \
  -f data/unit-price-observations.csv \
  -s UnitPriceObservation__c \
  -i External_Id__c \
  -o apex-cursor-demo \
  -w 10

# Infinite Loading tab
sf org open --path /lightning/n/InfiniteLoading

# Reporting drill-down tab (weekly chart + cursor detail)
sf org open --path /lightning/n/ReportingDrillingDown

# Adaptive Async tab (cursor + queueable + platform events)
sf org open --path /lightning/n/AdaptiveAsync

# Generate CSV, delete existing ADAPT-ORD-* orders, then bulk upsert accounts and orders
node scripts/generate-adaptive-orders.mjs --load

1. Open the Demo: The command above will take you directly to the Infinite Loading tab
2. View the Data Table: The component will automatically load the first batch of records
3. Test Infinite Scrolling: Scroll to the bottom of the table to load more records
4. Monitor Debug Info: Check the debug section at the bottom for real-time information

1. Choose RAM or CPU and wait for the monthly chart (mid-month values via pagination cursor fetchPage, no aggregate SOQL)
2. Drag across the chart to select a date range (snaps to days; drag handles to resize)
3. Daily rows for that range load below via pagination cursor fetchPage from session cache
4. Click Refresh to rebuild the session cursor and chart after data changes

1. Open Adaptive Async in the Apex Cursor Demo app (1500 demo orders after Step 7; rerunnable without reset)
2. Click Start run — each queueable scans the next cursor window and packs orders until Limits.getLimitCallouts() predicted HTTP callouts would be exceeded
3. Progress bar updates as AdaptiveAsyncChunk__e platform events arrive
4. Bar chart — bar color reflects callout pressure (green / amber / red); chunk size varies with order mix
5. Datatable lists each chunk (sequence, planned vs actual orders, predicted callouts, job id)