Making Data Dance
summary
Window functions like lead, lag, first_value, last_value and row_number are used to group and collapse chat messages by user and room, with examples that partition by fruit and by room. Raw SQL via the PG gem or ActiveRecord connection executes CTEs, views and materialized view refreshed concurrently to precompute hourly buckets and heatmaps, then index them for sub-second reads; a chord and sunburst built with D3 visualize movements from beacon UUID/major/minor data. Also covered: JSONB, ranges, UUIDs, upsert (9.5) and parallel scans (9.6).
AI-written summary · about
chapters (8)
- 0:00Start
- 5:26CAVEATS
- 7:02WINDOW FUNCTION GREATEST HITS
- 15:38QUESTIONS WE CAN NOW ANSWER
- 16:17SUBQUERY
- 27:12POSTGRES
- 30:05DATATYPES
- 32:24REFACTOR
related talks
31mMaking Data Dance
01h 06mPostgres Window Magic
50mJOIN "My Session" ON Modern = 'SQL'
30mData Rainbows - select * from cloud;
45mSELECT 'amazing_features' FROM "postgresql"
01mWindowing Functions for Database Analytics
21mA Hands-on Experience with Complex SQL
49mSELECT 'amazing features' FROM "postgresql"