dm: Analyze, build and deploy relational data models
summary
dm claims it cuts complex join-and-filter logic from 800→6 lines by representing multiple tables as a dm object that stores primary and foreign keys and works with local data frames or lazy tables via dbplyr. A Shiny UI generates reusable code and supports CRUD workflows; version 1.0.1 adds a model builder. New developer features include dm_deconstruct (deconstruct/reconstruct keyed tables) and dm_ptype (schema prototype) to write schemas before populating databases. Works with tidyverse, igraph, Arrow ideas, and converts between normalized and nested forms.
AI-written summary · about
chapters (18)
- 0:00dm package overview
- 1:11dm package history
- 2:38Install dm
- 3:20Relational models in R
- 4:46Keys in dm objects
- 5:52dm vs manual approaches
- 6:32Build dm with Shiny
- 7:40Define keys in data frames
- 8:46dm consulting cases
- 10:15CRUD in Shiny with dm
- 10:35Data analysis with dm
- 11:20Deconstruct and reconstruct models
- 13:10Adapt dm to workflows
- 13:32Transform whole models
- 14:38Create DB structures with dm_ptype
- 15:19Normalized to nested data
- 16:24Future of data transitions
- 16:46About cynkra
related talks
17mdm: Analyze, build and deploy relational data models
24mMake data central feature
34mDimensional Data Modeling in the Modern Era
44mOpenMDM - Data Management for SDV
01h 42mFall in Love with Data Transformation: DBT as Your Weapon
39mData Mesh: Why, What and How
18mduckplyr: Analyze large data with full dplyr compatibility
26mDataEngBytes 2023 - MEL-T1-01 - Joe Reis