Getting Started with Dolthub

The instructions here are excellent as an intro to Dolt's general usage. Here’s our dolt org.

Anyone can create a Pull Request to be merged into the database via the SQL UI at DoltHub, SQL editor integration, command line, or Python.

Supported statements

​https://docs.dolthub.com/interfaces/sql/sql-support/supported-statements These are limited, so dolt is not great for managing a bunch of schemas.

Useful statements

Spreadsheets

Dolt has a guide here for spreadsheets generally, including ODBC setup. Using the CLI, first ensure you’ve initialized the repo you’d like to pull from. Export a table like this:

dolt table export source_types > source_types.csv
dolt table export <table_name> > <file_name>

Add new files

dolt table import -c --pk <primary_key> <table_name> <file_name>
dolt table import -c --pk name format_types format_types.csv

Use dolt add . to start tracking all new files in the directory.

Run two Dolt repos with sql-server

Dolt's documentation on this is here.

There are two ways to run the dolt server with multiple databases served in one instance, in which you can use use db_name to switch databases.

Method 1

dolt sql-server -H 0.0.0.0 -P 3306 -u root -p root123 --multi-db-dir /path_to_dir_that_hosts_multi_repos/

Method 2

Create a config file, example.cnf:

log_level: info
​
behavior:
  read_only: false
  autocommit: true
​
user:
  name: root
  password: "root123"
​
listener:
  host: 0.0.0.0
  port: 3306
  max_connections: 10
  read_timeout_millis: 28800000
  write_timeout_millis: 28800000
  
databases: 
  - name: richardji_datasets
    path: /home/oracle/work/pdap/dolt/richardji_datasets
  - name: rji_test
    path: /home/oracle/work/pdap/dolt/rji_test
​
performance:
  query_parallelism: null

Then run it with:

dolt sql-server --config my2.cnf