ByteHouse CLI is a command-line application for the most direct way to interact with Bytehouse Services
To install the ByteHouse CLI, make sure you have Homebrew installed on your machine and run the following commands:
brew tap bytehouse-cloud/homebrew-core brew install bytehouse-cli
To upgrade the ByteHouse CLI, run the following command:
brew upgrade bytehouse-cli
You can get the most updated version here: https://github.com/bytehouse-cloud/cli/releases/
curl -o bytehouse-cli -L https://github.com/bytehouse-cloud/cli/releases/download/v126.96.36.199.1/bytehouse-v188.8.131.52.1-linux-amd64 chmod +x bytehouse-cli # You might want to add this binary executable to your `~/.bashrc` as alias, or `~/.zshrc`l echo "alias bytehouse-cli=\"$(pwd)/bytehouse-cli\"" > ~/.bashrc
Download the latest installer (bytehouse-vX.X.XX.X-windows-amd64) from https://github.com/bytehouse-cloud/cli/releases
There are few credentials you would need to get started
Note: The information you would need is the same as how you would log in with Web UI.
If you are unsure, you can also check your details on the top right corner of the Web Console.
The simplest way to start the application is to run it in your command line or PowerShell.
When specifying flag and its value when starting the application, the format is
--<flag> value , eg
Flags also have an alias, see in Reference: Alias
An example of starting bytehouse-cli is shown below:
bytehouse-cli --user <user name> --account <account name> --password <password> --region <region name> --secure # Example $ bytehouse-cli --user bob --account AWSXXX --password coolbob --region cn-north-1 --secure
--secureflag is needed when connecting to bytehouse's public domain
Sometimes it's neater and more manageable to keep all flags in a configuration file. With the configuration file, you can also specify query settings in it. You can use the
-cf flag with the path to configuration file as value.
For full usage of configuration File, see in Reference: Configuration File
An example of configuration file and usage is shown below:
$ cat bytehouse_conf.toml # Settings for connection account = "AWSXXXXX" user = "bob" password = "coolbob" region = "cn-north-1" secure = true # Settings for query Settings ansi_sql = true $ bytehouse-cli -cf bytehouse_conf.toml
Sometimes you could be writing shell script and it could be impractical to get into interactive mode. Bytehouse CLI allows the user to execute a SQL command and exit automatically.
If you launch bytehouse-cli with
--query flag, that SQL statement will be executed and bytehouse-cli will exit immediately after the execution.
$ bytehouse-cli -q "select 1"
Users can also allow bytehouse-cli to take in input from
$ echo "select 1" | bytehouse-cli
Users can also write a SQL script and pipe the input to bytehouse-cli
Queries are separated by
Queries will be run sequentially
Stops further execution when first query execution returns with an error
$ cat example.sql CREATE DATABASE bob_db; USE bob_db; CREATE TABLE bob_numbers ( i Int32 ) ENGINE = CnchMergeTree ORDER BY i; SHOW CREATE TABLE bob_numbers; $ bytehouse-cli < example.sql # This is also accepted $ cat example.sql | bytehouse-cli
It is very common to load data from a file, below shows some examples on how to do so.
Bytehouse » INSERT INTO bob_db.bob_number VALUES (1), (2), (3)
$ bytehouse-cli -q "INSERT INTO bob_db.bob_number VALUES (1), (2), (3)"
Bytehouse » INSERT INTO bob_db.bob_number FORMAT csv INFILE 'path/to/data.csv'
$ bytehouse-cli -q "INSERT INTO bob_db.bob_number FORMAT csv" < 'path/to/data.csv'
You can use the
INTO OUTFILE syntax after your query to save your results to a local file.
Bytehouse » SELECT * FROM bob_db.bob_number INTO OUTFILE 'out.csv' format csv
You can check the version of the ByteHouse CLI using the
--version flag. When flag is specified, ByteHouse CLI does not start
$ bytehouse-cli -v v1.5.2
You can show all the flags supported by using
-h flag or
#To display all options and their alias bytehouse-cli -h
Updated about 2 months ago