LakeSQL

LakeSQL is a command line tool that has been designed specifically for TiDB Cloud Lake. It allows users to establish a connection with TiDB Cloud Lake and execute queries directly from a CLI window.

LakeSQL is particularly useful for those who prefer a command line interface and need to work with TiDB Cloud Lake on a regular basis. With LakeSQL, users can easily and efficiently manage their databases, tables, and data, and perform a wide range of queries and operations with ease.

Installing LakeSQL

LakeSQL offers multiple installation options to suit different platforms and preferences. Choose your preferred method from the sections below or download the installation package from the LakeSQL release page to install it manually.

Shell Script

LakeSQL provides a convenient Shell script for installation. You can choose between two options:

Default Installation

Install LakeSQL to the user's home directory (~/.lakesql):

curl -fsSL https://lakesql-bin.tidbcloud.com/install/lakesql.sh | bash

# highlight-next-line
curl -fsSL https://lakesql-bin.tidbcloud.com/install/lakesql.sh | bash

                                  L A K E S Q L
                                    Installer

--------------------------------------------------------------------------------
Website: https://tidbcloud.com
Docs: https://docs.pingcap.com/tidbcloudlake/
Github: https://github.com/tidbcloud/lakesql
--------------------------------------------------------------------------------

>>> We'll be installing LakeSQL via a pre-built archive at https://lakesql-bin.tidbcloud.com/lakesql/v0.22.2/
>>> Ready to proceed? (y/n)

>>> Please enter y or n.
>>> y

--------------------------------------------------------------------------------

>>> Downloading LakeSQL archive via https://lakesql-bin.tidbcloud.com/lakesql/v0.22.2/lakesql-aarch64-apple-darwin.tar.gz ✓
>>> Unpacking archive to /Users/eric/.lakesql ... ✓
>>> Adding LakeSQL path to /Users/eric/.zprofile ✓
>>> Adding LakeSQL path to /Users/eric/.profile ✓
>>> Install succeeded! 🚀
>>> To start LakeSQL:

    lakesql --help

>>> More information at https://github.com/tidbcloud/lakesql

Custom Installation with `--prefix`

Install LakeSQL to a specified directory (e.g., /usr/local):

curl -fsSL https://lakesql-bin.tidbcloud.com/install/lakesql.sh | bash -s -- -y --prefix /usr/local

# highlight-next-line
curl -fsSL https://lakesql-bin.tidbcloud.com/install/lakesql.sh | bash -s -- -y --prefix /usr/local
                                  L A K E S Q L
                                    Installer

--------------------------------------------------------------------------------
Website: https://tidbcloud.com
Docs: https://docs.pingcap.com
Github: https://github.com/tidbcloud/lakesql
--------------------------------------------------------------------------------

>>> Downloading LakeSQL via https://lakesql-bin.tidbcloud.com/lakesql/v0.22.2/lakesql-aarch64-apple-darwin.tar.gz ✓
>>> Unpacking archive to /usr/local ... ✓
>>> Install succeeded! 🚀
>>> To start LakeSQL:

    lakesql --help

>>> More information at https://github.com/tidbcloud/lakesql

Homebrew (for macOS)

You can easily install LakeSQL on macOS using Homebrew with a simple command:

brew install tidbcloud/homebrew-tap/lakesql

Apt (for Ubuntu/Debian)

On Ubuntu and Debian systems, you can install LakeSQL using the Apt package manager:

curl -fsSL https://lakesql-bin.tidbcloud.com/keys/lakesql-archive-keyring.gpg \
  | sudo tee /usr/share/keyrings/lakesql-archive-keyring.gpg >/dev/null
echo "deb [signed-by=/usr/share/keyrings/lakesql-archive-keyring.gpg] https://lakesql-bin.tidbcloud.com/apt stable main" \
  | sudo tee /etc/apt/sources.list.d/lakesql.list >/dev/null
sudo apt-get update
sudo apt-get install -y lakesql

Cargo (Rust Package Manager)

To install LakeSQL using Cargo, utilize the cargo-binstall tool or build from source using the provided command.

Note

Before installing with Cargo, make sure you have the full Rust toolchain and the cargo command installed on your computer. If you don't, follow the installation guide at https://rustup.rs/.

Using cargo-binstall

Please refer to Cargo B(inary)Install - Installation to install cargo-binstall and enable the cargo binstall <crate-name> subcommand.

cargo binstall lakesql

Building from Source

When building from source, some dependencies may involve compiling C/C++ code. Ensure that you have the GCC/G++ or Clang toolchain installed on your computer.

cargo install lakesql

User Authentication

For connections to TiDB Cloud Lake, you can use the default cloudapp user or an SQL user created with the CREATE USER command. Note that the user account you use to log in to the TiDB Cloud Lake console cannot be used for connecting to TiDB Cloud Lake.

Connecting with LakeSQL

LakeSQL allows you to connect to both TiDB Cloud Lake instances.

Customize Connections with a DSN

A DSN (Data Source Name) is a simple yet powerful way to configure and manage your TiDB Cloud Lake connection in LakeSQL using a single URI-style string. This method allows you to embed your credentials and connection settings directly into your environment, streamlining the connection process.

DSN Format and Parameters

lake[+flight]://user[:password]@host[:port]/[database][?sslmode=disable][&arg1=value1]

Common DSN Parameters	Description
`tenant`	Tenant ID, TiDB Cloud Lake only.
`warehouse`	Warehouse name, TiDB Cloud Lake only.
`sslmode`	Set to `disable` if not using TLS.
`tls_ca_file`	Custom root CA certificate path.
`connect_timeout`	Connect timeout in seconds.

RestAPI Client Parameters	Description
`wait_time_secs`	Request wait time for page, default is `1`.
`max_rows_in_buffer`	Maximum rows for page buffer.
`max_rows_per_page`	Maximum response rows for a single page.
`page_request_timeout_secs`	Timeout for a single page request, default is `30`.
`presign`	Enable presign for data loading. Options: `auto`, `detect`, `on`, `off`. Default is `auto` (only enabled for TiDB Cloud Lake).

FlightSQL Client Parameters	Description
`query_timeout`	Query timeout in seconds.
`tcp_nodelay`	Defaults to `true`.
`tcp_keepalive`	TCP keepalive in seconds (default is `3600`, set to `0` to disable).
`http2_keep_alive_interval`	Keep-alive interval in seconds, default is `300`.
`keep_alive_timeout`	Keep-alive timeout in seconds, default is `20`.
`keep_alive_while_idle`	Defaults to `true`.

DSN Examples

# Local connection using HTTP API with presign detection
lake://root:@localhost:8000/?sslmode=disable&presign=detect

# TiDB Cloud Lake connection with tenant and warehouse info
lake://user1:password1@tnxxxx--default.gw.aws-us-east-2.default.tidbcloud.com:443/benchmark?enable_dphyp=1

# Local connection using FlightSQL API
lake+flight://root:@localhost:8900/database1?connect_timeout=10

Connect to TiDB Cloud Lake

The best practice for connecting to TiDB Cloud Lake is to obtain your DSN from TiDB Cloud Lake and export it as an environment variable. To obtain your DSN:

Log in to TiDB Cloud Lake and click Connect on the Overview page.
Select the database and warehouse you want to connect to.
Your DSN will be automatically generated in the Examples section. Below the DSN, you'll find a LakeSQL snippet that exports the DSN as an environment variable named LAKESQL_DSN and launches LakeSQL with the correct configuration. You can copy and paste it directly into your terminal.
```
export LAKESQL_DSN="lake://cloudapp:******@tn3ftqihs.gw.aws-us-east-2.default.tidbcloud.com:443/information_schema?warehouse=small-xy2t"
lakesql
```

LakeSQL Settings

LakeSQL provides a range of settings that allow you to define how query results are presented:

Setting	Description
`display_pretty_sql`	When set to `true`, SQL queries will be formatted in a visually appealing manner, making them easier to read and understand.
`prompt`	The prompt displayed in the command line interface, typically indicating the user, warehouse, and database being accessed.
`progress_color`	Specifies the color used for progress indicators, such as when executing queries that take some time to complete.
`show_progress`	When set to `true`, progress indicators will be displayed to show the progress of long-running queries or operations.
`show_stats`	If `true`, query statistics such as execution time, rows read, and bytes processed will be displayed after executing each query.
`max_display_rows`	Sets the maximum number of rows that will be displayed in the output of a query result.
`max_col_width`	Sets the maximum width in characters of each column's display rendering. A value smaller than 3 disables the limit.
`max_width`	Sets the maximum width in characters of the entire display output. A value of 0 defaults to the width of the terminal window.
`output_format`	Sets the format used to display query results (`table`, `csv`, `tsv`, `null`).
`expand`	Controls whether the output of a query is displayed as individual records or in a tabular format. Available values: `on`, `off`, and `auto`.
`multi_line`	Determines whether multi-line input for SQL queries is allowed. When set to `true`, queries can span multiple lines for better readability.
`replace_newline`	Specifies whether newline characters in the output of query results should be replaced with spaces. This can prevent unintended line breaks in the display.

For details of each setting, please refer to the reference information below:

`display_pretty_sql`

The display_pretty_sql setting controls whether SQL queries are displayed in a visually formatted manner or not. When set to false, as in the first query below, SQL queries are not formatted for visual appeal. In contrast, when set to true, as in the second query, SQL queries are formatted in a visually appealing manner, making them easier to read and understand.

// highlight-next-line
root@localhost:8000/default> !set display_pretty_sql false
root@localhost:8000/default> SELECT TO_STRING(ST_ASGEOJSON(ST_GEOMETRYFROMWKT('SRID=4326;LINESTRING(400000 6000000, 401000 6010000)'))) AS pipeline_geojson;
┌─────────────────────────────────────────────────────────────────────────┐
│                             pipeline_geojson                            │
│                                  String                                 │
├─────────────────────────────────────────────────────────────────────────┤
│ {"coordinates":[[400000,6000000],[401000,6010000]],"type":"LineString"} │
└─────────────────────────────────────────────────────────────────────────┘
1 row read in 0.063 sec. Processed 1 row, 1 B (15.76 rows/s, 15 B/s)

// highlight-next-line
root@localhost:8000/default> !set display_pretty_sql true
root@localhost:8000/default> SELECT TO_STRING(ST_ASGEOJSON(ST_GEOMETRYFROMWKT('SRID=4326;LINESTRING(400000 6000000, 401000 6010000)'))) AS pipeline_geojson;

SELECT
  TO_STRING(
    ST_ASGEOJSON(
      ST_GEOMETRYFROMWKT(
        'SRID=4326;LINESTRING(400000 6000000, 401000 6010000)'
      )
    )
  ) AS pipeline_geojson

┌─────────────────────────────────────────────────────────────────────────┐
│                             pipeline_geojson                            │
│                                  String                                 │
├─────────────────────────────────────────────────────────────────────────┤
│ {"coordinates":[[400000,6000000],[401000,6010000]],"type":"LineString"} │
└─────────────────────────────────────────────────────────────────────────┘
1 row read in 0.087 sec. Processed 1 row, 1 B (11.44 rows/s, 11 B/s)

`prompt`

The prompt setting controls the format of the command line interface prompt. In the example below, it was initially set to display the user and warehouse ({user}@{warehouse}). After updating it to {user}@{warehouse}/{database}, the prompt now includes the user, warehouse, and database.

// highlight-next-line
root@localhost:8000/default> !set prompt {user}@{warehouse}
root@localhost:8000 !configs
Settings {
    display_pretty_sql: true,
    prompt: "{user}@{warehouse}",
    progress_color: "cyan",
    show_progress: true,
    show_stats: true,
    max_display_rows: 40,
    max_col_width: 1048576,
    max_width: 1048576,
    output_format: Table,
    quote_style: Necessary,
    expand: Off,
    time: None,
    multi_line: true,
    replace_newline: true,
}
// highlight-next-line
root@localhost:8000 !set prompt {user}@{warehouse}/{database}
root@localhost:8000/default

`progress_color`

The progress_color setting controls the color used for progress indicators during query execution. In this example, the color has been set to blue:

// highlight-next-line
root@localhost:8000/default> !set progress_color blue

`show_progress`

When set to true, progress information is displayed during the execution of a query. The progress information includes the number of rows processed, the total number of rows in the query, the processing speed in rows per second, the amount of memory processed, and the processing speed in memory per second.

// highlight-next-line
root@localhost:8000/default> !set show_progress true
root@localhost:8000/default> select * from numbers(1000000000000000);
⠁ [00:00:08] Processing 18.02 million/1 quadrillion (2.21 million rows/s), 137.50 MiB/7.11 PiB (16.88 MiB/s) ░

`show_stats`

The show_stats setting controls whether query statistics are displayed after executing each query. When set to false, as the first query in the example below, query statistics are not displayed. In contrast, when set to true, as in the second query, query statistics such as execution time, rows read, and bytes processed are displayed after executing each query.

// highlight-next-line
root@localhost:8000/default> !set show_stats false
root@localhost:8000/default> select now();
┌────────────────────────────┐
│            now()           │
│          Timestamp         │
├────────────────────────────┤
│ 2024-04-23 23:27:11.538673 │
└────────────────────────────┘
// highlight-next-line
root@localhost:8000/default> !set show_stats true
root@localhost:8000/default> select now();
┌────────────────────────────┐
│            now()           │
│          Timestamp         │
├────────────────────────────┤
│ 2024-04-23 23:49:04.754296 │
└────────────────────────────┘
1 row read in 0.045 sec. Processed 1 row, 1 B (22.26 rows/s, 22 B/s)

`max_display_rows`

The max_display_rows setting controls the maximum number of rows displayed in the output of a query result. When set to 5 in the example below, only up to 5 rows are displayed in the query result. The remaining rows are indicated with (5 shown).

// highlight-next-line
root@localhost:8000/default> !set max_display_rows 5
root@localhost:8000/default> SELECT * FROM system.configs;
┌──────────────────────────────────────────────────────┐
│   group   │       name       │  value  │ description │
│   String  │      String      │  String │    String   │
├───────────┼──────────────────┼─────────┼─────────────┤
│ query     │ tenant_id        │ default │             │
│ query     │ cluster_id       │ default │             │
│ query     │ num_cpus         │ 0       │             │
│ ·         │ ·                │ ·       │ ·           │
│ ·         │ ·                │ ·       │ ·           │
│ ·         │ ·                │ ·       │ ·           │
│ storage   │ cos.endpoint_url │         │             │
│ storage   │ cos.root         │         │             │
│ 176 rows  │                  │         │             │
│ (5 shown) │                  │         │             │
└──────────────────────────────────────────────────────┘
176 rows read in 0.059 sec. Processed 176 rows, 10.36 KiB (2.98 thousand rows/s, 175.46 KiB/s)

`max_col_width` & `max_width`

The settings max_col_width and max_width specify the maximum permitted width in characters for individual columns and the entire display output, respectively. The following example sets column display width to 10 characters and the entire display width to 100 characters:

// highlight-next-line
root@localhost:8000/default> .max_col_width 10
// highlight-next-line
root@localhost:8000/default> .max_width 100
root@localhost:8000/default> select * from system.settings;
┌──────────────────────────────────────────────────────────────────────────────────────────────────┐
│    name    │  value  │ default │   range  │  level  │            description            │  type  │
│   String   │  String │  String │  String  │  String │               String              │ String │
├────────────┼─────────┼─────────┼──────────┼─────────┼───────────────────────────────────┼────────┤
│ acquire... │ 15      │ 15      │ None     │ DEFAULT │ Sets the maximum timeout in se... │ UInt64 │
│ aggrega... │ 0       │ 0       │ None     │ DEFAULT │ Sets the maximum amount of mem... │ UInt64 │
│ aggrega... │ 0       │ 0       │ [0, 100] │ DEFAULT │ Sets the maximum memory ratio ... │ UInt64 │
│ auto_co... │ 50      │ 50      │ None     │ DEFAULT │ Threshold for triggering auto ... │ UInt64 │
│ collation  │ utf8    │ utf8    │ ["utf8"] │ DEFAULT │ Sets the character collation. ... │ String │
│ ·          │ ·       │ ·       │ ·        │ ·       │ ·                                 │ ·      │
│ ·          │ ·       │ ·       │ ·        │ ·       │ ·                                 │ ·      │
│ ·          │ ·       │ ·       │ ·        │ ·       │ ·                                 │ ·      │
│ storage... │ 1048576 │ 1048576 │ None     │ DEFAULT │ Sets the byte size of the buff... │ UInt64 │
│ table_l... │ 10      │ 10      │ None     │ DEFAULT │ Sets the seconds that the tabl... │ UInt64 │
│ timezone   │ UTC     │ UTC     │ None     │ DEFAULT │ Sets the timezone.                │ String │
│ unquote... │ 0       │ 0       │ None     │ DEFAULT │ Determines whether TiDB Cloud Lake tr... │ UInt64 │
│ use_par... │ 0       │ 0       │ [0, 1]   │ DEFAULT │ This setting is deprecated        │ UInt64 │
│ 96 rows    │         │         │          │         │                                   │        │
│ (10 shown) │         │         │          │         │                                   │        │
└──────────────────────────────────────────────────────────────────────────────────────────────────┘
96 rows read in 0.040 sec. Processed 96 rows, 16.52 KiB (2.38 thousand rows/s, 410.18 KiB/s)

`output_format`

By setting the output_format to table, csv, tsv, or null, you can control the format of the query result. The table format presents the result in a tabular format with column headers, while the csv and tsv formats provide comma-separated values and tab-separated values respectively, and the null format suppresses the output formatting altogether.

// highlight-next-line
root@localhost:8000/default> !set output_format table
root@localhost:8000/default> show users;
┌────────────────────────────────────────────────────────────────────────────┐
│  name  │ hostname │  auth_type  │ is_configured │  default_role │ disabled │
│ String │  String  │    String   │     String    │     String    │  Boolean │
├────────┼──────────┼─────────────┼───────────────┼───────────────┼──────────┤
│ root   │ %        │ no_password │ YES           │ account_admin │ false    │
└────────────────────────────────────────────────────────────────────────────┘
1 row read in 0.032 sec. Processed 1 row, 113 B (31.02 rows/s, 3.42 KiB/s)

// highlight-next-line
root@localhost:8000/default> !set output_format csv
root@localhost:8000/default> show users;
root,%,no_password,YES,account_admin,false
1 row read in 0.062 sec. Processed 1 row, 113 B (16.03 rows/s, 1.77 KiB/s)

// highlight-next-line
root@localhost:8000/default> !set output_format tsv
root@localhost:8000/default> show users;
root % no_password YES account_admin false
1 row read in 0.076 sec. Processed 1 row, 113 B (13.16 rows/s, 1.45 KiB/s)

// highlight-next-line
root@localhost:8000/default> !set output_format null
root@localhost:8000/default> show users;
1 row read in 0.036 sec. Processed 1 row, 113 B (28.1 rows/s, 3.10 KiB/s)

`expand`

The expand setting controls whether the output of a query is displayed as individual records or in a tabular format. When the expand setting is set to auto, the system automatically determines how to display the output based on the number of rows returned by the query. If the query returns only one row, the output is displayed as a single record.

// highlight-next-line
root@localhost:8000/default> !set expand on
root@localhost:8000/default> show users;
-[ RECORD 1 ]-----------------------------------
         name: root
     hostname: %
    auth_type: no_password
is_configured: YES
 default_role: account_admin
     disabled: false

1 row read in 0.055 sec. Processed 1 row, 113 B (18.34 rows/s, 2.02 KiB/s)

// highlight-next-line
root@localhost:8000/default> !set expand off
root@localhost:8000/default> show users;
┌────────────────────────────────────────────────────────────────────────────┐
│  name  │ hostname │  auth_type  │ is_configured │  default_role │ disabled │
│ String │  String  │    String   │     String    │     String    │  Boolean │
├────────┼──────────┼─────────────┼───────────────┼───────────────┼──────────┤
│ root   │ %        │ no_password │ YES           │ account_admin │ false    │
└────────────────────────────────────────────────────────────────────────────┘
1 row read in 0.046 sec. Processed 1 row, 113 B (21.62 rows/s, 2.39 KiB/s)

// highlight-next-line
root@localhost:8000/default> !set expand auto
root@localhost:8000/default> show users;
-[ RECORD 1 ]-----------------------------------
         name: root
     hostname: %
    auth_type: no_password
is_configured: YES
 default_role: account_admin
     disabled: false

1 row read in 0.037 sec. Processed 1 row, 113 B (26.75 rows/s, 2.95 KiB/s)

`multi_line`

When the multi_line setting is set to true, allowing input to be entered across multiple lines. As a result, the SQL query is entered with each clause on a separate line for improved readability and organization.

// highlight-next-line
root@localhost:8000/default> !set multi_line true;
root@localhost:8000/default> SELECT *
> FROM system.configs;
┌──────────────────────────────────────────────────────┐
│   group   │       name       │  value  │ description │
│   String  │      String      │  String │    String   │
├───────────┼──────────────────┼─────────┼─────────────┤
│ query     │ tenant_id        │ default │             │
│ query     │ cluster_id       │ default │             │
│ query     │ num_cpus         │ 0       │             │
│ ·         │ ·                │ ·       │ ·           │
│ ·         │ ·                │ ·       │ ·           │
│ ·         │ ·                │ ·       │ ·           │
│ storage   │ cos.endpoint_url │         │             │
│ storage   │ cos.root         │         │             │
│ 176 rows  │                  │         │             │
│ (5 shown) │                  │         │             │
└──────────────────────────────────────────────────────┘
176 rows read in 0.060 sec. Processed 176 rows, 10.36 KiB (2.91 thousand rows/s, 171.39 KiB/s)

`replace_newline`

The replace_newline setting determines whether newline characters (\n) are replaced with the literal string (\n) in the output. In the example below, the replace_newline setting is set to true. As a result, when the string 'Hello\nWorld' is selected, the newline character (\n) is replaced with the literal string (\n). So, instead of displaying the newline character, the output displays 'Hello\nWorld' as 'Hello\nWorld':

// highlight-next-line
root@localhost:8000/default> !set replace_newline true
root@localhost:8000/default> SELECT 'Hello\nWorld' AS message;
┌──────────────┐
│    message   │
│    String    │
├──────────────┤
│ Hello\nWorld │
└──────────────┘
1 row read in 0.056 sec. Processed 1 row, 1 B (18 rows/s, 17 B/s)

// highlight-next-line
root@localhost:8000/default> !set replace_newline false;
root@localhost:8000/default> SELECT 'Hello\nWorld' AS message;
┌─────────┐
│ message │
│  String │
├─────────┤
│ Hello   │
│ World   │
└─────────┘
1 row read in 0.067 sec. Processed 1 row, 1 B (14.87 rows/s, 14 B/s)

Configuring LakeSQL Settings

You have the following options to configure a LakeSQL setting:

Use the !set <setting> <value> command. For more information, see Utility Commands.
Add and configure a setting in the configuration file ~/.config/lakesql/config.toml. To do so, open the file and add your setting under the [settings] section. The following example sets the max_display_rows to 10 and max_width to 100:

...
[settings]
max_display_rows = 10
max_width = 100
...

Configure a setting at runtime by launching LakeSQL and then specifying the setting in the format .<setting> <value>. Please note that settings configured in this way only take effect in the current session.

root@localhost:8000/default> .max_display_rows 10
root@localhost:8000/default> .max_width 100

Utility Commands

LakeSQL provides users with a variety of commands to streamline their workflow and customize their experience. Here's an overview of the commands available in LakeSQL:

Command	Description
`!exit`	Exits LakeSQL.
`!quit`	Exits LakeSQL.
`!configs`	Displays current LakeSQL settings.
`!set <setting> <value>`	Modifies a LakeSQL setting.
`!source <sql_file>`	Executes a SQL file.

For examples of each command, please refer to the reference information below:

`!exit`

Disconnects from TiDB Cloud Lake and exits LakeSQL.

➜  ~ lakesql
Welcome to LakeSQL 0.17.0-homebrew.
Connecting to localhost:8000 as user root.
Connected to TiDB Cloud Lake Query v1.2.427-nightly-b1b622d406(rust-1.77.0-nightly-2024-04-20T22:12:35.318382488Z)

// highlight-next-line
root@localhost:8000/default> !exit
Bye~

`!quit`

Disconnects from TiDB Cloud Lake and exits LakeSQL.

➜  ~ lakesql
Welcome to LakeSQL 0.17.0-homebrew.
Connecting to localhost:8000 as user root.
Connected to TiDB Cloud Lake Query v1.2.427-nightly-b1b622d406(rust-1.77.0-nightly-2024-04-20T22:12:35.318382488Z)

// highlight-next-line
root@localhost:8000/default> !quit
Bye~
➜  ~

`!configs`

Displays the current LakeSQL settings.

// highlight-next-line
root@localhost:8000/default> !configs
Settings {
    display_pretty_sql: true,
    prompt: "{user}@{warehouse}/{database}> ",
    progress_color: "cyan",
    show_progress: true,
    show_stats: true,
    max_display_rows: 40,
    max_col_width: 1048576,
    max_width: 1048576,
    output_format: Table,
    quote_style: Necessary,
    expand: Off,
    time: None,
    multi_line: true,
    replace_newline: true,
}

`!set <setting> <value>`

Modifies a LakeSQL setting.

root@localhost:8000/default> !set display_pretty_sql false

`!source <sql_file>`

Executes a SQL file.

➜  ~ more ./desktop/test.sql
CREATE TABLE test_table (
    id INT,
    name VARCHAR(50)
);

INSERT INTO test_table (id, name) VALUES (1, 'Alice');
INSERT INTO test_table (id, name) VALUES (2, 'Bob');
INSERT INTO test_table (id, name) VALUES (3, 'Charlie');
➜  ~ lakesql
Welcome to LakeSQL 0.17.0-homebrew.
Connecting to localhost:8000 as user root.
Connected to TiDB Cloud Lake Query v1.2.427-nightly-b1b622d406(rust-1.77.0-nightly-2024-04-20T22:12:35.318382488Z)

// highlight-next-line
root@localhost:8000/default> !source ./desktop/test.sql
root@localhost:8000/default> SELECT * FROM test_table;

SELECT
  *
FROM
  test_table

┌────────────────────────────────────┐
│        id       │       name       │
│ Nullable(Int32) │ Nullable(String) │
├─────────────────┼──────────────────┤
│               1 │ Alice            │
│               2 │ Bob              │
│               3 │ Charlie          │
└────────────────────────────────────┘
3 rows read in 0.064 sec. Processed 3 rows, 81 B (46.79 rows/s, 1.23 KiB/s)

LakeSQL

Installing LakeSQL

Shell Script

Default Installation

Custom Installation with --prefix

Homebrew (for macOS)

Apt (for Ubuntu/Debian)

Cargo (Rust Package Manager)

User Authentication

Connecting with LakeSQL

Customize Connections with a DSN

DSN Format and Parameters

DSN Examples

Connect to TiDB Cloud Lake

LakeSQL Settings

display_pretty_sql

prompt

progress_color

show_progress

show_stats

max_display_rows

max_col_width & max_width

output_format

expand

multi_line

replace_newline