Since the version 0.5.0, doteur is shipped in several packages, for more information check the features section. Soome CLI options might not be avalaible regarding which version you installed on your system

You will find in this section the instructions on how to use this tool on your computer. Following this guide, you should be able to use this tool at your will.

As a reminder, doteur is a CLI tool meaning that you need to run it through a shell or cmd session. There is no GUI available at the moment.


doteur 0.5.1
LABEYE Loïc <loic.labeye@pm.me>
Parse a SQL configuration and convert it into a .dot file, render the output if Graphviz is
installed

USAGE:
    doteur [OPTIONS] [INPUT]...

ARGS:
    <INPUT>...    Name of the sql file or database location if an URL arg is passed, can also be
                  a directory or several files

OPTIONS:
        --dark_mode            Wheter to render in dark mode or not
    -h, --help                 Print help information
    -i, --include <INCLUDE>    Filter to include only the given tables, accept simple regexs
        --legend               Includes hint about the relations type at the bottom of the output
                               file
    -o, --output <OUTPUT>      Name of the output file [default: output.dot]
    -V, --version              Print version information
    -x, --exclude <EXCLUDE>    Filter to exclude the given tables, accept simple regexs

Some functionnalities might not appear as they depend on which version this tool has been downloaded
or built for.


Example usage

Export a .sql file to a .dot

usr@pop-os:~$ doteur sample.sql

The output will be in the output.dot

Export a .sql file to a .png

usr@pop-os:~$ doteur sample.sql -o output.png

For the format supported, please refer to the graphviz documentation

[mysql_addons] Connect to a remote database without a dialog and export the file as .png

usr@ubuntu:~/doteur$ doteur mysql://newuser:password@localhost:3306/foo --url -o output.png

[mysql_addons] Connect to a remote database with dialog and export the file as .png

usr@ubuntu:~/doteur$ doteur --it -o output.png
Database url or ip: localhost
Database port: 3306
Database name: foo
Database user: username
Database password: [hidden]

The output will be in a png file.

[sqlite_addons] Connect to a sqlite3 database without a dialog and export the file as .png

doteur db.sqlite3 --sqlite -o output.png

Export a .sql file to a .png, render in dark mode

usr@pop-os:~$ doteur sample.sql -o output.png --dark-mode

Export a .sql file to a .png, include only tables who have either the name hello or world

usr@pop-os:~$ doteur sample.sql -o output.png -i hello world

Export a .sql file to a .png, include only tables that start with the name hello

usr@pop-os:~$ doteur sample.sql -o output.png -i hello*

Export a .sql file to a .png, exclude all tables that start with the name hello

usr@pop-os:~$ doteur sample.sql -o output.png -x hello*

See logs of a output

usr@pop-os:~$ RUST_LOG=ERROR doteur sample.sql

With available levels being DEBUG, INFO, WARN, ERROR


Copyright © 2021-2024 LABEYE Loïc. Distributed by an MIT Licence