README
¶
archer
Get help understanding your architecture.
Please keep in mind that this is a work in progress, but it's already useful.
Workspace
To use it, first you need to import the information into a workspace, than you can query and show graphs from it.
A workspace can be set using the -w cli arg. Default is ./.archer (if it exists) or ~/.archer.
Roots and projects
archer work with the idea of roots and projects. A project is a groping of source files or a table, for ex, depending on the importer.
A root is a groping of projects and allows to relate information from different sources.
Importing information
From gradle
Run
archer import gradle <root of your gradle project(s)>
All the projects will be imported. The root name is the name of the root gradle project.
From hibernate
Run
archer import hibernate <source paths> --root <root name>
Imports hibernate configuration from files inside the source paths. Currently only kotlin files are supported, and information is gathered from annotations.
The source paths can be a path on disk or a query (see below).
The root name is required in this case and should be the schema name where the tables live.
You can also use --glob to filter which files should be parsed. For ex: --glob '**/Db*.kt'.
From MySQL
Run
archer import mysql <connection string>
The connection string format is defined here.
All tables will be imported. The root name is the schema name.
Configuring things
You can use archer config set to add information to the projects.
Currently these are supported:
- ignore: does not output this project
- color: fixes the color to be used in graphs
Showing data
Textual output
Run
archer show
Graphs
For this you need to have graphviz dot installed and in your path.
Run
archer graph -o <output file.extension>
Selecting what to see
The simple version of the commands show all information available. This is usually too much, so there are some parameters/filters to select what should be show.
-r <name>
Only show information from one root
'-i '
Only show information that matches the query (see below)
'-e '
Don't show information that matches the query (see below)
Queries
Queries allows you to select which projects are interesting. The supported formats are:
Project query: <project name> or <project simple name>
You can use * inside there to avoid some typing.
Root query: root:<root name>
You can use * inside there to avoid some typing.
Any dependency query: <project 1 query> -> <project 2 query>
Shows any dependency chain that links project 1 to project 2
Max steps dependency query: <project 1 query> -N-> <project 2 query>
N is a number.
Shows any dependency chain that links project 1 to project 2, with at most N hops.
Invert query: !<query>
Inverts the matching.
Multiple matches required: <query> & <query>
To do an OR just pass different -i arguments.
Directories