dbt init and existing Python projects

[schlich]

Hello,

Very frequently I am working on a project and I decide that a dbt project would be a great addition to an existing (mature) project. However, the way that dbt init works seems to leave a lot to be desired in this workflow.

• dbt init seems to be designed to be run for greenfield projects or existing dbt projects and by default the project scaffold goes in a subdirectory instead of the root directory.
  • Desired behavior: A command line argument for dbt init that specifies that the files should be created in the current directory instead of a named directory, or the ability for dbt init to parse . as its argument.
• Naming your project the same name as an existing package results in an error
  • Desired behavior: Providing an existing folder as an argument to dbt init should create the scaffold files in that folder.
• dbt init doesn't look in subdirectories for dbt_project.yml. There may be some existing reason why this is the case, but when combined with the first bullet point, it means that, to my knowledge, i cant run a dbt project without cd-ing into the dbt project folder first (or specifying a command line argument every time... maybe this is configurable? But regardless, not ideal).
  • Desired behavior (complications likely): dbt cli commands should check subfolders for dbt-project.yml, or allow for a config that sets a different file e.g. pyproject.toml as the default root directory.

These are just the 3 most salient complications but overall I would love to see dbt take a less presumptuous approach when it comes to project structure :)

As such, I'm going to attempt to find the most painless way to configure a project like this and maybe start some documentation on it?