added option --adapter to dbt init, to create sample profiles.yml bas…

[brunomurino]

resolves #2533

Description

Added option '--adapter' to dbt init, to create sample profiles.yml based on chosen adapter.

It gets the sample profiles.yml from the adapter 'includes' folder

Related PRs:

• Added sample_profiles.yml so that adapter works with new --adapter option of dbt init dbt-presto#26
• Added sample_profiles.yml so that adapter works with new --adapter option of dbt init dbt-spark#98

Checklist

• I have signed the CLA
• I have run this code in development and it appears to resolve the stated issue
• This PR includes tests, or tests are not required/relevant for this PR
• I have updated the CHANGELOG.md and added information about my change to the "dbt next" section.

[brunomurino]

I think the test is failing because the container "fishtownanalytics/test-container:7" doesn't have the samples_profiles.yml on the postgres folder. How can I change that?

[beckjake]

I think the test is failing because the container "fishtownanalytics/test-container:7" doesn't have the samples_profiles.yml on the postgres folder. How can I change that?

I might be wrong about this, but I think the issue is that the setup.py file for each plugin needs to add the path to the sample_profiles.yml to package_data. This gets me basically every time I add a static file to dbt.

[brunomurino]

@beckjake Thanks for the heads up! It all worked now. Ready to be reviewed, I think!

[beckjake]

@brunomurino I (finally) got the branch this depends upon merged into dev/marian-anderson. Could you change the target branch for this PR to dev/marian-anderson please?

[jtcohen6]

@brunomurino This is a great change! I have a few comments around keeping the sample profiles consistent between the different adapters.

• The top-level profile name should be default for all of them, so that it matches the profile name specified in the starter project and just works from the get-go. I dislike that the starter project has profile: 'default' and name: 'my_new_project', but I think we would need to write a bunch more code to make that dynamic.
• Let's include bracketed entries (password: [password]) instead of fake entries (password: pa55word)
• Let's always include two outputs, dev and prod. I like how it demonstrates the options and extensibility available, especially for plugins that have multiple connection methods.
  • Postgres: dev and prod, with different usernames ([dev_username] and [prod_username]), passwords, and schemas
  • Redshift: dev (password-based), prod (iam)
  • Snowflake: dev (user-password), prod (keypair)
  • BigQuery: dev (oauth), prod (service_account) on BigQuery

I'm curious to hear what you think about the above. I'm flexible on the specifics, but definitely set on wanting consistency here and for the plugins as well (dbt-labs/dbt-spark#98 + dbt-labs/dbt-presto#26)

[brunomurino]

@brunomurino This is a great change! I have a few comments around keeping the sample profiles consistent between the different adapters.

• The top-level profile name should be default for all of them, so that it matches the profile name specified in the starter project and just works from the get-go. I dislike that the starter project has profile: 'default' and name: 'my_new_project', but I think we would need to write a bunch more code to make that dynamic.

• Let's include bracketed entries (password: [password]) instead of fake entries (password: pa55word)

• Let's always include two outputs, dev and prod. I like how it demonstrates the options and extensibility available, especially for plugins that have multiple connection methods.

  • Postgres: dev and prod, with different usernames ([dev_username] and [prod_username]), passwords, and schemas
  • Redshift: dev (password-based), prod (iam)
  • Snowflake: dev (user-password), prod (keypair)
  • BigQuery: dev (oauth), prod (service_account) on BigQuery

I'm curious to hear what you think about the above. I'm flexible on the specifics, but definitely set on wanting consistency here and for the plugins as well (fishtown-analytics/dbt-spark#98 + fishtown-analytics/dbt-presto#26)

I agree and have updated the sample profiles!

[brunomurino]

@jtcohen6 Updated the sample profiles on spark and presto plugins as well, let me me know what do you think.

[jtcohen6]

Hey @brunomurino, sorry for the delay! Thanks for humoring me on making the profiles more consistent + thorough. I left two quick comments on the redshift + snowflake profiles. Then we can merge this (+ spark + presto) for 0.18.0 :)

[brunomurino]

@jtcohen6 No worries at all and sorry for the delay. Feel free to give any more suggestions!

[jtcohen6]

Thanks @brunomurino! Very last ones, I promise.

[brunomurino]

@jtcohen6 Happy to help! I also checked the presto-spark repos and they're already in line with your suggestions on this repo!

[jtcohen6]

Looks good! I found the tiniest last change, I'm just going to add as suggestions. Approve at your convenience and I'll merge the PR. Thanks for your patience and for your contribution!

[jtcohen6]

Approved once suggestions are committed