Hashboard pull
command
The pull
command is a utility for getting resource configuration files from the web UI to a local workspace. You can use it to migrate resources to a code-based workflow. If you already have local files that correspond with Hashboard resources, hb pull
will update the local versions of those files.
Updating resources and dev workflow
The default behavior of pull
is to only update resources that were created through the Hashboard CLI (i.e. any hb
command).
For example: let's say you originally built four models with the CLI and another two models in the UI. If you run hb pull
from an empty directory, you'll only pull those four model files. You won't get the two UI-built models, or any additional resources built in the UI for that matter.
pull
resources originally built in the UI
You can also pull resources that were not built with the CLI. If you want to pull down a resource that was built in the UI, you have two options:
1. Pull all resources
You can use the --all
flag to get every single Hashboard resource regardless of whether it was created with the CLI.
hb pull --all
2. Pull a specific resource with a GRN
You can pull down a specific resource by its Global Resource Name.
# pull by a specific grn
hb pull m:s3AdinvBhsdJ:
# alternatively, pull by alias
hb pull m::my_sales_model
Config format used in pull
hb pull
uses the most verbose (and specific) possible format of a resource. Some configuration that is just defaulted or configured by Hashboard will be exported.
For example: if you export a resource, it will default to being exported with a pinned
GRN. You can remove the GRN to enable deploying the resource across different projects (or use an alias so that the resource is not bound to a specific instance in a single project).
dbt models and the pull
command
Currently, the Hashboard CLI does not automatically update Hashboard metadata inside of dbt files, so any dbt-based models are skipped when evaluating the pull
command.