fit.workflow() errors uninformatively if mode is required but not set

[EmilHvitfeldt]

library(tidymodels)

bag_spec <- decision_tree() %>%
  set_engine("C5.0")

workflow(mpg ~ ., bag_spec) %>% 
  fit(mtcars)
#> Error in `pull_workflow_spec_encoding_tbl()`:
#> ! Exactly 1 model/engine/mode combination must be located.
#> ℹ This is an internal error in the workflows package, please report it to the package authors.

[DavisVaughan]

pull_workflow_spec_encoding_tbl() is used by finalize_workflow(), which is only called from .fit_pre(). I think .fit_pre() could have another check after validate_has_model() called validate_model_has_mode(workflow) that checks that the model has set the mode to something other than "unknown".

We'd also update that comment to mention that the mode has to be finalized before finalize_workflow() is called too

workflows/R/fit.R

Lines 128 to 132 in 822f144

	# A model spec is required to ensure that we can always
	# finalize the blueprint, no matter the preprocessor
	validate_has_model(workflow)
	workflow <- finalize_blueprint(workflow)

[DavisVaughan]

With #164, the error happens as soon as you try to add the model spec with an unknown mode to the workflow

library(tidymodels)

bag_spec <- decision_tree() %>%
  set_engine("C5.0")

workflow(mpg ~ ., bag_spec)
#> Error in `add_model()` at workflows/R/workflow.R:71:4:
#> ! `spec` must have a known mode.
#> ℹ Set the mode of `spec` by using `parsnip::set_mode()` or by setting the mode directly in the parsnip specification function.

^{Created on 2022-08-08 by the reprex package (v2.0.1)}

[github-actions]

This issue has been automatically locked. If you believe you have found a related problem, please file a new issue (with a reprex: https://reprex.tidyverse.org) and link to this issue.