picks

[gogonzo]

Related issues in the milestone:

• https://github.com/insightsengineering/teal.transform/milestone/1

See related PRs:

• picks teal.transform#270
• picks teal#1642
• picks teal.modules.general#942

Installation

remotes::install_github("insightsengineering/teal.modules.general@redesign_extraction@main")

[github-actions]

✅ All contributors have signed the CLA
_{Posted by the CLA Assistant Lite bot.}

[gogonzo]

I have read the CLA Document and I hereby sign the CLA

[llrs-roche]

It looks like we get default columns with 70 less lines of code. Nice.

Why do most picks() on the app have values()? I understand why we want to specify a value for filtering on a teal_transform_filter() or why tm_g_distribution() some parameters use values() while others don't: those with values is to specify a given value as reference, response or for plot customization.
But on other modules, like tm_g_scatterplot() or tm_t_crosstable(), values() is used even for x and y where I don't expect specific values to be needed:

From the description of picks on insightsengineering/teal.transform#270 it is not apparent that we need to specify it for merging and keeping those columns.
Answer: The values() is added to be able to filter dynamically on each variable. It is also reset if the selection changes.

[llrs-roche]

Here we have list(picks()) instead of directly picks(). What does this add?

[gogonzo]

Here we have list(picks()) instead of directly picks(). What does this add?

scatterplotmatrix have unknown number of "arguments" (variables), having a list here allows to specify variables from multiple datasets.

---

those with values is to specify a given value as reference, response or for plot customization.

Yup, filter on an dimension-variable (x, y, color, size etc) is set by default in the modules. I think it is a nice default when app-user can filter axis after selecting a variable. On friday's meeting we discovered that picks-arguments which allow multiple-variables, it is wrong to set a values()

From the description of picks on insightsengineering/teal.transform#270 it is not apparent that we need to specify it for merging and keeping those columns

Yup, I should link merge_module in the picks documentation