Rename group/group_by arguments into by

[strengejacke]

easystats/easystats#404

[strengejacke]

hm, this error seems to be new, but unrelated to this PR. Why does this encoding-error suddenly occur?

https://github.com/easystats/datawizard/actions/runs/9109670481/job/25043053199?pr=502

[etiennebacher]

It might be due to an xfun update: yihui/xfun#85

[strengejacke]

We have some aliases, we should think about when to remove them?

#' @rdname text_format
#' @export
format_text <- text_format

#' @rdname recode_values
#' @export
change_code <- recode_values

[etiennebacher]

We have some aliases, we should think about when to remove them?

Yes, we could also address #265 at the same time

[etiennebacher]

Sorry @strengejacke I meant that we could address the text aliases you mentioned and the aliases mentioned in #265 at the same time but in another PR so that we keep this one strictly about renaming args. Sorry for the confusion 😬

[strengejacke]

I hope we don't need to add @rdname tags to the deprecated functions, but I fear, it must be... Let's see what the checks say.

[strengejacke]

@etiennebacher Why do we have this error for the selection-syntax vignette now?
https://github.com/easystats/datawizard/actions/runs/9127735513/job/25098644747?pr=502

[etiennebacher]

Sorry @strengejacke I meant that we could address the text aliases you mentioned and the aliases mentioned in #265 at the same time but in another PR so that we keep this one strictly about renaming args. Sorry for the confusion 😬

@strengejacke have you seen this?

[strengejacke]

Sorry @strengejacke I meant that we could address the text aliases you mentioned and the aliases mentioned in #265 at the same time but in another PR so that we keep this one strictly about renaming args. Sorry for the confusion 😬

Yeah, there are quite some files that have been changed... 🤣 But I think it's ok.

[strengejacke]

Sorry @strengejacke I meant that we could address the text aliases you mentioned and the aliases mentioned in #265 at the same time but in another PR so that we keep this one strictly about renaming args. Sorry for the confusion 😬

@strengejacke have you seen this?

I'm too unfamiliar with git to know how to split commits into separate PRs... 😬

[etiennebacher]

I'd really like to keep those separated so that we can point to a clear PR if necessary later, I can try to split this into two PRs a bit later

[strengejacke]

Actually, this PR can be merged. I'm not sure it's worth the effort splitting this PR (I just saw, in VSCode, I can add a new branch from the commit where I changed the aliases and then reset this PR to the commit before - should I do this? - but I don't think there's much benefit)

The only remaining mystery is the issue in the vignette:

Error in `map()`:
ℹ In index: 1.
ℹ With name: vignettes/selection_syntax.Rmd.
Caused by error:
! (converted from warning) No column names that matched the required search pattern were found.
Backtrace:
 1. my_function(iris, starts_with("Sep"))
 2. datawizard::extract_column_names(data, select = selection)
 3. insight::format_warning("No column names that matched the required search pattern were found.")
 4. insight::format_alert(..., type = "warning")

Quitting from lines 193-203 [unnamed-chunk-14] (selection_syntax.Rmd)

Not sure why this should suddenly fail.

[etiennebacher]

I'm not sure it's worth the effort splitting this PR (I just saw, in VSCode, I can add a new branch from the commit where I changed the aliases and then reset this PR to the commit before - should I do this?

If you talk about #503 I think it also includes the changes for the group_by args so merging it first wouldn't change anything I think

The only remaining mystery is the issue in the vignette:

I can reproduce this warning locally with this branch but not with main so it probably comes from the alias removing.

I will try to split this PR this weekend if it's fine for you, or are you in a hurry for this one?

[strengejacke]

No, no hurry.

[etiennebacher]

@strengejacke I fixed the error in the vignette here: b8ac8ea. FYI those instructions were helpful to know how to move the latest N commits to a new branch (still took some manual tweaking though)

pkgdown workflows still fail but it's something different on which we don't have much control I think

[etiennebacher]

We have three different test failures but all the rest is green (??):

• ubuntu-latest 3.6: will go away when we bump version to >= 4.0 anyway
• windows-latest 4.1: I think due to xfun but still super weird because only happens in this case
• ubuntu-latest 4.3: function 'chm_factor_ldetL2' not provided by package 'Matrix', I don't think we can do much on that

=> I'm merging