165 trim trace for qenv.error@main

[m7pr]

Close #165

[github-actions]

Unit Tests Summary

    1 files      9 suites   0s ⏱️
  62 tests   62 ✔️ 0 💤 0 ❌
134 runs  134 ✔️ 0 💤 0 ❌

Results for commit 71116d7.

♻️ This comment has been updated with latest results.

[github-actions]

Code Coverage Summary

Filename                 Stmts    Miss  Cover    Missing
---------------------  -------  ------  -------  ---------
R/qenv-concat.R             10       0  100.00%
R/qenv-constructor.R        16      13  18.75%   54-93
R/qenv-eval_code.R          52       2  96.15%   100, 109
R/qenv-get_code.R           20       1  95.00%   38
R/qenv-get_var.R            19       0  100.00%
R/qenv-get_warnings.R       24       0  100.00%
R/qenv-join.R               46       0  100.00%
R/qenv-show.R                1       1  0.00%    19
R/qenv-within.R              8       0  100.00%
R/utils.R                   10       0  100.00%
TOTAL                      206      17  91.75%

Diff against main

Filename      Stmts    Miss  Cover
----------  -------  ------  --------
TOTAL             0       0  +100.00%

Results for commit: 71116d7

Minimum allowed coverage is 80%

♻️ This comment has been updated with latest results

[chlebowa]

> devtools::load_all()
ℹ Loading teal.code
> qe <- qenv() |> within(i <- iris) |> within(stop("oh no"))
> qe
<qenv.error: oh no 
 when evaluating qenv code:
stop("oh no")>
> get_code(qe)
Error: oh no 
 when evaluating qenv code:
stop("oh no")

trace: 
 i <- iris
 stop("oh no")

This is what I intended 👍

As long as you're here, would you mind modifying this line in get_code like it was changed in teal_data::get_code?

# from
parse(text = paste(c("{", object@code, "}"), collapse = "\n"), keep.source = TRUE)
# to
parse(text = object@code, keep.source = TRUE)

I disabled auto-merge to give you chance.

[m7pr]

Hey @chlebowa why would we want to change this?

The line

parse(text = paste(c("{", object@code, "}"), collapse = "\n"), keep.source = TRUE)

was introduced in here #176 so that resulted expression (for deparse = FALSE) has a length 1, so it's consistent with the length of the result for deparse = TRUE #176 (comment)

[chlebowa]

before:

> devtools::load_all()
ℹ Loading teal.code
> qe <- qenv() |> within(i <- iris) |> within(m <- mtcars)
> get_code(qe, deparse = TRUE)
[1] "i <- iris\nm <- mtcars"
>

after:

> devtools::load_all()
ℹ Loading teal.code
> qe <- qenv() |> within(i <- iris) |> within(m <- mtcars)
> get_code(qe, deparse = TRUE)
[1] "i <- iris\nm <- mtcars"
> 

[m7pr]

Hey @chlebowa your before and after are the same. The line you mentioned is for deparse = FALSE

# parse(text = object@code, keep.source = TRUE) case
qe <- qenv() |> within(i <- iris) |> within(m <- mtcars)
get_code(qe, deparse = FALSE)
# > expression(i <- iris, m <- mtcars)
length(get_code(qe, deparse = FALSE))
# > [1] 2

# parse(text = paste(c("{", object@code, "}"), collapse = "\n"), keep.source = TRUE) case
qe <- qenv() |> within(i <- iris) |> within(m <- mtcars)
get_code(qe, deparse = FALSE)
# > expression({
# > i <- iris
# > m <- mtcars
# > })
length(get_code(qe, deparse = FALSE))
# > [1] 1

[m7pr]

If we are trying to be consistent with what's in https://github.com/insightsengineering/teal.data/blob/main/R/teal_data-get_code.R#L60 then I would actually modify what's in teal.data::get_code.teal_data so it returns output of length 1 for each deparse = FALSE

[m7pr]

get_code in teal_data is inconsistent in terms of the length of the output if you want to have a deparsed or non-deparsed output

library(teal.data)
tdata1 <- teal_data()
tdata1 <- within(tdata1, {
  a <- 1
  b <- a^5
  c <- list(x = 2)
})

length(get_code(tdata1, deparse = FALSE))
# > [1] 3
length(get_code(tdata1, deparse = TRUE))
# > [1] 1

[chlebowa]

I stand corrected, I ran my tests wrong.

Merge away.

[m7pr]

No worries, I made an issue in teal.code to follow-up on this one insightsengineering/teal.data#224