Replace table by plot on tm_missing_data

[llrs-roche]

Pull Request

Fixes #495

Replaces the table by a plot on "By Variables Levels" tab.
The colors match those in the other plots of the module.
It also required to add a "new" dependency, formatters, which is a dependency of tern, rlistings, rtable. I added it on Suggests.

There is an extra commit to update the lintr names

data <- within(teal_data(), {
  ADSL <- rADSL
  ADRS <- rADRS
})
join_keys(data) <- default_cdisc_join_keys[names(data)]

init(
  data = data,
  modules = modules(tm_missing_data())
) |> runApp()

[github-actions]

Code Coverage Summary

Filename                      Stmts    Miss  Cover    Missing
--------------------------  -------  ------  -------  -------------------------------------------
R/tm_a_pca.R                    876     876  0.00%    141-1147
R/tm_a_regression.R             762     762  0.00%    180-1041
R/tm_data_table.R               200     200  0.00%    100-348
R/tm_file_viewer.R              172     172  0.00%    47-254
R/tm_front_page.R               143     132  7.69%    77-246
R/tm_g_association.R            331     331  0.00%    161-561
R/tm_g_bivariate.R              681     417  38.77%   332-801, 842, 953, 970, 988, 999-1021
R/tm_g_distribution.R          1116    1116  0.00%    154-1414
R/tm_g_response.R               356     356  0.00%    179-608
R/tm_g_scatterplot.R            720     720  0.00%    261-1079
R/tm_g_scatterplotmatrix.R      283     264  6.71%    200-515, 576, 590
R/tm_missing_data.R            1167    1167  0.00%    126-1474
R/tm_outliers.R                1032    1032  0.00%    163-1348
R/tm_t_crosstable.R             273     273  0.00%    177-494
R/tm_variable_browser.R         788     783  0.63%    89-1025, 1063-1246
R/utils.R                       185     120  35.14%   87-249, 278-304, 316-325, 330, 344-363, 452
R/zzz.R                           2       2  0.00%    2-3
TOTAL                          9087    8723  4.01%

Diff against main

Filename               Stmts    Miss  Cover
-------------------  -------  ------  --------
R/tm_g_bivariate.R        -4      -4  +0.23%
R/tm_missing_data.R      +68     +68  +100.00%
TOTAL                    +64     +64  -0.03%

Results for commit: c296936

Minimum allowed coverage is 80%

♻️ This comment has been updated with latest results

[github-actions]

Unit Tests Summary

  1 files  23 suites   2s ⏱️
151 tests 36 ✅ 115 💤 0 ❌
194 runs  79 ✅ 115 💤 0 ❌

Results for commit c296936.

♻️ This comment has been updated with latest results.

[averissimo]

We don't need the quotes for all of these :-)

Suggested change

	teal.code::eval_code(obj, 'library("ggplot2");library("dplyr");library("tidyr")') # nolint: quotes_lintr
	teal.code::eval_code(obj, "library(ggplot2);library(dplyr);library(tidyr)")

[llrs-roche]

Do you want to update all the library calls of modules on this package? I think in the past we changed this and then we had to revert it back.

[averissimo]

Except here,

Can you break this into 2 eval_code?

Why? When concatennating "character" with "expression" (as here) it transforms it in an expression and the code that's run is a "string", not the actual contents.

I detected this while testing something else.

My suggestion:

      obj %>%
        teal.code::eval_code("library(ggplot2);library(dplyr)") %>% # nolint: quotes)
        teal.code::eval_code(
          c(
            as.expression(anl_merged_input()$expr)
          )
        )

[llrs-roche]

Well spotted! I saw that the code generated was:

require(nestcolor)
CO2 <- data.frame(CO2)
stopifnot(rlang::hash(CO2) == "09a462eb86fab9a952e092bf6253c333") # @linksto CO2
.raw_data <- list2env(list(CO2 = CO2))
lockEnvironment(.raw_data) # @linksto .raw_data
"library(\"ggplot2\");library(\"dplyr\")"
ANL_1 <- CO2 %>% dplyr::select(conc, uptake, Type, Treatment)
ANL <- ANL_1

With this modification we get:

require(nestcolor)
CO2 <- data.frame(CO2)
stopifnot(rlang::hash(CO2) == "09a462eb86fab9a952e092bf6253c333") # @linksto CO2
.raw_data <- list2env(list(CO2 = CO2))
lockEnvironment(.raw_data) # @linksto .raw_data
library("ggplot2")
library("dplyr")
ANL_1 <- CO2 %>% dplyr::select(conc, uptake, Type, Treatment)
ANL <- ANL_1

[averissimo]

As discussed, I don't think this is correct and the period is necessary

Note that on the first screenshot, I removed the period and changed the linter name and it passes.

On the second it fails

[llrs-roche]

Yes, this puzzles me. But note that if you search for nolint on the package there are some that end with a dot and some that don't. I also tried to create a reprex and I couldn't ...