library calls possible within a call

[gogonzo]

Possible to include library call in eval_code

#93

[github-actions]

Code Coverage Summary

Filename                     Stmts    Miss  Cover    Missing
-------------------------  -------  ------  -------  ------------------------
R/include_css_js.R               7       7  0.00%    12-20
R/qenv-concat.R                 10       0  100.00%
R/qenv-constructor.R            12       0  100.00%
R/qenv-eval_code.R              52       2  96.15%   100, 109
R/qenv-get_code.R               18       0  100.00%
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%    16
R/qenv-within.R                  7       0  100.00%
R/utils-code_dependency.R      190       7  96.32%   37, 42, 252-253, 317-320
R/utils.R                       12       0  100.00%
TOTAL                          398      17  95.73%

Diff against main

Filename              Stmts    Miss  Cover
------------------  -------  ------  -------
R/qenv-eval_code.R       +3       0  +0.24%
TOTAL                    +3       0  +0.03%

Results for commit: 5ea82b4

Minimum allowed coverage is 80%

♻️ This comment has been updated with latest results

[chlebowa]

What if someone detaches a package that is in the middle of the search path (for whatever reason)?

[gogonzo]

What if someone detaches a package that si in the middle of the search path (for whatever reason)?

as far as one uses eval_code to do so it's okey from reproducibility perspective

[github-actions]

Unit Tests Summary

    1 files    10 suites   1s ⏱️
  92 tests   92 ✔️ 0 💤 0 ❌
185 runs  185 ✔️ 0 💤 0 ❌

Results for commit 5ea82b4.

[chlebowa]

But you can then lose a function from scope and the condition won't be triggered. I can't decide if that's going to be a problem 🤔

[gogonzo]

But you can then lose a function from scope and the condition won't be triggered. I can't decide if that's going to be a problem 🤔

Removing namespace and then trying to obtain object from this namespace is a bug introduced by user. In the interactive session this would fail as well. It doesn't seem to be a qenv problem.

---

Trick I proposed also prevent us from the problem of detached parent. Parent will be reset again

[chlebowa]

Sounds good 👍

[chlebowa]

We haven't considered that using this feature forces us to call teal.data (in practice probably teal) before packages required for pre-processing.
Instead of

  library(scda)
  library(scda.2022)
  library(dplyr)
  library(teal.modules.clinical)
  library(teal.modules.general)
  library(teal.osprey)
  library(nestcolor)
  # optional libraries
  library(sparkline)

one would do

  library(teal)
  library(scda)
  library(scda.2022)
  library(dplyr)
  library(teal.modules.clinical)
  library(teal.modules.general)
  library(teal.osprey)
  library(nestcolor)
  # optional libraries
  library(sparkline)

This leaves us with a different search path:

                             old                           new
1                     .GlobalEnv                    .GlobalEnv
2              package:sparkline             package:sparkline
3              package:nestcolor             package:nestcolor
4            package:teal.osprey           package:teal.osprey
5                 package:osprey                package:osprey
6   package:teal.modules.general  package:teal.modules.general
7              package:shinyTree             package:shinyTree
8               package:ggmosaic              package:ggmosaic
9                package:ggplot2               package:ggplot2
10 package:teal.modules.clinical package:teal.modules.clinical
11                  package:tern                  package:tern
12               package:rtables               package:rtables
13            package:formatters            package:formatters
14                  package:teal                 package:dplyr
15        package:teal.transform             package:scda.2022
16              package:magrittr                  package:scda
17            package:teal.slice                  package:teal
18             package:teal.data        package:teal.transform
19                 package:shiny              package:magrittr
20                 package:dplyr            package:teal.slice
21             package:scda.2022             package:teal.data
22                  package:scda                 package:shiny
23                 tools:rstudio                 tools:rstudio
24                 package:stats                 package:stats
25              package:graphics              package:graphics
26             package:grDevices             package:grDevices
27                 package:utils                 package:utils
28              package:datasets              package:datasets
29               package:methods               package:methods
30                     Autoloads                     Autoloads
31                  package:base                  package:base

I don't think this should be a problem but it is something we should be aware of.