Closed m7pr closed 2 weeks ago
1 files 70 suites 1h 5m 38s :stopwatch: 720 tests 610 :white_check_mark: 110 :zzz: 0 :x: 1 983 runs 1 761 :white_check_mark: 222 :zzz: 0 :x:
Results for commit e9c52aac.
:recycle: This comment has been updated with latest results.
Results for commit f67d6885795b70ca9b46c417189893ed91a8fb8c
♻️ This comment has been updated with latest results.
@pawelru if logger
is able to guess the namespace inside a package, maybe we do not need to specify namespace = "teal.modules.clinical"
in every call of log_shiny_input_changes
namespace
string referring to the logger environment / config to be used to override the target of the message record to be used instead of the default namespace, which is defined by the R package name from which the logger was called, and falls back to a common, global namespace.
In this case namespace = "teal.modules.clinical"
is needed, becuase if we omit it, it falls back to namespace = "teal.logger"
: )
This I don't know. Definitely worth to check
I'm experiencing very funny issue in R CMD CHECK failures
* checking dependencies in R code ... WARNING
'::' or ':::' import not declared from: ‘jsonlite’
It's the jsonlite
package that is not included in tmc
but is used in
log_shiny_input_changes <- getFromNamespace("log_shiny_input_changes", "teal.logger")
Maybe we just need to export log_shiny_input_changes
in teal.logger
and regularly call teal.logger::log_shiny_input_changes
Part of https://github.com/insightsengineering/coredev-tasks/issues/553 and https://github.com/insightsengineering/coredev-tasks/issues/568