You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
We have been preparing a new release of ggplot2 and during a reverse dependency check, it became apparent that the prospective ggplot2 3.5.0 would break some functionality in constructive.
I wasn't able to thoroughly grasp what is the root cause of the breakage, but it would appear as if a scale's call field is (at least partially) responsible. As far is I understood the code, constructive is conconstituting a scale from its call. However, we have changed that field to more accurately reflect the user-level call, rather than the discrete_scale()/continuous_scale() intermediate plumbing. This helps ggplot2 to throw more informative errors and warnings, but unfortunately appears to depart from some assumptions in constructive.
library(ggplot2)
sc<- scale_colour_manual(values= c("blue", "green"))
sc$call#> scale_colour_manual(values = c("blue", "green"))constructive::construct(sc)
#> Error in `constructive::construct()`:#> ! {constructive} could not build the requested code.#> Caused by error in `endsWith()`:#> ! non-character object(s)sc<- xlim(c(10, 20))
sc$call#> xlim(c(10, 20))constructive::construct(sc)
#> Error in `constructive::construct()`:#> ! {constructive} could not build the requested code.#> Caused by error in `names(args_are_defaults)[args_are_defaults]`:#> ! invalid subscript type 'list'
The release of ggplot2 3.5.0 is scheduled for the 12th of Februari. The progress of the release can be tracked in tidyverse/ggplot2#5588. We hope that this issue serves as a timely heads up.
The text was updated successfully, but these errors were encountered:
This old thread has been automatically locked. If you think you have found something related to this, please open a new issue and link to this old issue if necessary.
Hi there!
We have been preparing a new release of ggplot2 and during a reverse dependency check, it became apparent that the prospective ggplot2 3.5.0 would break some functionality in constructive.
I wasn't able to thoroughly grasp what is the root cause of the breakage, but it would appear as if a scale's
call
field is (at least partially) responsible. As far is I understood the code, constructive is conconstituting a scale from its call. However, we have changed that field to more accurately reflect the user-level call, rather than thediscrete_scale()
/continuous_scale()
intermediate plumbing. This helps ggplot2 to throw more informative errors and warnings, but unfortunately appears to depart from some assumptions in constructive.Created on 2024-01-16 with reprex v2.0.2
To test with the release candidate, you can install it with the code below:
The release of ggplot2 3.5.0 is scheduled for the 12th of Februari. The progress of the release can be tracked in tidyverse/ggplot2#5588. We hope that this issue serves as a timely heads up.
The text was updated successfully, but these errors were encountered: