ggfortify fails with dev scales

sinhrks / ggfortify

Define fortify and autoplot functions to allow ggplot2 to handle some popular R packages.

Other

525 stars 65 forks source link

ggfortify fails with dev scales #197

Closed hadley closed 4 years ago

hadley commented 4 years ago

With test failures like:

ld$colour not equal to c("#F8766D", "#00BFC4", "#F8766D", "#F8766D").
4/4 mismatches
x[1]: "#FF0076"
y[1]: "#F8766D"

x[2]: "#00C3CA"
y[2]: "#00BFC4"

x[3]: "#FF0076"
y[3]: "#F8766D"

x[4]: "#FF0076"
y[4]: "#F8766D"

I suspect this is due to:

col_factor() now tries to avoid interpolating qualitative RColorBrewer palettes. Instead, it attempts to assign a palette color to each factor level. Interpolation will still be used if there are more factor levels than available colors, and a warning will be emitted in that case (@jcheng5, #191).

terrytangyuan commented 4 years ago

@hadley Thanks for informing! I think we could relax the tests a bit here.

hadley commented 4 years ago

Having investigated this a bit more, I now think it's due to the switch to farver in hue_pal(). I'm going to try and get to the bottom of it today, but it's possible we'll roll back to the previous implementation in which case you probably won't need to do anything.

terrytangyuan commented 4 years ago

@hadley Thanks. Please keep me posted then.

hadley commented 4 years ago

Yup, looks like this is fixed on my end 😄

terrytangyuan commented 4 years ago

@hadley I am seeing some mismatches in color here https://cran.r-project.org/web/checks/check_results_ggfortify.html that needs to be corrected by 2020-03-20. For example:

     ── 6. Failure: autoplot ts works for multivariate timeseries (@test-ts.R#150) ─
     ld$fill not equal to rep(c("#00BFC4", "#F8766D"), 4).
     4/8 mismatches
     x[1]: "#F8766D"
     y[1]: "#00BFC4"

     x[3]: "#F8766D"
     y[3]: "#00BFC4"

     x[6]: "#00BFC4"
     y[6]: "#F8766D"

     x[8]: "#00BFC4"
     y[8]: "#F8766D"

Is that related to this issue?