pathGrob usage failure

[paleolimbot]

This is a revdep issue (#3303) in which there is an installation warning about pathGrob() not having a pathId argument:

possible error in 'pathGrob(munched$x, munched$y, ': unused argument (pathId = munched$group)

This affects ggalt, ggforce, ggpol, ggpolypath, ggsolvencyii, and ggspatial. I'm guessing this has to do with #3128 (this is why ggspatial is using pathGrob...to put holes in polygons), but I don't know why.

[thomasp85]

I’ll take this once I get on top — this is my code all the way down

[thomasp85]

ok, so this is the same story as I had with sf, back when I added the new pathGrob capabilities... for weird reasons ggplot2 does not itself trigger this build warning, but packages subclassing polygonGrob will...

The fix is to mask pathGrob so it doesn't get called directly (this is super stupid as the current code is correct, but the static analyser is throwing a fit)... copying the sf approach should fix it

https://github.com/r-spatial/sf/blob/b860ad64ce3f7d1f5fc787048237fb952b226325/R/init.R#L31-L39

I hope someone else can take this up as my head is still not happy about prolonged computer use (reading is better than actively working)

[paleolimbot]

Just clarifying...this is something we don't have to fix in ggplot2, we just have to notify the downstream package developers? If so, I'd be happy to do this.

[thomasp85]

We should do this in ggplot2. ggforce doesn’t call pathGrob() directly, only through inheritance of GeomPolygon so it is something we should fix here by masking grid::pathGrob as I did for sf

[lock]

This old issue has been automatically locked. If you believe you have found a related problem, please file a new issue (with reprex) and link to this issue. https://reprex.tidyverse.org/