rlang::sym not recognized within label_bquote

[areisett]

I want to create facet labels dynamically. The labels come from a data frame and contain Greek letters. label_bquote alone fails to replace the Greek letter name stored in a variable with the symbol. Incorporating rlang::sym() around the variable names used to fix the issue, but now throws an error. However, this strategy still works when using bquote. The first piece of code below is from a solution to the stackoverflow post How can I make bquote replace the greek letter stored in a variable with the symbol?

### this works
library("tidyverse")

a <- "alpha"
b <- "beta"

ggplot(data.frame(x = c(1), y = c(1)), aes(x, y)) +
  geom_point() +
  labs(x = bquote(.(sym(a))[.(sym(b))])) +
  labs(y = bquote(alpha[beta]))

### this used to work, but now throws an error
library("tidyverse")
df <- data.frame(x = rep(1:5, 2), 
                 y = rnorm(10),
                 facet_var = rep(c("alpha", "beta"), times = c(5, 5)))

df
#>    x           y facet_var
#> 1  1 -1.14089312     alpha
#> 2  2  0.08233709     alpha
#> 3  3 -1.17177333     alpha
#> 4  4  1.00633917     alpha
#> 5  5 -0.01021232     alpha
#> 6  1 -1.48566764      beta
#> 7  2  0.32726416      beta
#> 8  3 -0.72468232      beta
#> 9  4  0.88635594      beta
#> 10 5  2.04821051      beta

ggplot(df, aes(x, y)) +
  geom_point() +
  facet_grid(facet_var ~ .,
             labeller = label_bquote(rows = .(sym(facet_var))))
#> Error in sym(facet_var): could not find function "sym"

### trying to be more explicit about sym doesn't seem to help
ggplot(df, aes(x, y)) +
  geom_point() +
  facet_grid(facet_var ~ .,
             labeller = label_bquote(rows = .(rlang::sym(facet_var))))
#> Error in rlang::sym: could not find function "::"

sessionInfo()
#> R version 4.0.2 (2020-06-22)
#> Platform: x86_64-w64-mingw32/x64 (64-bit)
#> Running under: Windows 10 x64 (build 18362)
#> 
#> Matrix products: default
#> 
#> locale:
#> [1] LC_COLLATE=English_United States.1252 
#> [2] LC_CTYPE=English_United States.1252   
#> [3] LC_MONETARY=English_United States.1252
#> [4] LC_NUMERIC=C                          
#> [5] LC_TIME=English_United States.1252    
#> 
#> attached base packages:
#> [1] stats     graphics  grDevices utils     datasets  methods   base     
#> 
#> other attached packages:
#> [1] forcats_0.5.0   stringr_1.4.0   dplyr_0.8.5     purrr_0.3.4    
#> [5] readr_1.3.1     tidyr_1.1.0     tibble_3.0.1    ggplot2_3.3.0  
#> [9] tidyverse_1.3.0
#> 
#> loaded via a namespace (and not attached):
#>  [1] tidyselect_1.1.0 xfun_0.14        haven_2.3.0      lattice_0.20-41 
#>  [5] colorspace_1.4-1 vctrs_0.3.0      generics_0.0.2   htmltools_0.4.0 
#>  [9] yaml_2.2.1       blob_1.2.1       rlang_0.4.6      pillar_1.4.4    
#> [13] glue_1.4.0       withr_2.2.0      DBI_1.1.0        dbplyr_1.4.4    
#> [17] modelr_0.1.8     readxl_1.3.1     lifecycle_0.2.0  munsell_0.5.0   
#> [21] gtable_0.3.0     cellranger_1.1.0 rvest_0.3.5      evaluate_0.14   
#> [25] labeling_0.3     knitr_1.28       fansi_0.4.1      highr_0.8       
#> [29] broom_0.5.6      Rcpp_1.0.4.6     scales_1.1.1     backports_1.1.6 
#> [33] jsonlite_1.6.1   farver_2.0.3     fs_1.4.1         hms_0.5.3       
#> [37] digest_0.6.25    stringi_1.4.6    grid_4.0.2       cli_2.0.2       
#> [41] tools_4.0.2      magrittr_1.5     crayon_1.3.4     pkgconfig_2.0.3 
#> [45] ellipsis_0.3.0   xml2_1.3.2       reprex_0.3.0     lubridate_1.7.8 
#> [49] assertthat_0.2.1 rmarkdown_2.1    httr_1.4.1       R6_2.4.1        
#> [53] nlme_3.1-148     compiler_4.0.2

^{Created on 2020-07-20 by the reprex package (v0.3.0)}

[yutannihilation]

Thanks, confirmed. The problem is here; since params is supplied, bquote() only searches in params not in the other environments. Probably we should create an environment contains params instead of a list so that bquote() can search in the parent environments.

ggplot2/R/labeller.r

Line 224 in a3019f9

eval(substitute(bquote(expr, params), list(expr = quoted)))

[yutannihilation]

Hmm, sorry. I don't know how to fix this properly atm...

Btw, I found another bug on label_bquote(). Maybe label_bquote() needs a major overhaul (I'm not sure if it's worth).

library(ggplot2)

df <- data.frame(x = 1:2, y = 1:2, f = c("alpha", "beta"))
p <- ggplot(df, aes(x, y)) + facet_wrap(~ f, labeller = label_bquote(.(f)))
p
#> Error in Map(f = function (...) : formal argument "f" matched by multiple actual arguments

[areisett]

Hmm... do you have any suggestions for a work-around?

[yutannihilation]

For this particular example, you can use label_parsed()