Could not find function "%<>%" with dplyr loaded

Question

[Disclaimer: similar questions have been asked many times. I don't believe this is the same as the many threads I've just read.]

I did:

library(dplyr)
colnames(LarvalSamples) %<>% 
  stringr::str_remove_all("_log") %>% 
  stringr::str_replace_all("Sea_Level", "Sea_Level_Height") %>% #sealevel, sealion, chinook, chl
  stringr::str_replace_all("SeaLion", "Sea_lion") %>% 
  stringr::str_replace_all("Chinook_Salmon", "Salmon") %>% 
  stringr::str_replace_all("Chlorophyll_a", "Chlorophyll_A")

Worked fine, no messages, outputs as expected/desired. Then I copy/pasted those first two lines, except the terminal pipe:

colnames(LarvalSamples) %<>%
  stringr::str_remove_all("_log")

Error in colnames(LarvalSamples) %<>% stringr::str_remove_all("_log") : could not find function "%<>%"

I realise there are other posts on here about not finding functions but dplyr is loaded AND worked on MORE code just two lines above. As it happens there are no "_log" patterns in the colnames but I tried a different character pattern which does exist and that failed the same so that's one potential source of error eliminated. Any thoughts/guesses appreciated, this feels like a bug more than a question tbh but would be good to get sharper eyes on it before raising it up the chain, if so required. Thanks.

> sessionInfo()
R version 3.5.0 (2018-04-23)
Platform: x86_64-w64-mingw32/x64 (64-bit)
Running under: Windows 10 x64 (build 17134)

Matrix products: default

locale:
[1] LC_COLLATE=English_United Kingdom.1252  LC_CTYPE=English_United Kingdom.1252   
[3] LC_MONETARY=English_United Kingdom.1252 LC_NUMERIC=C                           
[5] LC_TIME=English_United Kingdom.1252    

attached base packages:
[1] stats     graphics  grDevices utils     datasets  methods   base     

other attached packages:
[1] dplyr_0.8.0.1  beepr_1.3      gbm.auto_1.2.0

loaded via a namespace (and not attached):
 [1] Rcpp_1.0.0       compiler_3.5.0   pillar_1.3.1     shapefiles_0.7   tools_3.5.0      tibble_2.0.1    
 [7] gtable_0.2.0     lattice_0.20-35  pkgconfig_2.0.2  rlang_0.3.1      Matrix_1.2-14    DBI_1.0.0       
[13] rstudioapi_0.9.0 rgdal_1.4-2      gbm_2.1.5        dismo_1.1-4      gridExtra_2.3    stringr_1.4.0   
[19] raster_2.8-19    mapplots_1.5.1   rgeos_0.4-2      grid_3.5.0       tidyselect_0.2.5 glue_1.3.0      
[25] R6_2.4.0         survival_2.41-3  foreign_0.8-70   sp_1.3-1         purrr_0.3.1      magrittr_1.5    
[31] codetools_0.2-15 splines_3.5.0    maptools_0.9-5   assertthat_0.2.0 stringi_1.3.1    crayon_1.3.4    
[37] audio_0.1-5.1  

Update: Reproducible example below. This definitely seems to be a bug. With a completely fresh system:

Data <- data.frame(
    Name_Bad = sample(1:10),
    Name_Guud = sample(1:10)
  )

  colnames(Data) %<>%
    stringr::str_remove_all("_Bad") %>%
    stringr::str_replace_all("Guud", "Good")
  # Error: could not find function "%>%"

  install.packages("dplyr")
  library(dplyr)
  install.packages("stringr")
  library(stringr)

  colnames(Data) %<>%
    stringr::str_remove_all("_Bad") %>%
    stringr::str_replace_all("Guud", "Good")
# no error, worked

  colnames(Data) %<>%
    stringr::str_remove_all("_Bad")
  # Error: could not find function "%<>%"

Answer 1

%<>% isn’t exported by dplyr (only %>% is). You need to load magrittr instead.

Your reproducible example is running into a subtle magrittr bug, which causes the evaluation of pipeline expressions to search for some operators in magrittr’s scope, rather than in the calling scope. That way, x %<>% y %>% z, which evaluates as `%>%`(x %<>% y, z), finds and evaluates magrittr’s `%<>%` operator.