dplyr 'lag' does not respect its generic nature

[njcray]

'lag' is a generic function in R. Some packages extend it - for example xts does this.

So this works:

library(xts)
x=xts(1:32,order.by=seq(from=as.Date("2014-01-01"),to=as.Date("2014-02-01"),by="days"))
x
lag(x,k=4)
lag(x,k=4,na.pad=FALSE)
lag(x,4)

But if we now attach the dplyr library it does not extend the lag generic function. It replaces it with an explicit function which breaks xts functionality.

library(dplyr)
lag(x,k=4)
# Error in lag(x, k = 4) : unused argument (k = 4)
lag(x,k=4,na.pad=FALSE)
# Error in lag(x, k = 4, na.pad = FALSE) : 
#  unused arguments (k = 4, na.pad = FALSE)
lag(x,4)
# [1] NA NA NA NA  1  2  3  4  5  6  7  8  9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28
# (loses xts attributes)

> sessionInfo()
R version 3.0.2 (2013-09-25)
Platform: x86_64-w64-mingw32/x64 (64-bit)

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

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

other attached packages:
[1] dplyr_0.1.1 xts_0.9-7   zoo_1.7-10 

loaded via a namespace (and not attached):
[1] assertthat_0.1  grid_3.0.2      lattice_0.20-24 Rcpp_0.11.0     tools_3.0.2    

[Fluxionist]

I am facing exactly the same issue. Though I can write xts::lag() explicitly in my own code, there are tons of other packages already using "lag()" and it is hard to avoid them. What is the necessity to mask the default behavior of stats::lag and filter instead of choosing some other names for those functions?

Even if you ignore xts users, there are still lots of codes using stats::lag(x, k=xxx) which cause an error when dplyr is loaded.

[bwlewis]

I agree, and I think 'filter' is particularly inappropriate to use, since the available R 'Filter' function is a closer match to what dplyr is doing.

[hadley]

Agreed, I just haven't had the time to make the fix. Pull requests welcomed.