get reordering of columns on reference

[ColeMiller1]

Closes #4089.

Current behavior will re-order columns when using (m)get even when .SDcols is present. This can be dangerous when updating in place:

library(data.table)
dt <- data.table(x = 1, y = 2, z = 3)
cols <- c("y","x")

dt[, (cols) := lapply(.SD[get("x") == 1],function(x){x}), .SDcols = cols ,by = z][]
       x     y     z
   <num> <num> <num>
1:     2     1     3

Proposed will not re-order the columns when .SDcols argument is present.

dt[, (cols) := lapply(.SD[get("x") == 1],function(x){x}), .SDcols = cols ,by = z][]
       x     y     z
   <num> <num> <num>
1:     1     2     3

Also, some code was re-arranged to address a TODO comment related to this section of [.data.table.

[codecov]

Codecov Report

Merging #4212 into master will increase coverage by <.01%.
The diff coverage is 100%.

@@            Coverage Diff             @@
##           master    #4212      +/-   ##
==========================================
+ Coverage   99.61%   99.61%   +<.01%     
==========================================
  Files          72       72              
  Lines       13874    13904      +30     
==========================================
+ Hits        13820    13850      +30     
  Misses         54       54

Impacted Files	Coverage Δ
R/merge.R	100% <ø> (ø)	⬆️
src/subset.c	100% <ø> (ø)	⬆️
R/as.data.table.R	100% <ø> (ø)	⬆️
src/assign.c	100% <ø> (ø)	⬆️
R/xts.R	100% <100%> (ø)	⬆️
src/fifelse.c	100% <100%> (ø)	⬆️
src/fwriteR.c	100% <100%> (ø)	⬆️
src/chmatch.c	100% <100%> (ø)	⬆️
src/dogroups.c	100% <100%> (ø)	⬆️
R/setkey.R	100% <100%> (ø)	⬆️
... and 13 more

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 4d11fe0...bb5adf5. Read the comment docs.

[jangorecki]

Very nice!
Could we make this warning avoidable? see #3848

[ColeMiller1]

Thank you both for the great comments. Main changes

1. Removed the warning
2. Updated to cat(gettextf("New ansvar ...", brackify(ansvars), domain = "R-data.table")
3. Re-organized the comments and cat statements to better match the order of what I deleted

[MichaelChirico]

Looks great!

Can you add a test of the new behavior?

[mattdowle]

Quite an involved first time contribution! That part of the code is hard work, thank you. I've invited you to be project member; please accept invitation that shows in your github profile or projects tab. I'll add you to contributor list in a follow up commit.