IMPORTANT

search() path
library(A)
a
library(B)  # B provides a. So a now changes.

Finding functions versus variables/symbols.
find()

attach() - NOT UNDER ANY CIRCUMSTANCES (unless *never* change)

Define - parameters/formal arguments,  arguments, variables, symbols
Later:  Call frames and navigating the call stacks and evaluating in the parent frame.
F != FALSE .   Can change T but not TRUE

+ Everything is a function call
 1 + 2
 e = quote(1 + 2)
 typeof(e)
 e[[1]]
 e[[2]]
 e[[3]]


+ Difference between myList$var and myList[[ "var" ]] and myList [[ var ]] 
x = list(abc = 1, axy = 2)
x$a
NULL
x$ab
[1] 1
x$ax
[1] 2
x[["ab"]]
NULL
 [[  - no partial matching
 $ - literal name  or partial name.


+ myVector[idx] and myVector[[ ]]

+ What does [ return (assuming no methods)?

+ Partial matching - see above

+ How arguments are matched to parameters in a call.

myList[[ c(1, 2) ]] and myList[ c(1, 2) ]
a = list(abc = data.frame(x = 1:10, y = rnorm(10)), xyz = 20:29)
19> a[[1, 2]]
Error in a[[1, 2]] : incorrect number of subscripts
No suitable frames for recover()
19> a[[c(1, 2)]]
 [1] -0.54027727  0.70919693 -0.86538271  0.02580137 -1.57977740
 [6] -1.41737952 -0.43792000 -0.13447595 -0.26713443  1.06184319
20> a[c(1, 2)]
$abc
    x            y
1   1  1.026409341
2   2  1.398747184
3   3  1.363570447
4   4 -0.084175654
5   5  0.430476969
6   6 -0.518491540
7   7 -1.016935599
8   8  0.008144095
9   9  0.003179592
10 10 -0.712805619

$xyz
 [1] 20 21 22 23 24 25 26 27 28 29



c(x, y, z) and list(x, y, z)
x = 1:10
y = TRUE
z = seq(1.5, length = 20, by = 2)
a = list(x, y, z)
b = c(x, y, z)
class(a)
class(b)
length(b)


lapply() when you know you want a list.
sapply() when you expect it to collapse to  a vector/matrix reliably.
vapply
sapply()
See lapply_versus_sapply.R.  Come up with an example.

sapply() with 2 dimensions always seems to be the wrong orientation so t()



Computing on the language


+ force()