How much of the R package namespace is left?

pkg_names.R

# R only has one package namespace
# All packages exist in a single, non-hierarchical namespace
# Once a package name is used, it cannot be reused
# CRAN hosts most existing R packages; Bioconductor holds many more; once claimed on either service, a package name is reserved
# GitHub hosts many other packages with no "claim" to a given package name and there's no list of such packages
# 
# "Writing R Extensions" says this about names:
## "The mandatory ‘Package’ field gives the name of the package. 
## This should contain only (ASCII) letters, numbers and dot, have 
## at least two characters and start with a letter and not end in 
## a dot. If it needs explaining, this should be done in the 
## ‘Description’ field (and not the ‘Title’ field)."
#
# CRAN policy says this about names:
## Packages should be named in a way that does not conflict 
## (irrespective of case) with any current or past CRAN package 
## (the Archive area can be consulted), nor any current Bioconductor 
## package. Package maintainers give the right to use that package 
## name to CRAN when they submit, so the CRAN team may orphan a 
## package and allow another maintainer to take it over.
#
# So, how much of the package namespace have we used up (on CRAN)?
#

# Packages on CRAN (as of 2015-12-27)
options("repos" = "https://cran.rstudio.com")
a <- available.packages()
p1 <- a[,"Package"]
length(p1)
# [1] 7666

# Current packages is an underestimate because archived packages retain a "claim" to their name (permanently held by CRAN).
# get all Archived packages (current and former)
x <- paste0(readLines("http://cran.r-project.org/src/contrib/Archive/"), collapse = "")
# extract package names (this is a bit hacky)
p2 <- regmatches(x, gregexpr("(?<=\")[[:alnum:].]+(?=/\")", x, perl=TRUE))[[1]]
length(p2)
# [1] 7115

# combined names currently from CRAN and Archived
p3 <- unique(c(p1,p2))
# [1] 8920


# the total number of possible name combinations is:
# + all two character names
n2 <- (26*(37^1))
# + all three character names
n3 <- (26*(37^2))
# + all four character names (this is actually an overestimate)
n4 <- (26*(37^3))
# + all five character names (this is actually an overestimate)
n5 <- (26*(37^4))
# + all six character names (this is actually an overestimate)
n6 <- (26*(37^5))
# + all seven character names (this is actually an overestimate)
n7 <- (26*(37^6))
# + all eight character names (this is actually an overestimate)
n8 <- (26*(37^7))
# + all nine character names (this is actually an overestimate)
n9 <- (26*(37^8))
# + all ten character names (this is actually an overestimate)
n10 <- (26*(37^9))

# How many package names are available
possible <- c(n2, n3, n4, n5, n6, n7, n8, n9, n10)

# How many package names have we used?
table(nchar(p3))
#   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 
#  66  692 1010 1001 1171 1125  990  792  613  400  280  200  169  129   81   59   46   23   29   17   10    6    2    4    3    2
used <- as.vector(table(nchar(p3)))[1:9]

# percent of names used, by character length
cat(paste0("# ", 2:10, ": ", sprintf("%0.4f", 100*used/possible), "\n", collapse = ""))
# 2: 6.8607
# 3: 1.9441
# 4: 0.0766
# 5: 0.0021
# 6: 0.0001
# 7: 0.0000
# 8: 0.0000
# 9: 0.0000
# 10: 0.0000

# So, basically we're fine.