massage data

gistfile1.r

# Read in the TIOBE data
tiobe = read.csv("tiobe.csv", stringsAsFactors=FALSE)
tiobe_langs = tolower(tiobe[, "lang"])

# Looking at the TIOBE listings and some of the tweet data, massage some of the entries
# here. This won't be perfect but will help a little bit
replace_statuses = function(statuses, was, is) {
    gsub(was, is, statuses, ignore.case=TRUE)
}

replacements = list(c("objective c", "objective-c"), c("visual basic", "visual-basic"),
    c("emacs lisp", "emacs-lisp"), c("object pascal", "delphi/object-pascal"),
    c("delphi", "delphi/object-pascal"), c("common lisp", "common-lisp"),
    c("elisp", "emacs-lisp"))

for (pair in replacements) {
    statuses = replace_statuses(statuses, pair[1], pair[2])
}

tiobe_langs[7] = "visual-basic"
tiobe_langs[11] = "visual-basic"
tiobe_langs[20] = "delphi/object-pascal"
tiobe_langs[46] = "emacs-lisp"
tiobe_langs[41] = "common-lisp"

tiobe$lang = tiobe_langs
# we've got two visual-basic entries
tiobe[7, "rating"] = tiobe[7, "rating"] + tiobe[11, "rating"]
tiobe = tiobe[-11, ]