# Create_Fig6_1_Dec2020.R
# R Script accompanying _Sociophonetics_ by Tyler Kendall & Valerie Fridland
# Tyler Kendall & Valerie Fridland, 2020 (CC BY 4.0)
# 
# This simple script recreates Figure 6.1 in the book. In addition to recreating
#  the figure, it is meant as an example for using Vowels.R for vowel plotting.
#

# Install the vowels.R library if you do not have it installed already
# e.g. > install.packages("vowels", dependencies=T)
library(vowels)

# Set up graphical defaults, so it is easy to reset to the initial settings later
par(bg="white")
defaults<-par(no.readonly=TRUE)

# Set the path to the input file, which you can obtain from the _Sociophonetics_ 
#  website. Note this is output directly from FAVE vowel extraction software.
path.to.file <- "~/Documents/DCB_se3_ag3_f_01_1-out.txt"

# Read in the data
input <- read.delim(path.to.file, header=TRUE)

# Pull out the columns we want into a new data.frame, since the 
#  FAVE output includes much more than we need.
examp <- data.frame(Spkr = input$first_name, Vowel = input$vowel, Context = input$word, F1 = input$F1, F2 = input$F2, F3 = input$F3, F1gl = rep(NA, times=length(input$first_name)), F2gl = rep(NA, times=length(input$first_name)), F3gl = rep(NA, times=length(input$first_name)), Pre = input$pre_seg, Post = input$fol_seg)

# Specify preceding and following contexts that we will care about
pre.contexts <- c("B", "D", "G", "P", "T", "K", "S", "Z", "F", "V", "DH", "TH")
fol.contexts <- c("B", "D", "G", "P", "T", "K", "S", "Z", "F", "V", "DH", "TH", "M", "N", "NG")
vowels.of.int<- c("AE", "EH")

# Now pull out just the vowels and contexts of interest
vinterest <- droplevels(examp[examp$Vowel %in% vowels.of.int & examp$Pre %in% pre.contexts & examp$Post %in% fol.contexts,])
dim(vinterest) # 450 tokens

# And the other vowels that we might want for contextualizing the positions
#  of the vowels of interest
others <- droplevels(examp[examp$Vowel %in% c("AA", "AE", "AH", "AO", "AW", "AY", "EH", "EY", "IH", "IY"),])

# Relabel these using IPA characters; make sure the list of IPA characters lines up
#  with the original labels!
others$Vowel <- as.factor(others$Vowel)
levels(others$Vowel)
levels(others$Vowel) <- c("ɑ", "æ", "ʌ", "ɔ", "aw", "ay", "ɛ", "e", "ɪ", "i")

# Use vowels.R to compute means
mn.others <- compute.means(others[,1:9])

# Can check the number of tokens for the different vowels and contexts:
table(vinterest$Vowel, vinterest$Pre)
table(vinterest$Vowel, vinterest$Post)

# Set up plotting colors based on the number of Vowel+Following Environment pairings
vinterest.clrs <- as.factor(paste(vinterest$Vowel, vinterest$Post, sep=""))
clr.list <- rainbow(length(levels(vinterest.clrs)))
levels(vinterest.clrs) <- clr.list
vinterest.clrs <- as.character(vinterest.clrs)

# Set the file name, uncomment the next lines and the very bottom to save the plot to a file
# out.loc<-"~/Documents/DCB_se3_ag3_f_01_1-BOB-vowelplot.jpg"
# jpeg(filename = paste(out.loc, ".jpg", sep=""), width = 6, height = 5, units = "in", res = 300, pointsize = 11, quality=100, bg = "white", type = "quartz", antialias="none")

# Set up plotting space
par(mai=c(0.8, 0.8, 0.4, 0.2), family="Helvetica")

# Use vowelplot to first set up the plotting space, but just print in white at first
vowelplot(vinterest, color="speakers", color.choice="white", leg=NA, title=paste("CORAAL speaker ", unique(vinterest$Spkr), sep=""))

# Add a legend about the number of tokens for the vowel categories
ltab <- table(vinterest$Vowel)
names(ltab) <- c("/æ/", "/ɛ/") # If you change the vowels, be sure to update the labels here
legend("bottomleft", paste(names(ltab), ltab, sep=" N = "), cex=0.8, bty="n")
text(vinterest$F2, vinterest$F1, tolower(vinterest$Context), cex=0.7, col=vinterest.clrs)
text(mn.others$F2, mn.others$F1, mn.others$Vowel, col="black", cex=2)

# Return settings to their defaults
par(defaults)

# Uncomment to close the connection to the file, if writing to file
# dev.off()