Create wordcloud from mbox email archive [unfinish]

mbox.R

library(mboxr)
library(tm.plugin.webmining)
library(tm)
library(wordcloud)

fpath   <- path.expand("~/Downloads/jobs.mbox")
data    <- read_mbox(fpath)
content <- paste(unlist(data$content), collapse = " ")
test    <- extractHTMLStrip(content)
write.csv(test, "~/Downloads/jobs.csv")
test3 <- read.csv("~/Downloads/jobs.csv", stringsAsFactors = FALSE)$x

test2 <- removeWords(tolower(test3), c("research", "please", "skills", "applicants", 
                             "successful", "students", "submit", "andor", 
                             "review", "open", "campus", "begin", "join", 
                             "resources", "salary", "develop", "diverse", 
                             "education", "publication", "current", "changes", 
                             "experience", "consideration", "five", "pdfs", 
                             "opportunity", "background", "available", 
                             "textindent25in", "home", "computer", 
                             "annoucement", "document", "units", "hours", 
                             "this", "list", "located", "addresses", "letter", 
                             "project", "fontfamily", "send", "new", "located", 
                             "pdf", "search", "august", "across", 
                             "msostylepriority99", "without", "school", 
                             "benefits", "one", "email", "strong", "includes", 
                             "67698693", "baylor", "description", "north", "nov", 
                             "action", "gleonall", "schools", "skype", "continue", 
                             "relevant", "existing", "1st", "text", "thank", 
                             "references", "three", "full", "dear", "350", "506", 
                             "looking", "msolevelnumberformatbullet", "2018", 
                             "will", "contact", "include", "including", "sciences", 
                             "candidates", "department", "application", "program", 
                             "work", "related", "applications", "position", 
                             "candidate", "state", "msolevelnumberposition", 
                             "must", "indent25in", "time", "university", 
                             "can", "apply", "msoleveltabstopnone", "also", 
                             "institute", "within", "date", "qualifications", 
                             "may", "cover", "msolevelnumberleft", "center", 
                             "msolevelnumberformatbullet", "working", 
                             "undergraduate", "questions", "phd", "change"))
test2 <- stringr::str_remove(test2, "msolevel.*") %>%
  stringr::str_remove("information") %>%
  stringr::str_remove("msolevelnumberformatbullet") %>%
  stringr::str_remove("andor")
  

wordcloud(test2, max.words = 20)

clipr::write_clip(test)



cleanFun <- function(htmlString) {
  return(gsub("<.*?>", "", htmlString))
}


res <- list()
for(i in c(3:length(data$content))){
  print(i)
  res[[i]] <- html_text(read_html(data$content[2]))
}

res <- cleanFun(content)



setwd("~/Downloads")
test <- read.csv("~/Downloads/jobs.csv")
test <- merge_mbox_all(path = "jobs.mbox", 
                       out = "jobs.csv")