Mails data extraction.R

library(tm)
library(stringr)
library(stringi)
library(lubridate)

setwd("~/R/Mails data extraction/")

mbox_file <- "newsletter.mbox"

# Read the mbox file
mbox_content <- tolower(readLines(mbox_file, warn = FALSE))

# Function to extract information
extract_email_data <- function(mbox_content) {
  
  # Extract lines containing relevant information
  relevant_lines <- mbox_content[grep("^(date:|subject:|from:|content-type:)", mbox_content)]
  
  # Extract and format date and time
  dates_lines <- relevant_lines[grep("^date:", relevant_lines)]
  dates <- str_extract(dates_lines, "\\d{1,2} [a-z]+ \\d{4} \\d{2}:\\d{2}:\\d{2}")
  dates <- dmy_hms(dates)
  
  # Extract and clean From emails
  from_lines <- relevant_lines[grep("^from:", relevant_lines)]
  froms <- str_extract(from_lines, "<.+?>")
  froms <- str_replace_all(froms, "[<>]", "")
  
  # Extract Subject
  subject_lines <- relevant_lines[grep("^subject:", relevant_lines)]
  subjects <- str_replace(subject_lines, "^subject: ", "")
  
  # Extract Content-Type and related text
  content_text_lines <- mbox_content[grep("^content-type: text/html", mbox_content, fixed = TRUE)]
  content_texts <- str_extract_all(mbox_content, "<div dir=\"ltr\">.*?</div>")
  content_texts <- unlist(content_texts)
  
  # Remove HTML tags
  content_texts <- str_remove_all(content_texts, "<.*?>")  
  
  
  # Create a data frame
  email_df <- data.frame(Date_time = dates, From = froms, Subject = subjects, Text = content_texts, stringsAsFactors = FALSE)
  
  return(email_df)
}

# Apply the function 
emails_df <- extract_email_data(mbox_content)

# Write to a CSV file
timestamp <- format(Sys.time(), "%Y%m%d")
output_filename <- paste0("newsletter_data_", timestamp, ".csv")
write.table(emails_df, file = output_filename, sep = ",", row.names = FALSE, col.names = TRUE, quote = FALSE)