library(rvest)
library(dplyr)
library(tidyr)

# sudo crontab -e
# 5 9 * * * su matt -c "cd /home/matt/ham-radio-licenses/; Rscript /home/matt/ham-radio-licenses/scrape-arrl-fcc.R">/dev/null 2>&1

arrl_url <- "https://www.arrl.org/fcc-license-counts"

# Read the page
d_raw <- read_html(arrl_url)

# Get date
date_raw <- d_raw %>% 
            html_nodes(xpath = "/html/body/div[1]/div/div[2]/div/div[2]/div[1]/p[2]/em") %>% 
            # Get date
            html_text() %>% 
            gsub(".*, ","",.) %>% 
            as.Date(format = "%d-%b-%Y")

# Get table and clean up
tab <- d_raw %>%
    html_nodes(xpath="/html/body/div[1]/div/div[2]/div/div[2]/div[1]/table") %>%
    html_table() %>% 
    .[[1]] %>% 
    # Add date col
    bind_cols(Date = date_raw, .) %>% 
    # Insert Tech Plus for compatibility
    mutate("Tech-Plus" = NA,
           .before = General) %>% 
    mutate(a=NA, b=NA, c=NA, d=NA, e=NA, f=NA, g=NA,
           source_name = "ARRL FCC License Counts",
           source_detail = "http://www.arrl.org/fcc-license-counts")

# Append table
write.table(tab, file = "out/arrl-fcc-licenses-scraped.csv", sep = ",", 
            append = TRUE, quote = FALSE, 
            col.names = TRUE, row.names = FALSE,
            na = "") 

# Clean up table to remove any duplicates (sometimes the page isn't updated regularly)
# db <- read.csv("out/arrl-fcc-licenses-scraped.csv")
# db2 <- db %>% distinct(.keep_all = TRUE) %>% 
#         filter(Date != "Date")
# write.csv(db2, "out/arrl-fcc-licenses-scraped.csv",
#           quote = F,
#           row.names = F,
#           na = "")