BLS Occupation Code FTP Scrape Appleton, WI

oeu_scrape.r

library(blscrapeR)
library(dplyr)
library(stringr)
library(data.table)

# Using fread from data.table here becuse it's much faster and the first file is pretty huge.
doc <- data.table::fread("https://download.bls.gov/pub/time.series/oe/oe.data.0.Current")
titles <- data.table::fread("https://download.bls.gov/pub/time.series/oe/oe.occupation")
data_type <- data.table::fread("https://download.bls.gov/pub/time.series/oe/oe.datatype")

# Subset only to 11540, which I beleive is Appleton?
ids <- subset(doc, substr(doc$series_id, 1,11) == "OEUM0011540") %>%
    # Extranct occupation code from series ID
    mutate(occupation_code = str_sub(series_id, -8,-3)) %>%
    mutate(occupation_code = as.integer(occupation_code)) %>%
    # Extract data type code from series ID
    mutate(datatype_code = str_sub(series_id, -2,-1)) %>%
    mutate(datatype_code = as.integer(datatype_code))

# Join and produce final data set.
df <- inner_join(ids, titles) %>% inner_join(data_type) %>%
    select(year, period, series_id, occupation_name, datatype_name, value)

# Optionial: You can make it look exactly like the webpage table by using the tidyr "spread" function.
pretty_df <- select(df, occupation_name, datatype_name, value) %>%
    tidyr::spread(key=datatype_name, value = value)