## Creates single file containing median variables at the
## tract level for DC Area from the 2010 Census and 2010-2014 through
## 2013-2017 American Community Surveys for tracts in the DC Area

## Set up environment
library(tidyverse)

## Identify variables to keep representing age-race-sex variables
### Abbreviations for racial groups
races <- c(
    'blk'  ## Black alone
    , 'ami'  ## American Indian or Native Alaskan alone
    , 'asi'  ## Asian alone
    , 'pac'  ## Native Hawaiian or Other Pacific Islander alone
    , 'oth'  ## Other race alone
    , 'two'  ## Two or more races
    , 'hsp'  ## Hispanic or Latino (of any race)
    , 'nhw'  ## Non-Hispanic White
)

### Age group categories
agegrps <- c(
    '0004'
    , '0509'
    , '1014'
    , '1517'
    , '1819'
    , '2024'
    , '2529'
    , '3034'
    , '3544'
    , '4554'
    , '5564'
    , '6574'
    , '7584'
    , '85up'
)

### Sex categories
sexes <- c('m', 'f')

arsvars <- as.vector(sapply(races, paste0, sapply(sexes, paste0, agegrps)))


## Identify variables to keep that are not race variables
geovars <- c('GISJOIN', 'STATE', 'COUNTY')

## Set Functions
source('dcarea_functions.R')

## Create dataset that contains only DC-area tracts and constructed
## age-race-sex variables
select.arsvars <- function(dta) {
    dta <- select.dcarea(dta)
    dta$GISJOIN <- as.character(dta$GISJOIN)
    return(dta[, c(geovars, arsvars)])
}

## Load datasets containing median age variables from 2010 TIGER/Line files
load('../US/tracts/2010/tabular/age-race-sex/dataset/tracts-2010TIGER-age-race-sex.Rdata')

## Create list of data frames for each year and restrict to created
## age-race-sex variables
dtas <- list(trt10_ars, trt14_ars, trt15_ars, trt16_ars, trt17_ars)
arsdtas <- lapply(dtas, select.arsvars)

## Create single wide data frame with year appended to variable name
suffixes <- c('10', 14:17)
namelist <- lapply(suffixes, function(x) c(geovars, paste0(arsvars, x)))
arsdta <- mapply(setNames, arsdtas, namelist, SIMPLIFY = FALSE) %>%
    reduce(left_join, by='GISJOIN')
arsdta$STATE <- arsdta$STATE.x
arsdta$COUNTY <- arsdta$COUNTY.x
arsdta <- arsdta[, -(grep('^STATE.+|^COUNTY.+', names(arsdta), perl=TRUE))]

## Write file containing wide dataset to CSV format
write.csv(arsdta, 'tracts/2010/tabular/age-race-sex/dataset/tracts-2010TIGER-age-race-sex.csv')