app-mpt/scripts/generate-rdata.R

#!/usr/bin/env Rscript

library(readr)
library(dplyr)

# TODO rename functions
# This one generates a list for a tree or dendrogram
rsplit <- function(x) {
  if(ncol(x) > 1) {
      listSplit <- split(x[-c(1)], x[1], drop=F)
      lapply(names(listSplit), function(y) {list(name = y,
                                                 children = rsplit(listSplit[[y]]))})
  } else {
      lapply(seq(nrow(x[1])), function(y) {list(name = as.character(x[y,1]))})
  }
}

# This one generates a list for a treemap
makeList <- function(x) {
                                        #varia com n
    if(ncol(x) > 8) {
        listSplit <- split(x[-c(1)], x[1], drop=F)
        lapply(names(listSplit), function(y) {list(name = y,
                                        # varios n
                                                   escopo = colnames(x[1]),
                                                   value = as.integer( listSplit[[y]][1,1] ),
                                                   Hora = as.integer( listSplit[[y]][1,2] ),
                                                   Dia = as.character(listSplit[[y]][1,3]),
                                                   Idade = as.double( listSplit[[y]][1,4] ),
                                                   Sexo = as.double( listSplit[[y]][1,5] ),
                                                   Causador = as.character(listSplit[[y]][1,6]),
                                                   CNAE = as.character(listSplit[[y]][1,7]),
                                        # -n
                                        # children = listSplit[[y]][ -c(1:2) ] )})
                                                   children = makeList( listSplit[[y]][ -(1:7) ] ))})
    } else {
        lapply(seq(nrow(x[1])), function(y) {
            list(name = as.character(x[y,1]),
                 escopo = colnames(x[y,1]),
                 value = as.integer(x[y,2]),
                 Hora = as.integer( x[y,3] ),
                 Dia = as.character(x[y,4]),
                 Idade = as.double( x[y,5] ),
                 Sexo = as.double( x[y,6] ),
                 Causador = as.character(x[y,7]),
                 CNAE = as.character(x[y,8])

                 )})

    }
}

make_bardata <- function(x) {
      listSplit <- split(x[-1], x[1], drop=F)
      lapply(names(listSplit), function(y) {
          list(name = y,
               value = as.vector(filter(x, x[1] == y)$n))
      })
}

make_bardata2 <- function(x) {
      apply(unique(x[c(1, 2)]), 1, function(y) {
          list(name = c(as.character(y[1]), as.character(y[2])),
               value = as.vector(filter(x, x[1] == y[1], x[2] == y[2])$n))
      })
}

row_to_list  <- function(d) {
    apply(d, 1, function(x) {
        setNames(as.list(x), c("axis", "value"))
    })
}

make_radardata <- function(x) {
    listSplit <- split(x[-c(1)], x[1], drop=F)
    lapply(names(listSplit), function(y) {
        list(name = y,
             value = row_to_list(filter(x, x[1] == y)[-1]))
    })
}

make_radardata2 <- function(x) {
    apply(unique(x[c(1, 2)]), 1, function(y) {
        list(name = c(as.character(y[1]), as.character(y[2])),
             value = row_to_list(filter(x, x[1] == y[1], x[2] == y[2])[-c(1,2)]))
    })
}


print("loading data...")
data <- read_csv2("../app/data/amostra.csv")
brasil <- read_csv2("../app/data/brasil.csv")
acidentes <- read_csv2("../app/data/acidentes_victor.csv")

#barchart data
pais_bar <- read_csv2("../app/data/barchart/ac_pais_2012.csv")
regioes_bar <- read_csv2("../app/data/barchart/ac_regiao_2012.csv")
estados_bar <- read_csv2("../app/data/barchart/ac_uf_2012.csv")
meso_bar <- read_csv2("../app/data/barchart/ac_meso_2012.csv")
micro_bar <- read_csv2("../app/data/barchart/ac_micro_2012.csv")
municipios_bar <- read_csv2("../app/data/barchart/ac_mun_2012.csv")

#radarchart data
pais_radar <- read_csv2("../app/data/radarchart/pais_by_year.csv")
regioes_radar <- read_csv2("../app/data/radarchart/regiao_by_year.csv")
estados_radar <- read_csv2("../app/data/radarchart/uf_by_year.csv")
meso_radar <- read_csv2("../app/data/radarchart/meso_by_year.csv")
micro_radar <- read_csv2("../app/data/radarchart/micro_by_year.csv")
municipios_radar <- read_csv2("../app/data/radarchart/mun_by_year.csv")

print("Computing scatterplot data...")
sct_data <- as.matrix(data[ , c("hora_acidente", "idade_cat", "ds_agente_causador")])

print("Computing choropleth map data...")
choropleth_data <- read_csv2("../app/data/puf.csv")

print("Computing parallel coordinates data...")
parallelcoordinates_data <- read_csv2("../app/data/acidentes-sumarizado.csv")

print("Computing barchart data...")
pais_bar  <- make_bardata(pais_bar)
regioes_bar  <- make_bardata2(regioes_bar)
estados_bar  <- make_bardata2(estados_bar)
meso_bar  <- make_bardata2(meso_bar)
micro_bar  <- make_bardata2(micro_bar)
municipios_bar  <- make_bardata2(municipios_bar)

bar_data  <- c(pais_bar, regioes_bar, estados_bar,
               meso_bar, micro_bar, municipios_bar)

print("Computing radarchart data...")
pais_radar  <- make_radardata(pais_radar)
regioes_radar  <- make_radardata2(regioes_radar)
estados_radar  <- make_radardata2(estados_radar)
meso_radar  <- make_radardata2(meso_radar)
micro_radar  <- make_radardata2(micro_radar)
municipios_radar  <- make_radardata2(municipios_radar)

radar_data  <- c(pais_radar, regioes_radar, estados_radar,
                 meso_radar, micro_radar, municipios_radar)

print("Computing treemap data...")
treemap_data <- makeList(unique(acidentes))
tp <- unique(acidentes)

treemap_data_cities <-
data.frame(
                                tp["pais"],
                                tp["regiao"],
                                tp["uf"],
                                tp["mesorregiao"],
                                tp["microrregiao"],
                                tp["municipio"],
                                tp["acidentes_municipio"],
                                tp["hora_municipio"],
                                tp["diaSemana_municipio"],
                                tp["idadeMedia_municipio"],
                                tp["sexoPorcentagem_municipio"],
                                tp["dsAgenteModa_municipio"],
                                tp["classeCnaeModa_municipio"]
                                )

print("Computing dendrogram data...")
tree_data <- rsplit(unique(brasil))[[1]]

print("Writing out generated R objects...")
saveRDS(sct_data, "../app/data/rds/sct_data.rds")
saveRDS(choropleth_data, "../app/data/rds/choropleth_data.rds")
saveRDS(parallelcoordinates_data, "../app/data/rds/parallelcoordinates_data.rds")
saveRDS(bar_data, "../app/data/rds/bar_data.rds")
saveRDS(radar_data, "../app/data/rds/radar_data.rds")
saveRDS(treemap_data, "../app/data/rds/treemap_data.rds")
saveRDS(treemap_data_cities, "../app/data/rds/treemap_data_cities.rds")
saveRDS(tree_data, "../app/data/rds/tree_data.rds")

print("Done!")