R/graph_hospit_deces_tests.R

In SimonCoulombe/covidtwitterbot: What the Package Does (One Line, Title Case)

#' graph_deces_hospit_tests
#' @return
#' @export
#' @examples
#' graph_deces_hospit_tests()
#' @importFrom dplyr filter select mutate bind_rows coalesce group_by ungroup arrange lead lag case_when summarise
#' @importFrom tidyr gather spread drop_na
#' @importFrom ggplot2 '%+replace%'
#' @importFrom ggplot2 aes theme margin element_text element_line element_blank xlim ylim scale_y_continuous scale_x_continuous ggplot aes geom_col geom_line facet_wrap facet_grid labs xlim ylim scale_x_discrete geom_rect
#' @importFrom readr read_csv
#' @importFrom tibble tribble
#'
graph_deces_hospit_tests <- function() {
  rr <- get_inspq_manual_data_hospits() %>%
    dplyr::filter(date >= lubridate::ymd("20200315")) %>%
    dplyr::select(date,volumetrie, ) %>%
    tidyr::gather(key = type, value = nombre, volumetrie) %>%
    dplyr::bind_rows(get_inspq_covid19_hist() %>%
                       dplyr::filter(!is.na(date), groupe == "Ensemble du Québec") %>% # ajouter deces
                       dplyr::select(date, deces = deces_totaux_quotidien, cas = cas_quo_tot_n, soins_intensifs = hos_act_si_n,hospitalisations = hos_act_tot_n ) %>%
                       tidyr::gather(key= type, value = nombre, deces, cas, soins_intensifs, hospitalisations)
    ) %>%

    dplyr::group_by(type) %>%
    dplyr::arrange(date) %>%
    dplyr::mutate(moyenne7 = (nombre + dplyr::lag(nombre, 1) + dplyr::lag(nombre, 2) + dplyr::lag(nombre, 3) + dplyr::lead(nombre, 1) + dplyr::lead(nombre, 2) + dplyr::lead(nombre, 3)) / 7) %>%
    dplyr::ungroup() %>%
    mutate(
      type = factor(
        case_when(
          type == "cas" ~ "Nouveaux cas",
          type == "deces" ~ "Décès",
          type == "hospitalisations" ~ "Hospitalisations hors soins intensifs",
          type == "soins_intensifs" ~ "Hospitalisations aux soins intensifs",
          type == "volumetrie" ~ "Tests"
        ),
        levels = c("Nouveaux cas", "Décès", "Hospitalisations hors soins intensifs", "Hospitalisations aux soins intensifs", "Tests")
      )
    )

  p1 <-
    ggplot2::ggplot() +
    ggplot2::geom_col(data = rr, aes(x = date, y = nombre), width = 1, na.rm = TRUE, alpha = 0.8) +
    ggplot2::geom_line(data = rr, ggplot2::aes(x = date, y = moyenne7), color = palette_OkabeIto["blue"], size = 1, alpha = 0.8, na.rm = TRUE) +
    ggplot2::facet_wrap(~type, ncol = 1, scales = "free_y") +
    theme_simon(font_size = 12) +
    ggplot2::theme(axis.text.x = element_text(angle = 30, hjust = 1, vjust = 1)) + # rotate axis text 30 degrees
    ggplot2::labs(
      title = "Cas, Décès, hospitalisations, soins intensifs, tests",
      x = "date",
      y = "Nombre"
    ) +
    xlim(lubridate::ymd("20200315"), NA) +
    scale_y_continuous(expand = c(0, 0))


  p1 +
    geom_rect(
      data = rr,
      fill = palette_OkabeIto["vermillion"],
      alpha = 0.01,
      xmin = max(rr$date) - 3,
      xmax = max(rr$date),
      ymin = -Inf,
      ymax = Inf
    ) +
    labs(caption = "gossé par @coulsim")
}