data-raw/old_api/cws_bdmep.R

In italocegatta/inmetdown: An R package to import data from INMET's weather stations

# import_bdmep <- function(id, start, end, stations, proxy) {
#
#   n_row <- as.numeric(end - start + 1) * 2
#
#   if (is.null(stations)) {
#     stations <- cws_station() %>%
#       dplyr::filter(id %in% !!id)
#   } else {
#     stations <- stations %>%
#       dplyr::filter(id %in% !!id)
#   }
#
#   seq <- seq_along(id)
#   out <- vector("list", length(seq))
#
#   for (i in seq) {
#
#     table <- get_table_bdmep(id[i], start, end, n_row, "^.+instruções\n--------------------\n", proxy)
#
#     table <- table %>%
#       dplyr::select(
#         id = Estacao,
#         data = Data,
#         hora = Hora,
#         ppt = Precipitacao,
#         t_bs = TempBulboSeco,
#         t_bu = TempBulboUmido,
#         t_max = TempMaxima,
#         t_med = Temp.Comp.Media,
#         t_min = TempMinima,
#         ur_med = UmidadeRelativa,
#         ur_med_comp = Umidade.Relativa.Media,
#         pa_med = PressaoAtmEstacao,
#         pa_mar = PressaoAtmMar,
#         v_dir = DirecaoVento,
#         v_med = VelocidadeVento,
#         v_med_comp = Velocidade.do.Vento.Media,
#         ins = Insolacao,
#         neb = Nebulosidade,
#         evp_piche = Evaporacao.Piche
#       )
#
#     table <- suppressWarnings(dplyr::mutate_at(table, dplyr::vars(-id, -data), as.double))
#
#     table <- table %>%
#       dplyr::mutate(id = as.character(id)) %>%
#       dplyr::rowwise() %>%
#       dplyr::mutate(
#         data = as.Date(ifelse(lubridate::is.Date(data), data, lubridate::dmy(data)),  origin = "1970-01-01"),
#         date_time = lubridate::ymd_h(paste(data, stringr::str_pad(hora / 100, 2, pad = "0")))
#       ) %>%
#       dplyr::ungroup()
#
#     if (nrow(table) != as.numeric(end - start + 1) * 3) { # ponto fragil. deve testar para o hora final nas etapas 0, 12 e 18h
#
#       range_dttm <- range(table$date_time)
#       seq_dttm <- data.frame(date_time = seq.POSIXt(range_dttm[1], range_dttm[2], 'hour')) %>%
#         dplyr::filter(lubridate::hour(date_time) %in% c(0, 12, 18))
#
#       table <- dplyr::full_join(table, seq_dttm, by = "date_time")
#     }
#
#     out[[i]] <- table %>%
#       dplyr::mutate(
#         id = id[i],
#         data = lubridate::date(date_time)
#       ) %>%
#       dplyr::select(-date_time) %>%
#       dplyr::mutate_if(is.double, round, digits = 4) %>%
#       dplyr::arrange(id, data, hora)
#
#       # dplyr::group_by(id, data) %>%
#       # dplyr::summarise(
#       #   ppt = mean(ppt, na.rm = TRUE),
#       #   t_min = mean(t_min, na.rm = TRUE),
#       #   t_max = mean(t_max, na.rm = TRUE),
#       #   ur_med = mean(ur_med, na.rm = TRUE),
#       #   pa_med = mean(pa_med, na.rm = TRUE),
#       #   v_dir = mean(v_dir, na.rm = TRUE),
#       #   v_med = mean(v_med, na.rm = TRUE),
#       #   ins = mean(ins, na.rm = TRUE)
#       # ) %>%
#       # dplyr::mutate_if(is.double, round, digits = 1) %>%
#       # dplyr::rowwise() %>%
#       # dplyr::mutate(t_med = (t_max + t_min) / 2) %>%
#       # dplyr::ungroup() %>%
#       # tidyr::replace_na(list(
#       #   ppt = NA, t_med = NA, t_min = NA, t_max = NA,
#       #   ur_med = NA, pa_med = NA,
#       #   v_med = NA, v_dir = NA, ins = NA
#       # )) %>%
#       # dplyr::select(
#       #   id, data,
#       #   ppt,
#       #   t_max, t_med, t_min,
#       #   ur_med, pa_med,
#       #   ins,
#       #   v_dir, v_med
#       # ) %>%
#       # dplyr::arrange(id, data)
#   }
#
#   dplyr::bind_rows(out)
# }