In KWB-R/aquanes.report: Automated Reporting Tool for Water Suppliers

library(RMySQL)
library(ggplot2)
library(ggforce)
library(dplyr)
library(aquanes.report)

thresholds_file <- system.file("shiny/berlin_t/data/thresholds.csv",
                              package = "aquanes.report")

thresholds <- aquanes.report::get_thresholds(csv_path = thresholds_file )


if (params$run_as_standalone) {

### Raw data
dat <- aquanes.report::import_data_berlin_t()

conf_template <- aquanes.report::report_config_template(df = dat,
                                       temporal_aggregation = "raw", 
                                       output_timezone = "Europe/Berlin")
aquanes.report::report_config_to_txt(config_list = conf_template, "report_config_template.txt")

report_config_path <- file.path(getwd(), "report_config.txt")

if (file.exists(report_config_path)) {

### Import & use report configuration from "report_config.txt"
input <- aquanes.report::report_txt_to_config(report_config_path)

} else {


### Set a dummy configuration in case no "report_config.txt" exists 
input <- list(report_sitenames = unique(dat$SiteName),
              report_aggregation = "day",
              report_parameters_online = unique(dat$ParameterName[dat$Source == "online"])[1],
              report_parameters_offline = unique(dat$ParameterName[dat$Source == "offline"])[6],
              report_add_thresholds = FALSE,
              report_daterange = c("2017-06-01", "2017-06-30"),
              report_timezone = "CET")

}

### Aggregate to user defined time period

if (input$report_aggregation != "raw") {
dat <- aquanes.report::group_datetime(dat,
                                      by = input$report_aggregation)
}


date_idx <- as.Date(dat[,"DateTime"]) >= input$report_daterange[1] & as.Date(dat[,"DateTime"]) <= input$report_daterange[2]
site_idx <- dat[,"SiteName"] %in% input$report_sitenames
para_idx <- dat[,"ParameterName"] %in% c(input$report_parameters_online, input$report_parameters_offline)
row_idx <- date_idx & site_idx & para_idx


config <- list(run_as_standalone = TRUE,
               report_tz = aquanes.report::change_timezone(dat,tz = input$report_timezone)[date_idx,],
               report_data = aquanes.report::change_timezone(dat[row_idx,],tz = input$report_timezone), 
               report_aggregation = input$report_aggregation,
               report_sitenames = input$report_sitenames,
               report_parameters_online = input$report_parameters_online,
               report_parameters_offline = input$report_parameters_offline,
               report_parameters_calculated = unique(aquanes.report::calculate_operational_parameters(df = dat)$ParameterName),
               report_add_thresholds = input$report_add_thresholds,
               report_daterange = input$report_daterange,
               report_timezone = input$report_timezone
               )
} else {



 config <- list(run_as_standalone = params$run_as_standalone,
                report_tz = params$report_tz,
                report_data = params$report_data,
                report_aggregation = params$report_aggregation,
                report_sitenames = params$report_sitenames,
                report_parameters_online = params$report_parameters_online,
                report_parameters_offline = params$report_parameters_offline,
                report_parameters_calculated = params$report_parameters_calculated,
                report_add_thresholds = params$report_add_thresholds,
                report_daterange = params$report_daterange,
                report_timezone = params$report_timezone)
}


#table_thresholds <- aquanes.report::check_thresholds(df = config$report_tz,
#                                                     thresholds = thresholds)

Reporting period (`r config$report_daterange[1]` - `r config$report_daterange[2]`)

Here are the time series plots for the reporting period from r config$report_daterange[1] to r config$report_daterange[2] using the timezone "r config$report_timezone" with the following temporal aggregation level (i.e. median values): "r config$report_aggregation"

Thresholds

Nothing defined yet !

#knitr::kable(x = table_thresholds, 
#             row.names = FALSE)

Online parameters

You selected the following r length(config$report_parameters_online) parameter(s) for plotting:

r paste("* ", config$report_parameters_online,collapse = " \n")

At the following r length(config$report_sitenames) sampling location(s):

r paste("* ", config$report_sitenames,collapse = " \n")

   online <- config$report_data[config$report_data[,"ParameterName"] %in% config$report_parameters_online,] %>% 
             dplyr::mutate_(label = "sprintf('%s (%s)',ParameterName, ParameterUnit)")

    for (i in seq_along(config$report_parameters_online)) {

    sel_par1 <- config$report_parameters_online[order(config$report_parameters_online)][i]

    n_measurements <- nrow(online[online[,"ParameterName"] == sel_par1,])

    if (n_measurements > 0) {

    g1 <- ggplot2::ggplot(online, ggplot2::aes_string(x = "DateTime",
                                    y = "ParameterValue",
                                    col = "SiteName")) +
      ggforce::facet_wrap_paginate(~label,
                                   nrow = 1,
                                   ncol = 1,
                                   scales = "free_y",
                                   page = i) +
      ggplot2::geom_point() +  
      ggplot2::theme_bw(base_size = 20) +
      ggplot2::theme(legend.position = "top"
                     , strip.text.x = element_text(face = "bold")
                     , legend.title = element_blank()
                     ) +
      ggplot2::labs(x = "", y = "")

     if (config$report_add_thresholds == TRUE & sel_par1 %in% unique(thresholds$ParameterName)) {
     sel_threshold1 <- thresholds[thresholds$ParameterName == sel_par1,]
     g1 <- g1 + ggplot2::geom_hline(yintercept =  sel_threshold1$ParameterThreshold, 
                                   linetype = "dashed")   
     } 
     print(g1) 
    }
    }

Offline parameters

You selected the following r length(config$report_parameters_offline) parameter(s) for plotting:

r paste("* ", config$report_parameters_offline,collapse = " \n")

At the following r length(config$report_sitenames) sampling location(s):

r paste("* ", config$report_sitenames,collapse = " \n")

   offline <- config$report_data[config$report_data[,"ParameterName"] %in% config$report_parameters_offline,] %>% 
              dplyr::mutate_(label = "sprintf('%s (%s)',ParameterName, ParameterUnit)")

    for (i in seq_along(config$report_parameters_offline)) {

    sel_par2 <- config$report_parameters_offline[order(config$report_parameters_offline)][i]

    n_measurements <- nrow(offline[offline[,"ParameterName"] == sel_par2,])

    if (n_measurements > 0) {

    g2 <- ggplot2::ggplot(offline, ggplot2::aes_string(x = "DateTime",
                                    y = "ParameterValue",
                                    col = "SiteName")) +
      ggforce::facet_wrap_paginate(~label,
                                   nrow = 1,
                                   ncol = 1,
                                   scales = "free_y",
                                   page = i) +
      ggplot2::geom_point() +
      ggplot2::geom_line() +
      ggplot2::theme_bw(base_size = 20) +
      ggplot2::theme(legend.position = "top" 
                     , strip.text.x = element_text(face = "bold") 
                     , legend.title = element_blank()
                     #, axis.title = element_text(colour="black", size = 20)
                     #, axis.text = element_text(colour="black", size = 20)
                     ) +
      ggplot2::labs(x = "", y = "")

    if (config$report_add_thresholds == TRUE & sel_par2 %in% unique(thresholds$ParameterName)) {
     sel_threshold2 <- thresholds[thresholds$ParameterName == sel_par2,]
     g2 <- g2 + ggplot2::geom_hline(yintercept =  sel_threshold2$ParameterThreshold, 
                                   linetype = "dashed")   
     } 
    print(g2) 
    }
    }