File structure

Table of contents

1. /app
   1. /library
   2. /shiny
      1. /config
         1. /bldgHierarchy.csv
         2. /configApp.csv
         3. /configCsv.csv
         4. /configInfluxdb.csv
         5. /configTtn.csv
         6. /dataPoints.csv
         7. /etlAggFilterList.csv
      2. /csvScripts
      3. /data
      4. /R
      5. /www
2. /dataFetcher
3. /dist
4. /docs
5. /sampleData

/app

• This folder contains application specific things
• app.R is the starting script for R Portable
• config.cfg conatains configuration for R Portable
• packages.txt includes R libraries which get installed for R Portable. If you extend the application and require a new R package, then you have to edit this file.

/library

• Includes the downloaded R packages
• Only used with R Portable

/shiny

• This folder contains the Shiny application and all the saved time series data and app configurations
• If you work in a local R installation and don’t use R Portable, only this folder is interesting for you

/config

• All the configuration parameters for the app and project are saved here
• To keep things simple, all the configurations are saved as csv files

/bldgHierarchy.csv

• This file contains the project specific building hierarchy
• The file can be edited in the app under Menue > Settings > Building Hierarchy
• Every Datapoint respectively sensor gets assigned with a location. This information is used to label diagrams and create pull down menues
• The lcm app is made for one building with one or multiple flats resp apartments. So the building hierarchy includes the flats and its rooms
• Because some modules require the size in square meters and the amount of occupants of each flat analysis purposes, these parameters get saved as well in the hierarchy

Exemplary content:

flat;size;occupants;rooms
Flat A;120;3;Dormitory,Bathroom,Bathroom,Aisle,Kitchen,Living Room,Office

• Note that the different rooms of the flat are comma separated while the rest is separated with semicolons.

/configApp.csv

• This file contains application specific configuration settings
• The file can be edited in the app under Menue > Settings > App Configuration

/configCsv.csv

• This file contains information about the imported csv files
• The file can be edited in the app under Menue > Settings > Data Sources > csv Files

/configInfluxdb.csv

• This file contains information about the connected influxDB databases
• The file can be edited in the app under Menue > Settings > Data Sources > InfluxDB

/configTtn.csv

• This file contains information about the connected the things network applications
• Please see https://hslu-ige-laes.github.io/lora-devices-ttn/ to learn how to create an application and integrate LoRaWAN devices
• The file can be edited in the app under Menue > Settings > Data Sources > the things network

/dataPoints.csv

• This file contains information about the configured data points
• The file can be edited in the app under Menue > Settings > Data Points

/etlAggFilterList.csv

• This file contains information which datapoint types should get aggregated
• You only have to modify this file if you extend or adapt the application modules

/csvScripts

• In this folder you can save project specific csv parsing scripts
• For csv data sources the importer foresees the standard file formatting options with different settings of e.g. delimiter etc.
• Unfortunately some files come with special formatting and require special parsing. In such cases a specific user-defined R parsing script can be created
• Copy and paste the following code into a new R-file and save it in the folder /csvSripts. The script automatically appears in the pull-down menu of the csv importer under Menu > Data Sources > csv Files > Add New

Parsing script template:

myParsingScriptName <- function(filePath = NULL) {
  
  if(is.null(filePath)){
    return(NULL)
  }
  # read csv file
  df <- read.csv(filePath, stringsAsFactors=FALSE)
  
  # parse and transform data here
  # final data format should be in long format: time,value1,value2, etc.
  # time,value1,value2
  # 2020-07-30 13:00:00,53,10.5

  return(df)
}

Attention

• Make sure the returned timestamp of df is UTC and not local time!

/data

• This folder includes all the raw data from the data sources (except of InfluxDB) and the cached data
• To keep things simple, all raw time series data get saved as csv files
• Cached and aggregated values are saved as R .rds files (due to performance)

/R

• This folder contains all R files (server functions, modules and helper functions)
• The application is based on modules in order to keep everything in a maintainable and sustainable manner
• Basically every page in the application has its own module
• To learn how data can get exchanged between modules please consult the following link

/www

• Shiny application folder where files for the web server session can get saved, e.g. the favicon.ico

/dataFetcher

• This folder contains the R Data Fetcher script and a batch file which fetches the data from the things network apllications

/dist

• This folder contains the R Portable code to run “lcm” in Windows as self-contained desktop R application
• If you work with a local R installation you don’t need this folder

/docs

• This folder contains documentation.
• The documentation gets automatically generated with GithubPages and is published under https://hslu-ige-laes.github.io/lcm/
• The documentation is written in Markdown language

/sampleData

• Sample csv files for demonstration purposes