tests/testthat/test_that_startupSettings.R

In frumentum/foodcoop-storage: Look-up foodco-ops' storage of products

test_that(
  "startupSettings edits dataset correctly", {
    #### load dataset ####
    path <- system.file("data", package = "foodstorage")
    files <- list.files(path)
    # filter all backups (files which end up with .BAK)
    backup <- files[which(stringr::str_detect(files, ".BAK$"))]
    appDB <- dbConnect(SQLite(), file.path(path, backup))
    originalData <- dbGetQuery(appDB, '
SELECT strftime(\'%d/%m/%Y\',transactions.start/1000,\'unixepoch\') AS Tag,
      ROUND(SUM(transaction_products.quantity), 3) AS Menge, 
      transaction_products.unit AS Einheit,
      ROUND(transaction_products.price, 2) AS Preis, 
      transaction_products.title AS Produkt,
      ROUND(SUM(transaction_products.quantity * transaction_products.price), 3) AS Summe
FROM transaction_products 
LEFT JOIN transactions
      ON transactions._id = transaction_products.transaction_id 
WHERE transactions.status IS \'final\' AND transaction_products.account_guid IS \'lager\'
GROUP BY Tag, Produkt, Preis
ORDER BY transactions.start
'
    )
    dbDisconnect(appDB)
    
    kornDB <- files[which(stringr::str_detect(files, "kornInfo.sqlite"))]
    kornInfo <- dbConnect(SQLite(), file.path(path, kornDB))
    originalInfos <- dbReadTable(kornInfo, "productInfo")
    dbDisconnect(kornInfo)

    #### add three columns: ID, product summary and bulksize ####
    tmpData <- editDataset(originalData, originalInfos)
    newDataset <- tmpData$editData
    expect_is(newDataset, "data.frame")
    expect_equal(
      ncol(newDataset),
      ncol(originalData) + 6
    )
    expect_equal(
      colnames(newDataset),
      c(colnames(originalData), "ID", "Produkt_Zusammenfassung", 
        "Lieferant", "Lieferant2", "Produktgruppe", "VPE")
    )
    expect_is(newDataset$ID, "integer")
    expect_is(newDataset$VPE, "numeric")
    expect_is(newDataset$Produkt, "character")
    expect_is(newDataset$Produkt_Zusammenfassung, "character")
    expect_is(newDataset$Tag, "Date")
    expect_false(NA %in% newDataset$VPE)
    expect_false(0 %in% newDataset$ID)
    expect_false(NA %in% newDataset$ID)
    expect_gt( # expect greater than
      length(unique(newDataset$Produkt)),
      length(unique(newDataset$Produkt_Zusammenfassung))
    )
    # expect same number of rows in newDataset like in original Data
    expect_equal(
      nrow(newDataset),
      nrow(originalData)
    )
    
    #### expect no double dates and products
    duplicates <- identifyDuplicates(newDataset)
    summariseQuantity <- tmpData$summariseQuantity
    expect_equal(
      nrow(newDataset),
      length(duplicates) + nrow(summariseQuantity)
    )
    # dataset resulting from editDataset() without ID of duplicates = 
    #   dataset resulting from addCumulativeStorage()
    oldDataset <- newDataset
    newDataset <- addCumulativeStorage(oldDataset, duplicates, summariseQuantity)
    expect_equal(
      nrow(oldDataset %>% filter(!ID %in% duplicates)),
      nrow(newDataset)
    )
    expect_equal(
      oldDataset[!oldDataset$ID %in% duplicates, ]$ID,
      newDataset$ID
    )
    expect_is(newDataset$ID, "integer")
    # after addCumulativeStorage() no more duplicates are in the data
    testData <- newDataset %>%
      group_by(Produkt_Zusammenfassung, Tag) %>%
      select(Produkt_Zusammenfassung, Tag)
    expect_equal(
      nrow(testData[duplicated(testData), ]), 0
    )
    # after addCumulativeStorage() Bestand.Einheit is one more column
    expect_equal(
      colnames(newDataset),
      c(colnames(oldDataset), "Bestand.Einheit")
    )
    expect_is(newDataset$Bestand.Einheit, "numeric")
    ##### cumulative storage function does its job well #####
    # filter products "with duplicates"
    testProducts <- oldDataset %>%
      filter(ID %in% duplicates) %>%
      distinct(Produkt_Zusammenfassung) %>%
      unlist(use.names = F)
    # filter newDataset
    testData <- newDataset %>%
      filter(Produkt_Zusammenfassung %in% testProducts) %>%
      group_by(Produkt_Zusammenfassung) %>%
      mutate(testCumSum = round(ave(Menge, FUN = cumsum), 3)) %>%
      arrange(ID)
    # cumsums should be the same
    expect_equal(
      testData$testCumSum,
      testData$Bestand.Einheit
    )
    #### at least: test wrapper function startupSettings() ####
    testData <- startupSettings(originalData, originalInfos)
    expect_equal(
      testData,
      newDataset
    )
    
  }
)