Get results of study bug reporting, improvements

Author

Julius Fenn

1 Notes

2 global variables

Define your global variables:

3 create raw data files

# sets the directory of location of this script as the current directory
# setwd(dirname(rstudioapi::getSourceEditorContext()$path))

### load packages
# devtools::install_github('jdwilson4/multilayer_extraction')
require(pacman)
p_load('tidyverse', 'jsonlite', 'magrittr', 'xlsx',
       'stargazer', 'psych', 'jtools', 'DT', 'ggstatsplot', 
       'ggplot2',
       'base64enc')
 

### list data files
setwd("data")
dir()

[1] "jatos_results_data_20240430143549.txt"
[2] "jatos_results_data_20240514064500.txt"
[3] "jatos_results_data_20240718150114.txt"

suppressMessages(read_file('jatos_results_data_20240718150114.txt') %>%
  # ... split it into lines ...
  str_split('\n') %>% first() %>%
  # ... filter empty rows ...
  discard(function(x) x == '') %>%
  # ... parse JSON into a data.frame
  map_dfr(fromJSON, flatten=TRUE)) -> dat


### load functions
setwd("../functions")
for(i in 1:length(dir())){
  # print(dir()[i])
  source(dir()[i], encoding = "utf-8")
}

rm(i)

4 prepare data

4.1 set up data.frame

########################################
# create counter variable for data set
########################################
dat$ID <- NA

tmp_IDcounter <- 0
for(i in 1:nrow(dat)){
  if(!is.na(dat$sender[i]) && dat$sender[i] == "Report"){
    # tmp <- dat$prolific_pid[i]
    tmp_IDcounter = tmp_IDcounter + 1
  }
  dat$ID[i] <- tmp_IDcounter
}


########################################
# keep only complete data sets
########################################
sum(table(dat$ID) != max(table(dat$ID)))

[1] 0

sum(table(dat$ID) == max(table(dat$ID)))

[1] 26

dat <- dat[dat$ID %in% names(table(dat$ID))[table(dat$ID) == max(table(dat$ID))],]

########################################
# json (from JATOS) to 2D data.frame
########################################
tmp_notNumeric <- str_subset(string = colnames(dat), pattern = "^meta")
tmp_notNumeric <- str_subset(string = tmp_notNumeric, pattern = "labjs|location", negate = TRUE)

# "bug_browser_used_additional"
# tmp_notNumeric

vec_ques <- c("dummy_type",
              "bug_tool",
              
              "bug_CAMEL_element",
              "bug_CAMApp_part",
              "bug_Webpage_part",
              
              "bug_browser_used",
              "automatic_browser", "automatic_browser2",
              "automatic_browserVersion", "automatic_operatingSystem",
              
              "bug_text",
              "Base64Image",
              "bug_code",
              
              
              "improvement_tool",
              
              "improvement_CAMEL_element",
              "improvement_CAMEL_text",
              "improvement_CAMEL_importance",
              
              "improvement_CAMApp_part",
              "improvement_CAMApp_text",
              "improvement_CAMApp_importance",
              
              "improvement_Webpage_part",
              "improvement_Webpage_text",
              "improvement_Webpage_importance",
              
              
              "feedback_critic")

vec_notNumeric = vec_ques
questionnaire <- questionnairetype(dataset = dat, 
                                        listvars = vec_ques, 
                                        notNumeric = vec_notNumeric)

questionnaire$bug_browser_used_additional <- NA
for(i in 1:nrow(questionnaire)){
  tmp_dat <- dat[dat$ID == unique(questionnaire$ID)[i], ]
  questionnaire$bug_browser_used_additional[i] <- paste0(unlist(tmp_dat$bug_browser_used_additional), collapse = " \\ ")
}
questionnaire$bug_browser_used_additional[questionnaire$bug_browser_used_additional == ""] <- NA


dat_pictures <- questionnaire[, c("ID", "Base64Image")]


questionnaire$Base64Image <- NULL

5 overview

table(questionnaire$dummy_type)


        bug improvement 
         23           3

table(questionnaire$bug_tool)


         CAMApp   CAMEL Webpage 
      3       5       9       9

table(questionnaire$bug_tool, questionnaire$bug_CAMEL_element)

         
            concept connection other
          3       0          0     0
  CAMApp  5       0          0     0
  CAMEL   0       5          2     2
  Webpage 9       0          0     0

table(questionnaire$bug_tool, questionnaire$bug_CAMApp_part)

         
            analysis preprocessing upload
          3        0             0      0
  CAMApp  0        3             1      1
  CAMEL   9        0             0      0
  Webpage 9        0             0      0

table(questionnaire$bug_tool, questionnaire$bug_Webpage_part)

         
            afterLogIn beforeLogIn other
          3          0           0     0
  CAMApp  5          0           0     0
  CAMEL   9          0           0     0
  Webpage 0          4           3     2

6 reported bugs

6.1 for CAMEL

setwd("outputs")

## dynamic table for CAMEL
tmp_dat <- questionnaire[questionnaire$bug_tool == "CAMEL",]
tmp_dat <- tmp_dat[, str_detect(string = colnames(tmp_dat), pattern = "improvement", negate = TRUE)]
tmp_dat <- tmp_dat[, !colnames(tmp_dat) %in% c("bug_CAMApp_part", "bug_Webpage_part")]
nrow(tmp_dat)

[1] 9

DT::datatable(tmp_dat, options = list(pageLength = 5))

#> save as .xlsx
xlsx::write.xlsx2(x = tmp_dat, file = "bugs_CAMEL.xlsx")

6.2 for CAMApp

setwd("outputs")

## dynamic table for CAMApp
tmp_dat <- questionnaire[questionnaire$bug_tool == "CAMApp",]
tmp_dat <- tmp_dat[, str_detect(string = colnames(tmp_dat), pattern = "improvement", negate = TRUE)]
tmp_dat <- tmp_dat[, !colnames(tmp_dat) %in% c("bug_CAMEL_element", "bug_Webpage_part")]
nrow(tmp_dat)

[1] 5

DT::datatable(tmp_dat, options = list(pageLength = 5))

#> save as .xlsx
xlsx::write.xlsx2(x = tmp_dat, file = "bugs_CAMApp.xlsx")

6.3 for Webpage

setwd("outputs")

## dynamic table for Webpage
tmp_dat <- questionnaire[questionnaire$bug_tool == "Webpage",]
tmp_dat <- tmp_dat[, str_detect(string = colnames(tmp_dat), pattern = "improvement", negate = TRUE)]
tmp_dat <- tmp_dat[, !colnames(tmp_dat) %in% c("bug_CAMEL_element", "bug_CAMApp_part")]
nrow(tmp_dat)

[1] 9

DT::datatable(tmp_dat, options = list(pageLength = 5))

#> save as .xlsx
xlsx::write.xlsx2(x = tmp_dat, file = "bugs_Webpage.xlsx")

7 save screenshots

setwd("outputs/screenshots")
for(i in 1:nrow(dat_pictures)){
  if(!is.na(dat_pictures$Base64Image[i])){
    if(str_detect(string = dat_pictures$Base64Image[i], pattern = "data:application/pdf;base64,")){
  dat_pictures$Base64Image[i] <- str_remove_all(string = dat_pictures$Base64Image[i], pattern = "data:application/pdf;base64,")
  tmp_name <- "pdf"
}else if(str_detect(string = dat_pictures$Base64Image[i], pattern = "data:image/png;base64,")){
  dat_pictures$Base64Image[i] <- str_remove_all(string = dat_pictures$Base64Image[i], pattern = "data:image/png;base64,")
  tmp_name <- "png"
}else if(str_detect(string = dat_pictures$Base64Image[i], pattern = "data:image/jpeg;base64,")){
    dat_pictures$Base64Image[i] <- str_remove_all(string = dat_pictures$Base64Image[i], pattern = "data:image/jpeg;base64,")
    tmp_name <- "jpeg"
}else{
  print(i)
}
    binary_data <- base64enc::base64decode(what = dat_pictures$Base64Image[i])
    writeBin(binary_data, paste0(dat_pictures$ID[i], "_screenshot.", tmp_name))
  }
}

[1] 7

8 reported possible improvements

setwd("outputs")


## dynamic table for suggested improvements
tmp_dat <- questionnaire[questionnaire$dummy_type == "improvement",]
tmp_dat <- tmp_dat[, str_detect(string = colnames(tmp_dat), pattern = "bug", negate = TRUE)]
tmp_dat <- tmp_dat[, !colnames(tmp_dat) %in% c("bug")]
nrow(tmp_dat)

[1] 3

DT::datatable(tmp_dat, options = list(pageLength = 5))

#> save as .xlsx
xlsx::write.xlsx2(x = tmp_dat, file = "improvements.xlsx")