# setwd("C:/Users/nwickelmaier/Nextcloud/Documents/MDS/2023ss/60100_master_thesis/code")

# Read data

# dat0 <- read.table("../data/haum/event_logfiles_metadata_2023-09-23_01-31-30.csv",
#                   sep = ";", header = TRUE)
dat0 <- read.table("../data/haum/event_logfiles_small_metadata_2023-10-15_10-08-43.csv",
                  sep = ";", header = TRUE)
dat0$date <- as.Date(dat0$date)
dat0$date.start <- as.POSIXct(dat0$date.start)
dat0$date.stop <- as.POSIXct(dat0$date.stop)
dat0$artwork <- sprintf("%03d", dat0$artwork)

# TODO: Write a function that closes events spanning different log files
# OR: Remove openTopic and OpenPopup events that do not start with a
# flipCard (AND openPopup events without openTopic event beforehand)

table(dat0[!duplicated(dat0$trace), "event"])
# flipCard      move openPopup openTopic 
#    45528    247718       981      3457 
proportions(table(dat0[!duplicated(dat0$trace), "event"]))

tmp <- dat0[!duplicated(dat0$trace) & dat0$event %in% c("openTopic",
                                                        "openPopup"), ]

dat <- dat0
i <- 1
stop <- 1

while (stop > 0) {
  stop <- sum(!duplicated(dat$trace) & dat$event %in% c("openTopic", "openPopup"))
  dat <- dat[!(!duplicated(dat$trace) &
              dat$event %in% c("openTopic", "openPopup")), ]
  print(i)
  i <- i + 1
  print(table(dat[!duplicated(dat$trace), "event"]))
}

# Creating event logs

library(bupaverse)

names(dat)[names(dat) %in% c("date.start", "date.stop")] <- c("start",
                                                              "complete")

table(table(dat$start))
# --> hmm...

summary(aggregate(duration ~ trace, dat, mean))

# remove fragmented traces
tab <- xtabs( ~ trace + event, dat)

fragments <- NULL

for (i in seq_len(nrow(tab))) {
  if (tab[i, "openPopup"] != 0 & tab[i, "flipCard"] == 0) {
    fragments <- c(fragments, rownames(tab)[i])
  } else if (tab[i, "openTopic"] != 0 & tab[i, "flipCard"] == 0) {
    fragments <- c(fragments, rownames(tab)[i])
  } else if (tab[i, "openPopup"] != 0 & tab[i, "openTopic"] == 0) {
    fragments <- c(fragments, rownames(tab)[i])
  }
}
datrm <- dat[!dat$trace %in% fragments, ]

# TODO: Find trace that has flipCard --> openPopup --> openTopic
dato <- datrm[datrm$event != "move", ]
tmp <- lapply(unique(dato$trace), function(x) unique(dato[dato$trace == x, "event"]))
names(tmp) <- unique(dato$trace)

ids <- sapply(tmp, length) == 3
do.call(rbind, tmp[ids])

# TODO:
# fragmentary traces: for 4591 openTopic for topic 1 is in the raw log files, but gets
# probably removed in close_events(); how can I prevent that? How can I fix
# the traces and eventIds that do not match correctly ???
ct <- c(4591, 5937, 7080, 8412, 8279)
datrm[datrm$trace %in% ct, 1:10]

### WHY?????

alog <- activitylog(datrm,
                    case_id = "trace",
                    activity_id = "event",
                    #resource_id = "case",
                    resource_id = "artwork",
                    timestamps = c("start", "complete"))

process_map(alog)

process_map(alog, frequency("relative"))
process_map(alog, frequency("relative_consequent"))

library(processanimateR)

animate_process(to_eventlog(alog))

col_vector <- c("#7FC97F", "#BEAED4", "#FDC086", "#FFFF99", "#386CB0",
                "#F0027F", "#BF5B17", "#666666", "#1B9E77", "#D95F02",
                "#7570B3", "#E7298A", "#66A61E", "#E6AB02", "#A6761D",
                "#666666", "#A6CEE3", "#1F78B4", "#B2DF8A", "#33A02C",
                "#FB9A99", "#E31A1C", "#FDBF6F", "#FF7F00", "#CAB2D6",
                "#6A3D9A", "#FFFF99", "#B15928", "#FBB4AE", "#B3CDE3",
                "#CCEBC5", "#DECBE4", "#FED9A6", "#FFFFCC", "#E5D8BD",
                "#FDDAEC", "#F2F2F2", "#B3E2CD", "#FDCDAC", "#CBD5E8",
                "#F4CAE4", "#E6F5C9", "#FFF2AE", "#F1E2CC", "#CCCCCC",
                "#E41A1C", "#377EB8", "#4DAF4A", "#984EA3", "#FF7F00",
                "#FFFF33", "#A65628", "#F781BF", "#999999", "#66C2A5",
                "#FC8D62", "#8DA0CB", "#E78AC3", "#A6D854", "#FFD92F",
                "#E5C494", "#B3B3B3", "#8DD3C7", "#FFFFB3", "#BEBADA",
                "#FB8072", "#80B1D3", "#FDB462", "#B3DE69", "#FCCDE5",
                "#D9D9D9")

animate_process(to_eventlog(alog), mode = "relative", jitter = 10, legend = "color",
  mapping = token_aes(color = token_scale("artwork", 
    scale = "ordinal", 
    range = col_vector)))

elog <- to_eventlog(alog)
animate_process(elog[elog$artwork == "054", ])
animate_process(elog[elog$artwork == "080", ])
animate_process(elog[elog$artwork == "501", ])

process_map(alog[alog$artwork == "054", ])

animate_process(elog[elog$artwork %in% c("080", "054"), ],
                mode = "relative", jitter = 10, legend = "color",
                mapping = token_aes(color = token_scale("artwork",
                                                scale = "ordinal",
                                                range = c("black", "gray"))))
# --> not sure, yet, how to interpret this...

alog080 <- activitylog(dat[dat$artwork %in% "080", ],
                    #case_id = "case",
                    case_id = "trace",
                    activity_id = "event",
                    #resource_id = "trace",
                    resource_id = "case",
                    timestamps = c("start", "complete"))

process_map(alog080, frequency("relative"))



alog054 <- activitylog(dat[dat$artwork %in% "054", ],
                    #case_id = "case",
                    case_id = "trace",
                    activity_id = "event",
                    #resource_id = "trace",
                    resource_id = "case",
                    timestamps = c("start", "complete"))

process_map(alog054, frequency("relative"))