library(tidyverse)

splitsfile <- "~/Downloads/RSA 2018 - DDSS - Presentation (5).lss"

splits <- XML::xmlParse(splitsfile)
splits <- XML::xmlToList(splits)

class(splits[['AttemptHistory']][1][['Attempt']][['.attrs']][['ended']])

chunk2 <- do.call(rbind, lapply(1:length(splits[['AttemptHistory']]), function(i){
    # print(i)
    attempt <- splits[['AttemptHistory']][[i]]
    # print(names(splits[['AttemptHistory']][[i]]))
    if ('RealTime' %in% names(attempt) | 'PauseTime' %in% names(attempt)) {
        df <- tibble(attemptID=attempt[['.attrs']][['id']], 
               start=attempt[['.attrs']][['started']], 
               end=attempt[['.attrs']][['ended']])
    } else {
        df <- tibble(attemptID=attempt[['id']], 
               start=attempt[['started']], 
               end=attempt[['ended']])
    }
    if ('RealTime' %in% names(attempt)) {
        df$runtime <- attempt[['RealTime']]
    } else {
        df$runtime <- NA
    }
    df
}))

chunk2$runtime <- unlist(lapply(chunk2$runtime, function(t) { 
    dt <- unlist(stringr::str_split(t, ":"))
    as.difftime(as.double(dt[1])*60*60 + as.double(dt[2])*60 + as.double(dt[3]), units="secs")
})) %>% as.difftime(units="secs")
chunk2$start <- lubridate::mdy_hms(chunk2$start)
chunk2$end <- lubridate::mdy_hms(chunk2$end)

chunk <- do.call(rbind, lapply(splits[['Segments']], function(segments) {
    
    segments.df <- do.call(rbind, lapply(segments[['SegmentHistory']], function(segment) {
        if ('RealTime' %in% names(segment))
        data.frame(`attemptID` = segment$.attrs['id'], 
                   RealTime = segment$RealTime,
                   time = segment$S)
    }))
    segments.df$name <- rep(segments$Name, nrow(segments.df))
    segments.df
}))

chunk$RealTime <- unlist(lapply(chunk$RealTime, function(t) { 
    dt <- unlist(stringr::str_split(t, ":"))
    as.difftime(as.double(dt[1])*60*60 + as.double(dt[2])*60 + as.double(dt[3]), units="secs")
})) %>% as.difftime(units="mins")
    
chunk$name <- factor(chunk$name, levels=unique(chunk$name))
    
chunk <- chunk %>%
    group_by(name) %>%
    mutate(n=n()) %>%
    ungroup() 
 
labels <- unique(paste0(chunk$name, " - ",  chunk$n, " attempts"))
names(labels) <- unique(chunk$name)
units(chunk$RealTime) <- "mins"



glimpse(chunk)
glimpse(chunk2)

ggplot(chunk) +
    geom_line(aes(x=attemptID, y=RealTime, group=name, color=name)) +
    labs(x="Run", y="Time in minutes", title="Presentation section times") +
    scale_y_continuous(expand=c(0,0)) +
    scale_x_discrete(expand=c(0,0)) +
    scale_color_discrete(guide=guide_legend(title="Section", title.position="top")) +
    theme_classic() +
    theme(
       axis.title.y = element_text(hjust=0),
       axis.title.x = element_text(hjust=0),
       panel.grid.major.y = element_line(color="gray65", size=rel(0.5), linetype="dashed"),
       panel.grid.minor.y = element_line(color="gray85", size=rel(0.5), linetype="dashed"),
       axis.ticks = element_blank(),
       legend.position="bottom"
    )

chunk %>%
    filter(attemptID %in% 7:20) %>% # remove columns that are incomplete
    filter(!(name == "Introduction VMOS and Swot" & attemptID %in% c("9", "14", "18"))) %>% # filter outliers
  ggplot(aes(x=attemptID, y=RealTime, group=name, color=name)) +
    geom_point() +
    geom_smooth() +
    facet_wrap(~name) +
    labs(x="Run", y="Time in minutes", title="Presentation section times") +
    scale_y_continuous(expand=c(0,0)) +
    scale_x_discrete(expand=c(0,0)) +
    scale_color_discrete(guide=guide_legend(title="Section", title.position="top")) +
    theme_classic() +
    theme(
       axis.title.y = element_text(hjust=0),
       axis.title.x = element_text(hjust=0),
       panel.grid.major.y = element_line(color="gray65", size=rel(0.5), linetype="dashed"),
       panel.grid.minor.y = element_line(color="gray85", size=rel(0.5), linetype="dashed"),
       axis.ticks = element_blank(),
       legend.position="bottom"
    )

chunk %>%
    filter(attemptID %in% 7:20) %>% # remove columns that are incomplete
    filter(!(name == "Introduction VMOS and Swot" & attemptID %in% c("9", "14", "18"))) %>% # filter outliers
  ggplot(aes(x=attemptID, y=RealTime, group=forcats::fct_rev(name), color=name)) +
    geom_line(position="stack") +
    labs(x="Run", y="Time in minutes", title="Stacked presentation section times") +
    scale_y_continuous(expand=c(0,0)) +
    scale_x_discrete(expand=c(0,0)) +
    scale_color_discrete(guide=guide_legend(title="Section", title.position="top")) +
    theme_classic() +
    theme(
       axis.title.y = element_text(hjust=0),
       axis.title.x = element_text(hjust=0),
       panel.grid.major.y = element_line(color="gray65", size=rel(0.5), linetype="dashed"),
       panel.grid.minor.y = element_line(color="gray85", size=rel(0.5), linetype="dashed"),
       axis.ticks = element_blank(),
       legend.position="bottom"
    )


chunk2 %>%
    mutate(runtime = as.difftime(runtime/60, units="mins")) %>%
    filter(!is.na(runtime)) %>%
    mutate(attemptID = as.integer(attemptID)) %>%
    # filter(runtime > 20) %>%
    glimpse() %>%
  ggplot() +
    geom_line(aes(x=attemptID, y=runtime, group=1)) +
    # annotate("segment", x=-Inf, xend=Inf, y=0, yend=0) +
    labs(x="Run", y="Time in minutes", title="Overall run time") +
    scale_y_continuous(expand=c(0,0)) + #, limits=c(0,NA)) +
    scale_x_continuous(expand=c(0,0), limits=c(3,NA)) +
    theme_classic() +
    theme(
       axis.title.y = element_text(hjust=0),
       axis.title.x = element_text(hjust=0),
       panel.grid.major.y = element_line(color="gray65", size=rel(0.5), linetype="dashed"),
       panel.grid.minor.y = element_line(color="gray85", size=rel(0.5), linetype="dashed"),
       axis.ticks = element_blank()
    )
    

# compare runtime to derived time
## doesn't look like runtime is accurate and the NA's are when the run wasn't completed
chunk2 %>%
    mutate(runtime.derived = end-start) %>%
    mutate(diff = runtime.derived-runtime)

chunk2 %>%
    mutate(runtime = as.difftime(runtime/60, units="mins")) %>%
    filter(!is.na(runtime)) %>%
    mutate(attemptID = as.integer(attemptID)) %>%
    # filter(runtime > 20) %>%
    glimpse() %>%
  ggplot() +
    geom_line(aes(x=start, y=runtime, group=1)) +
    geom_point(aes(x=start, y=runtime, group=1)) +
    # annotate("segment", x=-Inf, xend=Inf, y=0, yend=0) +
    labs(x="Date", y="Time in minutes", title="Overall run time") +
    scale_y_continuous(expand=c(0,0)) + #, limits=c(0,NA)) +
    scale_x_datetime(expand=c(0,0)) +
    theme_classic() +
    theme(
       axis.title.y = element_text(hjust=0),
       axis.title.x = element_text(hjust=0),
       panel.grid.major.y = element_line(color="gray65", size=rel(0.5), linetype="dashed"),
       panel.grid.minor.y = element_line(color="gray85", size=rel(0.5), linetype="dashed"),
       axis.ticks = element_blank()
    )


chunk %>%
    left_join(chunk2, by="attemptID") %>%
    filter(attemptID %in% 7:20) %>% # remove columns that are incomplete
    filter(!(name == "Introduction VMOS and Swot" & attemptID %in% c("9", "14", "18"))) %>% # filter outliers
  ggplot(aes(x=start, y=RealTime, group=name, color=name)) +
    geom_point() +
    geom_smooth() +
    facet_wrap(~name) +
    labs(x="Date", y="Time in minutes", title="Presentation section times") +
    scale_y_continuous(expand=c(0,0)) +
    scale_x_datetime(expand=c(0,0)) +
    scale_color_discrete(guide=guide_legend(title="Section", title.position="top")) +
    theme_classic() +
    theme(
       axis.title.y = element_text(hjust=0),
       axis.title.x = element_text(hjust=0),
       panel.grid.major.y = element_line(color="gray65", size=rel(0.5), linetype="dashed"),
       panel.grid.minor.y = element_line(color="gray85", size=rel(0.5), linetype="dashed"),
       axis.ticks = element_blank(),
       legend.position="bottom"
    )

chunk %>%
    left_join(chunk2, by="attemptID") %>%
    filter(attemptID %in% 7:20) %>% # remove columns that are incomplete
    filter(!(name == "Introduction VMOS and Swot" & attemptID %in% c("9", "14", "18"))) %>% # filter outliers
  ggplot(aes(x=start, y=RealTime, group=forcats::fct_rev(name), color=name)) +
    geom_line(position="stack") +
    labs(x="Date", y="Time in minutes", title="Stacked presentation section times") +
    scale_y_continuous(expand=c(0,0)) +
    scale_x_datetime(expand=c(0,0)) +
    scale_color_discrete(guide=guide_legend(title="Section", title.position="top")) +
    theme_classic() +
    theme(
       axis.title.y = element_text(hjust=0),
       axis.title.x = element_text(hjust=0),
       panel.grid.major.y = element_line(color="gray65", size=rel(0.5), linetype="dashed"),
       panel.grid.minor.y = element_line(color="gray85", size=rel(0.5), linetype="dashed"),
       axis.ticks = element_blank(),
       legend.position="bottom"
    )