#----------------------------------------------------------------
#-------------------------General evolution----------------------
#----------------------------------------------------------------


#' Return graph representing the general attacks evolution by year
#'
#' @param Attacks data.frame with procesed source data
#' @param point.vjust Vector of vertical just to each point label
#' @param point.hjust Vector of Horizontal just to each point label
#' @param iso2013.y Vertical position to ISO27001:2013 label
#' @param iso2013.label.hjust Horizontal just to ISO27001:2013 label
#' @param smooth.x Horizontal position to smooth slope label
#' @param smooth.y Vertical position to smooth slope label
#' @param smooth.label.hjust Horizontal just to smooth slope label
#'
#' @return list(graph, slope)
#' @export
GetAttacksEvolution <- function(Attacks,
                                point.vjust = 0,
                                point.hjust = 0,
                                iso2013.y = 0,
                                iso2013.label.hjust = 0,
                                smooth.x = 0,
                                smooth.y = 0,
                                smooth.label.hjust = 0){

  #Extracting year from date
  attacks.evol <- dplyr::mutate(Attacks, Year = format(Attacks$Date, "%Y")) %>%
                  dplyr::group_by(Year)#Grouping by year
  #Counting attacks each year
  attacks.evol <- as.data.frame(table(attacks.evol$Year))
  colnames(attacks.evol) <- c("Year","Attacks")

  #Slope of smooth
  attacks.evol.slope <- lm(formula = Attacks ~ as.numeric(Year),
                           data = attacks.evol)$coef[[2]]
  #Graph
  attacks.evol.graph <- ggplot2::qplot(main = "Cyberattacks evolution",
                                       x = attacks.evol$Year,
                                       y = attacks.evol$Attacks,
                                       group = 1,
                                       xlab = "Years",
                                       ylab = "Attacks",
                                       data = attacks.evol,
                                       geom = "line")  +
                        ggplot2::geom_point() +
                        ggplot2::geom_label(aes(label=attacks.evol$Attacks),
                                   vjust = point.vjust,
                                   hjust = point.hjust)+
                        ggplot2::theme(plot.title = element_text(hjust = 0.5)) +
                        ggplot2::geom_smooth(method = "lm", se = FALSE) +
                        ggplot2::geom_label(aes(x = smooth.x, y = smooth.y,
                                       label = paste("Slope =",
                                                     signif(attacks.evol.slope))),
                                   color = "blue",
                                   hjust = smooth.label.hjust) +
                        ggplot2::geom_vline(xintercept = 3,
                                            color = "Red",
                                            linetype = "longdash") +
                        ggplot2::geom_label(aes(x = "2014", y = iso2013.y,
                                                label = "ISO27001:2013 effects"),
                                            color = "Red",
                                            hjust = iso2013.label.hjust)

  list(attacks.evol.graph, attacks.evol.slope)

}

#' Return graph representing the general ISO27001 evolution
#'
#' @param Cert_PerCountry data.frame with procesed source data
#' @param point.vjust Vector of vertical just to each point label
#' @param point.hjust Vector of Horizontal just to each point label
#' @param iso2013.y Vertical position to ISO27001:2013 label
#' @param iso2013.label.hjust Horizontal just to ISO27001:2013 label
#' @param smooth.x Horizontal position to smooth slope label
#' @param smooth.y Vertical position to smooth slope label
#' @param smooth.label.hjust Horizontal just to smooth slope label
#'
#' @return list(graph, slope)
#' @export
GetCertsEvolution <- function(Cert_PerCountry,
                              point.vjust = 0,
                              point.hjust = 0,
                              iso2013.y = 0,
                              iso2013.label.hjust = 0,
                              smooth.x = 0,
                              smooth.y = 0,
                              smooth.label.hjust = 0){

  #Years from columns to rows in 1 column
  Certs.evol <- tidyr::gather(Cert_PerCountry, "Year", "Count", 2:6)
  #Group by year
  Certs.evol <- dplyr::group_by(Certs.evol, Year)
  #Sum counts to 1 row for each year
  Certs.evol <- dplyr::summarise(Certs.evol, Count = sum(Count))
  #Removing X from years
  Certs.evol$Year <- substr(Certs.evol$Year, 2, 5)

  #Slope of smooth
  Certs.evol.slope <- lm(formula = Count ~ as.numeric(Year),
                           data = Certs.evol)$coef[[2]]
  #Graph
  Certs.evol.graph <- ggplot2::qplot(main = "ISO 27001 evolution",
                           x = Certs.evol$Year,
                           y = Certs.evol$Count,
                           group = 1,
                           xlab = "Years",
                           ylab = "Certifications",
                           data = Certs.evol,
                          geom = "line") +
            geom_point() +
            geom_label(aes(label=Certs.evol$Count),
                       vjust = point.vjust,
                       hjust = point.hjust)+
            theme(plot.title = element_text(hjust = 0.5)) +
            ggplot2::geom_smooth(method = "lm", se = FALSE) +
            ggplot2::geom_label(aes(x = smooth.x, y = smooth.y,
                                    label = paste("Slope =",
                                                  signif(Certs.evol.slope))),
                                color = "blue",
                                hjust = smooth.label.hjust)+
            ggplot2::geom_vline(xintercept = 3,
                                color = "Red",
                                linetype = "longdash") +
            ggplot2::geom_label(aes(x = "2014", y = iso2013.y,
                                    label = "ISO27001:2013 changes"),
                                color = "Red",
                                hjust = iso2013.label.hjust)

  list(Certs.evol.graph, Certs.evol.slope)

}

#' Return graph representing the general attacks evolution by month
#'
#' @param Attacks data.frame with procesed source data
#'
#' @return graph
#' @export
GetAttacksMonthEvolution <- function(Attacks){

  #Extract Year - Month from date
  attacks.evol <- mutate(Attacks, Year = format(Attacks$Date, "%Y-%m")) %>%
                  group_by(Year)
  #Group by Year - Month
  attacks.evol <- as.data.frame(table(attacks.evol$Year))
  colnames(attacks.evol) <- c("Year","Attacks")

  #Graph
  ggplot2::qplot(main = "Cyberattacks evolution",
                 x = attacks.evol$Year,
                 y = attacks.evol$Attacks,
                 group = 1,
                 xlab = "Months",
                 ylab = "Attacks",
                 data = attacks.evol,
                 geom = "line")  +
    geom_point() +
    theme(plot.title = element_text(hjust = 0.5)) +
    geom_smooth(method = 'loess') +
    scale_x_discrete(labels = c("2012/01", "", "", "", "", "", "2012/07", "", "", "", "", "",
                                "2013/01", "", "", "", "", "", "2013/07", "", "", "", "", "",
                                "2014/01", "", "", "", "", "", "2014/07", "", "", "", "", "",
                                "2015/01", "", "", "", "", "", "2015/07", "", "", "", "", "",
                                "2016/01", "", "", "", "", "", "2016/07", "", "", "", "", ""))


}

#----------------------------------------------------------------
#-------------------------Attack type evolution------------------
#----------------------------------------------------------------


#' Return pie graph to show % of each attack type
#'
#' @param Attacks data.frame with procesed source data
#' @param label.vjust Vector of vertical just to each portion label
#' @param label.hjust Vector of horizontal just to each portion label
#'
#' @return list(graph, AttackTypeShowedList)
#' @export
GetAttackTypePie <- function (Attacks,
                              label.vjust = 0,
                              label.hjust = 0){

  #Group by attack type
  attack.pie <- group_by(Attacks, Attack.standar)
  #Counting rows for each attack type
  attack.pie <- as.data.frame(table(attack.pie$Attack.standar))
  attack.pie <- setNames(attack.pie, c("Attack", "Count"))

  #Removing rows without attack type defined
  attack.pie <- attack.pie[attack.pie$Attack != "",]
  #Removing attacks with less than 1% of representation from total
  attack.pie <- attack.pie[attack.pie$Count > (sum(attack.pie$Count) * 0.01),]
  #Calc % of each attack type
  attack.pie <- mutate(attack.pie, Perc=paste(round(100 * attack.pie$Count / sum(attack.pie$Count), 2), "%"))


  #Graph
  graph1 <- ggplot(data=attack.pie,
                  aes(x=factor(1),
                   y=Count,
                   fill=Attack)) +
            geom_col(width = 1, color='black') +
            geom_label(aes(label = attack.pie$Perc),
                        vjust = label.vjust,
                        hjust = label.hjust) +
            coord_polar(theta="y") +
            scale_x_discrete(labels = c("")) +
            scale_y_discrete(labels = c("")) +
            theme(plot.title = element_text(hjust = 0.5),
                  axis.title.x=element_blank(),
                  axis.title.y=element_blank()) +
            ggtitle("Attack types")

  #Returning graph and attack list
  list(graph1, unique(attack.pie$Attack))
}


#' Return graph to show the evolution of a attack types list
#'
#' @param Attacks data.frame with procesed source data
#' @param TypeList List with attack types to show
#'
#' @return list(graph, data.frame with TypeList attacks data by year)
#' @export
GetAttackTypeEvolution <- function(Attacks, TypeList){
  #Obtaining year from date
  Attacks.pre <- dplyr::mutate(Attacks, Year = format(Attacks$Date, "%Y")) %>%
                group_by(Year, Attack.standar) #group by attack type
  #Count rows for each attack type and each year
  Attacks.pre <- as.data.frame(table(Attacks.pre$Year, Attacks.pre$Attack.standar))
  Attacks.pre <- setNames(Attacks.pre, c("Year", "Attack", "Count"))
  #Removing rows without an attack type specified
  Attacks.pre <- Attacks.pre[Attacks.pre$Attack %in% TypeList,]

  #Graph
  graph1 <- ggplot(data = Attacks.pre,
                   aes(x = Year,
                       y = Count,
                       color = Attack,
                       group = Attack)) +
            geom_point() +
            geom_line() +
            theme(plot.title = element_text(hjust = 0.5)) +
            ggtitle("Attack type evolution") +
            labs(colour = "Attack type") + xlab("Years") + ylab("Attacks")
  #Return graph and used data to separate representation
  list(graph1, Attacks.pre)

  }


#' Return graph to show the evolution of a single attack type and his smooth
#'
#' @param Attacks data.frame with procesed source data
#' @param AttackType Attack type to represent
#' @param smooth.x Horizontal position to smooth slope label
#' @param smooth.y Vertical position to smooth slope label
#' @param smooth.label.hjust Horizontal just to smooth slope label
#'
#' @return list(graph, slope)
#' @export
GetAttackTypeSigleEvolution <- function(Attacks,
                                        AttackType,
                                        smooth.x = 0,
                                        smooth.y = 0,
                                        smooth.label.hjust = 0){
  #Filtering for the AttackType specified
  attacks.evol <- Attacks[Attacks$Attack == AttackType,]

  #Cal slope
  slope1 <- lm(formula = Count ~ as.numeric(Year), data = attacks.evol)$coef[[2]]
  #Graph
  graph1 <- ggplot(data = attacks.evol,
                   aes(x = Year, y = Count, group = 1)) +
            geom_line() +
            geom_point() +
            theme(plot.title = element_text(hjust = 0.5)) +
            ggtitle(AttackType) +
            xlab("Years") + ylab("Attacks")+
            geom_smooth(method = "lm",
                        se = FALSE) +
            geom_label(aes(x = smooth.x, y = smooth.y,
                           label = paste("Slope =", signif(slope1, 5))),
                       color = "blue",
                       hjust = smooth.label.hjust)

  #Returning list(graph, slope)
  list(graph1, slope1)

}

#----------------------------------------------------------------
#-------------------------Geolocal evolution---------
#----------------------------------------------------------------

GetContinentPie <- function (Attacks,
                             Attacks.label.vjust = 0,
                             Attacks.label.hjust = 0,
                             Cert_PerCountry,
                             Certs.label.vjust = 0,
                             Certs.label.hjust = 0){

  #Group attacks by continent
  attack.pie <- group_by(Attacks, Continent)
  #Count attacks for each continent
  attack.pie <- as.data.frame(table(attack.pie$Continent))
  attack.pie <- setNames(attack.pie, c("Continent", "Count"))
  #Remove rows without Continent specified
  attack.pie <- attack.pie[attack.pie$Continent != "",]
  #Calc % of each continent
  attack.pie <- mutate(attack.pie, perc = round(100 * attack.pie$Count / sum(attack.pie$Count), 2))

  #Attacks graph
  graph1 <- ggplot(data=attack.pie,
                   aes(x=factor(1),
                       y=Count,
                       fill=Continent)) +
            geom_col(width = 1, color='black') +
            geom_label(aes(label=paste(attack.pie$perc, "%")),
                           vjust = Attacks.label.vjust,
                           hjust = Attacks.label.hjust) +
            coord_polar(theta="y") +
            scale_x_discrete(labels = c("")) +
            scale_y_discrete(labels = c("")) +
            theme(plot.title = element_text(hjust = 0.5),
                  axis.title.x=element_blank(),
                  axis.title.y=element_blank()) +
            ggtitle("Attacks")


  #Grouping certifications by continent
  cert.pie <- group_by(Cert_PerCountry, Continent)
  #Counting certificates for each continent
  cert.pie <- dplyr::summarise(cert.pie, Count = sum(X2011 + X2012 + X2013 + X2014 + X2015))
  #Remove rows without Continent specified
  cert.pie <- cert.pie[cert.pie$Continent != "",]
  #Calc % of each continent
  cert.pie <- mutate(cert.pie, perc = round(100 * cert.pie$Count / sum(cert.pie$Count), 2))

  #Certifications graph
  graph2 <- ggplot(data=cert.pie,
                   aes(x=factor(1),
                       y=Count,
                       fill=Continent)) +
            geom_col(width = 1, color='black') +
            geom_label(aes(label=paste(cert.pie$perc, "%")),
                           vjust = Certs.label.vjust,
                           hjust = Certs.label.hjust) +
            coord_polar(theta="y") +
            scale_x_discrete(labels = c("")) +
            scale_y_discrete(labels = c("")) +
            theme(plot.title = element_text(hjust = 0.5),
                  axis.title.x=element_blank(),
                  axis.title.y=element_blank()) +
            ggtitle("ISO 27001")

  list(graph1, unique(attack.pie[attack.pie$perc > 5,]$Continent),
       graph2, unique(cert.pie[cert.pie$perc > 5,]$Continent))
}

GetContinentAttacksEvolution <- function(Attacks){

  attacks.evol <- mutate(Attacks, Year = format(Attacks$Date, "%Y")) %>% group_by(Continent, Year)
  attacks.evol <- as.data.frame(table(attacks.evol$Continent, attacks.evol$Year))
  colnames(attacks.evol) <- c("Continent", "Year","Attacks")

  attacks.evol <- attacks.evol[attacks.evol$Continent != "Oceania",]
  attacks.evol <- attacks.evol[attacks.evol$Continent != "Africa",]

  graph1 <- ggplot2::qplot(main = "Cyberattacks evolution",
                           x = attacks.evol$Year,
                           y = attacks.evol$Attacks,
                           group = Continent,
                           xlab = "Years",
                           ylab = "Attacks",
                           data = attacks.evol,
                           geom = "point",
                           color = Continent)  +
    geom_line() +
    theme(plot.title = element_text(hjust = 0.5))

  graph1

}

GetContinentCertsEvolution <- function(Cert_PerCountry){

  certs.evol <- gather(Cert_PerCountry, "Year", "Certs", 2:6) %>% group_by(Continent, Year)

  certs.evol <- summarise(certs.evol,
                          #Continent = Continent,
                          #Year = Year,
                          Certs = sum(Certs))
  certs.evol$Year <- substr(certs.evol$Year,2,5)

  certs.evol <- certs.evol[certs.evol$Continent != "Oceania",]
  certs.evol <- certs.evol[certs.evol$Continent != "Africa",]

  graph1 <- ggplot2::qplot(main = "ISO 27001 evolution",
                           x = certs.evol$Year,
                           y = certs.evol$Certs,
                           group = Continent,
                           xlab = "Years",
                           ylab = "Certifications",
                           data = certs.evol,
                           geom = "point",
                           color = Continent)  +
    geom_line() +
    theme(plot.title = element_text(hjust = 0.5))

  graph1

}

GetContinentAttacksTopEvolution <- function(Attacks){
  attacks.evol <- mutate(Attacks, Year = format(Attacks$Date, "%Y")) %>% group_by(Year, Continent)
  attacks.evol <- as.data.frame(table(attacks.evol$Year, attacks.evol$Continent))
  attacks.evol <- setNames(attacks.evol, c("Year", "Continent", "Count"))
  attacks.evol <- attacks.evol[attacks.evol$Continent != "",]

  attacks.evol <- spread(attacks.evol, "Continent", "Count")

  graph1 <- ggplot(data = attacks.evol,
                   aes(x = Year, y = Americas, group = 1)) +

    geom_line() +
    geom_point() +

    theme(plot.title = element_text(hjust = 0.5)) +
    ggtitle("Americas") +
    xlab("Years") + ylab("Attacks")+
    stat_smooth(method = "lm", se = FALSE, aes(outfit=fit<<-..y..))

  graph2 <- ggplot(data = attacks.evol,
                   aes(x = Year, y = Asia, group = 1)) +

    geom_line() +
    geom_point() +

    theme(plot.title = element_text(hjust = 0.5)) +
    ggtitle("Asia") +
    xlab("Years") + ylab("Attacks")+
    stat_smooth(method = "lm", se = FALSE, aes(outfit=fit<<-..y..))

  graph3 <- ggplot(data = attacks.evol,
                   aes(x = Year, y = Europe, group = 1)) +

    geom_line() +
    geom_point() +

    theme(plot.title = element_text(hjust = 0.5)) +
    ggtitle("Europe") +
    xlab("Years") + ylab("Attacks")+
    stat_smooth(method = "lm", se = FALSE, aes(outfit=fit<<-..y..))


  list(graph1, graph2, graph3)

}

GetContinentCertsTopEvolution <- function(Attacks){

  certs.evol <- gather(Cert_PerCountry, "Year", "Certs", 2:6) %>% group_by(Continent, Year)

  certs.evol <- summarise(certs.evol,
                          #Continent = Continent,
                          #Year = Year,
                          Certs = sum(Certs))
  certs.evol$Year <- substr(certs.evol$Year,2,5)

  certs.evol <- spread(certs.evol, "Continent", "Certs")

  graph1 <- ggplot(data = certs.evol,
                   aes(x = Year, y = Americas, group = 1)) +

    geom_line() +
    geom_point() +

    theme(plot.title = element_text(hjust = 0.5)) +
    ggtitle("Americas") +
    xlab("Years") + ylab("Certifications")+
    stat_smooth(method = "lm", se = FALSE, aes(outfit=fit<<-..y..))

  graph2 <- ggplot(data = certs.evol,
                   aes(x = Year, y = Asia, group = 1)) +

    geom_line() +
    geom_point() +

    theme(plot.title = element_text(hjust = 0.5)) +
    ggtitle("Asia") +
    xlab("Years") + ylab("Certifications")+
    stat_smooth(method = "lm", se = FALSE, aes(outfit=fit<<-..y..))

  graph3 <- ggplot(data = certs.evol,
                   aes(x = Year, y = Europe, group = 1)) +

    geom_line() +
    geom_point() +

    theme(plot.title = element_text(hjust = 0.5)) +
    ggtitle("Europe") +
    xlab("Years") + ylab("Certifications")+
    stat_smooth(method = "lm", se = FALSE, aes(outfit=fit<<-..y..))


  list(graph1, graph2, graph3)

}

GetCountriesCol <- function(Attacks, Cert_PerCountry){

  certs.evol <- gather(Cert_PerCountry, "Year", "Certs", 2:6) %>% group_by(Continent, country_short)

  certs.evol <- summarise(certs.evol,
                          Certs = sum(Certs))

  graph1 <- ggplot2::ggplot(aes(x = reorder(country_short, Certs),
                                y = Certs),
                           data = certs.evol[certs.evol$Certs > (sum(certs.evol$Certs) * 0.02),]) +
          geom_col(aes(fill = Continent)) +
          theme(plot.title = element_text(hjust = 0.5)) +
          ggtitle("ISO 27001") +
          xlab("Country") + ylab("Certifications")

  attacks.evol <- mutate(Attacks, Year = format(Attacks$Date, "%Y")) %>% group_by(Year, Continent, Country)
  attacks.evol <- as.data.frame(table(attacks.evol$Year, attacks.evol$Continent, attacks.evol$Country))
  attacks.evol <- setNames(attacks.evol, c("Year", "Continent", "Country", "Count"))
  attacks.evol <- attacks.evol[attacks.evol$Continent != "",]
  attacks.evol <- attacks.evol[attacks.evol$Country != "",]
  attacks.evol <- attacks.evol[attacks.evol$Country != "AU",]

  attacks.evol <- attacks.evol[attacks.evol$Count > (sum((attacks.evol[attacks.evol$Country != "US",])$Count) * 0.01),]
  attacks.evol <- arrange(attacks.evol, Count)


  graph2 <- ggplot2::ggplot(aes(x = reorder(Country, Count),
                                y = Count),
                            data = attacks.evol) +
          geom_col(aes(fill = Continent)) +
          theme(plot.title = element_text(hjust = 0.5)) +
          ggtitle("Attacks") +
          xlab("Country") + ylab("Attacks")

  list(graph1, graph2)

  }

GetCountriesAttacksTopEvolution <- function(Attacks){
  attacks.evol <- mutate(Attacks, Year = format(Attacks$Date, "%Y")) %>% group_by(Year, Country)
  attacks.evol <- as.data.frame(table(attacks.evol$Year, attacks.evol$Country))
  attacks.evol <- setNames(attacks.evol, c("Year", "Country", "Count"))
  attacks.evol <- attacks.evol[attacks.evol$Country != "",]

  attacks.evol <- spread(attacks.evol, "Country", "Count")

  slope1 <- lm(formula = US ~ as.numeric(Year), data = attacks.evol)
  graph1 <- ggplot(data = attacks.evol,
                   aes(x = Year, y = US, group = 1)) +

    geom_line() +
    geom_point() +

    theme(plot.title = element_text(hjust = 0.5)) +
    ggtitle("US") +
    xlab("Years") + ylab("Attacks")+
    stat_smooth(method = "lm", se = FALSE, aes(outfit=fit<<-..y..)) +
    geom_label(aes(x = "2014", y = 700, label = signif(slope1$coef[[2]])), color = "blue")

  graph2 <- ggplot(data = attacks.evol,
                   aes(x = Year, y = GB, group = 1)) +

    geom_line() +
    geom_point() +

    theme(plot.title = element_text(hjust = 0.5)) +
    ggtitle("GB") +
    xlab("Years") + ylab("Attacks")+
    stat_smooth(method = "lm", se = FALSE, aes(outfit=fit<<-..y..))

  graph3 <- ggplot(data = attacks.evol,
                   aes(x = Year, y = IN, group = 1)) +

    geom_line() +
    geom_point() +

    theme(plot.title = element_text(hjust = 0.5)) +
    ggtitle("IN") +
    xlab("Years") + ylab("Attacks")+
    stat_smooth(method = "lm", se = FALSE, aes(outfit=fit<<-..y..))

  graph4 <- ggplot(data = attacks.evol,
                   aes(x = Year, y = JP, group = 1)) +

    geom_line() +
    geom_point() +

    theme(plot.title = element_text(hjust = 0.5)) +
    ggtitle("JP") +
    xlab("Years") + ylab("Attacks")+
    stat_smooth(method = "lm", se = FALSE, aes(outfit=fit<<-..y..))


  list(graph1, graph2, graph3, graph4)

}

GetCountriesCertsTopEvolution <- function(Attacks){

  certs.evol <- gather(Cert_PerCountry, "Year", "Certs", 2:6) %>% group_by(country_short, Year)

  certs.evol <- rbind(certs.evol[certs.evol$country_short == "US",],
                      certs.evol[certs.evol$country_short == "GB",],
                      certs.evol[certs.evol$country_short == "IN",],
                      certs.evol[certs.evol$country_short == "JP",])

  certs.evol <- summarise(certs.evol,
                          Certs = sum(Certs))
  certs.evol$Year <- substr(certs.evol$Year,2,5)

  certs.evol <- spread(certs.evol, "country_short", "Certs")

  graph1 <- ggplot(data = certs.evol,
                   aes(x = Year, y = US, group = 1)) +

    geom_line() +
    geom_point() +

    theme(plot.title = element_text(hjust = 0.5)) +
    ggtitle("US") +
    xlab("Years") + ylab("Certifications")+
    stat_smooth(method = "lm", se = FALSE, aes(outfit=fit<<-..y..))

  graph2 <- ggplot(data = certs.evol,
                   aes(x = Year, y = GB, group = 1)) +

    geom_line() +
    geom_point() +

    theme(plot.title = element_text(hjust = 0.5)) +
    ggtitle("GB") +
    xlab("Years") + ylab("Certifications")+
    stat_smooth(method = "lm", se = FALSE, aes(outfit=fit<<-..y..))

  graph3 <- ggplot(data = certs.evol,
                   aes(x = Year, y = IN, group = 1)) +

    geom_line() +
    geom_point() +

    theme(plot.title = element_text(hjust = 0.5)) +
    ggtitle("IN") +
    xlab("Years") + ylab("Certifications")+
    stat_smooth(method = "lm", se = FALSE, aes(outfit=fit<<-..y..))


  graph4 <- ggplot(data = certs.evol,
                   aes(x = Year, y = JP, group = 1)) +

    geom_line() +
    geom_point() +

    theme(plot.title = element_text(hjust = 0.5)) +
    ggtitle("JP") +
    xlab("Years") + ylab("Certifications")+
    stat_smooth(method = "lm", se = FALSE, aes(outfit=fit<<-..y..))


  list(graph1, graph2, graph3, graph4)

}


#----------------------------------------------------------------
#-------------------------gelocal/type---------------------------
#----------------------------------------------------------------

GetContinentAttackPie <- function (Attacks){

  attack.pie <- group_by(Attacks, Attack.standar, Country)
  attack.pie <- as.data.frame(table(attack.pie$Country, attack.pie$Attack.standar))
  attack.pie <- setNames(attack.pie, c("Country", "Attack", "Count"))

  attack.pie <- attack.pie[attack.pie$Country != "",]
  attack.pie <- attack.pie[attack.pie$Attack != "",]

  attack.pie$Attack <- as.character(attack.pie$Attack)
  attack.pie[attack.pie$Attack != "DDoS" &
               attack.pie$Attack != "Defacement" &
               attack.pie$Attack != "Injection",]$Attack <- "Otros"
  attack.pie$Attack <- as.factor(attack.pie$Attack)

  attack.pie <- group_by(attack.pie, Attack, Country)
  attack.pie <- summarise(attack.pie, Count = sum(Count))

  attack.pie.US <- attack.pie[attack.pie$Country == "US",]
  attack.pie.JP <- attack.pie[attack.pie$Country == "JP",]


  graph1 <- ggplot(data=attack.pie.US,
                   aes(x=factor(1),
                       y=Count,
                       fill=Attack)) +
    geom_col(width = 1, color='black') +
    geom_label(aes(label=paste(round(100 * attack.pie.US$Count / sum(attack.pie.US$Count), 2), "%")),
               vjust=c(0),
               hjust=c(0)) +
    coord_polar(theta="y") +
    scale_x_discrete(labels = c("")) +
    scale_y_discrete(labels = c("")) +
    theme(plot.title = element_text(hjust = 0.5),
          axis.title.x=element_blank(),
          axis.title.y=element_blank()) +
    ggtitle("US")

  graph2 <- ggplot(data=attack.pie.JP,
                   aes(x=factor(1),
                       y=Count,
                       fill=Attack)) +
    geom_col(width = 1, color='black') +
    geom_label(aes(label=paste(round(100 * attack.pie.JP$Count / sum(attack.pie.JP$Count), 2), "%")),
               vjust=c(0),
               hjust=c(0)) +
    coord_polar(theta="y") +
    scale_x_discrete(labels = c("")) +
    scale_y_discrete(labels = c("")) +
    theme(plot.title = element_text(hjust = 0.5),
          axis.title.x=element_blank(),
          axis.title.y=element_blank()) +
    ggtitle("JP")

  list(graph1, graph2)
}

GetContinentAttackEvolution <- function(Attacks){

  attack.evol <-mutate(Attacks, Year = format(Attacks$Date, "%Y")) %>%
              group_by(Attack.standar, Country, Year)
  attack.evol <- as.data.frame(table(attack.evol$Country, attack.evol$Attack.standar, attack.evol$Year))
  attack.evol <- setNames(attack.evol, c("Country", "Attack", "Year", "Count"))

  attack.evol <- attack.evol[attack.evol$Country != "",]
  attack.evol <- attack.evol[attack.evol$Attack != "",]

  attack.evol$Attack <- as.character(attack.evol$Attack)
  attack.evol[attack.evol$Attack != "DDoS" &
               attack.evol$Attack != "Defacement" &
               attack.evol$Attack != "Injection",]$Attack <- "Otros"
  attack.evol$Attack <- as.factor(attack.evol$Attack)

  attack.evol <- group_by(attack.evol, Attack, Country, Year)
  attack.evol <- summarise(attack.evol, Count = sum(Count))

  attack.evol.US <- attack.evol[attack.evol$Country == "US",]
  attack.evol.JP <- attack.evol[attack.evol$Country == "JP",]

  attack.evol.US.Otros <- attack.evol.US[attack.evol.US$Attack == "Otros",]
  attack.evol.JP.Otros <- attack.evol.JP[attack.evol.JP$Attack == "Otros",]


  graph1 <- ggplot2::qplot(main = "US",
                           x = attack.evol.US$Year,
                           y = attack.evol.US$Count,
                           group = Attack,
                           xlab = "Years",
                           ylab = "Certifications",
                           data = attack.evol.US,
                           geom = "point",
                           color = Attack)  +
            geom_line() +
            theme(plot.title = element_text(hjust = 0.5))

  graph2 <- ggplot2::qplot(main = "JP",
                           x = attack.evol.JP$Year,
                           y = attack.evol.JP$Count,
                           group = Attack,
                           xlab = "Years",
                           ylab = "Certifications",
                           data = attack.evol.JP,
                           geom = "point",
                           color = Attack)  +
            geom_line() +
            theme(plot.title = element_text(hjust = 0.5))

  graph3 <- ggplot2::qplot(main = "US",
                           x = attack.evol.US.Otros$Year,
                           y = attack.evol.US.Otros$Count,
                           group = 1,
                           xlab = "Years",
                           ylab = "Certifications",
                           data = attack.evol.US.Otros,
                           geom = "point")  +
            geom_line() +
            stat_smooth(method = "lm", se = FALSE, aes(outfit=fit<<-..y..)) +
            theme(plot.title = element_text(hjust = 0.5))

  graph4 <- ggplot2::qplot(main = "JP",
                           x = attack.evol.JP.Otros$Year,
                           y = attack.evol.JP.Otros$Count,
                           group = 1,
                           xlab = "Years",
                           ylab = "Certifications",
                           data = attack.evol.JP.Otros,
                           geom = "point")  +
            geom_line() +
            stat_smooth(method = "lm", se = FALSE, aes(outfit=fit<<-..y..)) +
            theme(plot.title = element_text(hjust = 0.5))

  list(graph1, graph2, graph3, graph4)

}

#----------------------------------------------------------------
#----------------------------------------------------------------
#----------------------------------------------------------------


GetBaseCertsGraph <- function(Cert_PerCountry, year){
  graph1 <- ggplot2::qplot(main = "Countries with above average number of companies certified with 27001 (2012)",
                           x = reorder(country_short,X2012),
                           y = X2012,
                           xlab = "Country",
                           ylab = "Number of certifications",
                           data = Cert_PerCountry[Cert_PerCountry$X2012 > mean(Cert_PerCountry$X2012),],
                           geom = "col",
                           fill = Continent)
  graph1
}




#' Return every graph used in the report file
#'
#' @param Cert_PerCountry data.frame with the processed data of ISO 27001 certifications
#' @param Attacks data.frame with the processed data of cyberattacks
#'
#' @return data.frame
#' @export
GetReportGraphs <- function(Cert_PerCountry,Attacks) {
  #2012
  graph1 <- ggplot2::qplot(main = "Countries with above average number of companies certified with 27001 (2012)",
                  x = reorder(country_short,X2012),
                  y = X2012,
                  xlab = "Country",
                  ylab = "Number of certifications",
                  data = Cert_PerCountry[Cert_PerCountry$X2012 > mean(Cert_PerCountry$X2012),],
                  geom = "col",
                  fill = Continent)

  attacks2k12 <- Attacks[Attacks$Date < "2013-01-01" & Attacks$Date >= "2012-01-01",]
  frameAttacks2k12 <- as.data.frame(table(attacks2k12$Country))
  colnames(frameAttacks2k12) <- c("Country","Attacks")
  graph2 <- ggplot2::qplot(main = "Countries with above average number of cyberattacks (2012)",
                  x = reorder(Country,Attacks),
                  y = Attacks,
                  xlab = "Country",
                  ylab = "Number of attacks",
                  data = frameAttacks2k12[frameAttacks2k12$Attacks > mean(frameAttacks2k12$Attacks),],
                  geom = "col",
                  fill = Continent)

  Attacks2012ByMonth <- mutate(attacks2k12, month = format(attacks2k12$Date, "%m")) %>% group_by(month)
  Attack2012FreqByMonth <- as.data.frame(table(Attacks2012ByMonth$month))
  colnames(Attack2012FreqByMonth) <- c("Month", "Attacks")
  graph3 <- ggplot2::qplot(x = as.numeric(Month),
                  y = Attacks,
                  main = "Global cyberattack progression by month (2012)",
                  data = Attack2012FreqByMonth,
                  geom = c("point", "smooth"),
                  xlim = c(1,12),
                  xlab = "Month") + ggplot2::scale_x_continuous(breaks = 1:12)

  #2013
  graph4 <- ggplot2::qplot(main = "Countries with above average number of companies certified with 27001 (2013)",
                  x = reorder(country_short,X2013),
                  y = X2013,
                  xlab = "Country",
                  ylab = "Number of certifications",
                  data = Cert_PerCountry[Cert_PerCountry$X2013 > mean(Cert_PerCountry$X2013),]
                  , geom = "col",
                  fill = Continent)
  attacks2k13 <- Attacks[Attacks$Date < "2014-01-01" & Attacks$Date >= "2013-01-01",]
  frameAttacks2k13 <- as.data.frame(table(attacks2k13$Country))
  colnames(frameAttacks2k13) <- c("Country","Attacks")
  graph5 <- ggplot2::qplot(main = "Countries with above average number of cyberattacks (2013)",
                  x = reorder(Country,Attacks),
                  y = Attacks,
                  xlab = "Country",
                  ylab = "Number of attacks",
                  data = frameAttacks2k13[frameAttacks2k13$Attacks > mean(frameAttacks2k13$Attacks),]
                  , geom = "col",
                  fill = Continent)

  Attacks2013ByMonth <- mutate(attacks2k13, month = format(attacks2k13$Date, "%m")) %>% group_by(month)
  Attack2013FreqByMonth <- as.data.frame(table(Attacks2013ByMonth$month))
  colnames(Attack2013FreqByMonth) <- c("Month", "Attacks")
  graph6 <- ggplot2::qplot(x = as.numeric(Month),
                  y = Attacks,
                  main = "Global cyberattack progression by month (2013)",
                  data = Attack2013FreqByMonth,
                  geom = c("point", "smooth"),
                  xlim = c(1,12),
                  xlab = "Month") + ggplot2::scale_x_continuous(breaks = 1:12)

  #2014
  graph7 <- ggplot2::qplot(main = "Countries with above average number of companies certified with 27001 (2014)",
                  x = reorder(country_short,X2014),
                  y = X2014,
                  xlab = "Country",
                  ylab = "Number of certifications",
                  data = Cert_PerCountry[Cert_PerCountry$X2014 > mean(Cert_PerCountry$X2014),]
                  , geom = "col",
                  fill = Continent)
  attacks2k14 <- Attacks[Attacks$Date < "2015-01-01" & Attacks$Date >= "2014-01-01",]
  frameAttacks2k14 <- as.data.frame(table(attacks2k14$Country))
  colnames(frameAttacks2k14) <- c("Country","Attacks")
  graph8 <- ggplot2::qplot(main = "Countries with above average number of cyberattacks (2014)",
                  x = reorder(Country,Attacks),
                  y = Attacks,
                  xlab = "Country",
                  ylab = "Number of attacks",
                  data = frameAttacks2k14[frameAttacks2k14$Attacks > mean(frameAttacks2k14$Attacks),]
                  , geom = "col",
                  fill = Continent)

  Attacks2014ByMonth <- mutate(attacks2k14, month = format(attacks2k14$Date, "%m")) %>% group_by(month)
  Attack2014FreqByMonth <- as.data.frame(table(Attacks2014ByMonth$month))
  colnames(Attack2014FreqByMonth) <- c("Month", "Attacks")
  graph9 <- ggplot2::qplot(x = as.numeric(Month),
                  y = Attacks,
                  main = "Global cyberattack progression by month (2014)",
                  data = Attack2014FreqByMonth,
                  geom = c("point", "smooth"),
                  xlim = c(1,12),
                  xlab = "Month") + ggplot2::scale_x_continuous(breaks = 1:12)

  #2015
  graph10 <- ggplot2::qplot(main = "Countries with above average number of companies certified with 27001 (2015)",
                  x = reorder(country_short,X2015),
                  y = X2015,
                  xlab = "Country",
                  ylab = "Number of certifications",
                  data = Cert_PerCountry[Cert_PerCountry$X2015 > mean(Cert_PerCountry$X2015),]
                  , geom = "col",
                  fill = Continent)
  attacks2k15 <- Attacks[Attacks$Date < "2016-01-01" & Attacks$Date >= "2015-01-01",]
  frameAttacks2k15 <- as.data.frame(table(attacks2k15$Country))
  colnames(frameAttacks2k15) <- c("Country","Attacks")
  graph11 <- ggplot2::qplot(main = "Countries with above average number of cyberattacks (2015)",
                  x = reorder(Country,Attacks),
                  y = Attacks,
                  xlab = "Country",
                  ylab = "Number of attacks",
                  data = frameAttacks2k15[frameAttacks2k15$Attacks > mean(frameAttacks2k15$Attacks),]
                  , geom = "col",
                  fill = Continent)

  Attacks2015ByMonth <- mutate(attacks2k15, month = format(attacks2k15$Date, "%m")) %>% group_by(month)
  Attack2015FreqByMonth <- as.data.frame(table(Attacks2015ByMonth$month))
  colnames(Attack2015FreqByMonth) <- c("Month", "Attacks")
  graph12 <- ggplot2::qplot(x = as.numeric(Month),
                  y = Attacks,
                  main = "Global cyberattack progression by month (2015)",
                  data = Attack2015FreqByMonth,
                  geom = c("point", "smooth"),
                  xlim = c(1,12),
                  xlab = "Month") + ggplot2::scale_x_continuous(breaks = 1:12)




  list(graph1,graph2,graph3,graph4,graph5,graph6,graph7,graph8,graph9,graph10,graph11,graph12)
}