Question

我有下表：

structure(list(Id = structure(c(1L, 1L, 2L, 2L, 1L, 3L, 3L, 3L
), .Label = c("a", "b", "c"), class = "factor"), stops = c(1, 
1, 1, 1, 1, 2, 2, 2)), .Names = c("Id", "stops"), row.names = c(NA, 
-8L), class = "data.frame")

我希望在止损没有改变时添加$停止新字符，但$ Id确实。

例如，我想得到：

structure(list(Id = structure(c(1L, 1L, 2L, 2L, 1L, 3L, 3L, 3L
), .Label = c("a", "b", "c"), class = "factor"), stops = structure(c(1L, 
1L, 2L, 2L, 3L, 4L, 4L, 4L), .Label = c("1", "1-1", "1-2", "2"
), class = "factor")), .Names = c("Id", "stops"), row.names = c(NA, 
-8L), class = "data.frame")

如果Id与前一个不同，我想这样做，如果Stops与前一个相同...

我尝试使用mutate（），但似乎我在这里工作得很远......

Answer 1

以下是使用data.table

的循环尝试

library(data.table)
setDT(df)[, `:=`(stops = as.character(stops), Idindx = rleid(Id))]
indx <- unique(df, by = "Idindx")[, counter := (1:.N) - 1L, by = rleid(stops)]
df[indx[counter > 0], stops := paste(stops, i.counter, sep = "-"), on = "Idindx"]
#    Id stops Idindx
# 1:  a     1      1
# 2:  a     1      1
# 3:  b   1-1      2
# 4:  b   1-1      2
# 5:  a   1-2      3
# 6:  c     2      4
# 7:  c     2      4
# 8:  c     2      4

第一步是为每个Id创建一个唯一索引（因为它们不是唯一的），并将stops转换为character（根据您所需的输出）

然后，对唯一索引进行操作可识别相同stops的计数并重新连接回原始数据

Answer 2

你可以写一个循环来解决你的问题：

# Original data
data <- structure(list(Id = structure(c(1L, 1L, 2L, 2L, 1L, 3L, 3L, 3L
), .Label = c("a", "b", "c"), class = "factor"), stops = c(1, 
                                                       1, 1, 1, 1, 2, 2, 2)), .Names = c("Id", "stops"), row.names = c(NA, 
                                                                                                                       -8L), class = "data.frame")

# Add new column, which will be converted in the following loop
data$stops_new <- as.character(data$stops)
new <- 1

for(i in 2:nrow(data)) {

  # Convert values of stops_new, if your specified conditions appear
  if(data$Id[i] != data$Id[i - 1] & data$stops[i] == data$stops[i - 1]) {
    data$stops_new[i] <- paste(data$stops_new[i], "-", new, sep = "")

    # Repeat the convertion for all values with the same ID and stop-value    
    j <- i + 1
    while(data$Id[i] == data$Id[j] & data$stops[i] == data$stops[j]) {
      data$stops_new[j] <- paste(data$stops[i], "-", new, sep = "")
      j <- j + 1
    }

    new <-  new + 1
  }
}

data

Answer 3

这是基础R解决方案。

创建指标，显示Id是否已更改（id.ind）以及stops是否已从上一行更改（stops.ind）（惯例是这些指标已设置to＆＃34; 0＆＃34;，即第一行没有变化）：

stops.ind <- c(0, diff(dat$stops))                                                                                                                                                                                                                                                                       
id.ind <- c(0, diff(as.numeric(dat$Id)))

创建新的停止向量：

stops <- new.stops <- dat$stops

逐行检查a）id中是否有变化而stops或b中没有变化）前一行中没有变化。如果a）将k增加1并附加＆＃34; -k＆＃34;到stops值b）使用stops的先前值：

k <- 0
for(i in 2 : nrow(dat)){
  if(id.ind[i] != 0 & stops.ind[i] == 0){
    k <- k + 1
    new.stops[i] <- paste0(stops[i], "-", k)
  }
   if(id.ind[i] == 0 & stops.ind[i] == 0)
    new.stops[i] <- new.stops[i - 1]
}

new.stops
# [1] "1"   "1"   "1-1" "1-1" "1-2" "2"   "2"   "2"  

new.dat <- data.frame(Id = dat$Id, stops = new.stops)

使用条件R向现有单元格添加字符

3 个答案: