Question

我有一个示例数据框act，其中包含两个看起来像这样的变量：

   activity_id          activity_ids
1          227 {227,32,33,34,35,252}
2           32 {227,32,33,34,35,252}
3           33 {227,32,33,34,35,252}
4           34 {227,32,33,34,35,252}
5           35 {227,32,33,34,35,252}
6          252 {227,32,33,34,35,252}
7          227 {227,32,33,34,35,252}
8           32 {227,32,33,34,35,252}
9           33 {227,32,33,34,35,252}
10          34 {227,32,33,34,35,252}

activity_id是整数变量，activity_ids是字符变量。

现在，我想添加一个新的布尔变量，说last_activity，它通过检查activity_id的值是否是{{1}的花括号中的数字集合中的最后一个数字来返回true或false。 1}}变量。对于此示例数据，此新变量activity_ids应仅针对第6行返回last_activity（因为252是最后一个数字），并且对于所有其他行返回TRUE。此外，在此示例数据中，FALSE变量在大括号内有6个数字。它可以在大括号内包含任意数量的值。所以，我需要一个可以为任意数量的值推广的代码。

谢谢！

Answer 1

使用基本R选项，sub可以在这里工作：

df <- data.frame(activity_id=c(227, 252),
                 activity_ids=c("{227,32,33,34,35,252}", "{227,32,33,34,35,252}"))

df$last_activity <- df$activity_id == sub(".*,(\\d+)\\}$", "\\1", df$activity_ids)
df

      activity_id          activity_ids last_activity
1             227 {227,32,33,34,35,252}         FALSE
2             252 {227,32,33,34,35,252}          TRUE

Demo

Answer 2

修改

我刚刚意识到，当activity_ids包含额外的信息时，原始方法存在问题。例如，

df$activity_ids[6] <- "{227,32,33,34,35,2521}" mapply(function(x, y) grepl(y, tail(x, 1), fixed = TRUE), strsplit(df$activity_ids, ","), df$activity_id) #[1] FALSE FALSE FALSE FALSE FALSE TRUE FALSE FALSE FALSE FALSE

仍然会返回错误的TRUE。

为了克服这一点，我们可以改为提取最后一个值的数字部分，然后将其与activity_id进行比较

mapply(function(x, y) y == sub("[^0-9]","",tail(x, 1)), strsplit(df$activity_ids, ","), df$activity_id) #[1] FALSE FALSE FALSE FALSE FALSE TRUE FALSE FALSE FALSE FALSE df$activity_ids[6] <- "{227,32,33,34,35,2521}" mapply(function(x, y) y == sub("[^0-9]","",tail(x, 1)), strsplit(df$activity_ids, ","), df$activity_id) #[1] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE

原始答案

非正则表达式选项将字符串拆分为＆＃34;，＆＃34;并使用tail获取最后一个值，并检查activity_id中是否存在grepl。

df$last_activity <- mapply(function(x, y) grepl(y, tail(x, 1), fixed = TRUE), strsplit(df$activity_ids, ","), df$activity_id) # activity_id activity_ids last_activity #1 227 {227,32,33,34,35,252} FALSE #2 32 {227,32,33,34,35,252} FALSE #3 33 {227,32,33,34,35,252} FALSE #4 34 {227,32,33,34,35,252} FALSE #5 35 {227,32,33,34,35,252} FALSE #6 252 {227,32,33,34,35,252} TRUE #7 227 {227,32,33,34,35,252} FALSE #8 32 {227,32,33,34,35,252} FALSE #9 33 {227,32,33,34,35,252} FALSE #10 34 {227,32,33,34,35,252} FALSE

Answer 3

正则表达式方法是使用stri_extract_last_regex包中的stringi从字符串中提取最后一个数字，并将其与activity_id进行比较

library(stringi)
df$activity_id == stri_extract_last_regex(df$activity_ids, "[0-9]+")

#[1] FALSE FALSE FALSE FALSE FALSE  TRUE FALSE FALSE FALSE FALSE

Answer 4

基地R：

transform(dat,s=Vectorize(grepl)(paste0(activity_id,"}"),activity_ids))
   activity_id          activity_ids     s
1          227 {227,32,33,34,35,252} FALSE
2           32 {227,32,33,34,35,252} FALSE
3           33 {227,32,33,34,35,252} FALSE
4           34 {227,32,33,34,35,252} FALSE
5           35 {227,32,33,34,35,252} FALSE
6          252 {227,32,33,34,35,252}  TRUE
7          227 {227,32,33,34,35,252} FALSE
8           32 {227,32,33,34,35,252} FALSE
9           33 {227,32,33,34,35,252} FALSE
10          34 {227,32,33,34,35,252} FALSE

为了加快计算速度，请使用包stringi

stringi::stri_detect_fixed(dat$activity_ids,paste0(dat$activity_id,"}"))
 [1] FALSE FALSE FALSE FALSE FALSE  TRUE FALSE FALSE FALSE FALSE

Answer 5

在基础apply()中使用R列方式的另一种方法：

cols <- c('activity_id', 'activity_ids')
df$last_activity <- apply(df[cols], 1, function(col) {
  x <- unlist(strsplit(col['activity_ids'], "[{},]"))
  return(col['activity_id'] == x[length(x)])
})

或使用mapply()：

df$last_activity <- mapply(function(x,y) {x == y[length(y)]}, 
       x = df$activity_id, 
       y = strsplit(df$activity_ids, "[{},]")
)

两者都会产生

   activity_id          activity_ids last_activity
1          227 {227,32,33,34,35,252}         FALSE
2           32 {227,32,33,34,35,252}         FALSE
3           33 {227,32,33,34,35,252}         FALSE
4           34 {227,32,33,34,35,252}         FALSE
5           35 {227,32,33,34,35,252}         FALSE
6          252 {227,32,33,34,35,252}          TRUE
7          227 {227,32,33,34,35,252}         FALSE
8           32 {227,32,33,34,35,252}         FALSE
9           33 {227,32,33,34,35,252}         FALSE
10          34 {227,32,33,34,35,252}         FALSE
11         212              somejunk         FALSE

如何在R中的花括号内提取整数？

5 个答案:

Demo