使用存在的关键字在 R 中对数据集进行子集化[关闭]

Question

我有这个数据集：

> dput(SampleEvents)
structure(list(Event = structure(c(10L, 5L, 6L, 11L, 10L, 7L, 
11L, 8L, 9L, 10L, 1L, 2L, 3L, 4L, 11L), .Label = c("e10", "e11", 
"e12", "e13", "e2", "e3", "e6", "e8", "e9", "Login", "Logout"
), class = "factor"), Transaction.ID = structure(c(NA, 1L, NA, 
2L, NA, NA, NA, NA, 3L, NA, NA, NA, NA, NA, NA), .Label = c("t1", 
"t4", "t5"), class = "factor"), User.ID = structure(c(1L, 1L, 
1L, 1L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L), .Label = c("kenn1", 
"kenn2"), class = "factor"), Event.Date = structure(c(1L, 1L, 
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L), .Label = "20/5/2017", class = "factor"), 
    Event.Time = structure(c(12L, 13L, 14L, 15L, 1L, 2L, 3L, 
    4L, 5L, 6L, 7L, 8L, 9L, 10L, 11L), .Label = c("10:01", "10:02", 
    "10:03", "10:04", "10:05", "10:06", "10:07", "10:08", "10:09", 
    "10:10", "10:11", "9:00", "9:30", "9:45", "9:50"), class = "factor")), .Names = c("Event", 
"Transaction.ID", "User.ID", "Event.Date", "Event.Time"), class = "data.frame", row.names = c(NA, 
-15L))

我想删除“事件”列下两个固定值内的所有行，即从“登录”到“注销”的行，这些行在“登录”和“注销”之间缺少所有交易 ID 值：

我还想保留数据集的当前顺序。

我如何在 R 中做到这一点？

Answer 1

您可以使用您提供的数据执行以下操作...

library(dplyr)

#add variables to mark login-logout blocks and number them
df <- df %>% mutate(session=cumsum(Event=="Login")-cumsum(Event=="Logout"),
                    block=c(0,cumsum(diff(session)!=0)),
                    block=ifelse(Event=="Logout",block-1,block))

#identify blocks to remove
df2 <- df %>% group_by(block) %>% 
              summarise(Login=first(session)>0,
                        noTrans=all(is.na(Transaction.ID))) %>% 
              filter(Login & noTrans)

#remove unwanted blocks and delete temporary variables
df <- df %>% filter(!(block %in% df2$block)) %>% 
             select(-c(session,block))

df
   Event Transaction.ID User.ID Event.Date Event.Time
1  Login           <NA>   kenn1  20/5/2017       9:00
2     e2             t1   kenn1  20/5/2017       9:30
3     e3           <NA>   kenn1  20/5/2017       9:45
4 Logout             t4   kenn1  20/5/2017       9:50
5     e8           <NA>   kenn2  20/5/2017      10:04
6     e9             t5   kenn2  20/5/2017      10:05