R：根據上面一行中的值更改連續值

Question

我之前問過類似的問題，但似乎我還不夠清楚。我有一個32列的data.frame。我想滿足這個特定標準

df$resting == "toolong". 

在從迪文我曾嘗試使用下面的代碼完成了這個建議，每行下面創建重複行。

df[ unlist(mapply(rep, rownames(df), 1+(df$resting=="toolong"))) , ] 

這工作得很好，但現在我需要改變一些「父」行和新創建重複行的值。變量行動和時間父行需要被設置爲

action <- "for" 
time <- 60 

重複的行需要有變量行動和時間設置爲

action <- "l" # which is what it is already so this can be ignored for now 
time <- "parent row time" - 60 # I am unsure how to code this. 

這裏是一個顯示數據結構的示例data.frame（原始數據庫中有更多列）。

id <- c(1,1,1,1,2,2,2,3,3,3) 
resting <- c("f","toolong","t","f","f","toolong","t","f","toolong","t") 
time <- c(23,145,34,16,17,134,67,89,123,12) 
act <- c("f","l","f","d","d","l","f","d","l","d") 
df <- data.frame(id, resting, time, act) 

這裏是最後的DF應該是什麼樣子。

id resting time act 
1 1  f 23 f 
2 1 toolong 60 for 
2.1 1 toolong 85 l 
3 1  t 34 f 
4 1  f 16 d 
5 2  f 17 d 
6 2 toolong 60 for 
6.1 2 toolong 74 l 
7 2  t 67 f 
8 3  f 89 d 
9 3 toolong 60 for 
9.1 3 toolong 63 l 
10 3  t 12 d 

謝謝，蒂姆

Answer 1

我用特徵向量（正如我在前面的回答中所述）：

df <- data.frame(id, resting, time, act, stringsAsFactors=FALSE) 

> df2 <- df[ unlist(mapply(rep, rownames(df), 1+(df$resting=="toolong"))) , ] 
> df2[ df2$resting=="toolong" & !duplicated(df2) , "act"] <- "for" 
> df2[ df2$resting=="toolong" & df2$act == "for" , "time"] <- 60 
> df2[ df2$resting=="toolong" & df2$act == "l" , "time"] <- df2[ df2$resting=="toolong" & df2$act == "l" , "time"] - 60 
> df2 
    id resting time act 
1 1  f 23 f 
2 1 toolong 60 for 
2.1 1 toolong 85 l 
3 1  t 34 f 
4 1  f 16 d 
5 2  f 17 d 
6 2 toolong 60 for 
6.1 2 toolong 74 l 
7 2  t 67 f 
8 3  f 89 d 
9 3 toolong 60 for 
9.1 3 toolong 63 l 
10 3  t 12 d