R 在数据帧中添加特定行

我试图将数据帧的特定行添加到一起

除了使用grepl查找行，然后再将它们定位到底部，我不确定是否有更好的方法

这是我的输入：

input = structure(list(
V1 = c("Sales", "Sales", "Sales", "Sales", "Sales","Sales"),
V2 = c("Johnny", "Meg", "Fred", "Johnny", "Meg", "Fred"),
V3 = c("Australia", "Australia", "Australia", "NZ", "NZ","NZ"), 
V4 = c(154L, 1898L, 175L, 1235L, 23L, 255L)), row.names = c(NA,6L),
 class = "data.frame")

这是我的预期输出：

structure(list(
V1 = c("Sales", "Sales", "Sales", "Sales", "Sales", 
"Sales", "Sales", "Sales", "Sales", "Sales", "Sales", "Sales"), 
V2 = c("Johnny", "Meg", "Fred", "Johnny", "Meg", "Fred", "Johnny + Fred", 
"Meg + Fred", "Johnny + Meg + Fred", "Johnny + Fred", "Meg + Fred", 
"Johnny + Meg + Fred"), 
V3 = c("Australia", "Australia", "Australia", "NZ",
 "NZ", "NZ", "Australia", "Australia", "Australia", "NZ", "NZ", "NZ"), 
V4 = c(154L, 1898L, 175L, 1235L, 23L, 255L, 329L, 2073L, 2227L, 1490L, 278L, 1513L)),
 class = "data.frame", row.names = c(NA, -12L)
)

我认为有一个更好的方法来添加这些行，过滤，然后添加，然后加入，等等

---

有人能给我指出我应该寻找的正确方向吗？

我使用

combn

数据输入部分 解决方案 结果

结果（comb2）
#一个tibble:6x4
V3 V1 V2 V4
1澳大利亚销售约翰尼+梅格2052
2新西兰销售约翰尼+梅格1258
3澳大利亚销售约翰尼+弗雷德329
4新西兰销售约翰尼+弗雷德1490
5澳大利亚销售Meg+Fred 2073
6新西兰销售Meg+Fred 278
结果（3）
#一个tibble:2x4
V3 V1 V2 V4
1澳大利亚销售约翰尼+梅格+弗雷德2227
2新西兰销售约翰尼+梅格+弗雷德1513
finalResult=bind_行（A、B、输入）%>%
选择（V1、V2、V3、V4）%>%过滤器（！V2%位于%c（'Johnny+Meg'））
>最终结果
#一个tibble:12x4
V1 V2 V3 V4
1销售约翰尼+弗雷德澳大利亚329
2销售约翰尼+弗雷德新西兰1490
3销售梅格+弗雷德澳大利亚2073
4销售Meg+Fred NZ 278
5销售约翰尼+梅格+弗雷德澳大利亚2227
6销售约翰尼+梅格+弗雷德新西兰1513
7销售约翰尼澳大利亚154
8销售Meg澳大利亚1898
9澳大利亚销售部175
10销售约翰尼新西兰1235
11新西兰销售Meg 23
12新西兰销售部255

使用

tidyverse

我们可以首先基于

V3

拆分数据帧，然后创建名称组合并添加

sum

以创建新的TIBLE并将其绑定到原始数据帧

library(tidyverse)

input %>%
     bind_rows(input %>%
                   group_split(V3) %>%
                   map_dfr(function(x) map_dfr(2:nrow(x), ~tibble(
                           V1 = first(x$V1), 
                           V2 = combn(x$V2, ., paste, collapse = " + "), 
                           V3 = first(x$V3),
                           V4 = combn(x$V4, .,sum)) %>% 
                           filter(grepl("\\bFred\\b", V2)))))


#      V1                  V2        V3   V4
#1  Sales              Johnny Australia  154
#2  Sales                 Meg Australia 1898
#3  Sales                Fred Australia  175
#4  Sales              Johnny        NZ 1235
#5  Sales                 Meg        NZ   23
#6  Sales                Fred        NZ  255
#7  Sales       Johnny + Fred Australia  329
#8  Sales          Meg + Fred Australia 2073
#9  Sales Johnny + Meg + Fred Australia 2227
#10 Sales       Johnny + Fred        NZ 1490
#11 Sales          Meg + Fred        NZ  278
#12 Sales Johnny + Meg + Fred        NZ 1513

---

使用相同的逻辑，但以R为基数，我们可以

rbind(input, do.call(rbind, lapply(split(input, input$V3), function(x) 
             do.call(rbind, lapply(2:nrow(x), function(y)
        subset(data.frame(V1 = x$V1[1],
                          V2 = combn(x$V2, y, paste, collapse = " + "), 
                          V3 = x$V3[1],
                          V4 = combn(x$V4, y, sum)), 
         grepl("\\bFred\\b", V2)))))))

---

道歉@RonakShah-感谢您发现错误！很好，你可以避免双重

do.call（rbind

使用

by

而不是

split

@jay.sf啊，是的，你是对的。

by

是我经常忘记的一个重要函数。它是

split

+

lapply

在一个函数中的组合。

result(comb2)
# A tibble: 6 x 4
  V3        V1    V2             V4
  <chr>     <chr> <chr>       <int>
1 Australia Sales Johnny+Meg   2052
2 NZ        Sales Johnny+Meg   1258
3 Australia Sales Johnny+Fred   329
4 NZ        Sales Johnny+Fred  1490
5 Australia Sales Meg+Fred     2073
6 NZ        Sales Meg+Fred      278

result(comb3)
# A tibble: 2 x 4
  V3        V1    V2                 V4
  <chr>     <chr> <chr>           <int>
1 Australia Sales Johnny+Meg+Fred  2227
2 NZ        Sales Johnny+Meg+Fred  1513


finalResult = bind_rows(A,B,input) %>%
  select(V1,V2,V3,V4) %>% filter(! V2 %in% c('Johnny+Meg'))

> finalResult 
# A tibble: 12 x 4
   V1    V2              V3           V4
   <chr> <chr>           <chr>     <int>
 1 Sales Johnny+Fred     Australia   329
 2 Sales Johnny+Fred     NZ         1490
 3 Sales Meg+Fred        Australia  2073
 4 Sales Meg+Fred        NZ          278
 5 Sales Johnny+Meg+Fred Australia  2227
 6 Sales Johnny+Meg+Fred NZ         1513
 7 Sales Johnny          Australia   154
 8 Sales Meg             Australia  1898
 9 Sales Fred            Australia   175
10 Sales Johnny          NZ         1235
11 Sales Meg             NZ           23
12 Sales Fred            NZ          255

library(tidyverse)

input %>%
     bind_rows(input %>%
                   group_split(V3) %>%
                   map_dfr(function(x) map_dfr(2:nrow(x), ~tibble(
                           V1 = first(x$V1), 
                           V2 = combn(x$V2, ., paste, collapse = " + "), 
                           V3 = first(x$V3),
                           V4 = combn(x$V4, .,sum)) %>% 
                           filter(grepl("\\bFred\\b", V2)))))


#      V1                  V2        V3   V4
#1  Sales              Johnny Australia  154
#2  Sales                 Meg Australia 1898
#3  Sales                Fred Australia  175
#4  Sales              Johnny        NZ 1235
#5  Sales                 Meg        NZ   23
#6  Sales                Fred        NZ  255
#7  Sales       Johnny + Fred Australia  329
#8  Sales          Meg + Fred Australia 2073
#9  Sales Johnny + Meg + Fred Australia 2227
#10 Sales       Johnny + Fred        NZ 1490
#11 Sales          Meg + Fred        NZ  278
#12 Sales Johnny + Meg + Fred        NZ 1513

rbind(input, do.call(rbind, lapply(split(input, input$V3), function(x) 
             do.call(rbind, lapply(2:nrow(x), function(y)
        subset(data.frame(V1 = x$V1[1],
                          V2 = combn(x$V2, y, paste, collapse = " + "), 
                          V3 = x$V3[1],
                          V4 = combn(x$V4, y, sum)), 
         grepl("\\bFred\\b", V2)))))))