R 从层次结构创建表

我有一个层次结构，我想创建一个表来捕获最后一个值、前一个值和列号

我不知道从哪里开始。我可以计算列数作为一个新列，然后我想我需要融化数据，但我不能计算出什么变量，我不知道搜索什么

我的输入如下所示：

input = structure(list(V1 = structure(c(1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 
1L, 1L), .Label = "ASIA PACIFIC", class = "factor"), V2 = structure(c(1L, 
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L), .Label = "AUSTRALIA", class = "factor"), 
    V3 = structure(c(1L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L), .Label = c("ACT", 
    "NEW SOUTH WALES"), class = "factor"), V4 = structure(c(1L, 
    3L, 3L, 3L, 3L, 3L, 2L, 2L, 2L, 4L), .Label = c("CANBERRA", 
    "NEWCASTLE", "SYDNEY", "WOLLONGONG"), class = "factor"), 
    V5 = structure(c(9L, 2L, 6L, 4L, 7L, 3L, 5L, 8L, 10L, 1L), .Label = c("###", 
    "BONDI", "CAMPBELLTOWN", "GEORGE ST", "MAIN ST", "NEWTOWN", 
    "PITT ST", "POKOLBIN", "SMITH ST", "STRANGE PDE"), class = "factor"), 
    V6 = structure(c(1L, 2L, 3L, 1L, 1L, 5L, 1L, 4L, 1L, 1L), .Label = c("###", 
    "CHARLES AVE", "FRANCIS ST", "TOM ST", "TONY LANE"), class = "factor"), 
    V7 = structure(c(1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L), .Label = "###", class = "factor")), class = "data.frame", row.names = c(NA, 
-10L))

我希望创建以下输出：

output = 
structure(list(V1 = structure(c(10L, 3L, 4L, 5L, 9L, 14L, 6L, 
13L, 11L, 15L, 2L, 12L, 8L, 7L, 1L), .Label = c("AUSTRALIA", 
"CANBERRA", "CHARLES AVE", "FRANCIS ST", "GEORGE ST", "MAIN ST", 
"NEW SOUTH WALES", "NEWCASTLE", "PITT ST", "SMITH ST", "STRANGE PDE", 
"SYDNEY", "TOM ST", "TONY LANE", "WOLLONGONG"), class = "factor"), 
    V2 = structure(c(6L, 4L, 9L, 11L, 11L, 5L, 8L, 10L, 8L, 7L, 
    1L, 7L, 7L, 3L, 2L), .Label = c("ACT", "ASIA PACIFIC", "AUSTRALIA", 
    "BONDI", "CAMPBELLTOWN", "CANBERRA", "NEW SOUTH WALES", "NEWCASTLE", 
    "NEWTOWN", "POKOLBIN", "SYDNEY"), class = "factor"), V3 = c(5L, 
    6L, 6L, 5L, 5L, 6L, 5L, 6L, 5L, 4L, 3L, 3L, 3L, 2L, 1L)), class = "data.frame", row.names = c(NA, 
-15L))

---

我不知道我应该搜索什么来开始。如果有人能建议我从哪里开始，我将不胜感激。

通常这种缺失的值用NA表示，但在这里用表示，我们可以从每行中删除这些值，并使用tail选择最后两个值并返回lengthnew_x

同样的，一个整洁的方法可能是

library(tidyverse)

input %>%
  mutate(row = row_number()) %>%
  gather(col, value, -row) %>%
  filter(value != "###") %>%
  group_by(row) %>%
  mutate(last_value = row_number()) %>%
  slice(c(n(), n() - 1)) %>%
  mutate(col = c("col1", "col2"), 
        last_value = max(last_value)) %>%
  spread(col, value) %>%
  ungroup() %>%
  select(-row)

# A tibble: 9 x 3
#  last_value col1        col2           
#       <int> <chr>       <chr>          
#1          6 CHARLES AVE BONDI          
#2          6 FRANCIS ST  NEWTOWN        
#3          5 GEORGE ST   SYDNEY         
#4          5 PITT ST     SYDNEY         
#5          6 TONY LANE   CAMPBELLTOWN   
#6          5 MAIN ST     NEWCASTLE      
#7          6 TOM ST      POKOLBIN       
#8          5 STRANGE PDE NEWCASTLE      
#9          4 WOLLONGONG  NEW SOUTH WALES

---

通常这种缺失值用NA表示，但在这里用表示，我们可以从每行中删除这些值，然后使用tail选择最后两个值，并返回lengthnew_x

同样的，一个整洁的方法可能是

library(tidyverse)

input %>%
  mutate(row = row_number()) %>%
  gather(col, value, -row) %>%
  filter(value != "###") %>%
  group_by(row) %>%
  mutate(last_value = row_number()) %>%
  slice(c(n(), n() - 1)) %>%
  mutate(col = c("col1", "col2"), 
        last_value = max(last_value)) %>%
  spread(col, value) %>%
  ungroup() %>%
  select(-row)

# A tibble: 9 x 3
#  last_value col1        col2           
#       <int> <chr>       <chr>          
#1          6 CHARLES AVE BONDI          
#2          6 FRANCIS ST  NEWTOWN        
#3          5 GEORGE ST   SYDNEY         
#4          5 PITT ST     SYDNEY         
#5          6 TONY LANE   CAMPBELLTOWN   
#6          5 MAIN ST     NEWCASTLE      
#7          6 TOM ST      POKOLBIN       
#8          5 STRANGE PDE NEWCASTLE      
#9          4 WOLLONGONG  NEW SOUTH WALES

---

谢谢我不认为我遵循了这个例子，如果我想把它一直扩展到顶部，我已经编辑了输出，因为我不能在这里发布，我如何避免重复？第一个值是多少？史密斯街，堪培拉哪个是数据框的列名？为什么没有选择列的最后两个值？抱歉@RonakShah-我已经塞满了，忘记在输入中添加header=F。@nicshah如果您想避免重复，那么为什么您的输出中会重复一些值，如新南威尔士、澳大利亚、悉尼？谢谢。我不认为我遵循了这个例子，如果我想把它一直扩展到顶部，我已经编辑了输出，因为我不能在这里发布，我如何避免重复？第一个值是多少？史密斯街，堪培拉哪个是数据框的列名？为什么没有选择列的最后两个值？抱歉@RonakShah-我已经塞满了，忘记在输入中添加header=F。@nicshah如果您想避免重复，那么为什么在输出中重复一些值，如新南威尔士、澳大利亚、悉尼？