r - 使用索引将多行变为单行

Question

根据这个输入:

structure(list(mid = c("text11", "text12", "text21", "text22", 
"text23"), term = c("test", "text", "section", "2", "sending"
)), class = "data.frame", row.names = c(NA, -5L))

怎么可能用mid来改造，把熔体排成单排。其中 text1, text2... text12 部分的中间显示行数和该行中存在的术语的新编号。用空格分隔合并它们。
数据框示例

data.frame(mid = c("text1", "text2"), term = c("test "text", "section 2 sending"
))

Answer 1

这应该工作

library(dplyr)
library(stringr)
df <- structure(list(mid = c("text11", "text12", "text21", "text22", 
                       "text23"), term = c("test", "text", "section", "2", "sending"
                       )), class = "data.frame", row.names = c(NA, -5L))

df %>% 
  mutate(mid = str_extract(mid, "text\\d")) %>% 
  group_by(mid) %>% 
  summarise(term = paste(term, collapse=" "))
# # A tibble: 2 x 2
#   mid   term             
#   <chr> <chr>            
# 1 text1 test text        
# 2 text2 section 2 sending

编辑 - 解决评论
针对评论中的问题，下面的函数适用于除最后一位以外的所有数字都标识组的任何情况(即下面示例中的 1 和 12)。

df <- structure(list(mid = c("text11", "text12", "text121", "text122",  "text123"), term = c("test", "text", "section", "2", "sending")), class = "data.frame", row.names = c(NA, -5L))

df %>% 
  mutate(mid = str_sub(mid, 1, (nchar(mid)-1))) %>% 
  group_by(mid) %>% 
  summarise(term = paste(term, collapse=" "))

# # A tibble: 2 x 2
#   mid    term             
#   <chr>  <chr>            
# 1 text1  test text        
# 2 text12 section 2 sending