如何使用包含两种语言的单独数据框将语言 1 中的单个单词替换为语言 2 中的单词？

Question

我有一个包含西班牙语单词的数据框，并想用他们的英文翻译替换它们。翻译保存在包含两个单词的单独字典数据框中。原始数据框中的单词可以以随机顺序出现多次。原始数据框中也可以有没有翻译的单词；它们应保持其原始形式。

来源

library(tidyverse)
df <- tibble(item = c("agua","guerra","pelea","hijo","grande","pelea","agua","guerra"))
dictionary <- tibble(word_ES=c("agua","guerra","pelea"), word_EN=c("water","war","fight"))

期望的结果

df2 <- tibble(item = c("water","war","fight","hijo","grande","fight","water","war"))

我对使用 tidyverse 包的建议特别感兴趣，但对其他解决方案也很满意。

Answer 1

一个选项是在data.table 中加入on

library(data.table)
setDT(df)[setDT(dictionary), item := word_EN, on = .(item = word_ES)]
df
#     item
#1:  water
#2:    war
#3:  fight
#4:   hijo
#5: grande
#6:  fight
#7:  water
#8:    war

---

或者，如果我们更喜欢tidyverse，则输出left_join 和coalesce

library(tidyverse)
df %>%
   left_join(dictionary, by = c('item' = 'word_ES')) %>% 
   transmute(item = coalesce(word_EN, item))

Answer 2

这可以在tidyverse 中使用dplyr 和stringr 轻松完成。

library(dplyr)
library(stringr)

df <- tibble(item = c("agua","guerra","pelea","hijo","grande","pelea","agua","guerra"))
dictionary <- tibble(word_ES=c("agua","guerra","pelea"), word_EN=c("water","war","fight"))

mutate(df, item = str_replace_all(item, setNames(dictionary$word_EN, dictionary$word_ES)))
#> # A tibble: 8 x 1
#>   item  
#>   <chr> 
#> 1 water 
#> 2 war   
#> 3 fight 
#> 4 hijo  
#> 5 grande
#> 6 fight 
#> 7 water 
#> 8 war