英文:
R | How to arrange in custom order character vectors of a df column?
问题
我有一个看起来像这样的数据框：
| 水果 | X | Y | Z |
|---|---|---|---|
| 苹果，香蕉，橙子，木瓜 | a | f | k |
| 香蕉，橙子，葡萄 | b | g | l |
| 橙子，香蕉 | c | h | m |
| 葡萄 | d | i | n |
| 香蕉，葡萄，橙子，苹果，木瓜 | e | j | o |
我想在每一行中设置自定义的出现顺序，如下所示：
- 苹果
- 橙子
- 木瓜
- 香蕉
- 葡萄
因此，该列将如下所示：
| 水果 | X | Y | Z |
|---|---|---|---|
| 苹果，橙子，木瓜，香蕉 | a | f | k |
| 橙子，香蕉，葡萄 | b | g | l |
| 橙子，香蕉 | c | h | m |
| 葡萄 | d | i | n |
| 苹果，橙子，木瓜，香蕉，葡萄 | e | j | o |
我该如何做？我尝试了其他帖子中的建议，但它们都是关于排列数据框行，这不是我需要的...
附言：是否有办法在管道内完成这个操作？
英文:
I have a dataframe that looks like this:
| Fruit | X | Y | Z |
|---|---|---|---|
| apple, banana, orange, papaya | a | f | k |
| banana, orange, grape | b | g | l |
| orange, banana | c | h | m |
| grape | d | i | n |
| banana, grape, orange, apple, papaya | e | j | o |
And I want to set a custom order of appearance in each row. Like:
- Apple
- Orange
- Papaya
- Banana
- Grape
So the column would look like:
| Fruit | X | Y | Z |
|---|---|---|---|
| apple, orange, papaya, banana | a | f | k |
| orange, banana, grape | b | g | l |
| orange, banana | c | h | m |
| grape | d | i | n |
| apple, orange, papaya, banana, grape | e | j | o |
How can I do this??? I've tried suggestions from other posts, but they're all about arranging dataframe rows, which isn't what I need...
P.S.: is there any way to do this inside a pipe?
答案1
得分: 4
以下是您的代码的中文翻译：
首先，我们可以使用下面的代码来实现：
library(dplyr)library(stringr)library(purrr)df1 <- df1 %>%mutate(Fruit = map_chr(strsplit(Fruit, ",\\s*"),~ toString(.x[order(match(.x,c("apple", "orange", "papaya", "banana", "grape")))])))
-输出
df1Fruit X Y Z1 apple, orange, papaya, banana a f k2 orange, banana, grape b g l3 orange, banana c h m4 grape d i n5 apple, orange, papaya, banana, grape e j o
或者使用 separate_longer_delim：
library(tidyr)df1 <- df1 %>%mutate(rn = row_number()) %>%separate_longer_delim(Fruit, delim = regex(",\\s*")) %>%arrange(rn, factor(Fruit,levels = c("apple", "orange", "papaya", "banana", "grape"))) %>%reframe(Fruit = str_c(Fruit, collapse = ", "),.by = c("rn", "X", "Y", "Z")) %>%select(-rn) %>%relocate(Fruit, .before = 1)
-输出
df1Fruit X Y Z1 apple, orange, papaya, banana a f k2 orange, banana, grape b g l3 orange, banana c h m4 grape d i n5 apple, orange, papaya, banana, grape e j o
如果列是 list 类型，我们可以不使用 strsplit，而是使用下面的代码：
df1 <- df1 %>%mutate(Fruit = map(Fruit,~ .x[order(match(.x, c("apple", "orange", "papaya", "banana", "grape")))]))
或者使用 unnest：
df1 <- df1 %>%mutate(rn = row_number()) %>%unnest(Fruit) %>%arrange(rn, factor(Fruit,levels = c("apple", "orange", "papaya", "banana", "grape"))) %>%reframe(Fruit = list(Fruit),.by = c("rn", "X", "Y", "Z")) %>%select(-rn) %>%relocate(Fruit, .before = 1)
-输出
df1# A tibble: 5 × 4Fruit X Y Z1 <chr [4]> a f k2 <chr [3]> b g l3 <chr [2]> c h m4 <chr [1]> d i n5 <chr [5]> e j o
最后，这是您的数据：
df1 <- structure(list(Fruit = c("apple, banana, orange, papaya", "banana, orange, grape","orange, banana", "grape", "banana, grape, orange, apple, papaya"), X = c("a", "b", "c", "d", "e"), Y = c("f", "g", "h", "i","j"), Z = c("k", "l", "m", "n", "o")), class = "data.frame", row.names = c(NA,-5L))
英文:
We could do
library(dplyr)library(stringr)library(purrr)df1 <- df1 %>%mutate(Fruit = map_chr(strsplit(Fruit, ",\\s*"),~ toString(.x[order(match(.x,c("apple", "orange", "papaya", "banana", "grape")))])))
-output
df1Fruit X Y Z1 apple, orange, papaya, banana a f k2 orange, banana, grape b g l3 orange, banana c h m4 grape d i n5 apple, orange, papaya, banana, grape e j o
Or using separate_longer_delim
library(tidyr)df1 <- df1 %>%mutate(rn = row_number()) %>%separate_longer_delim(Fruit, delim = regex(",\\s*")) %>%arrange(rn, factor(Fruit,levels = c("apple", "orange", "papaya", "banana", "grape"))) %>%reframe(Fruit = str_c(Fruit, collapse = ", "),.by = c("rn", "X", "Y", "Z")) %>%select(-rn) %>%relocate(Fruit, .before = 1)
-output
df1Fruit X Y Z1 apple, orange, papaya, banana a f k2 orange, banana, grape b g l3 orange, banana c h m4 grape d i n5 apple, orange, papaya, banana, grape e j o
If the column is list, we don't need the strsplit, instead
df1 <- df1 %>%mutate(Fruit = map(Fruit,~ .x[order(match(.x, c("apple", "orange", "papaya", "banana", "grape")))]))
Or with unnest
df1 <- df1 %>%mutate(rn = row_number()) %>%unnest(Fruit) %>%arrange(rn, factor(Fruit,levels = c("apple", "orange", "papaya", "banana", "grape"))) %>%reframe(Fruit = list(Fruit),.by = c("rn", "X", "Y", "Z")) %>%select(-rn) %>%relocate(Fruit, .before = 1)
-output
df1# A tibble: 5 × 4Fruit X Y Z<list> <chr> <chr> <chr>1 <chr [4]> a f k2 <chr [3]> b g l3 <chr [2]> c h m4 <chr [1]> d i n5 <chr [5]> e j o
data
df1 <- structure(list(Fruit = c("apple, banana, orange, papaya", "banana, orange, grape","orange, banana", "grape", "banana, grape, orange, apple, papaya"), X = c("a", "b", "c", "d", "e"), Y = c("f", "g", "h", "i","j"), Z = c("k", "l", "m", "n", "o")), class = "data.frame", row.names = c(NA,-5L))
答案2
得分: 3
以下是翻译好的部分：
主要特点是使用 separate_rows，然后创建具有以下级别的 factor 类：
library(dplyr)library(tidyr)df %>%group_by(group = row_number()) %>%separate_rows(Fruit) %>%mutate(Fruit= factor(Fruit, levels = c("apple", "orange", "papaya", "banana", "grape"))) %>%arrange(Fruit, .by_group = TRUE) %>%summarise(Fruit = toString(Fruit)) %>%bind_cols(df[2:4]) %>%select(-group)
Fruit X Y Z<chr> <chr> <chr> <chr>1 apple, orange, papaya, banana a f k2 orange, banana, grape b g l3 orange, banana c h m4 grape d i n5 apple, orange, papaya, banana, grape e j o
英文:
Here is one more (a tidyverse solution):
Main feature is to use separate_rows and then create factor class with the levels:
library(dplyr)library(tidyr)df %>%group_by(group = row_number()) %>%separate_rows(Fruit) %>%mutate(Fruit= factor(Fruit, levels = c("apple", "orange", "papaya", "banana", "grape"))) %>%arrange(Fruit, .by_group = TRUE) %>%summarise(Fruit = toString(Fruit)) %>%bind_cols(df[2:4]) %>%select(-group)
Fruit X Y Z<chr> <chr> <chr> <chr>1 apple, orange, papaya, banana a f k2 orange, banana, grape b g l3 orange, banana c h m4 grape d i n5 apple, orange, papaya, banana, grape e j o
通过集体智慧和协作来改善编程学习和解决问题的方式。致力于成为全球开发者共同参与的知识库，让每个人都能够通过互相帮助和分享经验来进步。
评论