CODEWITHSUNDEEP
r
Slaapkamer
Slaapkamer

Reputation: 75

R: extract vector name from value in dataframe

I would like to create vectors from dataframe rows and use one of the cell values as the vector name. For example:

df <- data.frame(name=c("michel","nadia","carole"),
                 cities=c("Paris,Montreal","Chicago,Denver","Beijing,Bankok"))

...and I would like to get a vector (for example from the first row), like this:

michel <- c("Paris","Montreal")

How can we do this in R?

Thanks!

Michel

Upvotes: 0

Views: 589

Answers (3)

IceCreamToucan
IceCreamToucan

Reputation: 28675

It's worth noting that data frames in R can have list columns, so you could just convert the cities column to a list of vectors. That way everything is still organized neatly in a data.frame and you can access vectors for specific names by subsetting the data frame 

df$cities <- strsplit(as.character(df$cities), ',')

df
#     name          cities
# 1 michel Paris, Montreal
# 2  nadia Chicago, Denver
# 3 carole Beijing, Bankok

df$cities[[which(df$name == 'michel')]]
# [1] "Paris"    "Montreal"

Upvotes: 1

akrun
akrun

Reputation: 886938

We can use split into a list of named vectors and use list2env for creating multiple objects in the global env (not recommended)

list2env(split(as.character(df$cities), df$name), envir = .GlobalEnv)

If we need to separate the 'cities'

library(dplyr)
library(tidyr)
df %>%
    separate_rows(cities) %>% 
    {split(as.character(.$cities), .$name)} %>%
     list2env(.GlobalEnv)

and now check the ls() for the 'name' objects created in the global env

Or another option is to extract the elements into a list, then convert it to a named list with deframe and use list2env

library(purrr)
library(tibble)
library(stringr)
df %>% 
     mutate(cities = str_extract_all(cities, "\\w+")) %>% 
     deframe %>%
     list2env(.GlobalEnv)

Or for a single row

assign(as.character(df$name)[1], 
      scan(text = as.character(df$cities)[1], what = "", sep=","))

Now, check the ls() and the 'michel' object would be found

michel
#[1] "Paris"    "Montreal"

Upvotes: 2

Till
Till

Reputation: 6628

First we split the city names into vectors:

cities <- strsplit(as.character(df$cities), ",")

And we give the list entries the names and use list2env() to copy each entry to the global environment:

names(cities) <- df$name
list2env(cities, envir = .GlobalEnv)

The result are vectors containing the city names in separated entries, for each name:

> michel
[1] "Paris"    "Montreal"
> nadia
[1] "Chicago" "Denver" 
> carole
[1] "Beijing" "Bankok"

Upvotes: 1

Related Questions