3

I am given a character vector:

tibble(H = c("6'2\"", "5'10\"", "5'5\"", "5'1\"", "5'5\"", "5'4\""))

and I want to convert it to be in cm.

Please advise how can I do this?

Sotos
  • 51,121
  • 6
  • 32
  • 66
SteveS
  • 3,789
  • 5
  • 30
  • 64
  • 2
    Cool, what have you tried so far? Starting with just the math, how many centimeters are in an inch, and how many inches are in a foot? – camille Mar 19 '19 at 15:11
  • 2
    Try `as.numeric(gsub('"', "", gsub("'", '.', df1$H))) * 2.54` – akrun Mar 19 '19 at 15:13
  • 2
    Or use `library(measurements);library(tidyverse);df1 %>% mutate(H = str_remove(H, '"') %>% str_replace("'", ".") %>% as.numeric %>% conv_unit(., 'inch', 'cm'))` – akrun Mar 19 '19 at 15:16
  • @akrun I could not find a dupe (I found [this](https://stackoverflow.com/questions/40171796/creating-a-function-in-r-to-convert-a-unit-of-measurement) which is the opposite) but you should add your comment as answer. The `measurements` package is very good to know – Sotos Mar 19 '19 at 15:27
  • 1
    Try with one liner `library(gsubfn);gsubfn("(\\d)'(\\d+)", ~ as.numeric(x) * 30.48 + as.numeric(y) * 2.54, df1$H)` – akrun Mar 19 '19 at 15:28

5 Answers5

2

There are couple of methods to use

1) Read with fread after pasting into a single string

library(data.table)
fread(paste(sub('"', "", df1$H), collapse="\n"), sep="'")[, 
               as.matrix(.SD) %*% c(30.48, 2.54)][,1]
 #[1] 187.96 177.80 165.10 154.94 165.10 162.56

2) Using gsubfn

library(gsubfn)
as.numeric(gsubfn("(\\d)'(\\d+)", ~ as.numeric(x) * 30.48 + 
            as.numeric(y) * 2.54, sub('"', '', df1$H)))
 #[1] 187.96 177.80 165.10 154.94 165.10 162.56

3) with separate

library(tidyverse)
df1 %>% 
    separate(H, into = c("H1", "H2"), convert = TRUE) %>%
    transmute(H = H1 * 30.48 + H2 * 2.54)
# A tibble: 6 x 1
#      H
#  <dbl>
#1  188.
#2  178.
#3  165.
#4  155.
#5  165.
#6  163.

4) with measurements

library(measurements)
library(tidyverse)
df1 %>% 
   separate(H, into = c("H1", "H2"), convert = TRUE) %>%
   transmute(H = conv_unit(H1, "ft", "cm") + conv_unit(H2, "inch", "cm"))
akrun
  • 874,273
  • 37
  • 540
  • 662
2

One option is to extract all the numbers and convert it into a matrix and then perform the calculation.

mat <- stringr::str_extract_all(df$H, "\\d+", simplify = TRUE)

as.numeric(mat[, 1]) * 30.48 + as.numeric(mat[, 2]) * 2.54
#[1] 187.96 177.80 165.10 154.94 165.10 162.56

where mat is

#     [,1] [,2]
#[1,] "6"  "2" 
#[2,] "5"  "10"
#[3,] "5"  "5" 
#[4,] "5"  "1" 
#[5,] "5"  "5" 
#[6,] "5"  "4"

First column is feet and second column inches.

And just for my own curiosity I wanted to solve this in base R

sapply(strsplit(sub("(\\d+)'(\\d+).*", "\\1-\\2", df$H), "-"), function(x) 
       as.numeric(x[1]) * 30.48 + as.numeric(x[2]) * 2.54)

#[1] 187.96 177.80 165.10 154.94 165.10 162.56

This follows similar logic that extract 2 numbers from the string using sub, split them using strsplit and then for each of them convert it into numeric and perform the calculation.

Ronak Shah
  • 377,200
  • 20
  • 156
  • 213
2

Using the stringi package to extract the relevant units:

library(stringi)

Raw <- c("6'2\"", "5'10\"", "5'5\"", "5'1\"", "5'5\"", "5'4\"")

## Extract Feet units by regex searching for 1 or more digits followed by a '
Feet <- stri_extract_first_regex(Raw, "[[:digit:]]+(?=')"))

## Extract Inch units by regex searching for 1 or 2 digits followed by a "
Inches <- stri_extract_first_regex(Raw, "[[:digit:]]{1,2}(?=\")"))

## Combine Feet and Inches
TotalInches <- 12 * as.numeric(Feet) + as.numeric(Inches)

## Convert to cm
CM <- 2.54 * TotalInches

print(CM)
# [1] 187.96 177.80 165.10 154.94 165.10 162.56

If you need to do this for multiple columns, it might make sense to define the steps as a function at the top of your script so you can call it more concisely and don't need to store intermediate results in the global environment.

One other consideration taken in the function version here is to replace NA matches with 0 values so that valid measurements like 1' or 11"return valid results instead of NA as well.

FtInToCm <- function(x){
  Feet <- as.numeric(stringi::stri_extract_first_regex(Raw, "[[:digit:]]+(?=')"))
  Inches <- as.numeric(stringi::stri_extract_first_regex(Raw, "[[:digit:]]{1,2}(?=\")"))
  return(2.54 * (12 * ifelse(is.na(Feet),0,Feet) + ifelse(is.na(Inches),0,Inches)))
}

FtInToCm(Raw)
#[1] 187.96 177.80 165.10 154.94 165.10 162.56
Matt Summersgill
  • 4,054
  • 18
  • 47
1
> dat <- tibble(H = c("6'2\"", "5'10\"", "5'5\"", "5'1\"", "5'5\"", "5'4\""))
> dat$inches <- gsub("[\"]", "",dat$H) %>% 
    strsplit(., "'") %>% 
    lapply(., function(x) {
      x <- as.numeric(x);
      (x[1]*30.48) + (x[2]/12)*30.48
    }) %>% 
   unlist
> dat
# A tibble: 6 x 2
  H        inches
  <chr>     <dbl>
1 "6'2\""    188.
2 "5'10\""   178.
3 "5'5\""    165.
4 "5'1\""    155.
5 "5'5\""    165.
6 "5'4\""    163.

You can even use map instead of lapply

> gsub("[\"]", "",dat$H) %>% 
    strsplit(., "'") %>% 
    map_dbl(function(x){
      x <- as.numeric(x)
      (x[1]*30.48) + (x[2]/12)*30.48
    })
[1] 187.96 177.80 165.10 154.94 165.10 162.56
Jilber Urbina
  • 58,147
  • 10
  • 114
  • 138
1

I add another answer, just to give you another option and because I already wrote it before I saw the other answers.

I convert the string to numeric first and then convert units:

library(dplyr)
library(stringr)
df <- tibble(H = c("6'2\"", "5'10\"", "5'5\"", "5'1\"", "5'5\"", "5'4\""))

df %>% 
  mutate(foot = str_extract(H, "^\\d+'"), 
         inch = str_extract(H, "\\d+\"$")) %>% # split foot from inch
  mutate(foot = as.numeric(str_remove(foot, "[^\\d]")),
         inch = as.numeric(str_remove(inch, "[^\\d]"))) %>% # convert to numeric
  mutate(H_new = cm(foot * 12) + cm(inch)) # convert units
# A tibble: 6 x 4
  H         foot  inch H_new
  <chr>    <dbl> <dbl> <dbl>
1 "6'2\""      6     2  188.
2 "5'10\""     5    10  178.
3 "5'5\""      5     5  165.
4 "5'1\""      5     1  155.
5 "5'5\""      5     5  165.
6 "5'4\""      5     4  163.
JBGruber
  • 11,727
  • 1
  • 23
  • 45