Skip to content

Instantly share code, notes, and snippets.

@laurakwiley

laurakwiley/jt_extract_numbers_from_string.R

Last active December 4, 2015 21:12
Download ZIP

Revisions

  1. laurakwiley revised this gist Dec 4, 2015. 1 changed file with 11 additions and 0 deletions.
    11 changes: 11 additions & 0 deletions jt_extract_numbers_from_string.R
    View file
    Original file line number Diff line number Diff line change
    @@ -1,3 +1,4 @@
    library(magrittr)
    library(tidyr)
    library(dplyr)

    @@ -15,3 +16,13 @@ data %>%
    library(stringr)
    data %>%
    mutate(numeric_val = str_match(Var1, "= ?([0-9]+\\.?[0-9]?)")[,2] %>% as.numeric())

    # In case your data really is a matrix
    data <- matrix(c("MHSMOCA:MOCATOTAL = 24", "MHSMOCA:MOCATOTAL = 24.5","MHSMOCA:MOCA7TOTAL = 24", "MsdHSMOCA:MOCA7TOTAL = 26.54654"), nrow = 2, ncol = 2)
    # Note this will make all of your strings factors,
    data %<>%
    as.data.frame() %>%
    tbl_df()
    # Can use mutate_each to apply as.character() to every column
    data %>%
    mutate_each(funs(as.character(.)))
  2. laurakwiley created this gist Dec 4, 2015.
    17 changes: 17 additions & 0 deletions jt_extract_numbers_from_string.R
    View file
    Original file line number Diff line number Diff line change
    @@ -0,0 +1,17 @@
    library(tidyr)
    library(dplyr)

    data <- data.frame(Var1 = c("MHSMOCA:MOCATOTAL = 24", "MHSMOCA:MOCATOTAL = 24.5","MHSMOCA:MOCA7TOTAL = 24")) %>% tbl_df()

    # Option 1: Use a variable splitting function from tidyr
    ## tidyr::separate takes the column you want to split on, what you want the new column names to be, what the separating text is, and whether you want to change data types after separation (so here rendering the number as a dbl or int)
    data %>%
    separate(col = Var1, into = c("Text","Number"), sep = "=", convert = TRUE)

    # Option 2: Extract Number using regex
    ## Here I am using a regex to extract an equals sign, possible whitespace, and then I have a capturing group for [0-9](at least once) possibly followed by a period and more numbers. I used the "[,2]" to get the capturing group from the regex
    ## Note you can pipe within the function to perform the column type change, otherwise it keeps the number as a string
    ##
    library(stringr)
    data %>%
    mutate(numeric_val = str_match(Var1, "= ?([0-9]+\\.?[0-9]?)")[,2] %>% as.numeric())