Last active
September 12, 2018 15:11
-
-
Save expersso/b3da2c02edbd4a3c9c5c9c7d44ddff27 to your computer and use it in GitHub Desktop.
Wrangle complicated spreadsheet into tidy dataframe
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
# Inspired by http://www.brodrigues.co/blog/2018-09-11-human_to_machine/ | |
library(tidyverse) | |
library(readxl) | |
# Extract all data from one sheet into tidy dataframe | |
get_sheet <- function(sheet, path = "time-use.xlsx") { | |
ranges <- list( | |
col = "C3:BS5", | |
row = "A6:B58", | |
data = "C6:BS58" | |
) | |
# Read the three different ranges into a list | |
x <- imap(ranges, ~read_excel(path, sheet, ., FALSE)) | |
# Extract column names | |
x$col <- x$col %>% | |
discard(~all(is.na(.))) %>% | |
t() %>% | |
as.data.frame() %>% | |
set_names(c("population", "subgroup", "unit")) %>% | |
fill(population, subgroup, unit) | |
# Extract row names | |
x$row <- x$row %>% | |
unite(row, 1:2, sep = "_") %>% | |
pull() | |
# Extract actual data | |
x$data <- x$data %>% | |
discard(~all(is.na(.))) %>% | |
t() | |
# Combine into tibble | |
cbind(x$col, x$data) %>% | |
set_names(c(names(x$col), x$row)) %>% | |
gather(id, value, -c(1:3)) %>% | |
separate(id, c("activity_id", "activity"), sep = "_") %>% | |
tbl_df() | |
} | |
# Read data from all relevant sheets into one dataframe | |
df_full <- excel_sheets("time-use.xlsx")[7:10] %>% | |
set_names() %>% | |
imap_dfr(~get_sheet(.), .id = "day") | |
# Extract a subset of the data to match original blog post | |
df_sub <- df_full %>% | |
filter(subgroup == "Total") %>% | |
select(day, population, activity, time = value) |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment