| #' Check for integers | |
| #' | |
| #' @param df data frame to check | |
| #' @param select_int data frame returns just the columns of data that are | |
| #' integer64. The default returns all columns. | |
| #' @param tidy logical set to TRUE to change set any integer or integer64 to | |
| #' integer and give a message. FALSE will return the original data with no changes | |
| #' and message. | |
| #' | |
| #' @returns data frame and log where integer64 found |
| library(dplyr) | |
| library(tidyr) | |
| # Create the data - reduced the dates in timescale to make them easier to run and see the results | |
| data <- tibble::tribble( | |
| ~Doctor_name, ~Start_date, ~End_date, ~Weekly_hours, | |
| "Tom", "2023-07-01", "2023-07-06", 40, | |
| "Dick", "2023-06-01", "2023-08-30", 20, | |
| "Harry", "2023-05-01", "2023-07-14", 8 |
| library(tidyverse) | |
| library(httr) # to load from the web | |
| library(readxl) # to read in the file | |
| library(janitor) # to clean the spreadsheet | |
| # load data directly from the webpage --------------------------- | |
| tmp <- tempfile(fileext = ".xlsx") | |
| url <- GET(url = "https://www.ons.gov.uk/file?uri=%2fpeoplepopulationandcommunity%2fpopulationandmigration%2fpopulationestimates%2fdatasets%2fclinicalcommissioninggroupmidyearpopulationestimates%2fmid2020sape23dt6a/sape23dt6amid2020ccg2021estimatesunformatted.xlsx", | |
| write_disk(tmp)) |
| library(tidyverse) | |
| library(lubridate) | |
| # Fix the random records according to this 'seed' so everytime the sequence is | |
| # generated it is always the same | |
| set.seed(123) | |
| # create data and repeat 34 x 3 = 102 times | |
| data <- data.frame( | |
| date = sample(seq(as.Date('1999/01/01'), as.Date('2000/01/01'), by = "day"), 102), |
| # Suicides # | |
| # code based mostly on https://www.trafforddatalab.io/charticles/2020-11-16-suicides/ | |
| # Source: Office for National Statistics | |
| # URL: https://www.ons.gov.uk/peoplepopulationandcommunity/birthsdeathsandmarriages/deaths/datasets/suicidesbylocalauthority | |
| # Licence: OGL 3.0 | |
| # load libraries --------------------------- | |
| library(tidyverse) ; library(httr) ; library(readxl) ; library(ggtext) ; library(stringr) ; library(janitor) |
| library(lubridate) | |
| library(tidyverse) | |
| # create random start and end dates and ids which can be repeated | |
| data <- data.frame( | |
| start_date = sample(seq(as.Date('2019/01/01'), as.Date('2021/01/01'), by = "day"), 300), | |
| end_date = sample(seq(as.Date('2019/01/01'), as.Date('2021/01/01'), by = "day"), 300), | |
| patient_id = floor(runif(300, min = 1, max = 300)) | |
| ) |
| tidy_imd <- function(data, tidy_imd_vars = NULL){ | |
| # First rename all variables | |
| data <- data %>% | |
| dplyr::select(lsoa_code_2011 = LSOAcode2011, | |
| lsoa_name_2011 = LSOAname2011, | |
| everything()) %>% | |
| janitor::clean_names() %>% | |
| dplyr::select(lsoa_code_2011, | |
| lsoa_name_2011, |
| # create a tibble of month names (generated by copying over from Excel using {datapasta} :) | |
| birthmonth <- tibble::tribble( | |
| ~month_full, ~month_short, | |
| "January", "Jan", | |
| "February", "Feb", | |
| "March", "Mar", | |
| "April", "Apr", | |
| "May", "May", | |
| "June", "Jun", |
| Post was removed from RStudio: https://community.rstudio.com/t/how-to-remove-similar-kind-of-data-from-a-csv-file/89706 | |
| I would create a look up table with the categories listed out, apply a filter flag and then use it to filter. Having the separate table of unique categories means you can check your assumptions and it may highlight any data quality issues in the original data around these names. | |
| ```{r reprex} | |
| library(dplyr) | |
| library(stringr) | |
| # Create reprex | |
| df <- tibble::tribble( | |
| ~start, ~end, ~weight, |
