"Please can anyone suggest some dplyr code that can turn my sample data_in into my sample data_out?"
data_in <- tibble::tribble(
~patient, ~delivery_date, ~obs_point, ~obs_date,
"A", "2024-01-28", "antenatal", "2024-01-01",
"A", "2024-01-28", "at_delivery", "2024-01-28",
"B", "2024-01-28", "at_delivery", "2024-01-27",
"B", "2024-01-28", "at_delivery", "2024-01-28",
"C", "2024-01-28", "at_delivery", "2024-01-27",
| library(ggplot2) | |
| library(lubridate, warn.conflicts = FALSE) | |
| t <- lubridate::as_datetime("2021-01-01") | |
| e1 <- lubridate::as_datetime("2021-12-31 23:59:59") | |
| n1 <- as.numeric(lubridate::as.duration(e1 - t)) | |
| s1 <- sample(seq.int(n1), 300L) | |
| imd_api_url <- paste0( | |
| "https://services3.arcgis.com/ivmBBrHfQfDnDf8Q/arcgis/rest/services/", | |
| "Indices_of_Multiple_Deprivation_(IMD)_2019/FeatureServer/0/query" | |
| ) | |
| req <- httr2::request(imd_api_url) |> | |
| httr2::req_url_query(f = "json") | |
| # just get IDs only from API initially (no maxRecordCount for ID queries) | |
| ids <- req |> |
| # https://digital.nhs.uk/data-and-information/data-collections-and-data-sets/data-collections/gp-data-available-through-sdcs | |
| zip_url <- "https://digital.nhs.uk/binaries/content/assets/website-assets/data-and-information/data-collections/general-practice-data-collections/catchment-area-8.zip" | |
| dl <- tempfile("catchment_files", tempdir(), ".zip") | |
| download.file(zip_url, dl, mode = "wb") | |
| out_dir <- paste0(tempdir(), "\\catchment_files") | |
| if (!dir.exists(out_dir)) dir.create(out_dir) | |
| utils::unzip(dl, junkpaths = TRUE, exdir = out_dir) | |
| gp_data <- dir(out_dir, full.names = TRUE) |> |
| { | |
| "breadcrumbs.enabled": false, | |
| "diffEditor.codeLens": true, | |
| "editor.acceptSuggestionOnEnter": "off", | |
| "editor.bracketPairColorization.enabled": false, | |
| "editor.cursorBlinking": "phase", | |
| "editor.cursorStyle": "line", |
| # Apr 2024: stripped out several packages from the list that were loaded as dependencies anyway | |
| pkgs <- c( | |
| "usethis", | |
| "tidyterra", | |
| "devtools", | |
| "tmap", | |
| "janitor", | |
| "afcolours", | |
| "arrow", | |
| "assertthat", |
| [ | |
| { | |
| "key": "ctrl+d", | |
| "command": "editor.action.deleteLines", | |
| "when": "textInputFocus && !editorReadonly" | |
| }, | |
| { | |
| "key": "ctrl+shift+k", | |
| "command": "-editor.action.deleteLines", | |
| "when": "textInputFocus && !editorReadonly" |
| options( | |
| usethis.full_name = "Fran Barton", | |
| usethis.protocol = "ssh", | |
| usethis.description = list( | |
| `Authors@R` = 'person( | |
| "Fran", "Barton", | |
| email = "francis.barton@nuh.nhs.uk", | |
| role = c("aut", "cre"), | |
| comment = c(ORCID = "0000-0002-5650-1176") | |
| )', |
| # originally written at | |
| # https://community.rstudio.com/t/unlist-columns-with-lists-of-different-length/ | |
| # with grateful acknowledgement of those who worked it out before me :-) | |
| ## improved (well, single pipe) version, to summarise values in a df into a summary table | |
| rectangularise <- function(df) { | |
| df %>% | |
| purrr::map(unique) %>% | |
| purrr::map(sort) %>% | |
| purrr::map(~ `length<-`(., max(lengths(map(df, ~ unique(.)))))) %>% |
| # https://gist.github.com/francisbarton/a5907ec17154a59d5c9058d64b3012f5 | |
| # I use this pattern regularly in chatty packages | |
| snippet inf | |
| usethis::ui_info( | |
| stringr::str_glue("") | |
| ) | |
| # set up a testthat block | |
| snippet testt |