version 1.0.0

DESCRIPTION

@@ -1,11 +1,13 @@
 Package: codec
 Title: Community Data Explorer for Cincinnati
-Version: 0.7.2
+Version: 1.0.0
 Authors@R: c(
     person("Cole", "Brokamp",
            email = "[email protected]",
 	   role = c("aut", "cre")),
     person("Erika", "Manning",
+           role = "aut"),
+    person("Andrew", "Vancil",
            role = "aut")
 	   )
 Description: codec provides tools for working with metadata in R and storing it alongside data in a YAML file. This package serves as the definition of the CoDEC data specifications and provides helpers to contribute and validate CoDEC data.
@@ -18,33 +20,29 @@ Suggests:
     roxygen2,
     knitr,
     rmarkdown,
+    dplyr,
     curl,
     glue,
-    mapview,
-    gh,
-    DT,
-    leaflet,
     callr,
     downloadthis,
-    cincy (>= 1.0.2),
+    tibble,
     bsplus
 Remotes:
-    geomarker-io/cincy
+    geomarker-io/cincy,
+    cole-brokamp/fr
 Config/testthat/edition: 3
 URL: https://github.com/geomarker-io/codec,
     http://geomarker.io/codec/
 BugReports: https://github.com/geomarker-io/codec/issues
 Imports: 
-    dplyr,
-    forcats,
     fs,
     purrr (>= 1.0.0),
-    readr,
-    rlang (>= 0.4.11),
     stringi,
     stringr,
-    tibble,
-    yaml, 
+    yaml,
+    vroom,
+    cincy (>= 1.1.0),
+    fr (>= 0.4.0),
     sf
 VignetteBuilder: knitr
 Depends: 

NAMESPACE

@@ -1,13 +1,5 @@
 # Generated by roxygen2: do not edit by hand
 
-export(":=")
-export(.data)
-export(add_attr_from_tdr)
-export(add_attrs)
-export(add_col_attrs)
-export(add_type_attrs)
-export(as_label)
-export(as_name)
 export(check_codec_tdr)
 export(check_codec_tdr_csv)
 export(check_files)
@@ -16,18 +8,4 @@ export(check_tdr_path)
 export(codec_colors)
 export(codec_data)
 export(codec_tdr)
-export(enquo)
-export(enquos)
-export(glimpse_attr)
-export(glimpse_schema)
-export(glimpse_tdr)
-export(read_tdr)
-export(read_tdr_csv)
-export(write_tdr)
-export(write_tdr_csv)
-importFrom(rlang,":=")
-importFrom(rlang,.data)
-importFrom(rlang,as_label)
-importFrom(rlang,as_name)
-importFrom(rlang,enquo)
-importFrom(rlang,enquos)
+importFrom(cincy,interpolate)

R/codec_check.R

@@ -9,63 +9,51 @@
 #' - the data contains a year (or year and month) column(s)
 #' - all fields in the CSV data are described in the metadata and vice-versa
 #' See `vignette("codec-specs")` for the CoDEC specifications.
-#' @param tdr a codec tabular-data-resource
-#' @param tdr_md a codec tabular-data-resource metadata list object
+#' @param x a codec fr_tdr object (or data frame for check_census_tract_id(), check_date()
+#' and a list for check_codec_tdr())
 #' @param path path to tdr folder
 #' @param name the name field from tabular-data-resource.yaml
 #' @return for `check_codec_tdr_csv`, a tibble with added
 #' tabular-data-resource attributes (equivalent to read_tdr_csv with `codec = TRUE`)
+#' @importFrom cincy interpolate
 #' @export
 check_codec_tdr_csv <- function(path) {
   check_files(path)
-  tdr <- read_tdr(path)$tdr
-  check_codec_tdr(tdr)
+  d <- fr::read_fr_tdr(fs::path(path, "tabular-data-resource.yaml"))
 
-  md_fields <- names(tdr$schema$fields)
-  d_fields <- names(readr::read_csv(read_tdr(path)$csv_file, n_max = 0, show_col_types = FALSE))
-  if(! all(d_fields %in% md_fields)) {
-    stop("the metadata does not describe all fields in the data", call. = FALSE)
-  }
-  if(! all(md_fields %in% d_fields)) {
-    stop("the metadata describes fields that are not in the data", call. = FALSE)
-  }
-
-  tdr_d <- read_tdr_csv(path)
-  check_data(tdr_d)
-  return(invisible(tdr_d))
-}
+  check_codec_tdr(as.list(d))
+  check_census_tract_id(as.data.frame(d))
+  check_date(as.data.frame(d))
 
-#' Check data
-#' @rdname check_codec_tdr_csv
-check_data <- function(tdr) {
-  check_census_tract_id(tdr)
-  check_date(tdr)
+  return(invisible(d))
 }
 
 #' Check census tract id column
 #' @rdname check_codec_tdr_csv
-check_census_tract_id <- function(tdr) {
+check_census_tract_id <- function(x) {
   census_tract_id_names <- paste0("census_tract_id", c("_2000", "_2010", "_2020"))
+  tdr_data <- as.data.frame(x)
+  tdr_data_names <- names(tdr_data)
 
   # has census_tract_id_{year} or census_tract_id column
-  if (!any(names(tdr) %in% census_tract_id_names)) {
+  if (!any(tdr_data_names %in% census_tract_id_names)) {
     stop("must contain a census tract id column called census_tract_id_2000, census_tract_id_2010, or census_tract_id_2020", call. = FALSE)
   }
 
   # make sure only one tract column
-  if (sum(names(tdr) %in% census_tract_id_names) > 1) {
+  if (sum(tdr_data_names %in% census_tract_id_names) > 1) {
     stop("must contain only one census tract id column", call. = FALSE)
     }
 
-  census_tract_id_name <- census_tract_id_names[census_tract_id_names %in% names(tdr)]
+  census_tract_id_name <- census_tract_id_names[census_tract_id_names %in% tdr_data_names]
   census_tract_id_year <- stringr::str_extract(census_tract_id_name, "[0-9]+")
 
   required_census_tract_ids <-
     parse(text = paste0("cincy::tract_tigris_", census_tract_id_year)) |>
     eval() |>
     purrr::pluck(paste0("census_tract_id_", census_tract_id_year))
 
-  if (!all(required_census_tract_ids %in% tdr[[census_tract_id_name]])) {
+  if (!all(required_census_tract_ids %in% tdr_data[[census_tract_id_name]])) {
     stop("the census tract id column, ",
       census_tract_id_name,
       ", does not contain every census tract in ",
@@ -74,28 +62,31 @@ check_census_tract_id <- function(tdr) {
     )
   }
 
-  return(invisible(tdr))
+  return(invisible(x))
 }
 
-#' Check date
+#' Check year or year-month column
 #' @rdname check_codec_tdr_csv
-check_date <- function(tdr) {
+check_date <- function(x) {
+
+  tdr_data <- as.data.frame(x)
+  tdr_data_names <- names(tdr_data)
 
-  if (! "year" %in% names(tdr)) {
+  if (! "year" %in% tdr_data_names) {
     stop("must contain a 'year' column", call. = FALSE)
   }
 
-  years <- unique(tdr$year)
-  if (! identical(years, as.integer(years))) {
-    stop("the 'year' field must only contain integer years", call. = FALSE)
+  years <- unique(tdr_data$year)
+  if (! all(years %in% 1970:2099)) {
+    stop("the 'year' field must only contain integer years between 1970 and 2099", call. = FALSE)
   }
 
-  if ("month" %in% names(tdr)) {
-    if (! all(tdr$month %in% 1:12)) {
+  if ("month" %in% tdr_data_names) {
+    if (! all(tdr_data$month %in% 1:12)) {
       stop("the 'month' field  must only contain integer values 1-12", call. = FALSE)
     }
   }
-  return(invisible(tdr))
+  return(invisible(x))
 }
 
 #' Check files
@@ -129,17 +120,18 @@ check_files <- function(path) {
 
   # try to read (first 100 lines of) CSV file
   test_read_csv_file <-
-    purrr::safely(readr::read_csv)(
+    purrr::safely(vroom::vroom)(
       file = tdr_csv,
+      delim = ",",
       n_max = 100,
       col_names = TRUE,
       show_col_types = FALSE,
-      locale = readr::locale(
+      locale = vroom::locale(
         encoding = "UTF-8",
         decimal_mark = ".",
         grouping_mark = "",
       ),
-      name_repair = "check_unique",
+      .name_repair = "check_unique",
     )
 
   if (!is.null(test_read_csv_file$error)) {
@@ -152,7 +144,9 @@ check_files <- function(path) {
 #' check CoDEC tdr
 #' @rdname check_codec_tdr_csv
 #' @export
-check_codec_tdr <- function(tdr_md) {
+check_codec_tdr <- function(x) {
+
+  tdr_md <- as.list(x)
 
   # must have "name" and "path" descriptors
   if (!purrr::pluck_exists(tdr_md, "name")) stop("`name` property descriptor is required", call. = FALSE)
@@ -212,7 +206,7 @@ check_codec_tdr <- function(tdr_md) {
     )
   }
 
-  return(invisible(tdr_md))
+  return(invisible(x))
 }
 
 
@@ -242,6 +236,8 @@ check_tdr_path <- function(path) {
   # path ends with .csv
   if (! fs::path_ext(path) == "csv") stop("'path' must end with '.csv'", call. = FALSE)
   # path can be a URL
+
+  is_url <- function(.x) grepl("^((http|ftp)s?|sftp)://", .x)
   if (is_url(path)) return(invisible(NULL))
   # if not URL, check for absolute path
   if (fs::is_absolute_path(path)) stop("'path' must be a relative file path")