remove additional_metadata from archive as well

Author: dajmcdon

R/archive.R

@@ -212,9 +212,6 @@ NULL
 #'   that should be considered key variables (in the language of `data.table`)
 #'   apart from "geo_value", "time_value", and "version". Typical examples
 #'   are "age" or more granular geographies.
-#' @param additional_metadata List of additional metadata to attach to the
-#'   `epi_archive` object. The metadata will have the `geo_type` field; named
-#'   entries from the passed list or will be included as well.
 #' @param compactify Optional; Boolean. `TRUE` will remove some
 #'   redundant rows, `FALSE` will not, and missing or `NULL` will remove
 #'   redundant rows, but issue a warning. See more information at `compactify`.
@@ -295,7 +292,6 @@ new_epi_archive <- function(
     geo_type,
     time_type,
     other_keys,
-    additional_metadata,
     compactify,
     clobberable_versions_start,
     versions_end,
@@ -353,7 +349,6 @@ new_epi_archive <- function(
       geo_type = geo_type,
       time_type = time_type,
       other_keys = other_keys,
-      additional_metadata = additional_metadata,
       clobberable_versions_start = clobberable_versions_start,
       versions_end = versions_end
     ),
@@ -426,7 +421,6 @@ is_locf <- function(vec, tolerance) { # nolint: object_usage_linter
 validate_epi_archive <- function(
     x,
     other_keys,
-    additional_metadata,
     compactify,
     clobberable_versions_start,
     versions_end) {
@@ -437,9 +431,6 @@ validate_epi_archive <- function(
   if (any(c("geo_value", "time_value", "version") %in% other_keys)) {
     cli_abort("`other_keys` cannot contain \"geo_value\", \"time_value\", or \"version\".")
   }
-  if (any(names(additional_metadata) %in% c("geo_type", "time_type"))) {
-    cli_warn("`additional_metadata` names overlap with existing metadata fields \"geo_type\" or \"time_type\".")
-  }
 
   # Conduct checks and apply defaults for `compactify`
   assert_logical(compactify, len = 1, any.missing = FALSE, null.ok = TRUE)
@@ -489,7 +480,6 @@ as_epi_archive <- function(
     geo_type = deprecated(),
     time_type = deprecated(),
     other_keys = character(0L),
-    additional_metadata = list(),
     compactify = NULL,
     clobberable_versions_start = NA,
     .versions_end = max_version_with_row_in(x), ...,
@@ -521,11 +511,10 @@ as_epi_archive <- function(
   time_type <- guess_time_type(x$time_value)
 
   validate_epi_archive(
-    x, other_keys, additional_metadata,
-    compactify, clobberable_versions_start, versions_end
+    x, other_keys, compactify, clobberable_versions_start, versions_end
   )
   new_epi_archive(
-    x, geo_type, time_type, other_keys, additional_metadata,
+    x, geo_type, time_type, other_keys,
     compactify, clobberable_versions_start, versions_end
   )
 }
@@ -554,7 +543,7 @@ print.epi_archive <- function(x, ..., class = TRUE, methods = TRUE) {
     c(
       ">" = if (class) "An `epi_archive` object, with metadata:",
       "i" = if (length(setdiff(key(x$DT), c("geo_value", "time_value", "version"))) > 0) {
-        "Non-standard DT keys: {setdiff(key(x$DT), c('geo_value', 'time_value', 'version'))}"
+        "Other DT keys: {setdiff(key(x$DT), c('geo_value', 'time_value', 'version'))}"
       },
       "i" = if (nrow(x$DT) != 0L) {
         "Min/max time values: {min(x$DT$time_value)} / {max(x$DT$time_value)}"
@@ -690,7 +679,8 @@ print.epi_archive <- function(x, ..., class = TRUE, methods = TRUE) {
 #' @export
 #'
 #' @aliases grouped_epi_archive
-group_by.epi_archive <- function(.data, ..., .add = FALSE, .drop = dplyr::group_by_drop_default(.data)) {
+group_by.epi_archive <- function(.data, ..., .add = FALSE, 
+                                 .drop = dplyr::group_by_drop_default(.data)) {
   # `add` makes no difference; this is an ungrouped `epi_archive`.
   detailed_mutate <- epix_detailed_restricted_mutate(.data, ...)
   assert_logical(.drop)

R/methods-epi_archive.R

@@ -114,9 +114,6 @@ epix_as_of <- function(x, max_version, min_time_value = -Inf, all_versions = FAL
       as_of = max_version,
       other_keys = other_keys
     )
-  if (length(x$additional_metadata) != 0L) {
-    attr(as_of_epi_df, "additional_metadata") <- x$additional_metadata
-  }
 
   return(as_of_epi_df)
 }
@@ -240,9 +237,8 @@ epix_fill_through_version <- function(x, fill_versions_end,
 #'   Default here is `TRUE`.
 #' @return the resulting `epi_archive`
 #'
-#' @details In all cases, `additional_metadata` will be an empty list, and
-#'   `clobberable_versions_start` will be set to the earliest version that could
-#'   be clobbered in either input archive.
+#' @details In all cases, `clobberable_versions_start` will be set to the 
+#'   earliest version that could be clobbered in either input archive.
 #'
 #' @examples
 #' # Example 1
@@ -331,18 +327,6 @@ epix_merge <- function(x, y,
     cli_abort("`x` and `y` must share data type on their `time_value` column.")
   }
 
-  if (length(x$additional_metadata) != 0L) {
-    cli_warn("x$additional_metadata won't appear in merge result",
-      class = "epiprocess__epix_merge_ignores_additional_metadata"
-    )
-  }
-  if (length(y$additional_metadata) != 0L) {
-    cli_warn("y$additional_metadata won't appear in merge result",
-      class = "epiprocess__epix_merge_ignores_additional_metadata"
-    )
-  }
-  result_additional_metadata <- list()
-
   result_clobberable_versions_start <-
     if (all(is.na(c(x$clobberable_versions_start, y$clobberable_versions_start)))) {
       NA # (any type of NA is fine here)
@@ -508,7 +492,6 @@ epix_merge <- function(x, y,
   return(as_epi_archive(
     result_dt[], # clear data.table internal invisibility flag if set
     other_keys = setdiff(key(result_dt), c("geo_value", "time_value", "version")),
-    additional_metadata = result_additional_metadata,
     # It'd probably be better to pre-compactify before the merge, and might be
     # guaranteed not to be necessary to compactify the merge result if the
     # inputs are already compactified, but at time of writing we don't have

man/epi_archive.Rd

@@ -77,14 +77,6 @@ test_that("other_keys cannot contain names geo_value, time_value or version", {
   )
 })
 
-test_that("Warning thrown when other_metadata contains overlapping names with geo_type field", {
-  expect_warning(as_epi_archive(archive_data, additional_metadata = list(geo_type = 1), compactify = FALSE),
-    regexp = "`additional_metadata` names overlap with existing metadata fields"
-  )
-  expect_warning(as_epi_archive(archive_data, additional_metadata = list(time_type = 1), compactify = FALSE),
-    regexp = "`additional_metadata` names overlap with existing metadata fields"
-  )
-})
 
 test_that("epi_archives are correctly instantiated with a variety of data types", {
   d <- as.Date("2020-01-01")
@@ -98,22 +90,22 @@ test_that("epi_archives are correctly instantiated with a variety of data types"
 
   ea1 <- as_epi_archive(df, compactify = FALSE)
   expect_equal(key(ea1$DT), c("geo_value", "time_value", "version"))
-  expect_equal(ea1$additional_metadata, list())
+  expect_null(ea1$additional_metadata)
 
-  ea2 <- as_epi_archive(df, other_keys = "value", additional_metadata = list(value = df$value), compactify = FALSE)
+  ea2 <- as_epi_archive(df, other_keys = "value", compactify = FALSE)
   expect_equal(key(ea2$DT), c("geo_value", "time_value", "value", "version"))
-  expect_equal(ea2$additional_metadata, list(value = df$value))
+  expect_null(ea2$additional_metadata)
 
   # Tibble
   tib <- tibble::tibble(df, code = "x")
 
   ea3 <- as_epi_archive(tib, compactify = FALSE)
   expect_equal(key(ea3$DT), c("geo_value", "time_value", "version"))
-  expect_equal(ea3$additional_metadata, list())
+  expect_null(ea3$additional_metadata)
 
-  ea4 <- as_epi_archive(tib, other_keys = "code", additional_metadata = list(value = df$value), compactify = FALSE)
+  ea4 <- as_epi_archive(tib, other_keys = "code", compactify = FALSE)
   expect_equal(key(ea4$DT), c("geo_value", "time_value", "code", "version"))
-  expect_equal(ea4$additional_metadata, list(value = df$value))
+  expect_null(ea4$additional_metadata)
 
   # Keyed data.table
   kdt <- data.table::data.table(
@@ -128,12 +120,12 @@ test_that("epi_archives are correctly instantiated with a variety of data types"
   ea5 <- as_epi_archive(kdt, compactify = FALSE)
   # Key from data.table isn't absorbed when as_epi_archive is used
   expect_equal(key(ea5$DT), c("geo_value", "time_value", "version"))
-  expect_equal(ea5$additional_metadata, list())
+  expect_null(ea5$additional_metadata)
 
-  ea6 <- as_epi_archive(kdt, other_keys = "value", additional_metadata = list(value = df$value), compactify = FALSE)
+  ea6 <- as_epi_archive(kdt, other_keys = "value", compactify = FALSE)
   # Mismatched keys, but the one from as_epi_archive overrides
   expect_equal(key(ea6$DT), c("geo_value", "time_value", "value", "version"))
-  expect_equal(ea6$additional_metadata, list(value = df$value))
+  expect_null(ea6$additional_metadata)
 
   # Unkeyed data.table
   udt <- data.table::data.table(
@@ -146,11 +138,11 @@ test_that("epi_archives are correctly instantiated with a variety of data types"
 
   ea7 <- as_epi_archive(udt, compactify = FALSE)
   expect_equal(key(ea7$DT), c("geo_value", "time_value", "version"))
-  expect_equal(ea7$additional_metadata, list())
+  expect_null(ea7$additional_metadata)
 
-  ea8 <- as_epi_archive(udt, other_keys = "code", additional_metadata = list(value = df$value), compactify = FALSE)
+  ea8 <- as_epi_archive(udt, other_keys = "code", compactify = FALSE)
   expect_equal(key(ea8$DT), c("geo_value", "time_value", "code", "version"))
-  expect_equal(ea8$additional_metadata, list(value = df$value))
+  expect_null(ea8$additional_metadata)
 
   # epi_df
   edf1 <- jhu_csse_daily_subset %>%
@@ -159,11 +151,11 @@ test_that("epi_archives are correctly instantiated with a variety of data types"
 
   ea9 <- as_epi_archive(edf1, compactify = FALSE)
   expect_equal(key(ea9$DT), c("geo_value", "time_value", "version"))
-  expect_equal(ea9$additional_metadata, list())
+  expect_null(ea9$additional_metadata)
 
-  ea10 <- as_epi_archive(edf1, other_keys = "code", additional_metadata = list(value = df$value), compactify = FALSE)
+  ea10 <- as_epi_archive(edf1, other_keys = "code", compactify = FALSE)
   expect_equal(key(ea10$DT), c("geo_value", "time_value", "code", "version"))
-  expect_equal(ea10$additional_metadata, list(value = df$value))
+  expect_null(ea10$additional_metadata)
 
   # Keyed epi_df
   edf2 <- data.frame(
@@ -180,11 +172,11 @@ test_that("epi_archives are correctly instantiated with a variety of data types"
 
   ea11 <- as_epi_archive(edf2, compactify = FALSE)
   expect_equal(key(ea11$DT), c("geo_value", "time_value", "version"))
-  expect_equal(ea11$additional_metadata, list())
+  expect_null(ea11$additional_metadata)
 
-  ea12 <- as_epi_archive(edf2, other_keys = "misc", additional_metadata = list(value = df$misc), compactify = FALSE)
+  ea12 <- as_epi_archive(edf2, other_keys = "misc", compactify = FALSE)
   expect_equal(key(ea12$DT), c("geo_value", "time_value", "misc", "version"))
-  expect_equal(ea12$additional_metadata, list(value = df$misc))
+  expect_null(ea12$additional_metadata)
 })
 
 test_that("`epi_archive` rejects nonunique keys", {

man/epix_merge.Rd

@@ -177,6 +177,8 @@ test_that("epix_merge forbids and warns on metadata and naming issues", {
     ),
     regexp = "overlapping.*names"
   )
+  
+  skip("Additional metadata is no longer used, not tested.")
   expect_warning(
     epix_merge(
       as_epi_archive(tibble::tibble(geo_value = "ak", time_value = test_date, version = test_date + 1L, x_value = 1L),

tests/testthat/test-archive.R

@@ -150,7 +150,7 @@ test_that("Renaming columns gives appropriate colnames and metadata", {
   expect_identical(attr(renamed_edf1, "metadata")$other_keys, c("age_group"))
   # renaming using select
   renamed_edf2 <- edf %>%
-    as_epi_df(additional_metadata = list(other_keys = "age")) %>%
+    as_epi_df(other_keys = "age") %>%
     select(geo_value, time_value, age_group = age, value)
   expect_identical(renamed_edf1, renamed_edf2)
 })

tests/testthat/test-epix_merge.R

@@ -119,7 +119,6 @@ The following pieces of metadata are included as fields in an `epi_archive`
 object:
 
 * `geo_type`: the type for the geo values.
-* `additional_metadata`: list of additional metadata for the data archive.
 
 Metadata for an `epi_archive` object `x` can be accessed (and altered) directly,
 as in `x$geo_type`, etc. Just like `as_epi_df()`, the function