doc: docs

NEWS.md

@@ -32,7 +32,10 @@ Pre-1.0.0 numbering scheme: 0.x will indicate releases, while 0.x.y will indicat
 ## Breaking changes
 
 - Switched `epi_df`'s `other_keys` default from `NULL` to `character(0)`; PR #390
-- Refactor `epi_archive` to use S3 instead of R6 for its object model. The calls to some functions will change, but the functionality will remain the same. It will also help us maintain the package better in the future. (#340)
+- Refactor `epi_archive` to use S3 instead of R6 for its object model. The
+  functionality stay the same, but it will break the member function interface.
+  For migration, convert `epi_archive$merge` to `epi_archive %>% epix_merge` and
+  similar for `slide` and `fill_through_version` (#340).
 
 # epiprocess 0.7.0
 

R/grouped_epi_archive.R

@@ -1,3 +1,9 @@
+# At time of writing, roxygen parses content in collation order, impacting the
+# presentation of .Rd files that document multiple functions (see
+# https://github.com/r-lib/roxygen2/pull/324). We use @include tags (determining
+# `Collate:`) below to get the desired ordering.
+
+
 #' Get var names from select-only `tidy_select`ing `...` in `.data`
 #'
 #' Convenience function for performing a `tidy_select` on dots according to its
@@ -420,13 +426,6 @@ epix_slide.grouped_epi_archive <- function(x, f, ..., before, ref_time_values,
 }
 
 
-# At time of writing, roxygen parses content in collation order, impacting the
-# presentation of .Rd files that document multiple functions (see
-# https://github.com/r-lib/roxygen2/pull/324). Use @include tags (determining
-# `Collate:`) and ordering of functions within each file in order to get the
-# desired ordering.
-
-
 #' @include methods-epi_archive.R
 #' @rdname group_by.epi_archive
 #'