Merge branch 'ndefries/epix-slide-pass-reftimevalue-without-env' into ndefries/epix-slide-dots-rtv

Author: nmdefries

.github/workflows/R-CMD-check.yaml

@@ -4,7 +4,7 @@ on:
   push:
     branches: [main, master]
   pull_request:
-    branches: [main, master]
+    branches: [main, master, dev]
 
 name: R-CMD-check
 

.github/workflows/pkgdown.yaml

@@ -4,7 +4,7 @@ on:
   push:
     branches: [main, master]
   pull_request:
-    branches: [main, master]
+    branches: [main, master, dev]
   release:
     types: [published]
   workflow_dispatch:

DESCRIPTION

@@ -1,7 +1,7 @@
 Type: Package
 Package: epiprocess
 Title: Tools for basic signal processing in epidemiology
-Version: 0.6.0
+Version: 0.6.0.9999
 Authors@R: c(
     person("Jacob", "Bien", role = "ctb"),
     person("Logan", "Brooks", role = "aut"),
@@ -21,7 +21,8 @@ Description: This package introduces a common data structure for epidemiological
     work with revisions to these data sets over time, and offers associated
     utilities to perform basic signal processing tasks.
 License: MIT + file LICENSE
-Imports: 
+Imports:
+    cli,
     data.table,
     dplyr (>= 1.0.0),
     fabletools,
@@ -48,7 +49,7 @@ Suggests:
     knitr,
     outbreaks,
     rmarkdown,
-    testthat (>= 3.0.0),
+    testthat (>= 3.1.5),
     waldo (>= 0.3.1),
     withr
 VignetteBuilder: 
@@ -61,7 +62,7 @@ Config/testthat/edition: 3
 Encoding: UTF-8
 LazyData: true
 Roxygen: list(markdown = TRUE)
-RoxygenNote: 7.2.1
+RoxygenNote: 7.2.3
 Depends: 
     R (>= 2.10)
 URL: https://cmu-delphi.github.io/epiprocess/

NAMESPACE

@@ -6,6 +6,7 @@ S3method(as_epi_df,data.frame)
 S3method(as_epi_df,epi_df)
 S3method(as_epi_df,tbl_df)
 S3method(as_epi_df,tbl_ts)
+S3method(as_tibble,epi_df)
 S3method(as_tsibble,epi_df)
 S3method(dplyr_col_modify,col_modify_recorder_df)
 S3method(dplyr_col_modify,epi_df)
@@ -17,6 +18,7 @@ S3method(group_by,epi_archive)
 S3method(group_by,epi_df)
 S3method(group_by,grouped_epi_archive)
 S3method(group_by_drop_default,grouped_epi_archive)
+S3method(group_modify,epi_df)
 S3method(groups,grouped_epi_archive)
 S3method(next_after,Date)
 S3method(next_after,integer)
@@ -30,7 +32,6 @@ export(archive_cases_dv_subset)
 export(arrange)
 export(as_epi_archive)
 export(as_epi_df)
-export(as_slide_computation)
 export(as_tsibble)
 export(detect_outlr)
 export(detect_outlr_rm)
@@ -85,38 +86,36 @@ importFrom(dplyr,ungroup)
 importFrom(lubridate,days)
 importFrom(lubridate,weeks)
 importFrom(magrittr,"%>%")
+importFrom(purrr,map_lgl)
 importFrom(rlang,"!!!")
 importFrom(rlang,"!!")
 importFrom(rlang,.data)
 importFrom(rlang,.env)
 importFrom(rlang,arg_match)
-importFrom(rlang,call2)
 importFrom(rlang,caller_arg)
 importFrom(rlang,caller_env)
 importFrom(rlang,check_dots_empty0)
 importFrom(rlang,enquo)
 importFrom(rlang,enquos)
-importFrom(rlang,env)
-importFrom(rlang,eval_tidy)
 importFrom(rlang,f_env)
 importFrom(rlang,f_rhs)
+importFrom(rlang,global_env)
 importFrom(rlang,is_environment)
 importFrom(rlang,is_formula)
 importFrom(rlang,is_function)
+importFrom(rlang,is_missing)
 importFrom(rlang,is_quosure)
 importFrom(rlang,is_string)
 importFrom(rlang,missing_arg)
 importFrom(rlang,new_function)
-importFrom(rlang,pairlist2)
-importFrom(rlang,quo_get_env)
-importFrom(rlang,quo_get_expr)
 importFrom(rlang,quo_is_missing)
-importFrom(rlang,quo_set_env)
 importFrom(rlang,sym)
 importFrom(rlang,syms)
 importFrom(stats,cor)
 importFrom(stats,median)
+importFrom(tibble,as_tibble)
 importFrom(tidyr,unnest)
 importFrom(tidyselect,eval_select)
 importFrom(tidyselect,starts_with)
 importFrom(tsibble,as_tsibble)
+importFrom(utils,tail)

NEWS.md

@@ -4,6 +4,45 @@ Note that `epiprocess` uses the [Semantic Versioning
 ("semver")](https://semver.org/) scheme for all release versions, but any
 inter-release development versions will include an additional ".9999" suffix.
 
+## Breaking changes:
+
+* Changes to `epix_slide`:
+  * The `f` computation is now required to take at least three arguments. `f`
+    must take an `epi_df` with the same column names as the archive's `DT`,
+    minus the `version` column; followed by a one-row tibble containing the
+    values of the grouping variables for the associated group; followed by a
+    Date containing the reference time value that was used; followed by any
+    number of named arguments.
+
+## New features:
+
+* `epix_slide` has been made more like `dplyr::group_modify`. It will no longer
+  perform element/row recycling for size stability, accepts slide computation
+  outputs containing any number of rows, and no longer supports `all_rows`.
+  * To keep the old behavior, manually perform row recycling within `f`
+    computations, and/or `left_join` a data frame representing the desired
+    output structure with the current `epix_slide()` result to obtain the
+    desired repetitions and completions expected with `all_rows = TRUE`.
+* `epix_slide` will only output grouped or ungrouped tibbles. Previously, it
+  would sometimes output `epi_df`s, but not consistently, and not always with
+  the metadata desired. Future versions will revisit this design, and consider
+  more closely whether/when/how to output an `epi_df`.
+  * To keep the old behavior, convert the output of `epix_slide()` to `epi_df`
+    when desired and set the metadata appropriately.
+* `epix_slide` `f` computations passed as functions or formulas now have
+  access to the reference time value. If `f` is a function, it is passed a
+  Date containing the reference time value as the third argument. If a
+  formula, `f` can access the reference time value via `.z` or
+  `.ref_time_value`.
+
+## Improvements:
+
+* `epi_slide` and `epix_slide` now support `as_list_col = TRUE` when the slide
+  computations output atomic vectors, and output a list column in "chopped"
+  format (see `tidyr::chop`).
+* `epi_slide` now works properly with slide computations that output just a
+  `Date` vector, rather than converting `slide_value` to a numeric column.
+
 # epiprocess 0.6.0
 
 ## Breaking changes:
@@ -23,6 +62,13 @@ inter-release development versions will include an additional ".9999" suffix.
   * Slide functions now keep any grouping of `x` in their results, like
     `mutate` and `group_modify`.
     * To obtain the old behavior, `dplyr::ungroup` the slide results immediately.
+* Additional `epi_slide` changes:
+  * When using `as_list_col = TRUE` together with `ref_time_values` and
+    `all_rows=TRUE`, the marker for excluded computations is now a `NULL` entry
+    in the list column, rather than a `NA`; if you are using `tidyr::unnest()`
+    afterward and want to keep these missing data markers, you will need to
+    replace the `NULL` entries with `NA`s. Skipped computations are now more
+    uniformly detectable using `vctrs` methods.
 * Additional`epix_slide` changes:
   * `epix_slide`'s `group_by` argument has been replaced by `dplyr::group_by` and
     `dplyr::ungroup` S3 methods. The `group_by` method uses "data masking" (also

R/archive.R

@@ -307,7 +307,7 @@ epi_archive =
               Abort("compactify must be boolean or null.")
             }
 
-            # Apply defaults and conduct checks and apply defaults for
+            # Apply defaults and conduct checks for
             # `clobberable_versions_start`, `versions_end`:
             if (missing(clobberable_versions_start)) {
               clobberable_versions_start <- NA
@@ -640,7 +640,7 @@ epi_archive =
           slide = function(f, ..., before, ref_time_values, 
                            time_step, new_col_name = "slide_value",
                            as_list_col = FALSE, names_sep = "_",
-                           all_rows = FALSE, all_versions = FALSE) {
+                           all_versions = FALSE) {
             # For an "ungrouped" slide, treat all rows as belonging to one big
             # group (group by 0 vars), like `dplyr::summarize`, and let the
             # resulting `grouped_epi_archive` handle the slide:
@@ -649,7 +649,7 @@ epi_archive =
               before = before, ref_time_values = ref_time_values,
               time_step = time_step, new_col_name = new_col_name,
               as_list_col = as_list_col, names_sep = names_sep,
-              all_rows = all_rows, all_versions = all_versions
+              all_versions = all_versions
             ) %>%
               # We want a slide on ungrouped archives to output something
               # ungrouped, rather than retaining the trivial (0-variable)