Data science platform changes

.Rbuildignore

@@ -14,7 +14,7 @@ README.md
 ^docs$
 ^pkgdown$
 ^codecov\.yml$
-inst/shiny-examples/SCD-effect-sizes/rsconnect
+inst/shiny-examples/SCD-effect-sizes/rsconnect/shinyapps.io/jepusto/SCD-effect-sizes.dcf
 ^CRAN-SUBMISSION$
 ^init.R$
 ^run.R$

DESCRIPTION

@@ -6,8 +6,7 @@ Authors@R: c(
   person(c("James","E."), "Pustejovsky", email = "jepusto@gmail.com", role = c("aut", "cre")),
   person("Man", "Chen", email = "manchen9005@gmail.com", role = "aut"),
   person(c("Daniel","M."), "Swan", email = "dms031000@gmail.com", role = "aut"),
-  person("Paulina", "Grekov", email = "grekov.paulina@gmail.com", role = "ctb"),
-  person("Amy", "Cai", email = "amycai2023@gmail.com", role = "ctb")
+  person("Paulina", "Grekov", email = "grekov.paulina@gmail.com", role = "ctb")
   )
 Description: 
   Provides R functions for calculating basic effect size indices for 
@@ -20,8 +19,6 @@ Description:
 URL: https://jepusto.github.io/SingleCaseES/
 BugReports: https://github.com/jepusto/SingleCaseES/issues
 License: GPL-3
-Depends: 
-  R (>= 4.1.0)
 Imports:
   stats, 
   purrr,
@@ -36,6 +33,7 @@ Suggests:
     shinytest2 (>= 0.5.0),
     stringr,
     rvest,
+    xml2,
     ggplot2,
     purrrlyr,
     testthat,

R/calculate-effect-sizes.R

@@ -456,46 +456,62 @@ batch_calc_ES <- function(dat,
     if (utils::packageVersion("dplyr") >= '1.1.0') {
       dat |>
       dplyr::group_by(!!!rlang::syms(c(grouping, aggregate))) |>
-      dplyr::reframe(
-        calc_ES(
-          condition = .data[[condition]],
-          outcome = .data[[outcome]],
-          baseline_phase = baseline_phase,
-          intervention_phase = intervention_phase,
-          ES = ES_names,
-          improvement = .data[[improvement]],
-          scale =  .data[[scale]],
-          intervals = .data[[intervals]],
-          observation_length = .data[[observation_length]],
-          goal = if ("PoGO" %in% ES_names) .data[[goal]] else NULL,
-          confidence = confidence,
-          format = "long",
-          ...,
-          warn = warn
+        dplyr::reframe(
+          tryCatch(
+            calc_ES(
+              condition = .data[[condition]],
+              outcome = .data[[outcome]],
+              baseline_phase = baseline_phase,
+              intervention_phase = intervention_phase,
+              ES = ES_names,
+              improvement = .data[[improvement]],
+              scale =  .data[[scale]],
+              intervals = .data[[intervals]],
+              observation_length = .data[[observation_length]],
+              goal = if ("PoGO" %in% ES_names) .data[[goal]] else NULL,
+              confidence = confidence,
+              format = "long",
+              ...,
+              warn = warn
+            ),
+            error = function(e) {
+              stop(
+                paste0("Effect size calculation failed: ", conditionMessage(e)),
+                call. = FALSE
+              )
+            }
+          )
         )
-      ) 
     } else {
       dat |>
-      dplyr::group_by(!!!rlang::syms(c(grouping, aggregate))) |>
-      dplyr::summarise(
-        calc_ES(
-          condition = .data[[condition]],
-          outcome = .data[[outcome]],
-          baseline_phase = baseline_phase,
-          intervention_phase = intervention_phase,
-          ES = ES_names,
-          improvement = .data[[improvement]],
-          scale =  .data[[scale]],
-          intervals = .data[[intervals]],
-          observation_length = .data[[observation_length]],
-          goal = if ("PoGO" %in% ES_names) .data[[goal]] else NULL,
-          confidence = confidence,
-          format = "long",
-          ...,
-          warn = warn
-        ),
-        .groups = "drop"
-      )
+        dplyr::group_by(!!!rlang::syms(c(grouping, aggregate))) |>
+        dplyr::summarise(
+          tryCatch(
+            calc_ES(
+              condition = .data[[condition]],
+              outcome = .data[[outcome]],
+              baseline_phase = baseline_phase,
+              intervention_phase = intervention_phase,
+              ES = ES_names,
+              improvement = .data[[improvement]],
+              scale = .data[[scale]],
+              intervals = .data[[intervals]],
+              observation_length = .data[[observation_length]],
+              goal = if ("PoGO" %in% ES_names) .data[[goal]] else NULL,
+              confidence = confidence,
+              format = "long",
+              ...,
+              warn = warn
+            ),
+            error = function(e) {
+              stop(
+                paste0("Effect size calculation failed: ", conditionMessage(e)),
+                call. = FALSE
+              )
+            }
+          ),
+          .groups = "drop"
+        )
     }
 
   ES_long_names <- names(ES_ests_long)
@@ -548,15 +564,24 @@ batch_calc_ES <- function(dat,
       n_weights <- 
         dat |> 
         dplyr::filter(!!rlang::sym(condition) %in% c(baseline_phase, intervention_phase)) |> 
-        dplyr::mutate(!!rlang::sym(condition) := ifelse(!!rlang::sym(condition) == baseline_phase, "A", "B")) |> 
+        dplyr::mutate(
+          !!rlang::sym(condition) := ifelse(
+            !!rlang::sym(condition) == baseline_phase,
+            "A",
+            "B"
+          )
+        ) |> 
         dplyr::group_by(!!!rlang::syms(c(grouping, aggregate, condition))) |> 
         dplyr::summarise(n = dplyr::n(), .groups = "drop") |> 
-        tidyr::pivot_wider(names_from = !!rlang::sym(condition), values_from = n) |>
+        tidyr::pivot_wider(
+          names_from = !!rlang::sym(condition),
+          values_from = n
+        ) |>
         dplyr::ungroup() |> 
         dplyr::mutate(
           weights = calc_ES_weights(weighting, A = A, B = B)
         ) |> 
-        dplyr::select(-c(A,B))
+        dplyr::select(-dplyr::any_of(c("A", "B")))
 
       ES_weights <- 
         ES_ests_long |>