fix barchart and forest_rsp as noticed by @mhallal1

R/tm_g_barchart_simple.R

@@ -334,13 +334,13 @@ srv_g_barchart_simple <- function(id,
       })
       quo <- teal.code::new_quosure(env = data)
       quo <- teal.code::eval_code(quo, as.expression(merge_inputs()$expr))
-      teal::validate_has_data(quo[["ANL"]], 2)
       quo
     })
 
     count_q <- shiny::reactive({
       req(merged_data_q())
       quo <- merged_data_q()
+      teal::validate_has_data(quo[["ANL"]], 2)
       groupby_vars <- r_groupby_vars()
 
       # count

R/tm_g_forest_rsp.R

@@ -523,10 +523,15 @@ srv_g_forest_rsp <- function(id,
         input[[extract_input("paramcd", paramcd$filter[[1]]$dataname, filter = TRUE)]]
       ),
       handlerExpr = {
+        req(anl_merged_q())
         anl <- anl_merged_q()[["ANL"]]
         aval_var <- anl_merged()$columns_source$aval_var
+
+        paramcd_level <- unlist(anl_merged()$filter_info$paramcd[[1]]$selected)
+        if (length(paramcd_level) == 0) return(NULL)
+
         sel_param <- if (is.list(default_responses)) {
-          default_responses[[input[[extract_input("paramcd", paramcd$filter[[1]]$dataname, filter = TRUE)]]]]
+          default_responses[[paramcd_level]]
         } else {
           default_responses
         }
@@ -558,6 +563,7 @@ srv_g_forest_rsp <- function(id,
 
     # Prepare the analysis environment (filter data, check data, populate envir).
     validate_checks <- shiny::reactive({
+      req(anl_merged_q())
       q1 <- anl_merged_q()
       adsl_filtered <- q1[[parentname]]
       anl_filtered <- q1[[dataname]]