[Bug]: Error creating a Teal app - Validation fail in FilteredData constructor

[vedhav]

What happened?

Note: Using the current refactor now
When running the example app for teal.modules.hermes::adtteSpecServer the validation fails with this error:

Listening on http://127.0.0.1:6616
Warning: Error in self$get_data: Assertion on 'dataname' failed: Must be a subset of the empty set, i.e. also empty.
  3: runApp
  2: print.shiny.appobj
  1: <Anonymous>

On the main branch the error is:

Warning: Error in validate_dataset_args: Assertion on 'dataset_args' failed: Must be of type 'list', not 'NULL'.

Example app

ui <- function(id) {
  ns <- NS(id)

  teal.widgets::standard_layout(
    encoding = uiOutput(ns("encoding_ui")),
    output = verbatimTextOutput(ns("summary"))
  )
}

server <- function(id, data, filter_panel_api) {
  checkmate::assert_class(data, "reactive")
  checkmate::assert_class(shiny::isolate(data()), "teal_data")
  moduleServer(id, function(input, output, session) {
    output$asdf <- renderUI({
      div(
        experimentSpecInput(session$ns("experiment"), data, mae_name = "MAE"),
        assaySpecInput(session$ns("assay")),
        geneSpecInput(session$ns("genes"), funs = list(Mean = colMeans)),
        adtteSpecInput(session$ns("adtte"))
      )
    })
    experiment <- experimentSpecServer(
      "experiment",
      data = data,
      filter_panel_api = filter_panel_api,
      mae_name = "MAE"
    )
    assay <- assaySpecServer(
      "assay",
      assays = experiment$assays
    )
    genes <- geneSpecServer(
      "genes",
      funs = list(Mean = colMeans),
      gene_choices = experiment$genes
    )
    adtte <- adtteSpecServer(
      "adtte",
      data = data,
      adtte_name = "ADTTE",
      mae_name = "MAE",
      adtte_vars = list(
        aval = "AVAL",
        avalu = "AVALU",
        is_event = "is_event",
        paramcd = "PARAMCD",
        usubjid = "USUBJID"
      ),
      experiment_data = experiment$data,
      experiment_name = experiment$name,
      assay = assay,
      genes = genes,
      probs = reactive({
        0.5
      })
    )
    output$summary <- renderPrint({
      binned_adtte_subset <- adtte$binned_adtte_subset()
      summary(binned_adtte_subset)
    })
  })
}

my_app <- function() {
  data <- teal_data()
  data <- within(data, {
    ADTTE <- teal.modules.hermes::rADTTE %>%
      dplyr::mutate(is_event = .data$CNSR == 0)
    MAE <- hermes::multi_assay_experiment
  })
  datanames <- c("ADTTE", "MAE")
  datanames(data) <- datanames
  join_keys(data) <- default_cdisc_join_keys[datanames]

  app <- init(
    data = data,
    modules = modules(
      module(
        label = "adtteSpec example",
        server = server,
        ui = ui,
        datanames = "all"
      )
    )
  )
  shinyApp(app$ui, app$server)
}

if (interactive()) {
  my_app()
}

sessionInfo()

R version 4.3.0 (2023-04-21)
Platform: aarch64-apple-darwin20 (64-bit)
Running under: macOS 14.1.1

Matrix products: default
BLAS:   /Library/Frameworks/R.framework/Versions/4.3-arm64/Resources/lib/libRblas.0.dylib 
LAPACK: /Library/Frameworks/R.framework/Versions/4.3-arm64/Resources/lib/libRlapack.dylib;  LAPACK version 3.11.0

locale:
[1] en_US.UTF-8/UTF-8/en_US.UTF-8/C/en_US.UTF-8/en_US.UTF-8

time zone: Asia/Kolkata
tzcode source: internal

attached base packages:
[1] stats     graphics  grDevices utils     datasets  methods   base     

other attached packages:
 [1] teal.modules.hermes_0.1.5.9007 teal_0.14.0.9024              
 [3] teal.transform_0.4.0.9007      teal.slice_0.4.0.9024         
 [5] teal.data_0.3.0.9015           shinytest2_0.3.1              
 [7] ggplot2_3.4.4                  magrittr_2.0.3                
 [9] testthat_3.2.0                 shiny_1.8.0                   

loaded via a namespace (and not attached):
  [1] splines_4.3.0               later_1.3.1                
  [3] hermes_1.6.0                filelock_1.0.2             
  [5] bitops_1.0-7                tibble_3.2.1               
  [7] R.oo_1.25.0                 XML_3.99-0.15              
  [9] lifecycle_1.0.4             Rdpack_2.6                 
 [11] edgeR_4.0.1                 doParallel_1.0.17          
 [13] rprojroot_2.0.4             processx_3.8.2             
 [15] lattice_0.22-5              MultiAssayExperiment_1.28.0
 [17] backports_1.4.1             limma_3.58.1               
 [19] plotly_4.10.3               sass_0.4.7                 
 [21] rmarkdown_2.25              jquerylib_0.1.4            
 [23] yaml_2.3.7                  remotes_2.4.2.1            
 [25] httpuv_1.6.12               sessioninfo_1.2.2          
 [27] pkgbuild_1.4.2              DBI_1.1.3                  
 [29] RColorBrewer_1.1-3          abind_1.4-5                
 [31] pkgload_1.3.3               zlibbioc_1.48.0            
 [33] R.cache_0.16.0              GenomicRanges_1.54.1       
 [35] purrr_1.0.2                 R.utils_2.12.3             
 [37] BiocGenerics_0.48.1         RCurl_1.98-1.13            
 [39] styler_1.10.2               rappdirs_0.3.3             
 [41] circlize_0.4.15             GenomeInfoDbData_1.2.11    
 [43] IRanges_2.36.0              S4Vectors_0.40.1           
 [45] ggrepel_0.9.4               teal.widgets_0.4.1.9008    
 [47] teal.reporter_0.2.1.9012    shinyvalidate_0.1.3        
 [49] codetools_0.2-19            DelayedArray_0.28.0        
 [51] xml2_1.3.5                  DT_0.30                    
 [53] tidyselect_1.2.0            shape_1.4.6                
 [55] shinyRadioMatrix_0.2.1      shinyWidgets_0.8.0         
 [57] BiocFileCache_2.10.1        matrixStats_1.1.0          
 [59] stats4_4.3.0                jsonlite_1.8.7             
 [61] GetoptLong_1.0.5            ellipsis_0.3.2             
 [63] survival_3.5-7              iterators_1.0.14           
 [65] foreach_1.5.2               progress_1.2.2             
 [67] tools_4.3.0                 rtables_0.6.5.9013         
 [69] Rcpp_1.0.11                 glue_1.6.2                 
 [71] gridExtra_2.3               teal.code_0.4.1.9009       
 [73] SparseArray_1.2.2           xfun_0.41                  
 [75] MatrixGenerics_1.14.0       usethis_2.2.2              
 [77] ggfortify_0.4.16            GenomeInfoDb_1.38.1        
 [79] dplyr_1.1.4                 withr_2.5.2                
 [81] formatR_1.14                fastmap_1.1.1              
 [83] fansi_1.0.5                 shinyjs_2.1.0              
 [85] callr_3.7.3                 digest_0.6.33              
 [87] R6_2.5.1                    mime_0.12                  
 [89] colorspace_2.1-0            biomaRt_2.58.0             
 [91] RSQLite_2.3.3               R.methodsS3_1.8.2          
 [93] utf8_1.2.4                  tidyr_1.3.0                
 [95] generics_0.1.3              data.table_1.14.8          
 [97] prettyunits_1.2.0           httr_1.4.7                 
 [99] htmlwidgets_1.6.3           S4Arrays_1.2.0             
[101] pkgconfig_2.0.3             gtable_0.3.4               
[103] blob_1.2.4                  ComplexHeatmap_2.18.0      
[105] XVector_0.42.0              brio_1.1.3                 
[107] htmltools_0.5.7             teal.logger_0.1.3.9007     
[109] profvis_0.3.8               clue_0.3-65                
[111] scales_1.2.1                Biobase_2.62.0             
[113] png_0.1-8                   knitr_1.45                 
[115] rstudioapi_0.15.0           rjson_0.2.21               
[117] curl_5.1.0                  checkmate_2.3.0            
[119] cachem_1.0.8                GlobalOptions_0.1.2        
[121] stringr_1.5.1               parallel_4.3.0             
[123] miniUI_0.1.1.1              shinycssloaders_1.0.0      
[125] AnnotationDbi_1.64.1        desc_1.4.2                 
[127] pillar_1.9.0                grid_4.3.0                 
[129] logger_0.2.2                vctrs_0.6.4                
[131] urlchecker_1.0.1            promises_1.2.1             
[133] tern_0.9.2.9017             dbplyr_2.4.0               
[135] xtable_1.8-4                cluster_2.1.4              
[137] evaluate_0.23               cli_3.6.1                  
[139] locfit_1.5-9.8              compiler_4.3.0             
[141] rlang_1.1.2                 crayon_1.5.2               
[143] formatters_0.5.4.9003       ps_1.7.5                   
[145] forcats_1.0.0               fs_1.6.3                   
[147] stringi_1.8.2               viridisLite_0.4.2          
[149] assertthat_0.2.1            Biostrings_2.70.1          
[151] munsell_0.5.0               lazyeval_0.2.2             
[153] devtools_2.4.5              Matrix_1.6-3               
[155] hms_1.1.3                   bit64_4.0.5                
[157] KEGGREST_1.42.0             statmod_1.5.0              
[159] SummarizedExperiment_1.32.0 rbibutils_2.2.16           
[161] fontawesome_0.5.2           broom_1.0.5                
[163] memoise_2.0.1               bslib_0.6.0                
[165] bit_4.0.5

Relevant log output

No response

Code of Conduct

• I agree to follow this project's Code of Conduct.

Contribution Guidelines

• I agree to follow this project's Contribution Guidelines.

Security Policy

• I agree to follow this project's Security Policy.

[donyunardi]

Please check again as update was made and we don't have the validate_dataset_args function anymore in the refactor branch.

[vedhav]

After using the refactor branch. We get a new error and I've edited the issue description with it.

[chlebowa]

I get a different one 🤔

Error in module(label = "adtteSpec example", server = server, ui = ui,  : 
  Called from module(label = "adtteSpec example", ...)
  `ui` with `data` or `datasets` argument is no longer accepted.
  If some `ui` inputs depend on data, please move the logic to your `server` instead.
  Possible solutions are renderUI() or updateXyzInput() functions.

As expected.

[chlebowa]

Right. After modifying the ui function in the example (not exactly current refactor, is it?), I get the same error.

Warning: Error in self$get_data: Assertion on 'dataname' failed: Must be a subset of the empty set, i.e. also empty.

It is raised here:
teal::srv_teal() > teal:::teal_data_to_filtered_data() > teal.slice::init_filtered_data() > teal.slice:::FilteredData$new() > teal.slice:::FilteredData$set_dataset()

In the last call one of the arguments is passed self$get_data(parent_dataname, filtered = TRUE) and therein lies the problem.

[chlebowa]

All of this is because the datasets are ADTTE and MAE.

The error is not triggered when ADSL is included in the app.

Now, who messed with examples in tmh? 😠

[chlebowa]

Then there's another bug in experimentSpecInput. Do I have to look at blame?

[vedhav]

Then there's another bug in experimentSpecInput. Do I have to look at blame?

Well looks like I'm the guilty one. In my defense, this was back when we were passing teal_data into UI and reactive(<teal_data>) into the server, so by convention, this function was a module UI function.

[vedhav]

btw I'm using this cool VS code extension GitLens that does blame on every active line which is so cool and helpful.

[donyunardi]

@vedhav Can we close this now?

[vedhav]

@donyunardi insightsengineering/teal.modules.hermes#351 should close it. Not sure how this was missed during example app changes (probably because they are non tm_* examples)