Validate survey-referenced list names between two XLSForms

Checks that every list name referenced in target's survey questions is also referenced in dev's survey questions. Returns a tibble row for each list name actively used by target's survey that is absent from dev's survey.

Usage

validate_survey_list_names(target, dev)

Arguments

target

An xlsform object representing the authoritative reference form.

dev

An xlsform object representing the form being validated.

Value

A tibble with columns check, severity, name, list_name, and detail. Has zero rows when all list names referenced by target's survey are also referenced by dev's survey.

Details

How it differs from validate_list_names()

validate_list_names() compares the lists defined in each form's choices sheet. validate_survey_list_names() compares the lists actively used by survey questions — the second token in type values like select_one list_a.

The two checks are complementary. A list can be defined in choices but never used in the survey (orphaned list), or — after a question type change from select_one list_a to text — it may still be defined in choices while no longer referenced in any survey question. This check surfaces the latter case.

See also

validate_xlsform() to run all checks together; validate_list_names() for the complementary choices-side check; xlsform_referenced_list_names() to extract referenced list names from a form.

Examples

target <- read_xlsform(system.file("extdata/form.xlsx", package = "idem"))

# No issues: all lists target's survey uses are also used in dev's survey
validate_survey_list_names(target, target)
#> # A tibble: 0 × 5
#> # ℹ 5 variables: check <chr>, severity <chr>, name <chr>, list_name <chr>,
#> #   detail <chr>

# Issues found: dev's survey has all select questions replaced by text,
# so none of target's referenced lists appear in dev's survey
dev_no_selects <- xlsform(
  survey = data.frame(
    type = rep("text", nrow(target$survey)),
    name = target$survey$name
  ),
  choices = target$choices
)
validate_survey_list_names(target, dev_no_selects)
#> # A tibble: 171 × 5
#>    check             severity name               list_name detail               
#>    <chr>             <chr>    <chr>              <chr>     <chr>                
#>  1 survey_list_names error    l_survey_modality  NA        List 'l_survey_modal…
#>  2 survey_list_names error    l_enum_id          NA        List 'l_enum_id' is …
#>  3 survey_list_names error    l_gender           NA        List 'l_gender' is r…
#>  4 survey_list_names error    l_admin1           NA        List 'l_admin1' is r…
#>  5 survey_list_names error    l_admin2           NA        List 'l_admin2' is r…
#>  6 survey_list_names error    l_admin3           NA        List 'l_admin3' is r…
#>  7 survey_list_names error    l_admin4           NA        List 'l_admin4' is r…
#>  8 survey_list_names error    l_cluster_id       NA        List 'l_cluster_id' …
#>  9 survey_list_names error    l_yn               NA        List 'l_yn' is refer…
#> 10 survey_list_names error    l_hoh_civil_status NA        List 'l_hoh_civil_st…
#> # ℹ 161 more rows