The package test suite checks implementation details with mocked responses and small deterministic files. For real projects, it is also useful to inspect output shape: Are expected columns present? Are rows non-empty where they should be? Are list endpoints returning the fields users need?
onet2r includes a practical validation script that
maintainers can run from the source tree with
Rscript inst/examples/validate-outputs.R. From an installed
package, source the file returned by:
system.file("examples", "validate-outputs.R", package = "onet2r")
#> [1] "/home/runner/work/_temp/Library/onet2r/examples/validate-outputs.R"What the Script Checks
The validation script always checks deterministic local outputs:
- reading bundled OEWS sample data;
- reading a local O*NET archive-format table;
- creating a reference-SOC OEWS weight panel;
- aggregating an occupation score with coverage and provenance;
- configuring cache and rate-limit settings.
When ONET_API_KEY is available, it also performs live
smoke checks across the main exported O*NET functions. Each result is
inspected for expected type, minimum row count, and important
columns.
Example: Local Validation without an API Key
The local validation below uses package functions rather than copied output.
oews <- onet_oews_national(2024, path = sample_oews)
weights <- onet_weight_panel_oews(oews, year = 2024)
abilities <- onet_archive_read(
"30.3",
"Abilities",
path = archive_303,
release_date = "2026-05-01"
)
oral_scores <- abilities |>
filter(element_id == "1.A.1.a.1") |>
transmute(onet_soc_code, measure_score = data_value)
aggregate <- onet_measure_aggregate(
oral_scores,
weights,
measure_id = "oral_comprehension_fixture"
)
tibble::tibble(
check = c(
"OEWS output is a tibble",
"archive output has source dates",
"weight panel has reference SOC codes",
"aggregate has provenance",
"aggregate has coverage"
),
passed = c(
inherits(oews, "tbl_df"),
all(c("source_date", "data_value") %in% names(abilities)),
all(c("reference_soc_code", "employment") %in% names(weights)),
nrow(onet_provenance(aggregate)) == 1,
nrow(onet_coverage(aggregate)) == 1
)
) |>
onet_kable()| check | passed |
|---|---|
| OEWS output is a tibble | TRUE |
| archive output has source dates | TRUE |
| weight panel has reference SOC codes | TRUE |
| aggregate has provenance | TRUE |
| aggregate has coverage | TRUE |
weights |>
head(5) |>
onet_kable()| reference_soc_code | year | employment | weight_share | source | source_taxonomy | reference_taxonomy |
|---|---|---|---|---|---|---|
| 11-1011 | 2024 | 211230 | 0.040 | OEWS | 2018 SOC | 2018 SOC |
| 15-1252 | 2024 | 1847900 | 0.353 | OEWS | 2018 SOC | 2018 SOC |
| 29-1141 | 2024 | 3175400 | 0.607 | OEWS | 2018 SOC | 2018 SOC |
aggregate |>
select(-coverage, -provenance) |>
onet_kable()| measure_id | aggregate | total_employment | covered_employment | employment_coverage_share | n_occupations | n_reference_soc |
|---|---|---|---|---|---|---|
| oral_comprehension_fixture | 4.574 | 5234530 | 5234530 | 1 | 4 | 4 |
Example: Live Validation Status
Live validation only runs when a user supplies an API key. The site still executes this chunk and reports whether live checks are available in the current environment.
tibble::tibble(
check = "ONET_API_KEY available for live API validation",
passed = nzchar(Sys.getenv("ONET_API_KEY"))
) |>
onet_kable()| check | passed |
|---|---|
| ONET_API_KEY available for live API validation | FALSE |
This does not replace unit tests. It gives maintainers and users a repeatable way to confirm that live API outputs still match the shapes used in examples, vignettes, and downstream analyses.