Getting started with risAT

This vignette walks through the main features of {risAT} using concrete examples. For a full parameter reference, see the function reference.

{risAT} provides a tidyverse-friendly interface to the Austrian RIS (Rechtsinformationssystem) Open Government Data API v2.6. It covers the /Judikatur endpoint — case law from all nine supported court applications — and is designed for reproducible legal research.

First search

The fastest way in is a court-specific wrapper. Each wrapper targets a single court and provides typed, validated arguments. Let’s retrieve Administrative Court (VwGH) decisions on housing law (Baurecht):

library(risAT)

results <- ris_search_vwgh(query = "Baurecht")
results

#> # A tibble: 9,637 × 5
#>    id                 document_type case_number decision_date vwgh_decision_type
#>    <chr>              <chr>         <chr>       <chr>         <chr>             
#>  1 JWT_2025050017_20… Text          Ro 2025/05… 2026-05-06    Beschluss         
#>  2 JWR_2025050017_20… Rechtssatz    Ro 2025/05… 2026-05-06    Beschluss         
#>  3 JWR_2025050017_20… Rechtssatz    Ro 2025/05… 2026-05-06    Beschluss         
#>  4 JWT_2025070002_20… Text          Ro 2025/07… 2026-04-22    Erkenntnis        
#>  5 JWT_2023050255_20… Text          Ra 2023/05… 2026-04-10    Beschluss         
#>  6 JWR_2023050255_20… Rechtssatz    Ra 2023/05… 2026-04-10    Beschluss         
#>  7 JWR_2023050255_20… Rechtssatz    Ra 2023/05… 2026-04-10    Beschluss         
#>  8 JWT_2025050071_20… Text          Ra 2025/05… 2026-04-10    Beschluss         
#>  9 JWR_2025050071_20… Rechtssatz    Ra 2025/05… 2026-04-10    Beschluss         
#> 10 JWT_2025050145_20… Text          Ra 2025/05… 2026-03-27    Erkenntnis        
#> # ℹ 9,627 more rows

All search functions return a tibble — one row per document. The most frequently used columns are:

Column	Description
`id`	Unique RIS document identifier
`case_number`	Business number (Geschäftszahl)
`decision_date`	Date of the decision
`document_type`	`Rechtssatz` or `Text` (full decision)
`content_urls`	List-column of download links
`app_metadata`	List-column of court-specific and response metadata

Filtering results

All wrappers share a common set of filter arguments. Date ranges follow ISO 8601 ("YYYY-MM-DD"), and decision_type accepts the German API values and lowercase aliases.

results <- ris_search_vwgh(
  query              = "Asyl",
  decision_date_from = "2023-01-01",
  decision_date_to   = "2024-12-31",
  decision_type      = "Erkenntnis"
)

nrow(results)

#> [1] 584

The in_ris_since argument is a convenient shortcut for recently published documents:

recent <- ris_search_vwgh(in_ris_since = "one_month")
nrow(recent)

#> [1] 855

Decisions over time

Because results come back as a tidy tibble, downstream analysis with {dplyr} and {ggplot2} is straightforward. Below we retrieve VwGH asylum decisions from 2019 to 2024 and visualise the yearly count.

df_asyl <- ris_search_vwgh(
  query              = "Asyl",
  decision_date_from = "2019-01-01",
  decision_date_to   = "2024-12-31",
  decision_type      = "Erkenntnis"
)

library(dplyr)
library(ggplot2)
library(lubridate)

df_asyl |>
  mutate(year = year(as.Date(decision_date))) |>
  count(year) |>
  ggplot(aes(x = year, y = n)) +
  geom_col(fill = "#2E4057") +
  scale_x_continuous(breaks = 2019:2024) +
  scale_y_continuous(expand = expansion(mult = c(0, 0.05))) +
  labs(
    title    = "VwGH decisions on asylum law by year",
    subtitle = "Search term: 'Asyl', decision type: Erkenntnis, 2019–2024.",
    caption  = "Data: RIS OGD API v2.6; retrieved via {risAT}.",
    x = NULL, y = NULL
  ) +
  theme_minimal() +
  theme(
    plot.title          = element_text(face = "bold"),
    plot.title.position = "plot",
    plot.subtitle       = element_text(color = "grey40", margin = margin(b = 10)),
    plot.caption        = element_text(color = "grey50", size = rel(0.8)),
    panel.grid.major.x  = element_blank(),
    panel.grid.minor    = element_blank()
  )

Court-specific wrappers

risAT provides wrappers for all nine Judikatur applications:

Court	Function
VfGH (Constitutional Court)	`ris_search_vfgh()`
VwGH (Administrative Court)	`ris_search_vwgh()`
Justiz (OGH, OLG, LG, BG)	`ris_search_justiz()`
BVwG (Federal Administrative Court)	`ris_search_bvwg()`
LVwG (State Administrative Courts)	`ris_search_lvwg()`
DSK / DSB (Data Protection)	`ris_search_dsk()`
DOK (Disciplinary Bodies)	`ris_search_dok()`
PVAK (Staff Representation)	`ris_search_pvak()`
GBK (Equal Treatment Commission)	`ris_search_gbk()`

Some wrappers expose court-specific parameters. ris_search_justiz() lets you filter by court and legal area:

results_ogh <- ris_search_justiz(
  query      = "Schadenersatz",
  court      = "OGH",
  legal_area = "Zivilrecht"
)
nrow(results_ogh)

#> [1] 9409

ris_search_lvwg() restricts results to a single state administrative court via federal_state. English state names are accepted:

results_wien <- ris_search_lvwg(
  federal_state = "Vienna",       # also accepted: "Wien"
  decision_type = "Erkenntnis"
)
nrow(results_wien)

#> [1] 9602

ris_search_gbk() filters Equal Treatment Commission decisions by commission, senate, and discrimination ground — English aliases are supported:

results_gbk <- ris_search_gbk(
  discrimination_ground = "gender",         # "Geschlecht"
  commission            = "private_sector"  # "Gleichbehandlungskommission"
)
nrow(results_gbk)

#> [1] 284

VfGH: constitutional court decisions

The VfGH wrapper defaults to legal principles (Rechtssätze) rather than full decision texts — the most common use case for constitutional court research. Toggle via search_legal_principles and search_decision_text:

# Default: Rechtssätze (RS)
results_vfgh <- ris_search_vfgh(query = "Grundrecht")
nrow(results_vfgh)

# Request full decision texts instead
results_vfgh_et <- ris_search_vfgh(
  query                   = "Grundrecht",
  search_legal_principles = FALSE,
  search_decision_text    = TRUE
)

#> [1] 1530

Verifying results with echo

Setting echo = TRUE prints the search parameters and a direct link to the same results on the official RIS website. This is useful for:

Verification — open the URL in a browser and compare with the API results.
Sharing — paste the URL to share a query with colleagues who don’t use R.
Exploration — browse results on the RIS website before refining your query.

results <- ris_search_vwgh(
  query              = "Baurecht",
  decision_date_from = "2024-01-01",
  echo               = TRUE
)

The printed output shows the total hit count and a clickable RIS URL that preserves all filter parameters.

Working with list-columns

Two columns contain richer nested data.

`content_urls` — download links

Each element of content_urls is a character vector of URLs pointing to the document in different formats (XML, HTML, RTF, PDF):

results$content_urls[[1]]

#> [1] "https://www.ris.bka.gv.at/Dokumente/Vwgh/JWT_2025050017_20260506J00/JWT_2025050017_20260506J00.xml" 
#> [2] "https://www.ris.bka.gv.at/Dokumente/Vwgh/JWT_2025050017_20260506J00/JWT_2025050017_20260506J00.html"
#> [3] "https://www.ris.bka.gv.at/Dokumente/Vwgh/JWT_2025050017_20260506J00/JWT_2025050017_20260506J00.rtf" 
#> [4] "https://www.ris.bka.gv.at/Dokumente/Vwgh/JWT_2025050017_20260506J00/JWT_2025050017_20260506J00.pdf"

To extract all PDF URLs across results, filter by file extension:

library(purrr)

pdf_urls <- results |>
  mutate(
    pdf_url = map_chr(
      content_urls,
      \(x) {
        hit <- x[grepl("\\.pdf$", x, ignore.case = TRUE)]
        if (length(hit)) hit[[1]] else NA_character_
      }
    )
  ) |>
  filter(!is.na(pdf_url)) |>
  select(case_number, decision_date, pdf_url)

pdf_urls

#> # A tibble: 9,637 × 3
#>    case_number     decision_date pdf_url                                        
#>    <chr>           <chr>         <chr>                                          
#>  1 Ro 2025/05/0017 2026-05-06    https://www.ris.bka.gv.at/Dokumente/Vwgh/JWT_2…
#>  2 Ro 2025/05/0017 2026-05-06    https://www.ris.bka.gv.at/Dokumente/Vwgh/JWR_2…
#>  3 Ro 2025/05/0017 2026-05-06    https://www.ris.bka.gv.at/Dokumente/Vwgh/JWR_2…
#>  4 Ro 2025/07/0002 2026-04-22    https://www.ris.bka.gv.at/Dokumente/Vwgh/JWT_2…
#>  5 Ra 2023/05/0255 2026-04-10    https://www.ris.bka.gv.at/Dokumente/Vwgh/JWT_2…
#>  6 Ra 2023/05/0255 2026-04-10    https://www.ris.bka.gv.at/Dokumente/Vwgh/JWR_2…
#>  7 Ra 2023/05/0255 2026-04-10    https://www.ris.bka.gv.at/Dokumente/Vwgh/JWR_2…
#>  8 Ra 2025/05/0071 2026-04-10    https://www.ris.bka.gv.at/Dokumente/Vwgh/JWT_2…
#>  9 Ra 2025/05/0071 2026-04-10    https://www.ris.bka.gv.at/Dokumente/Vwgh/JWR_2…
#> 10 Ra 2025/05/0145 2026-03-27    https://www.ris.bka.gv.at/Dokumente/Vwgh/JWT_2…
#> # ℹ 9,627 more rows

`app_metadata` — court-specific fields and response context

Court-specific fields and pagination context are nested in app_metadata. Use tidyr::unnest_wider() to bring them to the top level:

library(tidyr)

results |>
  select(id, case_number, app_metadata) |>
  unnest_wider(app_metadata) |>
  glimpse()

#> Rows: 9,637
#> Columns: 6
#> $ id          <chr> "JWT_2025050017_20260506J00", "JWR_2025050017_20260506J02"…
#> $ case_number <chr> "Ro 2025/05/0017", "Ro 2025/05/0017", "Ro 2025/05/0017", "…
#> $ response    <list> [NA, 9637, 1, 100], [NA, 9637, 1, 100], [NA, 9637, 1, 100…
#> $ technisch   <list> ["JWT_2025050017_20260506J00", "Vwgh", "Verwaltungsgerich…
#> $ allgemein   <list> ["2026-06-02", "2026-06-02", "https://www.ris.bka.gv.at/D…
#> $ request     <list> ["/Judikatur", "Vwgh", 1, "OneHundred", "https://www.ris.…

The response sub-list contains pagination context (total hits, page number) and request records the exact API parameters sent:

# Total hits reported by the API
results$app_metadata[[1]]$response$hits

# Page context
results$app_metadata[[1]]$response$page_number
results$app_metadata[[1]]$response$page_size

#> 9637
#> 1
#> 100

The two-step pattern

Under the hood every wrapper calls two lower-level functions:

ris_req_case_law() — builds an httr2 request object (no network call).
ris_perform_case_law() — executes the request with automatic pagination.

Use these directly for more control — to inspect the URL before sending, or to target application codes not covered by a named wrapper:

# Build — no network call yet
req <- ris_req_case_law(
  application = "Vwgh",
  query       = "Baurecht"
)

# Inspect the URL before sending
req$url

# Execute — fetches all pages automatically
results <- ris_perform_case_law(req)
nrow(results)

#> https://data.bka.gv.at/ris/api/v2.6/Judikatur?Applikation=Vwgh&Suchworte=Baurecht&SortierungSortDirection=Descending&SortierungSortedByColumn=Datum&DokumenttypSucheInEntscheidungstexten=true&DokumenttypSucheInRechtssaetzen=true&Seitennummer=1&DokumenteProSeite=OneHundred