domstolar

# Icelandic Courts (dómstólar) **Requires:** Tier 0 (core) via RSS + pagination; Tier 2 (browser) only for JS-rendered search. Three independent court sites — no unified API, all HTML scraping. ## Sites | Court | Domain | Rulings from | |-------|--------|-------------| | Héraðsdómstólar (8 district courts) | `heradsdomstolar.is` | ~2015+ | | Landsréttur (Court of Appeals) | `landsrettur.is` | 2018+ | | Hæstiréttur (Supreme Court) | `haestirettur.is` | 1999+ | Case number prefixes: `E-` = einkamál (civil), `S-` = sakamál (criminal). ## Héraðsdómstólar (District Courts) ### RSS Feeds (best entry point, no JS needed) ``` https://www.heradsdomstolar.is/heradsdomstolar/{slug}/domar/rss/ ``` Courts: `reykjavik`, `reykjanes`, `vesturland`, `vestfirdir`, `nordurland-vestra`, `nordurland-eystra`, `austurland`, `sudurland` Each RSS item has: `<title>` (case number), `<description>` (reifun/summary), `<link>`, `<guid>` (UUID), `<pubDate>`. ### PDF Download (direct, no JS) ``` https://www.heradsdomstolar.is/Cache/Verdicts/{guid}.pdf ``` ### Search (GET params) ``` https://www.heradsdomstolar.is/default.aspx?SearchAction=Search&pageid=fd8e17eb-6e70-11e5-80c3-005056bc50d4&Verdict={text}&Court={court}&CaseNumber=&CaseType=&LawParagraphs=&Keywords=&Parties=&FromDate={DD.MM.YYYY}&ToDate={DD.MM.YYYY} ``` **Caveat**: Search results render client-side (JS required → Playwright). Use RSS + pagination for non-JS scraping. ### Pagination (AJAX, returns HTML fragments, no JS needed) ``` GET /default.aspx?pageitemid={pageitemid}&offset={n}&count={batch_size} ``` Reykjavík pageitemid: `98619c0d-86c7-11e5-80c6-005056bc6a40` (initial count=22, then batches of 6). Returns `.verdict-box` divs with date, case number, judge, parties, reifun, link. ### Individual Ruling ``` https://www.heradsdomstolar.is/default.aspx?pageid=347c3bb1-8926-11e5-80c6-005056bc6a40&id={guid} ``` ## Landsréttur (Court of Appeals) ### Search ``` https://www.landsrettur.is/search?Text={text}&CaseNumber=&LawCitations=&Keywords=&Parties=&CaseType=&FromDate={DD.MM.YYYY}&ToDate={DD.MM.YYYY} ``` ### Pagination ``` GET /domar-og-urskurdir/$Verdicts/Index/?pageitemid=5cf6e850-20b6-11e9-85de-94b86df896cb&offset={n}&count={n} ``` Returns HTML fragments. Default count=12. ### Individual Ruling ``` https://www.landsrettur.is/domar-og-urskurdir/domur-urskurdur/?Id={case_uuid}&verdictid={verdict_uuid} ``` ## Hæstiréttur (Supreme Court) ### Search ``` https://www.haestirettur.is/default.aspx?SearchAction=Search&pageid=fd8e17eb-6e70-11e5-80c3-005056bc50d4&tab=1&Verdict={text}&CaseNumber=&LawParagraphs=&Keywords=&Parties=&FromDate=&ToDate= ``` ### Pagination ``` GET /default.aspx?pageitemid=a31e2c30-9510-11e5-80c6-005056bc6a40&offset={n}&count={n} ``` ### Individual Ruling ``` https://www.haestirettur.is/default.aspx?pageid=0f2f6428-7b6a-11eb-947c-005056bc0bdb&id={guid} ``` ## Metadata Per Ruling - **Málsnúmer** (case number): e.g. `E-679/2023`, `36/2020` - **Dagsetning** (date) - **Dómari** (judge name + title) - **Málsaðilar** (parties) with lawyers - **Reifun** (summary/abstract) - **Lykilorð** (keywords/tags) - **Lagagreinar** (legal statutes cited) - **Full text** (HTML on page, PDF for district courts) ## Scraping Strategy 1. **Discovery**: Poll RSS feeds (8 district courts) + paginate listing endpoints 2. **Full text**: Fetch PDFs from `/Cache/Verdicts/{guid}.pdf` (district) or scrape HTML (upper courts) 3. **Search**: Use Playwright for JS-rendered search results, or stick to pagination endpoints 4. **Cross-references**: Upper court rulings link to lower court decisions — useful for case tracking ## Relevant Search Terms for Launaþjófnaður - `launaþjófnaður` (wage theft) - `vangreidd laun` / `vangoldin laun` (unpaid wages) - `vinnulaun` (work wages) - `kjarasamningur` (collective agreement) - `ólögmæt launakjör` (illegal wage conditions) - `mansal` (human trafficking — severe cases) - `vinnuréttindi` (labor rights) ## Example: Fetch + Parse Script ```python """Search Icelandic court rulings across all three court levels.""" import httpx from bs4 import BeautifulSoup COURTS = { "héraðsdómstólar": { "base": "https://www.heradsdomstolar.is", "search": "/default.aspx", "params": { "SearchAction": "Search", "pageid": "fd8e17eb-6e70-11e5-80c3-005056bc50d4", }, "text_param": "Verdict", "pagination_pageitemid": "98619c0d-86c7-11e5-80c6-005056bc6a40", }, "landsréttur": { "base": "https://www.landsrettur.is", "search": "/search", "params": {}, "text_param": "Text", "pagination_pageitemid": "5cf6e850-20b6-11e9-85de-94b86df896cb", }, "hæstiréttur": { "base": "https://www.haestirettur.is", "search": "/default.aspx", "params": { "SearchAction": "Search", "pageid": "fd8e17eb-6e70-11e5-80c3-005056bc50d4", "tab": "1", }, "text_param": "Verdict", "pagination_pageitemid": "a31e2c30-9510-11e5-80c6-005056bc6a40", }, } def paginate_results(court_key: str, offset: int = 0, count: int = 20): """Fetch a page of rulings via the AJAX pagination endpoint (no JS needed).""" court = COURTS[court_key] url = f"{court['base']}/default.aspx" params = { "pageitemid": court["pagination_pageitemid"], "offset": offset, "count": count, } r = httpx.get(url, params=params, timeout=30) r.raise_for_status() return parse_verdict_boxes(r.text) def parse_verdict_boxes(html: str) -> list[dict]: """Parse .verdict-box elements from HTML fragment.""" soup = BeautifulSoup(html, "html.parser") results = [] for box in soup.select(".verdict-box, .row"): # Extract what we can — structure varies by court links = box.select("a[href]") title = box.get_text(strip=True)[:200] href = links[0]["href"] if links else None results.append({"title": title, "url": href}) return results def fetch_rss(court_slug: str) -> list[dict]: """Fetch RSS feed for a district court.""" url = f"https://www.heradsdomstolar.is/heradsdomstolar/{court_slug}/domar/rss/" r = httpx.get(url, timeout=30) r.raise_for_status() soup = BeautifulSoup(r.text, "xml") items = [] for item in soup.find_all("item"): items.append({ "case_number": item.title.text if item.title else "", "summary": item.description.text if item.description else "", "url": item.link.text if item.link else "", "guid": item.guid.text if item.guid else "", "date": item.pubDate.text if item.pubDate else "", }) return items def download_pdf(guid: str, output_path: str): """Download district court ruling PDF.""" url = f"https://www.heradsdomstolar.is/Cache/Verdicts/{guid}.pdf" r = httpx.get(url, timeout=60) r.raise_for_status() with open(output_path, "wb") as f: f.write(r.content) # Usage: # items = fetch_rss("reykjavik") # download_pdf(items[0]["guid"], "ruling.pdf") # results = paginate_results("hæstiréttur", offset=0, count=20) ``` ## robots.txt Note All three sites disallow `/domar*` in robots.txt. Search and pagination endpoints are not blocked. Use responsibly for personal/research purposes.