Wabenhain/src/lib/legal-blocks.test.ts

import { describe, expect, it } from "vitest";
import { extractRevision, parseLegalParagraph, sectionAnchors, splitSentences, tokenizeInline } from "./legal-blocks";

describe("parseLegalParagraph", () => {
  it("keeps ordinary prose as a text block", () => {
    expect(parseLegalParagraph("Cookies sind kleine Textdateien. Sie speichern Einstellungen.")).toEqual({
      kind: "text",
      text: "Cookies sind kleine Textdateien. Sie speichern Einstellungen.",
    });
  });

  it("turns a paragraph whose lines all start with a dash into a list", () => {
    expect(parseLegalParagraph("- Auskunft\n- Berichtigung\n• Löschung")).toEqual({
      kind: "list",
      items: ["Auskunft", "Berichtigung", "Löschung"],
    });
  });

  it("does not treat a single dashed line mixed with prose as a list", () => {
    expect(parseLegalParagraph("Einleitung\n- Punkt").kind).toBe("text");
  });

  it("recognises a storage entry and pulls labelled facts out of it", () => {
    const block = parseLegalParagraph(
      "wabenhain_session – Hält dich angemeldet. Inhalt: signierte Kennung, keine Passwörter. Speicherdauer: 30 Tage; wird beim Abmelden gelöscht."
    );
    expect(block).toEqual({
      kind: "entry",
      name: "wabenhain_session",
      note: null,
      text: "Hält dich angemeldet.",
      facts: [
        { label: "Inhalt", value: "signierte Kennung, keine Passwörter." },
        { label: "Speicherdauer", value: "30 Tage; wird beim Abmelden gelöscht." },
      ],
    });
  });

  it("keeps abbreviations such as Art. and Abs. inside one fact", () => {
    const block = parseLegalParagraph(
      "theme – Speichert die Darstellung. Rechtsgrundlage: § 25 Abs. 2 Nr. 2 TDDDG und Art. 6 Abs. 1 lit. f DSGVO. Wird sonst gelöscht."
    );
    expect(block.kind === "entry" && block.facts).toEqual([
      { label: "Rechtsgrundlage", value: "§ 25 Abs. 2 Nr. 2 TDDDG und Art. 6 Abs. 1 lit. f DSGVO." },
    ]);
    expect(block.kind === "entry" && block.text).toBe("Speichert die Darstellung. Wird sonst gelöscht.");
  });

  it("recognises a provider entry written as 'Name (Zweck): Text'", () => {
    const block = parseLegalParagraph("Stripe (Zahlungsabwicklung): Lädt das Formular. Anbieter: Stripe Payments Europe, Ltd., Dublin.");
    expect(block).toMatchObject({ kind: "entry", name: "Stripe", note: "Zahlungsabwicklung", text: "Lädt das Formular." });
    expect(block.kind === "entry" && block.facts).toEqual([{ label: "Anbieter", value: "Stripe Payments Europe, Ltd., Dublin." }]);
  });

  it("recognises English fact labels", () => {
    const block = parseLegalParagraph("locale – Stores your language. Storage period: 1 year.");
    expect(block.kind === "entry" && block.facts).toEqual([{ label: "Storage period", value: "1 year." }]);
  });

  it("does not mistake a sentence with a spaced dash for an entry", () => {
    expect(parseLegalParagraph("Wir versenden schnell – meist in zwei Tagen.").kind).toBe("text");
  });
});

describe("splitSentences", () => {
  it("does not split after z. B. or Art.", () => {
    expect(splitSentences("Etwa z. B. Chrome. Nach Art. 6 DSGVO. Ende.")).toEqual(["Etwa z. B. Chrome.", "Nach Art. 6 DSGVO.", "Ende."]);
  });
});

describe("extractRevision", () => {
  it("moves a trailing 'Stand:' line out of the sections", () => {
    const { sections, revision } = extractRevision([
      { heading: "A", paragraphs: ["Text"] },
      { heading: "B", paragraphs: ["Mehr", "Stand: September 2026"] },
    ]);
    expect(revision).toBe("September 2026");
    expect(sections).toEqual([
      { heading: "A", paragraphs: ["Text"] },
      { heading: "B", paragraphs: ["Mehr"] },
    ]);
  });

  it("drops a section that only held the revision line", () => {
    const { sections, revision } = extractRevision([
      { heading: "A", paragraphs: ["Text"] },
      { heading: "Stand", paragraphs: ["Last updated: May 2026"] },
    ]);
    expect(revision).toBe("May 2026");
    expect(sections).toHaveLength(1);
  });

  it("returns null when no revision is given", () => {
    expect(extractRevision([{ heading: "A", paragraphs: ["Text"] }]).revision).toBeNull();
  });
});

describe("sectionAnchors", () => {
  it("builds readable, unique anchors from headings", () => {
    expect(sectionAnchors(["1. Verantwortlicher", "Deine Rechte", "Deine Rechte", "Über uns & mehr"])).toEqual([
      "verantwortlicher",
      "deine-rechte",
      "deine-rechte-2",
      "ueber-uns-mehr",
    ]);
  });

  it("falls back to a numbered anchor for headings without letters", () => {
    expect(sectionAnchors(["§§", ""])).toEqual(["abschnitt-1", "abschnitt-2"]);
  });
});

describe("tokenizeInline", () => {
  it("links web addresses with a path and e-mail addresses", () => {
    expect(tokenizeInline("Siehe stripe.com/de/privacy oder schreib an [email protected].")).toEqual([
      { kind: "text", value: "Siehe " },
      { kind: "link", value: "stripe.com/de/privacy", href: "https://stripe.com/de/privacy" },
      { kind: "text", value: " oder schreib an " },
      { kind: "link", value: "[email protected]", href: "mailto:[email protected]" },
      { kind: "text", value: "." },
    ]);
  });

  it("keeps explicit https links and never links other schemes", () => {
    expect(tokenizeInline("https://example.org/a javascript:alert(1)")).toEqual([
      { kind: "link", value: "https://example.org/a", href: "https://example.org/a" },
      { kind: "text", value: " javascript:alert(1)" },
    ]);
  });

  it("leaves a bare domain without path as text", () => {
    expect(tokenizeInline("wabenhain.com")).toEqual([{ kind: "text", value: "wabenhain.com" }]);
  });
});