strip.test.ts

import { describe, expect, it } from "vitest";
import { stripMarkdown } from "./strip.js";

describe("stripMarkdown", () => {
  describe("edge cases", () => {
    it("returns empty string as-is", () => {
      expect(stripMarkdown("")).toBe("");
    });

    it("returns plain text unchanged", () => {
      expect(stripMarkdown("Just some normal text")).toBe("Just some normal text");
    });

    it("preserves emojis", () => {
      expect(stripMarkdown("Hello 🌍🎉")).toBe("Hello 🌍🎉");
    });

    it("preserves newlines", () => {
      expect(stripMarkdown("Line 1\n\nLine 2")).toBe("Line 1\n\nLine 2");
    });
  });

  describe("inline formatting", () => {
    it("strips bold **text**", () => {
      expect(stripMarkdown("This is **bold** text")).toBe("This is bold text");
    });

    it("strips bold __text__", () => {
      expect(stripMarkdown("This is __bold__ text")).toBe("This is bold text");
    });

    it("strips italic *text*", () => {
      expect(stripMarkdown("This is *italic* text")).toBe("This is italic text");
    });

    it("strips italic _text_", () => {
      expect(stripMarkdown("This is _italic_ text")).toBe("This is italic text");
    });

    it("strips bold+italic ***text***", () => {
      expect(stripMarkdown("This is ***bold italic*** text")).toBe(
        "This is bold italic text",
      );
    });

    it("strips strikethrough ~~text~~", () => {
      expect(stripMarkdown("This is ~~deleted~~ text")).toBe("This is deleted text");
    });

    it("strips inline code", () => {
      expect(stripMarkdown("Run `npm install` now")).toBe("Run npm install now");
    });

    it("does not strip mid-word underscores (snake_case)", () => {
      expect(stripMarkdown("use my_variable_name here")).toBe(
        "use my_variable_name here",
      );
    });

    it("handles nested bold inside italic", () => {
      expect(stripMarkdown("This is ***nested*** text")).toBe("This is nested text");
    });
  });

  describe("links and images", () => {
    it("strips links, keeps text", () => {
      expect(stripMarkdown("[OpenClaw](https://openclaw.ai)")).toBe("OpenClaw");
    });

    it("strips images, keeps alt text", () => {
      expect(stripMarkdown("![Screenshot](./img.png)")).toBe("Screenshot");
    });

    it("strips reference-style links", () => {
      expect(stripMarkdown("[OpenClaw][1]")).toBe("OpenClaw");
    });

    it("removes reference definitions", () => {
      expect(stripMarkdown("[1]: https://openclaw.ai")).toBe("");
    });
  });

  describe("block elements", () => {
    it("strips h1-h6 headers", () => {
      expect(stripMarkdown("# Title")).toBe("Title");
      expect(stripMarkdown("## Subtitle")).toBe("Subtitle");
      expect(stripMarkdown("### Section")).toBe("Section");
      expect(stripMarkdown("###### Deep")).toBe("Deep");
    });

    it("strips blockquotes", () => {
      expect(stripMarkdown("> This is a quote")).toBe("This is a quote");
    });

    it("strips nested blockquotes", () => {
      expect(stripMarkdown(">> Nested quote")).toBe("Nested quote");
    });

    it("strips horizontal rules", () => {
      expect(stripMarkdown("---")).toBe("");
      expect(stripMarkdown("***")).toBe("");
      expect(stripMarkdown("___")).toBe("");
    });

    it("strips fenced code blocks, keeps content", () => {
      const input = "```javascript\nconst x = 1;\n```";
      expect(stripMarkdown(input)).toBe("const x = 1;");
    });

    it("strips fenced code blocks without language", () => {
      const input = "```\nhello world\n```";
      expect(stripMarkdown(input)).toBe("hello world");
    });
  });

  describe("lists", () => {
    it("strips unordered list markers with -", () => {
      expect(stripMarkdown("- Item one\n- Item two")).toBe("Item one\nItem two");
    });

    it("strips unordered list markers with *", () => {
      expect(stripMarkdown("* Item one\n* Item two")).toBe("Item one\nItem two");
    });

    it("strips unordered list markers with +", () => {
      expect(stripMarkdown("+ Item one")).toBe("Item one");
    });

    it("preserves ordered list numbers", () => {
      expect(stripMarkdown("1. First\n2. Second")).toBe("1. First\n2. Second");
    });

    it("strips task list checkboxes", () => {
      expect(stripMarkdown("[ ] Todo\n[x] Done")).toBe("Todo\nDone");
    });

    it("preserves indentation for nested lists", () => {
      expect(stripMarkdown("Top\n  - Nested item\n  - Another")).toBe(
        "Top\n  Nested item\n  Another",
      );
    });
  });

  describe("escaped characters", () => {
    it("removes backslash escapes", () => {
      expect(stripMarkdown("Price: \\$10")).toBe("Price: $10");
      expect(stripMarkdown("Not \\*bold\\*")).toBe("Not *bold*");
    });
  });

  describe("HTML comments", () => {
    it("removes HTML comments", () => {
      expect(stripMarkdown("Text <!-- hidden --> more")).toBe("Text  more");
    });
  });

  describe("whitespace cleanup", () => {
    it("collapses 3+ blank lines to 2", () => {
      expect(stripMarkdown("A\n\n\n\nB")).toBe("A\n\nB");
    });

    it("trims leading/trailing whitespace", () => {
      expect(stripMarkdown("  hello  \n\n")).toBe("hello");
    });
  });

  describe("real-world LLM output", () => {
    it("handles a typical Claude morning briefing", () => {
      const input = `### Morgenbriefing

**Wetter:** 12°C, bewölkt, kein Regen erwartet.

**Termine:**
- 10:00 Standup
- 14:00 1:1 mit Lisa

**E-Mails:**
- _3 ungelesene_ in iCloud
- **1 wichtige** von HR

> Zitat des Tages: "The best way to predict the future is to invent it."

---

Guten Morgen! ☀️`;

      const expected = `Morgenbriefing

Wetter: 12°C, bewölkt, kein Regen erwartet.

Termine:
10:00 Standup
14:00 1:1 mit Lisa

E-Mails:
3 ungelesene in iCloud
1 wichtige von HR

Zitat des Tages: "The best way to predict the future is to invent it."

Guten Morgen! ☀️`;

      expect(stripMarkdown(input)).toBe(expected);
    });
  });
});