LLM Wiki is a cross-platform desktop application that turns your documents into an organized, interlinked knowledge base — automatically. Instead of traditional RAG (retrieve-and-answer from scratch every time), the LLM incrementally builds and maintains a persistent wiki from your sources。
TypeScript
import { describe, expect, it } from "vitest"
import { MAX_BATCH_URLS, fetchImportUrl, parseImportUrls, urlSourceFileName } from "./url-source-import"
describe("parseImportUrls", () => {
it("normalizes fragments, removes duplicates, and preserves order", () => {
expect(parseImportUrls(" https://example.com/a#part\nhttps://example.com/a\nhttp://例子.测试/b ")).toEqual([
"https://example.com/a",
"http://xn--fsqu00a.xn--0zwm56d/b",
])
})
it("rejects malformed and non-http URLs", () => {
expect(() => parseImportUrls("not-a-url")).toThrow("Invalid URL")
expect(() => parseImportUrls("file:///tmp/secret")).toThrow("Unsupported URL scheme")
expect(() => parseImportUrls("https://user:[email protected]/page")).toThrow("embedded credentials")
})
it("caps batch size", () => {
const urls = Array.from({ length: MAX_BATCH_URLS + 1 }, (_, index) => `https://example.com/${index}`)
expect(() => parseImportUrls(urls.join("\n"))).toThrow(`at most ${MAX_BATCH_URLS}`)
})
})
describe("urlSourceFileName", () => {
it("prefers a safe HTML title", () => {
expect(urlSourceFileName(
"https://example.com/post/123",
"text/html; charset=utf-8",
"<html><head><title>安全 / Useful: Guide</title></head></html>",
)).toBe("安全-Useful-Guide.html")
})
it("uses the URL leaf for plain text", () => {
expect(urlSourceFileName("https://example.com/docs/readme.md", "text/plain", "hello"))
.toBe("readme.txt")
})
it("avoids Windows reserved names and tolerates malformed escapes", () => {
expect(urlSourceFileName("https://example.com/AUX", "text/plain", "hello"))
.toBe("AUX-web.txt")
expect(urlSourceFileName("https://example.com/bad%escape", "text/plain", "hello"))
.toBe("bad-escape.txt")
})
})
describe("fetchImportUrl", () => {
it("blocks a public URL redirect before requesting a private target", async () => {
const fetch = async (url: string | URL | Request, init?: RequestInit & { maxRedirections?: number }) => {
expect(String(url)).toBe("https://example.com/start")
expect(init).toMatchObject({ redirect: "manual", maxRedirections: 0 })
return new Response(null, {
status: 302,
headers: { location: "http://169.254.169.254/latest/meta-data" },
})
}
await expect(fetchImportUrl(fetch as typeof globalThis.fetch, "https://example.com/start", new AbortController().signal))
.rejects.toThrow("cannot redirect")
})
it("blocks IPv4-mapped IPv6 private redirect targets", async () => {
const fetch = async () => new Response(null, {
status: 302,
headers: { location: "http://[::ffff:169.254.169.254]/metadata" },
})
await expect(fetchImportUrl(fetch as typeof globalThis.fetch, "https://example.com", new AbortController().signal))
.rejects.toThrow("cannot redirect")
})
it("allows an explicitly requested private URL and follows relative redirects", async () => {
const seen: string[] = []
const fetch = async (url: string | URL | Request, init?: RequestInit & { maxRedirections?: number }) => {
seen.push(String(url))
expect(init).toMatchObject({ redirect: "manual", maxRedirections: 0 })
return seen.length === 1
? new Response(null, { status: 302, headers: { location: "/page" } })
: new Response("ok", { status: 200 })
}
const response = await fetchImportUrl(
fetch as typeof globalThis.fetch,
"http://192.168.1.50/start",
new AbortController().signal,
)
expect(await response.text()).toBe("ok")
expect(seen).toEqual(["http://192.168.1.50/start", "http://192.168.1.50/page"])
})
})