Skip to content

Never stop coding. Free MIT AI gateway: one endpoint, 375 providers (150+ free), 1,400+ models Kimi, Claude, GPT, Gemini, GLM, DeepSeek, MiniMax. Works with Claude Code, Codex, Cursor, OpenCode, Cline & Copilot. Quota-aware auto-fallback, RTK+Caveman compression saves 15-95% tokens, MCP/A2A, Desktop/PWA. Built by 900+ contributors

2.3 KB
Raw Blame
// #5486 — Split a plain string into text + URL segments so messages that embed a
// URL (e.g. the GitLab Duo OAuth setup instructions pointing at
// https://gitlab.com/-/profile/applications) can render clickable links instead of
// dead red text in the OAuth error step. Pure (no React) so it is unit-testable.
//
// The matcher is a single unbounded repetition over a NEGATED character class
// (`[^\s...]+`) — linear time, no catastrophic backtracking (ReDoS-safe per the
// bounded-regex rule). A trailing sentence punctuation char is peeled back out of
// the URL so "…restart." keeps its period as text and the link stays valid.

export type TextSegment = { text: string; href?: string };

const URL_RE = /https?:\/\/[^\s"'<>)]+/g;

// Defense-in-depth scheme allowlist. URL_RE already requires an http(s):// prefix, but
// validate the scheme EXPLICITLY before exposing `href` so the http(s)-only guarantee is
// enforced on the value itself — not merely implied by the regex. A non-parseable or
// non-http(s) match degrades to plain text (no `href`). This also makes the sink safe to
// static analysis: `href` provably can never carry a javascript:/data:/vbscript: scheme
// (CodeQL js/xss + js/client-side-unvalidated-url-redirection).
function safeHttpHref(url: string): string | undefined {
  let protocol: string;
  try {
    protocol = new URL(url).protocol;
  } catch {
    return undefined;
  }
  return protocol === "http:" || protocol === "https:" ? url : undefined;
}

export function linkifyText(input: string): TextSegment[] {
  if (!input) return [];
  const segments: TextSegment[] = [];
  let last = 0;
  let match: RegExpExecArray | null;
  URL_RE.lastIndex = 0;
  while ((match = URL_RE.exec(input)) !== null) {
    if (match.index > last) {
      segments.push({ text: input.slice(last, match.index) });
    }
    let url = match[0];
    let trailing = "";
    while (url.length > 0 && /[.,;:!?]$/.test(url)) {
      trailing = url.slice(-1) + trailing;
      url = url.slice(0, -1);
    }
    const href = safeHttpHref(url);
    segments.push(href ? { text: url, href } : { text: url });
    if (trailing) segments.push({ text: trailing });
    last = match.index + match[0].length;
  }
  if (last < input.length) segments.push({ text: input.slice(last) });
  return segments;
}