pipeline/packages/collect/src/collectors/repo-detail.ts

import { formatCount, type RepoMeta } from "@pipeline/shared";

/**
 * Shared single-repo detail helpers — the fetch + format logic behind the
 * github-repo collector, parameterized by README truncation length so
 * github-daily-pick can ask for a deeper excerpt without duplicating code.
 */

/** Format a repo-detail API payload as markdown (the LLM's fact source),
 *  plus the typed RepoMeta the text module treats as authoritative. */
export function formatRepoDetail(
  queryOwner: string,
  queryRepo: string,
  data: any,
  readmeLimit = 3000,
): { text: string; meta: RepoMeta | null } {
  // The API returns the REAL owner (transferred repos report their current
  // owner). Prefer data.fullName; fall back to the query params.
  const fullName = data.fullName || `${queryOwner}/${queryRepo}`;
  const [detailOwner, detailName] = fullName.split("/");

  const meta: RepoMeta = {
    owner: detailOwner || queryOwner,
    name: detailName || queryRepo,
    fullName,
    language: data.language ?? "",
    languageColor: data.languageColor ?? "",
    stars: data.stars,
    forks: data.forks,
    license: data.license ?? "",
  };

  const lines: string[] = [`# ${fullName}\n`];
  if (data.description) lines.push(`${data.description}\n`);

  // Structured metadata block — LLM copies verbatim into scene.github.repo.
  lines.push(`<!-- repo-meta: ${JSON.stringify(meta)} -->`);

  const meta2: string[] = [];
  if (data.language) meta2.push(`Language: ${data.language}`);
  if (data.stars != null) meta2.push(`Stars: ${formatCount(data.stars)}`);
  if (data.forks != null) meta2.push(`Forks: ${formatCount(data.forks)}`);
  if (data.openIssues != null) meta2.push(`Open Issues: ${formatCount(data.openIssues)}`);
  if (data.topics?.length) meta2.push(`Topics: ${data.topics.join(", ")}`);
  if (meta2.length) {
    lines.push("");
    lines.push(meta2.map((m) => `- ${m}`).join("\n"));
  }

  if (data.readme) {
    const readme = data.readme.length > readmeLimit
      ? data.readme.slice(0, readmeLimit) + "\n..."
      : data.readme;
    lines.push("\n## README\n", readme);
  }

  return { text: lines.join("\n"), meta };
}

/** Fetch one repo's detail payload and format it. Returns null on any failure
 *  so callers can decide how to degrade. Retries transient upstream errors
 *  (5xx / network) up to 2 times with a short backoff — a blip shouldn't burn
 *  the daily pick's topic. */
export async function fetchRepoDetail(
  repoUrlTemplate: string,
  owner: string,
  repo: string,
  readmeLimit = 3000,
): Promise<{ text: string; meta: RepoMeta | null } | null> {
  if (!repoUrlTemplate) return null;
  const url = repoUrlTemplate
    .replace(":owner", encodeURIComponent(owner))
    .replace(":repo", encodeURIComponent(repo));

  const sleep = (ms: number) => new Promise((r) => setTimeout(r, ms));
  for (let attempt = 0; attempt < 3; attempt++) {
    try {
      const response = await fetch(url);
      if (response.ok) {
        const raw = (await response.json()) as Record<string, any>;
        const data = (raw.data ?? raw) as Record<string, any>;
        return formatRepoDetail(owner, repo, data, readmeLimit);
      }
      // 4xx is a real answer (repo gone etc.) — do not retry.
      if (response.status < 500) return null;
    } catch {
      // network error — retry
    }
    if (attempt < 2) await sleep(1000 * (attempt + 1));
  }
  return null;
}