feat: add repo clone job

apps/worker/src/jobs/clone.ts

@@ -0,0 +1,87 @@
+import { simpleGit } from "simple-git";
+import { mkdtemp, readdir, stat } from "node:fs/promises";
+import { join } from "node:path";
+import { tmpdir } from "node:os";
+import type { CloneResult } from "@codeboard/shared";
+
+async function countFiles(dir: string): Promise<{ files: number; lines: number }> {
+  let files = 0;
+  let lines = 0;
+
+  const entries = await readdir(dir, { withFileTypes: true });
+  for (const entry of entries) {
+    if (entry.name === ".git" || entry.name === "node_modules") continue;
+    const fullPath = join(dir, entry.name);
+    if (entry.isDirectory()) {
+      const sub = await countFiles(fullPath);
+      files += sub.files;
+      lines += sub.lines;
+    } else {
+      files++;
+      const fileStat = await stat(fullPath);
+      lines += Math.ceil(fileStat.size / 40);
+    }
+  }
+
+  return { files, lines };
+}
+
+export async function cloneRepository(repoUrl: string): Promise<CloneResult> {
+  const tmpDir = await mkdtemp(join(tmpdir(), "codeboard-"));
+  const git = simpleGit();
+
+  await git.clone(repoUrl, tmpDir, ["--depth", "1", "--single-branch"]);
+
+  const localGit = simpleGit(tmpDir);
+  const log = await localGit.log({ maxCount: 1 });
+  const lastCommit = log.latest?.hash ?? "unknown";
+
+  const repoName = repoUrl
+    .replace(/\.git$/, "")
+    .split("/")
+    .slice(-1)[0] ?? "unknown";
+
+  const { files: totalFiles, lines: totalLines } = await countFiles(tmpDir);
+
+  const languageCounts: Record<string, number> = {};
+  const extMap: Record<string, string> = {
+    ".ts": "TypeScript", ".tsx": "TypeScript",
+    ".js": "JavaScript", ".jsx": "JavaScript",
+    ".py": "Python", ".go": "Go",
+    ".rs": "Rust", ".java": "Java",
+    ".rb": "Ruby", ".php": "PHP",
+  };
+
+  async function scanLanguages(dir: string) {
+    const entries = await readdir(dir, { withFileTypes: true });
+    for (const entry of entries) {
+      if (entry.name.startsWith(".") || entry.name === "node_modules") continue;
+      const fullPath = join(dir, entry.name);
+      if (entry.isDirectory()) {
+        await scanLanguages(fullPath);
+      } else {
+        const ext = entry.name.slice(entry.name.lastIndexOf("."));
+        const lang = extMap[ext];
+        if (lang) {
+          const fileStat = await stat(fullPath);
+          languageCounts[lang] = (languageCounts[lang] ?? 0) + fileStat.size;
+        }
+      }
+    }
+  }
+  await scanLanguages(tmpDir);
+
+  return {
+    localPath: tmpDir,
+    metadata: {
+      name: repoName,
+      description: "",
+      defaultBranch: "main",
+      languages: languageCounts,
+      stars: 0,
+      lastCommit,
+      totalFiles,
+      totalLines,
+    },
+  };
+}