@tea-agent/loop-agent 0.40.0-next.12 → 0.40.0-next.14

package/dist/worker/console/chat/repo-browser.js

@@ -4,8 +4,12 @@ import { lstat, open, readdir } from "node:fs/promises";
 import path from "node:path";
 import { guardReadPath, isSensitivePath, scrubSecrets, } from "./explore-tools.js";
 const TREE_LIMIT = 500;
-const READ_LIMIT = 256 * 1024;
-const RESPONSE_LIMIT = 128 * 1024;
+/** DSH-style bounded source read: large text files return this leading slice
+ * instead of being rejected solely because the complete file is larger. */
+export const TEXT_PREVIEW_LIMIT = 512 * 1024;
+/** Full-file ceiling for the sandboxed HTML preview route and its assets. */
+export const HTML_PREVIEW_ASSET_LIMIT = 20 * 1024 * 1024;
+export const REPO_HTML_PREVIEW_ROUTE_PREFIX = "/api/operator/v1/chat/repo/html/";
 const SKIP_DIRS = new Set([
     ".git",
     ".harness",
@@ -32,6 +36,45 @@ export class RepoBrowserError extends Error {
 function guardedCode(code) {
     return code === "PATH_OUTSIDE_REPO" ? "OUT_OF_REPO" : code;
 }
+/** Match the path-encoded HTML preview route. Keeping workspace + file path in
+ * the URL path lets browser-relative CSS/image references retain their scope. */
+export function matchRepoHtmlPreviewRoute(pathname) {
+    if (!pathname.startsWith(REPO_HTML_PREVIEW_ROUTE_PREFIX))
+        return null;
+    const rest = pathname.slice(REPO_HTML_PREVIEW_ROUTE_PREFIX.length);
+    const [workspaceId = "", ...pathSegments] = rest.split("/");
+    return { workspaceId, rawPath: pathSegments.join("/") };
+}
+export function repoHtmlAssetContentType(rawPath) {
+    const extension = path.extname(rawPath).toLowerCase();
+    const types = {
+        ".html": "text/html; charset=utf-8",
+        ".htm": "text/html; charset=utf-8",
+        ".css": "text/css; charset=utf-8",
+        ".js": "text/javascript; charset=utf-8",
+        ".mjs": "text/javascript; charset=utf-8",
+        ".json": "application/json; charset=utf-8",
+        ".txt": "text/plain; charset=utf-8",
+        ".svg": "image/svg+xml",
+        ".png": "image/png",
+        ".jpg": "image/jpeg",
+        ".jpeg": "image/jpeg",
+        ".gif": "image/gif",
+        ".webp": "image/webp",
+        ".avif": "image/avif",
+        ".bmp": "image/bmp",
+        ".ico": "image/x-icon",
+        ".woff": "font/woff",
+        ".woff2": "font/woff2",
+        ".ttf": "font/ttf",
+        ".otf": "font/otf",
+        ".mp3": "audio/mpeg",
+        ".wav": "audio/wav",
+        ".mp4": "video/mp4",
+        ".webm": "video/webm",
+    };
+    return types[extension] ?? "application/octet-stream";
+}
 export async function listRepoDirectory(repoRoot, rawPath) {
     const root = rawPath === "";
     const guard = await guardReadPath(repoRoot, root ? "." : rawPath);
@@ -92,8 +135,6 @@ export async function readRepoPreview(repoRoot, rawPath) {
             const info = await handle.stat();
             if (!info.isFile())
                 throw new RepoBrowserError(400, "NOT_A_FILE", "path is not a regular file");
-            if (info.size > READ_LIMIT)
-                throw new RepoBrowserError(413, "TOO_LARGE", `file exceeds ${READ_LIMIT} byte preview limit`);
             const rechecked = await guardReadPath(repoRoot, rawPath);
             if (!rechecked.ok)
                 throw new RepoBrowserError(403, guardedCode(rechecked.code), "file path changed while it was being opened");
@@ -103,9 +144,10 @@ export async function readRepoPreview(repoRoot, rawPath) {
                 current.ino !== info.ino)
                 throw new RepoBrowserError(403, "OUT_OF_REPO", "file identity changed while it was being opened");
             size = info.size;
-            bytes = Buffer.alloc(size);
-            const bytesRead = size
-                ? (await handle.read(bytes, 0, size, 0)).bytesRead
+            const readSize = Math.min(size, TEXT_PREVIEW_LIMIT);
+            bytes = Buffer.alloc(readSize);
+            const bytesRead = readSize
+                ? (await handle.read(bytes, 0, readSize, 0)).bytesRead
                 : 0;
             bytes = bytes.subarray(0, bytesRead);
         }
@@ -114,17 +156,14 @@ export async function readRepoPreview(repoRoot, rawPath) {
         }
         if (bytes.includes(0))
             throw new RepoBrowserError(415, "BINARY_UNSUPPORTED", "binary files cannot be previewed");
-        let end = Math.min(bytes.length, RESPONSE_LIMIT);
-        while (end > 0 && end < bytes.length && (bytes[end] & 0xc0) === 0x80)
-            end -= 1;
-        const content = scrubSecrets(bytes.subarray(0, end).toString("utf8")).scrubbed;
+        const content = scrubSecrets(bytes.toString("utf8")).scrubbed;
         return {
             path: guard.repoRelative,
             kind: "text",
             content,
             language: path.extname(guard.repoRelative).slice(1).toLowerCase() || "text",
             bytes: size,
-            truncated: bytes.length > RESPONSE_LIMIT,
+            truncated: size > bytes.length,
             sha256: createHash("sha256").update(bytes).digest("hex"),
         };
     }
@@ -138,3 +177,46 @@ export async function readRepoPreview(repoRoot, rawPath) {
         throw error;
     }
 }
+/** Read one complete file for the sandboxed HTML iframe. Unlike source reads,
+ * this route needs complete HTML/CSS/image bytes so documents and relative
+ * assets render correctly; the separate 20 MiB cap keeps memory bounded. */
+export async function readRepoHtmlAsset(repoRoot, rawPath) {
+    const guard = await guardReadPath(repoRoot, rawPath);
+    if (!guard.ok)
+        throw new RepoBrowserError(403, guardedCode(guard.code), guard.message);
+    try {
+        const handle = await open(guard.abs, constants.O_RDONLY | constants.O_NOFOLLOW);
+        try {
+            const info = await handle.stat();
+            if (!info.isFile())
+                throw new RepoBrowserError(400, "NOT_A_FILE", "path is not a regular file");
+            if (info.size > HTML_PREVIEW_ASSET_LIMIT)
+                throw new RepoBrowserError(413, "TOO_LARGE", `file exceeds ${HTML_PREVIEW_ASSET_LIMIT} byte HTML preview limit`);
+            const rechecked = await guardReadPath(repoRoot, rawPath);
+            if (!rechecked.ok)
+                throw new RepoBrowserError(403, guardedCode(rechecked.code), "file path changed while it was being opened");
+            const current = await lstat(rechecked.abs);
+            if (current.isSymbolicLink() ||
+                current.dev !== info.dev ||
+                current.ino !== info.ino)
+                throw new RepoBrowserError(403, "OUT_OF_REPO", "file identity changed while it was being opened");
+            return {
+                body: await handle.readFile(),
+                contentType: repoHtmlAssetContentType(guard.repoRelative),
+                bytes: info.size,
+            };
+        }
+        finally {
+            await handle.close();
+        }
+    }
+    catch (error) {
+        if (error instanceof RepoBrowserError)
+            throw error;
+        if (error.code === "ENOENT")
+            throw new RepoBrowserError(404, "NOT_FOUND", "file does not exist");
+        if (error.code === "ELOOP")
+            throw new RepoBrowserError(403, "OUT_OF_REPO", "symbolic links are not previewable");
+        throw error;
+    }
+}