From 9712c49b72d477d8b5759a837619de5c60d1ccaa Mon Sep 17 00:00:00 2001
From: Pepijn <pepijn@huggingface.co>
Date: Fri, 17 Apr 2026 09:36:49 +0100
Subject: [PATCH] fix(docker): make VLABench asset download tolerant + HF
 mirror option

Google Drive returns HTTP 429 ("too many users") on VLABench's shared
academic asset file, blocking CI builds unpredictably.

- Add VLABENCH_ASSETS_REPO build arg: when set, snapshot_download the
  given HF Hub dataset repo straight into VLABench/assets/. This is the
  reliable path for CI once a mirror is uploaded.
- Keep the gdown script as fallback but make it best-effort (|| echo WARN)
  so quota errors don't block the image build.

If neither source succeeds, the constant.py patch (empty-dir guard)
keeps module import working; only task-build at runtime will surface
missing-mesh errors.

Made-with: Cursor
---
 docker/Dockerfile.benchmark.vlabench | 29 +++++++++++++++++++++++-----
 1 file changed, 24 insertions(+), 5 deletions(-)

diff --git a/docker/Dockerfile.benchmark.vlabench b/docker/Dockerfile.benchmark.vlabench
index 0a5dbd6c7..ac5843afe 100644
--- a/docker/Dockerfile.benchmark.vlabench
+++ b/docker/Dockerfile.benchmark.vlabench
@@ -41,11 +41,30 @@ p.write_text(t.replace( \
       mujoco==3.2.2 dm-control==1.0.22 \
       open3d colorlog scikit-learn openai gdown
 
-# Download VLABench mesh assets from Google Drive. Task configs reference
-# object meshes (obj/meshes/fruit/, containers/basket/, etc.); without them
-# the task builder crashes with IndexError on random.choice([]).
-# gdown handles Google Drive's confirmation tokens for large files.
-RUN python ~/VLABench/scripts/download_assets.py --choice all
+# Download VLABench mesh assets. Task configs reference object meshes
+# (obj/meshes/fruit/, containers/basket/, etc.); without them the task builder
+# picks from an empty mesh list and crashes at task-build time.
+#
+# Preferred source: an HF Hub mirror. Set VLABENCH_ASSETS_REPO at build time
+# (e.g. --build-arg VLABENCH_ASSETS_REPO=lerobot/vlabench-assets) and we'll
+# snapshot_download the repo into VLABench's assets dir. This is the reliable
+# path for CI — Google Drive frequently returns HTTP 429 ("Too many users have
+# viewed or downloaded this file recently") on shared academic files.
+#
+# Fallback: VLABench's own gdown-based script. Best-effort only; the build
+# will NOT fail if gdown hits a Drive quota, letting us ship an image where
+# task-build may still IndexError but the rest of the pipeline is exercised.
+ARG VLABENCH_ASSETS_REPO=""
+RUN ASSETS_DIR="$HOME/VLABench/VLABench/assets" && \
+    if [ -n "${VLABENCH_ASSETS_REPO}" ]; then \
+        echo "Downloading VLABench assets from HF Hub: ${VLABENCH_ASSETS_REPO}" && \
+        python -c "from huggingface_hub import snapshot_download; \
+snapshot_download(repo_id='${VLABENCH_ASSETS_REPO}', repo_type='dataset', local_dir='${ASSETS_DIR}')"; \
+    else \
+        echo "No VLABENCH_ASSETS_REPO set — falling back to gdown (best-effort)" && \
+        python ~/VLABench/scripts/download_assets.py --choice all || \
+          echo "WARN: VLABench asset download failed (likely Google Drive quota). Env task-build may crash."; \
+    fi
 
 # Overlay the PR's source code on top of the nightly image.
 COPY --chown=user_lerobot:user_lerobot . .