Skip to content

Skip e2e/integration tests by default and gracefully skip gated tokenizer test #4757

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
rjpower merged 1 commit into from
Apr 14, 2026
+12 −7
Merged

Skip e2e/integration tests by default and gracefully skip gated tokenizer test #4757

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
6 changes: 4 additions & 2 deletions pyproject.toml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -223,5 +223,7 @@ markers = [
]
testpaths = ["tests", "experiments"]

# Don't run TPU or slow tests by default
addopts = "--session-timeout=480 -m 'not tpu_ci and not slow'"
# Don't run TPU, slow, or integration tests by default. Integration tests
# require external infrastructure (Iris cluster, GCS, gated HF repos, etc.)
# and are run from dedicated CI workflows.
addopts = "--session-timeout=480 -m 'not tpu_ci and not slow and not integration'"
1 change: 1 addition & 0 deletions tests/test_integration_test.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -8,6 +8,7 @@
import pytest


@pytest.mark.integration
@pytest.mark.skipif(os.getenv("CI") == "true", reason="Skip this test in CI, since we run it as a separate worflow.")
def test_integration_test_run():
MARIN_ROOT = os.path.abspath(os.path.join(os.path.dirname(__file__), ".."))
Expand Down
12 changes: 7 additions & 5 deletions tests/test_marin_tokenizer.py
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,7 +1,6 @@
# Copyright The Marin Authors
# SPDX-License-Identifier: Apache-2.0

import os
import tempfile

import pytest
Expand All @@ -16,13 +15,16 @@

@pytest.fixture
def marin_tokenizer():
"""Fixture that provides a configured marin tokenizer for testing."""
"""Fixture that provides a configured marin tokenizer for testing.

The base llama3 tokenizer lives in a gated Hugging Face repo. When the
current environment lacks credentials (or network access), skip rather
than fail - this test exercises our tokenizer surgery, not HF auth.
"""
try:
llama3_tokenizer = load_llama3_tokenizer()
except Exception as e:
if os.getenv("CI", False) in ["true", "1"]:
pytest.skip("Llama 3 tokenizer repository is gated")
raise e
pytest.skip(f"Llama 3 tokenizer is unavailable (gated repo or no network): {e}")
Comment on lines 26 to +27
Copy link
Copy Markdown

@chatgpt-codex-connector chatgpt-codex-connector Bot Apr 14, 2026

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

P2 Badge Restrict tokenizer test skip to expected HF access errors

Catching Exception here causes the test to be skipped for any failure in load_llama3_tokenizer(), including real regressions (for example, a bad model ID, local API misuse, or transformer-side breakage) rather than only missing credentials/network. That turns genuine failures into skips and can let tokenizer changes merge without coverage; this should only skip for the specific access-related exceptions documented by load_llama3_tokenizer.

Useful? React with 👍 / 👎.

tokenizer = create_marin_tokenizer(llama3_tokenizer)

# Roundtrip write-read to ensure consistency
Expand Down
Loading