fix: handle tiktoken network errors in offline environments (issue #264)

In air-gapped / offline Docker deployments, tiktoken.get_encoding() tries
to download the encoding file from openaipublic.blob.core.windows.net.
When that request fails it raises a URLError / OSError — not an ImportError
— so the previous except clause silently missed it and the crash surfaced in
the UI.

Widened `except ImportError` to `except Exception` so all failures —
"not installed" and "network unreachable" — fall through to the word-count
fallback (words × 1.3). Added a loguru WARNING so operators can see when
the fallback is active.

TIKTOKEN_CACHE_DIR now reads from the environment with a blank-safe
fallback (`or` guard prevents os.makedirs("") on empty env var). This lets
Docker images redirect the cache to a path outside /app/data/ so user-data
volume mounts cannot shadow the pre-baked encoding.

Both images now pre-download the o200k_base encoding during the builder
stage (internet is available at build time) and copy it into the runtime
image at /app/tiktoken-cache. ENV TIKTOKEN_CACHE_DIR=/app/tiktoken-cache
is set in the runtime stage so no network call is ever needed at runtime.

Added test_token_count_network_error_fallback in tests/test_utils.py:
patches tiktoken.get_encoding with a URLError and asserts token_count()
returns a positive int instead of raising.

Fixes #264

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
This commit is contained in:
orihatav 2026-02-21 22:54:15 +02:00 committed by Luis Novo
parent 38d9ed5986
commit d0bbe4a921
5 changed files with 61 additions and 3 deletions

View file

@ -151,6 +151,28 @@ class TestTokenUtilities:
assert isinstance(count, int)
assert count > 0
def test_token_count_network_error_fallback(self):
"""Test fallback when tiktoken raises a network error (issue #264).
In offline environments tiktoken.get_encoding() tries to download the
encoding file and raises a URLError/OSError, not an ImportError.
The except clause must catch Exception (not only ImportError) so that
these network failures also fall through to the word-count estimate.
"""
import urllib.error
from unittest.mock import patch
with patch(
"tiktoken.get_encoding",
side_effect=urllib.error.URLError("No network (simulated offline)"),
):
text = "one two three four five"
count = token_count(text)
# Must not raise; must return a positive int via the fallback
assert isinstance(count, int)
assert count > 0
# ============================================================================
# TEST SUITE 3: Version Utilities