hermes-agent/tests/gateway/test_api_server_media_data_urls.py
Tarun Ravikumar 2068754d6f feat(api-server): inline MEDIA: image tags as base64 data URLs for remote frontends
Salvage of the surviving piece of #2696 by @tarunravi. The PR's other two
changes (tool progress streaming, SSE None-sentinel fix) were independently
superseded on main by the structured hermes.tool.progress SSE events and the
rewritten queue-drain loop.

Remote OpenAI-compatible frontends can't read server-local file paths, so
MEDIA:<path> tags (browser screenshots, generated images) were dead text.
_resolve_media_to_data_urls() now inlines small (<=5MB) local images as
markdown data URLs across all four response surfaces: chat completions
(non-streaming), session chat, session chat stream final event, and the
Responses API. Non-image, missing, or oversized paths pass through
untouched.
2026-07-02 03:23:44 -07:00

77 lines
2.5 KiB
Python

"""MEDIA: tag → base64 data-URL resolution for the API server (salvage of #2696).
Remote OpenAI-compatible frontends can't read local file paths, so
``MEDIA:<path>`` image tags in final responses are inlined as markdown
data URLs before crossing the HTTP boundary.
"""
import base64
import unittest
import pytest
pytest.importorskip("aiohttp")
from gateway.platforms.api_server import _resolve_media_to_data_urls # noqa: E402
# 1x1 transparent PNG
_PNG_BYTES = base64.b64decode(
"iVBORw0KGgoAAAANSUhEUgAAAAEAAAABCAYAAAAfFcSJAAAADUlEQVR4nGNgYGBgAAAABQAB"
"h6FO1AAAAABJRU5ErkJggg=="
)
class TestResolveMediaToDataUrls(unittest.TestCase):
def _write_png(self, tmpdir_name="hermes_media_test"):
import tempfile
from pathlib import Path
d = Path(tempfile.mkdtemp(prefix=tmpdir_name))
p = d / "shot.png"
p.write_bytes(_PNG_BYTES)
return p
def test_media_tag_inlined(self):
p = self._write_png()
out = _resolve_media_to_data_urls(f"Here you go: MEDIA:{p}")
self.assertIn("data:image/png;base64,", out)
self.assertNotIn("MEDIA:", out)
def test_backtick_wrapped_tag(self):
p = self._write_png()
out = _resolve_media_to_data_urls(f"See `MEDIA:{p}` above")
self.assertIn("data:image/png;base64,", out)
def test_missing_file_left_untouched(self):
text = "MEDIA:/nonexistent/path/shot.png"
self.assertEqual(_resolve_media_to_data_urls(text), text)
def test_non_image_left_untouched(self):
text = "MEDIA:/tmp/archive.zip"
self.assertEqual(_resolve_media_to_data_urls(text), text)
def test_text_without_media_passthrough(self):
self.assertEqual(_resolve_media_to_data_urls("plain text"), "plain text")
self.assertEqual(_resolve_media_to_data_urls(""), "")
def test_oversized_image_skipped(self):
from gateway.platforms import api_server as mod
p = self._write_png()
orig = mod._MEDIA_DATA_URL_MAX_BYTES
mod._MEDIA_DATA_URL_MAX_BYTES = 1
try:
text = f"MEDIA:{p}"
self.assertEqual(_resolve_media_to_data_urls(text), text)
finally:
mod._MEDIA_DATA_URL_MAX_BYTES = orig
def test_multiple_tags(self):
p1 = self._write_png()
p2 = self._write_png("hermes_media_test2")
out = _resolve_media_to_data_urls(f"MEDIA:{p1}\nand MEDIA:{p2}")
self.assertEqual(out.count("data:image/png;base64,"), 2)
if __name__ == "__main__":
unittest.main()