mirror of
https://github.com/tiennm99/DocsGPT.git
synced 2026-10-03 22:13:01 +00:00
Alex
7da46c2bea
- Ship tiktoken's cl100k_base inside the package and build the encoding from it, so token counting never downloads anything. - Default EMBEDDINGS_CACHE_DIR to <data home>/models instead of FastEmbed's temp dir, and read tokenizer.json and repo metadata from that cache, so a model downloads once and survives reboots. - TTS_PROVIDER=none and STT_PROVIDER=none switch the speech features off: the endpoints return 404, audio files fail to ingest with a clear message, /api/config reports tts_available/stt_available, and the UI hides the Speak and microphone buttons. - Drop the Google Fonts Roboto import from the web UI. - prefetch-models fills the cache the app reads; verify-offline checks the packaged encoding. - Docs: new Air-Gapped Deployment guide, settings and cache notes.
89 lines
2.9 KiB
Python
89 lines
2.9 KiB
Python
from unittest.mock import MagicMock, patch
|
|
|
|
import pytest
|
|
|
|
from docsgpt.parser.file.audio_parser import AudioParser
|
|
from docsgpt.parser.file.base_parser import DocumentParseError
|
|
from docsgpt.parser.file.bulk import get_default_file_extractor
|
|
from docsgpt.stt.upload_limits import AudioFileTooLargeError
|
|
|
|
|
|
def test_audio_init_parser():
|
|
parser = AudioParser()
|
|
assert isinstance(parser._init_parser(), dict)
|
|
assert not parser.parser_config_set
|
|
parser.init_parser()
|
|
assert parser.parser_config_set
|
|
|
|
|
|
@patch("docsgpt.stt.upload_limits.settings")
|
|
@patch("docsgpt.parser.file.audio_parser.STTCreator.create_stt")
|
|
@patch("docsgpt.parser.file.audio_parser.settings")
|
|
def test_audio_parser_transcribes_file(
|
|
mock_settings, mock_create_stt, mock_limit_settings, tmp_path
|
|
):
|
|
mock_settings.STT_PROVIDER = "openai"
|
|
mock_settings.STT_LANGUAGE = "en"
|
|
mock_settings.STT_ENABLE_TIMESTAMPS = False
|
|
mock_settings.STT_ENABLE_DIARIZATION = False
|
|
mock_limit_settings.STT_MAX_FILE_SIZE_MB = 25
|
|
|
|
mock_stt = MagicMock()
|
|
mock_stt.transcribe.return_value = {"text": "Transcript from audio"}
|
|
mock_create_stt.return_value = mock_stt
|
|
audio_file = tmp_path / "meeting.wav"
|
|
audio_file.write_bytes(b"audio-bytes")
|
|
|
|
parser = AudioParser()
|
|
result = parser.parse_file(audio_file)
|
|
|
|
assert result == "Transcript from audio"
|
|
mock_create_stt.assert_called_once_with("openai")
|
|
mock_stt.transcribe.assert_called_once_with(
|
|
audio_file,
|
|
language="en",
|
|
timestamps=False,
|
|
diarize=False,
|
|
)
|
|
|
|
|
|
@patch("docsgpt.parser.file.audio_parser.STTCreator.create_stt")
|
|
@patch("docsgpt.parser.file.audio_parser.settings")
|
|
def test_audio_parser_refuses_when_speech_to_text_is_disabled(
|
|
mock_settings, mock_create_stt, tmp_path
|
|
):
|
|
mock_settings.STT_PROVIDER = "none"
|
|
audio_file = tmp_path / "meeting.wav"
|
|
audio_file.write_bytes(b"audio-bytes")
|
|
|
|
with pytest.raises(DocumentParseError, match="STT_PROVIDER"):
|
|
AudioParser().parse_file(audio_file)
|
|
mock_create_stt.assert_not_called()
|
|
|
|
|
|
@patch("docsgpt.stt.upload_limits.settings")
|
|
def test_audio_parser_rejects_oversized_files(mock_limit_settings, tmp_path):
|
|
mock_limit_settings.STT_MAX_FILE_SIZE_MB = 1
|
|
|
|
audio_file = tmp_path / "meeting.wav"
|
|
audio_file.write_bytes(b"x" * (2 * 1024 * 1024))
|
|
|
|
parser = AudioParser()
|
|
|
|
try:
|
|
parser.parse_file(audio_file)
|
|
except AudioFileTooLargeError as exc:
|
|
assert "exceeds" in str(exc)
|
|
else:
|
|
raise AssertionError("Expected oversized audio file to be rejected")
|
|
|
|
|
|
def test_default_file_extractor_supports_audio_extensions():
|
|
extractor = get_default_file_extractor()
|
|
|
|
assert isinstance(extractor[".wav"], AudioParser)
|
|
assert isinstance(extractor[".mp3"], AudioParser)
|
|
assert isinstance(extractor[".m4a"], AudioParser)
|
|
assert isinstance(extractor[".ogg"], AudioParser)
|
|
assert isinstance(extractor[".webm"], AudioParser)
|