| """ |
| Tests for TextProcessor. |
| |
| All LLM and embedding calls are mocked so no API keys are needed. |
| """ |
|
|
| import sys |
| import os |
|
|
| sys.path.insert(0, os.path.join(os.path.dirname(__file__), "..")) |
|
|
| import pytest |
| from unittest.mock import patch, MagicMock |
|
|
| from omni_memory.core.config import OmniMemoryConfig |
| from omni_memory.processors.text_processor import TextProcessor |
| from omni_memory.triggers.base import TriggerDecision |
|
|
|
|
| |
| |
| |
|
|
| FAKE_EMBEDDING = [0.1] * 384 |
|
|
|
|
| def _make_processor(**kwargs): |
| """Create a TextProcessor with mocked cold storage.""" |
| config = OmniMemoryConfig() |
| cold_storage = MagicMock() |
| return TextProcessor(config=config, cold_storage=cold_storage, **kwargs) |
|
|
|
|
| |
| |
| |
|
|
| class TestRedundancyDetection: |
| def test_first_text_is_accepted(self): |
| proc = _make_processor() |
| result = proc._check_redundancy("This is brand new text") |
| assert result.decision == TriggerDecision.ACCEPT |
|
|
| def test_identical_text_is_rejected(self): |
| proc = _make_processor() |
| proc._recent_texts.append("the quick brown fox jumps over the lazy dog") |
| result = proc._check_redundancy("the quick brown fox jumps over the lazy dog") |
| assert result.decision == TriggerDecision.REJECT |
|
|
| def test_novel_text_is_accepted(self): |
| proc = _make_processor() |
| proc._recent_texts.append("the quick brown fox jumps over the lazy dog") |
| result = proc._check_redundancy("quantum physics explains particle behavior") |
| assert result.decision == TriggerDecision.ACCEPT |
|
|
| def test_partially_overlapping_text(self): |
| proc = _make_processor() |
| proc._recent_texts.append("the quick brown fox jumps over the lazy dog near the river") |
| |
| result = proc._check_redundancy( |
| "the quick brown fox went to the market and bought some apples" |
| ) |
| assert result.decision in (TriggerDecision.ACCEPT, TriggerDecision.UNCERTAIN) |
|
|
|
|
| |
| |
| |
|
|
| class TestLengthValidation: |
| @patch("omni_memory.processors.text_processor.TextProcessor.generate_embedding", return_value=FAKE_EMBEDDING) |
| @patch("omni_memory.processors.text_processor.TextProcessor.generate_summary", side_effect=lambda t: t[:200]) |
| def test_short_text_rejected(self, mock_summary, mock_embed): |
| proc = _make_processor(min_length=10) |
| result = proc.process("hi") |
| assert result.success is False |
| assert result.skipped is True |
|
|
| @patch("omni_memory.processors.text_processor.TextProcessor.generate_embedding", return_value=FAKE_EMBEDDING) |
| @patch("omni_memory.processors.text_processor.TextProcessor.generate_summary", side_effect=lambda t: t[:200]) |
| def test_short_text_accepted_with_force(self, mock_summary, mock_embed): |
| proc = _make_processor(min_length=10) |
| result = proc.process("hi", force=True) |
| assert result.success is True |
| assert result.mau is not None |
|
|
| @patch("omni_memory.processors.text_processor.TextProcessor.generate_embedding", return_value=FAKE_EMBEDDING) |
| @patch("omni_memory.processors.text_processor.TextProcessor.generate_summary", side_effect=lambda t: t[:200]) |
| def test_long_text_truncated(self, mock_summary, mock_embed): |
| proc = _make_processor(max_length=50) |
| long_text = "x" * 200 |
| result = proc.process(long_text) |
| assert result.success is True |
| |
| assert result.mau.details["full_text"].endswith("...") |
|
|
| @patch("omni_memory.processors.text_processor.TextProcessor.generate_embedding", return_value=FAKE_EMBEDDING) |
| @patch("omni_memory.processors.text_processor.TextProcessor.generate_summary", side_effect=lambda t: t[:200]) |
| def test_normal_text_accepted(self, mock_summary, mock_embed): |
| proc = _make_processor(min_length=5) |
| result = proc.process("This is a perfectly normal piece of text for testing.") |
| assert result.success is True |
| assert result.mau is not None |
| assert result.mau.modality_type.value == "text" |
|
|
|
|
| |
| |
| |
|
|
| class TestSummaryGeneration: |
| def test_short_text_used_as_summary(self): |
| proc = _make_processor() |
| summary = proc.generate_summary("Short text under 200 chars") |
| assert summary == "Short text under 200 chars" |
|
|
| @patch.object(TextProcessor, "_call_llm", return_value="A concise summary.") |
| def test_long_text_calls_llm(self, mock_llm): |
| proc = _make_processor() |
| long_text = "word " * 100 |
| summary = proc.generate_summary(long_text) |
| assert summary == "A concise summary." |
| mock_llm.assert_called_once() |
|
|
| @patch.object(TextProcessor, "_call_llm", return_value="") |
| def test_llm_failure_returns_truncated(self, mock_llm): |
| proc = _make_processor() |
| long_text = "word " * 100 |
| summary = proc.generate_summary(long_text) |
| |
| assert len(summary) <= 200 |
|
|
|
|
| |
| |
| |
|
|
| class TestProcessPipeline: |
| @patch("omni_memory.processors.text_processor.TextProcessor.generate_embedding", return_value=FAKE_EMBEDDING) |
| @patch("omni_memory.processors.text_processor.TextProcessor.generate_summary", side_effect=lambda t: t[:200]) |
| def test_process_creates_mau(self, mock_summary, mock_embed): |
| proc = _make_processor() |
| result = proc.process( |
| "A detailed document about artificial intelligence and machine learning.", |
| session_id="test_session", |
| ) |
| assert result.success is True |
| mau = result.mau |
| assert mau is not None |
| assert mau.embedding == FAKE_EMBEDDING |
| assert mau.metadata.session_id == "test_session" |
|
|
| @patch("omni_memory.processors.text_processor.TextProcessor.generate_embedding", return_value=[]) |
| @patch("omni_memory.processors.text_processor.TextProcessor.generate_summary", side_effect=lambda t: t[:200]) |
| def test_process_fails_on_empty_embedding(self, mock_summary, mock_embed): |
| proc = _make_processor() |
| result = proc.process("Some valid text content for testing purposes.") |
| assert result.success is False |
| assert result.error is not None |
|
|
| def test_reset_clears_recent_texts(self): |
| proc = _make_processor() |
| proc._recent_texts = ["a", "b", "c"] |
| proc.reset() |
| assert proc._recent_texts == [] |
|
|